Skip to main content

Full text of "A manual of marine engineering: comprising the design, construction, and working of marine machinery"

See other formats


lecAyr.Jr* 7\>r&7 




A MANUAL 



OF 



MARINE ENGINEERING 



STANDARD WORKS FOR ENGINEERS. 



Iim.TEENTH EDITION, Thoroughly Revised. In Pocket Size. Leather. 
A POCKET-BOOK OF 

Marine Engineering Rules and Tables, 

FOR THE USE OF MARINE ENGINEERS, NAVAL ARCHITECTS. DESIGNERS, 
DRAUGHTSMEN, SUPERINTENDENTS. AND OTHERS. 
l.v \ B. BEATON, U , M IM-liE.. aud BL M. ROUNTHWAITE, M.I.Mech.E., M.IN.A. 

" Admirably fullila its porpo —Marine Engineer. 

In L.-. Handsome Cloth. With Frontispiece, 6 Plates, 65 other Illustration*, and 60 Tables. 

The Screw Propeller 

And other Competing Instruments fop Marine Propulsion. 
By a. 1. BEATON, M.Inst.C.E., M.I.Mech.E., M.I.N.A. 
u-eful t" know about the screw propeller. . . . Thorouglily up-to-date."— Steamship. 



HON. In Large 8to. 



Handsome Cloth. 
3 Plates. 



Pp. i-xv + 425. With 377 Illustrations aud 



The Theory of the Steam Turbine. 

A TREATISE ON THE PRINCIPLES OF CONSTRUCTION OF THE STEAM TUREINE, 
WITH HISTORICAL NOTES ON ITS DEVELOPMENT. 
By ALEXANDER Jl'DE. 
"The most satisfactory book on the Theory of the Steam Turbine we have yet seen."— Engineering. 

. iLdition. Large Crown 8vo. With Numerous Illustiations. 

Engine-Room Practice. 

A Handbook for Engineers and Officers in the Royal Navy and 

Mercantile Marine, including the Management cf the 

Main and Auxiliary Engines on Board Ship. 

By JUHN G. LIVERSIDGE, Eug. Capt. R.N., A.M.I.C.E. 
" r e contents cannot fail TO BE APPRECIATED."— The Steamship. 

In Large Crown thro. Cloth. Pp. i-xxviii -i- 244. With 25 Illustrations. 

Sea Water Distillation. 

By FRANK NORMANDY, of the Middle Temple, Barrister-at-Law. 
"Tin- analytical treatment of the problem is concise and comprehensive in its scope."— Marine Engineer. 



VALVES AND VALVE GEARING. A 

Practical Text Book for the use of Engineers, 
Draughtsmen, and Students. By ('has. Hurst. 
I ! i ii Ki'iTioN, Revised aud Enlarged. 
Fully Illustrated. 

"Wiil prove a very valuable aid. "— Marine En- 

Qm'T 



BOILERS, MARINE AND LAND : 

Their Construction and Strength. By T W 
Traill, M.Inst.C.E, F.E.R.N. Fourth 
Edition, Revised. Pocket Size. Leather. 

"The most valuable work on Boilers publish, d in 
England." -Shipping World. 



THE THERMO-DYNAMIC PRINCI- 
PLES 01 BNOINB DESIGN. By LlONEl 
M. BOBBS, K.V. Instructor in Applied 
Mechanic* and Marine Engine Design at 
\ai College, Greenwich. In 
Large Crown 8vo. 

" Like]] to become » atandnrd reference." 

—Mechanica linnineer. j 

roUBTB RditiobT. Illustrated with Plates, Diagrams, and Figures. 

Steel Ships: 

THEIR CONSTRUCTION AND MAINTENANCE. 

By ITIOMAS WALTON, Naval Architect, 

" E ' written Ii every chapter in the book that it is difficult to select any of them as being 

the trork la excellent, and will prove of great value to thoeTfof whom 



ELECTRICAL RULES AND TABLES 

(A Pocket-Book of). By J. Munro, C.E., and 
Andrew jamieson, .m.i.e.e. twentieth 
Edition. 

" Worthy of the highest commendation we can give- 
it. —Eleetrtdau. 



PRESENT-DAY SHIPBUILDING. For 
Shipyard Student*, Bhipa Officeri and En- 

W ILTOR. ln< loth. \\ ui, 

a Plata* and 108 other Ulustratlona. 

inmend n to a I who have to do 

With 



KNOW YOUR OWN SHIP. By Thos. 

w ai.ton. Fourteenth Edition. Fullv 
Illustrated. 

"Mr Walton's wort will obtain lasting success 
because of it- unique Btneaa for those lor whom it 
baa been written."— Shipping World 



LONDON: CHARLES CRIFFIN & CO., LTD., EXETER STREET, STRAND, W.C. 2. 

NEW YORK : D. VAN NOSTRAND CO. 



I 

3 



■ 




o 






1 


a 




-*-> 


w 


g 






-/: 


"« 


















5 


E-> 


r. 


4) 
















1 -■ 




CO 



?: o 



- 


— 


o 

3 




•A 


.. 


'O 




x" 


M 


~^ 


GO 


X 


J ~; 


tfl 


on 


%. 


GO 

S3 


r 


■g 
















z 




^j 




»_ 




~ 


w 


™ 


. — 


> 


=: 


CO 


o 


a 


a. 


~ 


_ 




i: 










-*■ 


a 






i_j 


o3 


Sh 


<j 




0) 


IB 




0Q 


-U 


09 






V 




tf 


— 


— > 

a 


to 
0? 


W 


D 


e3 


3> 

oq 


j 


a 

H 


^3 

01 



~ .S3 
5 "- '■*■ 



a o 
a •- 






- a 
a 



O ~*A 



w 

x. 



A M ANTIAL 



OF 



MARINE ENGINEERING: 



COMPRISING 



THE DESIGN, CONSTRUCTION, AND WORKING OF 
MARINE MACHINERY. 



BY 



A. E. SEATOX, 



70RMERLY LECTURER ON MARINE ENGINEERING TO THE RoYAL NAVAL COLLEGE, GREENWICH; 

MEMBER OF THE INSTITUTION OF CIVIL ENGINEERS J MEMBER OF THE INSTITUTION 

OF NAVAL ARCHITECTS; MEMBER OF THE INSTITUTION OF MECHANICAL 

ENGINEERS: MEMBER OF THE NORTH-EAST COAST INSTITUTION 

OF SHIPBUILDERS AND ENGINEERS ; MEMBER OF 

THE INSTITUTE OF METALS, ETC. 



TUlitb numerous Cables ano illustrations rccuceb from "CdorKfiig 

IDrawings ano pbotoiuapbs. 



EIGHTEENTH EDITION, THOROUGHLY REVISED, GREATLY ENLARGED, 
AND MOSTLY RE- WRIT J EN TO DATE. 



■ 




LONDON: 
CHARLES GRIFFIN & COMPANY, LIMITED. 

NEW ?ORK: D. VAN NOSTRAND CO. 

19 18. 

[All Sights Reserved.] 







Engineering 
Library 



/l/7<zc/i* . Z^e/o/. 



. 



• . ■ ■ ■ . • .-• 



PREFACE TO THE EIGHTEENTH EDITION. 



The demand for this new Edition came when the War, with all 
its pomps and circumstances, has made it difficult to produce- — 
both to Author and Publishers. Since the issue of the last Edition 
the changes in engineering practice have been many and great 
In some measure this is due to the special demands arising out 
of the "War conditions, but it is largely due to the advance that 
goes on now, day by day, from the better knowledge of science 
gained by diligent research, and by the better application of it, 
whereby that experience is gained which engenders confidence as 
well as stimulates invention, and this produces improvements of 
many kinds. 

The economic side of engineering, however, is asserting itself 
to a degree that never obtained in pre- War times, and it will 
surely remain as a predominating factor in all our every-day 
calculations for many years to come, so that we are compelled 
now to approach and to determine problems on lines not con- 
templated formerly. To be successful it will be necessary to 
cast aside all prejudices, to treat lightly the precedents, and to 
concentrate the solving of them — each on its own merits — by 
giving full heed to the physical and economic conditions only. 

Necessity has ever been the mother of invention. To-day it 
will be likewise the remover of prejudice as well as the alma 

41 15 



VI PREFACE. 

mater of research to all her children, so that they may thrive 
in a way they never have done hitherto in this country. 

D. O. R A. exercises a powerful influence over authors and 
publishers, whereby they are restrained from making public any 
of the wonderful advances achieved during the past four years, 
or to allude to the inventions whereby so much has been accom- 
plished on the sea by the genius that otherwise might have 
remained dormant. Nevertheless this Edition does contain much 
that is new, and what was old has been renovated and brought 
up to date. 

Special Appendices have been added which deal with the 
Heavy Oil Engine, Geared Turbines and Superheaters, as now in 
general use, and altogether the attempt is made to maintain the 
character of the Manual as far as circumstances will permit. 

A. E. SEATON. 

Westminster, September, 1918. 



ORIGINAL PKEFACE. 



The following Work has been prepared to supply the existing want of a 
Manual showing the application of Theoretical Principles to the Design 
and Construction of Marine Machinery, as determined by the experience 
of leading engineers, and carried out in the most recent successful practice. 
The data on which it is based, now first thrown into form for publication, 
have been collected during many years of study and practical work. It 
is hoped that the volume will be found useful by the engineer and draughts- 
man engaged in practice as a Handbook of Reference, and by the student, 
launched for the first time on the intricacies of Marine Construction, as 
a guide, supplying to some extent bis lack of experience. 

The rules and formulas introduced (which have been divested as far as 
possible of complexity, and given in the simplest form attainable) may be 
used by any one who designs with some regard to theory, and, by varying 
the constants, be made to suit his own ideas of strength and stiffness. It 
may, perhaps, be thought by some that in certain instances details have been 
entered into with unnecessary minuteness ; but it should be remembered, 
on the other hand, that not every engineer has the contents of a well-filled 
drawing-office to fall back upon in cases of doubt and difficulty. 

It is hardly necessary to premise that it is wholly impossible to reconcile 
the practice of the naval designer, who thinks more of efficiency and weight 
than of cost, with that of the mercantile engineer, who studies efficiency 
and cost with but small regard to weight, and, therefore, few rules can be 
driven which shall absolutely suit both. However, the manufacturer of 
machinery for the Merchant Service might follow with advantage much that 
has been proved to be good in naval practice, and the Naval Authorities 
might again, on their part, borrow from the Mercantile Marine a few 
suggestions which would render a warship, while no less efficient than at 
present, perhaps somewhat less intricate for those who have to work her. 

In conclusion, the author can but express a hope that the publication of 
these notes, imperfect as they necessarily are, may tend to make a little 
clearer some of the technicalities of Marine Design and Construction, and 
so help forward, in however slight a degree, the application of scientittc 
investigation to those problems which the marine engineer is called upon, 
day by day, to solve. 

Hull, January With, 1883. 



GENERAL CONTENT S 



CHAPTER I. 

General Introduction, . . . . pp. 1-18 

Fundamental Principles — Paddle-Wheels — The Screw — Hydromotors — Motive Power 
—Steam used Expansively — Early Marino Engines — Propellers — Multiple Screws. 

CHAPTER II. 

Resistance of Ships and Indicated Horse-Power Necessary for Speed, pp. 19-55 

Value of Trial Trips — Resistance of a Ship — Chief Cause of Resistance — Residual 
Resistance — Coefficient of Fineness — Skin Resistance — Wetted Skin of a Ship — Kirk's 
Analysis — Mumford's Calculation of Wetted Skin — Seaton's Modification — Form for 
Required Speed — Seaton's Rule for Limitation of Speed — Power necessary — Tank 
Experiments — Rankine's Rule — Kirk's Analysis — Progressive Trials — Curve of Revolu- 
tions — Curve of Slip — Sea Performances of Steamers — Progressive Trials for Effect of 
Depth of Water — Taylor's Formula for Depths of Water on Trials. 

CHAPTER III. 

Marine Engines : Their Types and Variations of Design, . pp. 56-108 

Various Types — Paddle-wheel Engines — Beam Engines — Side Lever Engines — 
Oscillating Engines — Vertical Direct-acting Engines — Twin-cylinder Engines — Diagonal 
Paddle Engines — Screw Engines — Horizontal — Trunk Engines — Return Connecting-rod 
Engines — Vertical Direct-acting Engines — Comparison of Present and Past Practice — 
Arrangement of Cylinders — Compound Engine — Three-, Four-, and Six-cylinder Com- 
pounds — Single and Two-crank Engines — Three-, Four-, Five-, Six-, and Eight-crank 
Engines — Oil Engines — Diesel — Semi-Diesel — Reversal of Oil Engines — Of Propeller — ■ 
Turbine Machinery — Experiments with Turbines — Combination of Turbines with Reci- 
procators — Geared Turbines — Design of Oil Engines — Double-acting Diesel — Two-cycle 
Diesel — Fuel Consumption of the latter — Gas Engines. 

CHAPTER IV. 

Steam used Expansively, . . . pp. 109-135 

Reciprocating Engines — Turbines — Expansion in Stages — Moist Steam — Adiabatic 
Expansion — Clearance — Effect of Clearance and Cushioning — Mean Pressure in a Com- 
pound Engine — Actual Mean Pressure in Practice — Frictional Resistance in Stop Valves 
and Pipes — Wire-drawing — Liquefaction during Expansion — Timing of h'xhaust — Lead 
— Calculation of Expected Mean Pressure — Graphic Method. 

CHAPTER V. 

Steam used after Expansion— Turbines, . . pp. 136-151 

Modern Reaction Turbine — Modern Impulse Turbine — Combination Turbines- 
Shape of Passages — Compound Turbines — Design of Screw for Turbine Ships — Efficiency 



X CONTENTS. 

of Turbine — Power developed by Turbine — Rateau's Formula for Steam Consumption — 
Sea Experiences with Turbines— Torsion Meters — Amsler's Meter — Fottinger's Meter — 
Hopkinson-Thring Meter — Bevis-Gibson Meter — Collie's Meter — Denny-Johnson Meter — 
Shaft Horse-power — Steam Engine Indicator. 

CHAPTER VI. - 

Efficiency of Marine Engines, . . pp. 152-173 

Water consumed per Shaft Horse-power — Mechanical Efficiency — Vertical Engines — 
Efficiency of Turbines and Reciprocators — Results of Trials with Triple-compound 
Engine — Yarrow's Experiments with Torpedo Boat — Steam Efficiency — High Pressure : 
Its Advantages and Disadvantages — Efficiency of the Engine as a Machine — Friction of 
Piston — Of Stuffing-boxes, Guides, and Slides — Friction at Shaft Journals — Of Valve 
Motions — Loss from Pumps — Inertia of Moving Parts — Losses due to Mechanical Defects 
and Physical Causes — Experiments with Superheated Steam — Resistance of Pipes and 
Passages. 

CHAPTER VII, 

Engines — Simple and Compound, . . -pp. 174-194 

Elementary Steam Engine — Genesis of Compound Engine — Expansive Engine — 
Effects of Increase of Pressure — Progress made by Early Engineers — Receiver Compound 
Engine — Expansive and Compound Engines Compared — Division of Work — Direct- 
expansion Compound Engine — Requisites in the Marine Engine — Comparative Theoretical 
Efficiency of Various Marine Engines — Further Comparison of Efficiency of Engines — 
Experiments in the Mercantile Marine — Triple-expansion Compound Engine — Increased 
Pressure of Steam — Compound System of Cylinders. 

CHAPTER VIIL 

Horse-power — Nominal, Indicated, and Shaft or Brake. . pp. 195-209 

Nominal Horse-power — Lloyd's N.H.P. — Estimated Horse-power — Indicated Horse- 
power — Indicator Diagram — Mean Pressure — -Shaft Horse-power— -Thrust Horse- power 
— Brake Horse-power — Indicated Thrust — Tow-rope Horse-power- -Net Horse-power — 
Piston Speed and Revolutions — Rate of Revolution of Marino Engines — Revolutions — 
Length of Stroke. 

CHAPTER IX. 

General Design and the Influences which effect It, . . pp. 210-227 

General Design and Arrangements — The Condenser — Inclination — Effect of External 
Causes— Supply of Materials — Influence of Tonnage Laws — Of the Board of Trade — 
Balancing and Avoidance of Vibration — Pitch of Screw and Vibration — Suction, Cavi- 
tation, etc. — Steel Castings — Aluminium — Duralumin — Future of the Reciprocator. 



CHAPTER X. 

The Cylinder and its Fittings, . . -pp. 228-261 

Size — Rates of Expansion — Diameter of Cylinders — Back Pressure in L.P. Cylinder — 
Intermediate M.P. Cylinder — Arrangement of Cylinder — Size of Steam Ports — Triple and 
Quadruple Engines — Ratio of Cylinders — Drop Valves — Main Steam Pipe — Area through 
Valves— Steam Ports and Passages — Opening of Port to Steam — Exhaust Passages — 
Cylinder Liner — Width of Steam Ports — Piston Valves — "Drop" Valves — Double- 



CONTENTS. xi 

ported Valves — Steam Jackets — Boring Holes — Auxiliary Valves — Escape or Relit i 
Valves — Drain Cocks — Receiver Space — Column Pacings and Feet -Holding-down Bolts 
— Horizontal Cylinders — Oscillating Cylinders — Cylinder Covers — Cylinder Cover Studs 
and Bolts — Cylinder Flanges — Clearance of Piston — Valve- box Covers — small Doors 
and Covers — Lagging and Clothing of Cylinders— L. P. Cylinder Body — Stufling-boxcs and 
Glands. 

CHAPTER XI. 

The Piston — Piston-rod — Connecting-rod, . . pp. 262-285 

Piston — Ramsbottom's Rings — Common Piston Rings — Piston Springs — Cameron's 
I 'a tent — Mather & Piatt's Patent — Buckley's Patent — Qualter & Hall's Patent — 
Rowan's — Restrained Packings — Body of Piston — Pistons of Ordinary Marine Engines — 
Details of Construction — Junk-ring Bolts— Safety Rings and Lock Bolts — Solid Packings 
— Diameter of Piston-rod — Piston-rod Ends — Cast-steel Piston-rod Crosshead — Piston-rod 
Guides — Surface 01 Guide Block — Crossheads and Gudgeons — Connecting-rods — Con- 
necting-rod Bolts and Brasses — Caps of Connecting-rod Brasses — Gudgeon End of Rod. 

CHAPTER XII. 

Shafting, Cranks and Crank-shafts, etc., . pp. 286-333 

Shafting of Modern Engine — Crank-shaft of Oil Engine — Tunnel or Intermediate 
Shafting — Alternating Stresses — Effect of Stresses on Materials — Safe Working Stress — 
Twisting Moment — Resistance to Twisting — Torsional Stiffness of a Shaft — Bending 
Moment — Equivalent Twisting Moment — Crank-shafts — Curve of Twisting Moments — 
Momentum of Moving Parts — Overhung Crank — Paddle Shafts — Crank-shaft of Screw 
Engine — Built Crank-shaft — Couplings — Surface of Crank-pins and Shaft Journals — 
Drivers — Taper Bolts — Cross Keys — Propeller Shafts — Outer Bearing — Screw-shaft End 
— Stern Bush — Stern Tube — Thrust Shaft — Thrust — Friction Loss on Thrust Block — 
Revolutions per Square Inch on Thrust Block — Diameter of Thrust Collars — Length of 
the Bearings of Tunnel Shafting — Diameter of Shafts (Rules for Lloyd's, B.O.T., Bureau 
Veritas) — Shafts for Paddle Steamers — Turbine Shafting — Shafts of Oil Engine — Steel 
Shafts— Diameter of Propeller Shaft — Details of Crank-shafts — Hollow Shafts — Shafts 
for Oil Engines. 

CHAPTER XIII. 

Foundations, Bed-plates, Columns, Guides, and Framing, pp. 334-343 

Bed-plates and Foundations — Main Bearings — Caps or Keeps for Main Bearings — 
Main-bearing Bolts — Brasses — Columns — Guide-plates — Framing — Entablature of Oscil- 
lating and Steeple Engines. 

CHAPTER XIV. 

The Condenser, . . pp. 344-372 

The Common or Jet Condenser — The Amount of Injection Water — The Area of 
Injection Orifice — Snifting Valve — Surface Condenser — Condenser Tubes — Condenser 
Efficiency — The Effect on Economy of Consumption — Surface Condenser Efficiency — 
Effect of Air mixed with Steam — Air Leaks to Condenser — Flow of Vapour and Water — 
Cooling Surface — Allowances of Cooling Surface — Cooling Surface per Horse-power — 
Greatest Length of Tube — Condenser Tubes — Tube-plates — Tube Packings — Steam Side 
and Water Side of Tubes — Spacing of Tubes — Body of the Condenser — Construction of 
the Surface Condenser — Shape of the Modern Condenser — Stiffness of Flat Surfaces — 
Quantity of Cooling Water — Passage of Circulating Water — Size of Circulating Pump 
Pipes — Size of Inlet and Discharge Pipes — Extra Supply Cock — Manholes — Drain Cocks 
— Testing — Cementing — Evaporators. 



x ii CONTENTS. 

CHAPTER XV. 

Pumps pp. 373-411 

Air-pump — Single-acting Vertical — Edwards' Air-pump — Double-acting Air-pump — 
Weir's Dual Pump — Air-pump Indicator Diagrams — Efficiency of Air-pumps — Air-pumps 
with and without Foot Valves — Size of Air-pump — Capacity — Air-pump for Jet-condenser 
— For Surface Condenser — Rotary Air-pumps — Parsons' Vacuum-augmentor — Pump 
Rods — Pump Buckets — Valves — Coe & Kinghorn's Patent — Thompson's — Beldam's — 
Area through Valve Seats — Suction Pipes from the Condenser to the Air-pump — Circu- 
lating Pumps — Single-acting Pump — Double-acting Pump — Size of Circulating Pump — 
Circulating Pump Rods, Buckets, and Valves — Valve Area — Diameter of Suction Pipes — 
Rotary Pumps — Centrifugal Pumps — Pumps of R.M.S. " Mauretania " — Feed Pumps — 
Sea Water — Net Feed Water — Relief Valves — Valves and Valve-boxes — Air Vessels — 
Feed Pipes — " Pet Valves " — Feed Tank — Feed-pump Rod — Bilge Pumps — Capacity of 
Bilge Pumps — Directing Boxes — Mud Boxes — Sanitary Pump — Ejectors — Power for 
Circulating Pumps. 

CHAPTER XVL 

Valves and Valve Gear, . • . pp. 412-443 

Seaward's Valves — Locomotive, Trick, Double-ported and Treble-ported Valves — 
Travel of Flat Valves — Piston Valves — Relief Frames — Double Valves — Dawe & Holt's 
Patent — Common Relief Frame — Martin and Andrews' Valve — Through Exhaust Valve 
Back Guides and Springs — Balance Pistons— Joy's Assistant Cylinder — Valve Rods or 
Spindles — Valve-rod Bolts — Valve-rod Guides — Proportions of Slide Valves — Double- 
ported Valve — Link Motion — Slot Link — Position of Suspension Pin — Size of Slot Link — 
Single-bar Link — Double-bar Links — Size of Bar Links — Single Eccentric Gear — Hack- 
worth's Dynamic Valve Gear — Marshall's Valve Gear — Joy's Valve Gear — Sell's Valve 
Gear — Expansion Valves — Grid-iron Expansion Valves — Outside Cut-off Valves — Inside 
Cut-off Valves — Piston Expansion Valves— Expansion Valves for Compound Engines — 
Eccentrics — Eccentric Straps — Eccentric Rods — Reversing Gear — Diameter of Weigh- 
shaft. 

CHAPTER XVII. 
Valve Diagrams, * . • . pp. 444-456 

Motion of the Piston — Lead — Inside Lap— Effect of " Notching-up " — Expansion 
Valve on an Independent Face, Central Position Ports closed — Ditto. Open — Expansion 
Valve working on the Back of Main Valve, Cutting-off at Inside Edge, Variation in Cut-off 
by Varying Travel — Expansion Valve on Back of Main Valve, Cutting-off at Outside Edge 
— Construction of Valve Diagrams — Effect on Indicator Diagrams of Crank Sequence. 



CHAPTER XVIII. 

Propellers, . . . . .pp. 457-501 

Fundamental Principle — -Stream Water — Effects of Wake Currents on Screw — The 
Paddle Wheel — Nominal or Apparent Slip — Real Slip — Path of Blade Tips — Dimensions 
and Form of Propeller — Paddle with Fixeil Floats — Feathering Floats — Control of Floats 
— Diameter of Feathering Wheel — Area of Float — Number and Proportions of Floats — 
Immersion of Float — Bosses — Paddle Arms — Position of Gudgeons — Outer Bearing — 
The Screw Propeller — Surface — Smith's Original Screw — Developed and Projected 
Surfaces — Diameter of Screw — Indicated Thrust — Rankinc's Rule for Thrust — Pressure 
per Square Inch on Propeller Blade — Diameter of Screw suited to Ship — Limit to Diameter 
— Pitch of Screw — Pitch Ratio — Surface Ratio — Acting Surface — Thrust — Thickness of 



CONTENTS. Xlll 

Blade — Propeller Boss — Blades— Number of Blades — Shape of Blades — Section of Blades 
— Studs or Bolts — Diameter of Bolts — Material for Blades — Weight — Feathering Screws 
— Bevis' Patent — Lifting Screws. 

CHAPTER XIX. 
Sea Cocks and Valves, . . .pp. 502-510 

Kingston Valve — Discharge Valves — Bilge Valves — Communication Boxes — Bilge 
Suction Piping — Water Service — Expansion Joints — Safety Collars — Flanges. 



CHAPTER XX. 

Auxiliary Machinery, .... pp. 511-537 

Classes of Auxiliary Machinery — Exhaust Steam — Intermittent and Constant Working 
— Electric Light Engines — Use of Electric Current for all Auxiliary Work — Loss of Steam 
— Feed Pumps — Engine - room Pumps — Reversing Gears — Ventilating and Forced 
Draught Fans — Hvdraulic Pumps and Accumulators — Air Compressor — Brown's Steam 
Tiller — Refrigerating Machines — Auxiliary Condenser. 



CHAPTER XXI. 

Boilers, Fuel, etc., Evaporation, . . .pp. 538-556 

Efficiency of Furnace — Chimney Draught — Fuel — Patent Fuels — Oil Fuel — Oil 
Burners — Value of a Fuel — Rate of Combustion — Artificial Draught — Howden's System 
of Forced Draught — Ellis and Eaves' System — Quantity of Fuel burnt on Grate — Size 
of Funnel — Combustion Heating Surface — Size and Height of Funnel — Evaporation — 
Tube Surface — Evaporative Power — Efficiency of Boiler. 



CHAPTER XXII. 

Boilers— Tank Boiler Design and Details, . , pp. 557-585 

Forms of Tank Boiler — Cylindrical Boilers with Two Furnaces — With Three Furnaces- 
■ — With Four Furnaces — Naval Boilers of To-day — Small Boilers — Double-ended Boiler — 
Oval Boilers — Holt's Boiler — Dry Combustion Chamber Boiler — Gunboat Boilers - 
Vertical Cylindrical Boiler — Locomotive Boiler — Wet Bottom Locomotive Boilers — 
Double-ended Locomotive Boilers — Seaton's Locomotive Boiler — Dimensions of a Boiler 
— Area of Fire Grate — Consumption of Fuel — Heating Surface — Tube Surface — Total 
Heating Surface — Efficiency of Heating Surface — Efficiency of Boilers — Area through. 
Tubes — Capacity of Boiler Shell — Steam Space — Area of Uptake or Funnel Sections — 
Capacity of Funnels — Examples of Boiler's Details. 

CHAPTER XXIII. 

Water-tube Boilers, . . . . pp. 586-633- 

Development of Grate Area — Tubes — Circulating Tubes — Stays — Steam Drums— The 
Two Types — Herreshoff Boiler — Babcock & Wilcox — Normand — Ferguson & Fleming 
— Seaton's Boiler — Yarrow — Blechynden-Wnite-Forster — Thornycroft — Mumford — Du 
Temple — Reed — Belleville — Diirr — Niclausse — Stirling — Hohenstein— Miyabara— Thorny- 
croft-Marshall — Amount of Water in Water-tube Boiler — Feed Arrangements— Fire 
Places — Consumption Trials. 



X1V CONTENTS. 

CHAPTER XXIV. 

Boilers— Construction and Detail. . . pp. 634-668 

Testing by Hydraulic Pressure — Admiralty Rules, Bureau Veritas, German Govern- 
ment Rules for Testing — Boiler Shell, Cylindrical — Riveting — General Rules for Riveted 
Joints — Treble-riveting — Butt Joints with Double Butt Straps — Treble-riveted Butt 
Joint — Thickness and Breadth of Butt Straps — Treble-riveted Lap Joints — Double Butt 
Straps and Double-riveting — Butt Joints with Double Straps and Single-riveted — Do. 
Double-riveted — Butt Joints with Double Straps, Treble-riveted — Treble-riveted (zig- 
zag) Butt Joint — Do. Quadruple-riveted — Welded Joints — Circumferential Seams — 
Methods of Work— Material — Allowance for Wear — Boiler Ends — Riveting — Quality of 
Plate — Thickness of End Plates — Furnaces — Rule for Thickness of a Plain Furnace — 
Methods of Stiffening Furnaces — Methods of Connecting Furnaces to End Plates — Com- 
bustion Chambers — Tubes — Stay Tubes — Stays — Continental and British Practice — 
Water Spaces — Manholes — Weight of Boilers. 

CHAPTER XXV. 

Boiler Mountings and Fittings, . , pp. 669-690 

Smoke-box — Funnel — Furnace Fronts and Doors — Fire Bars — Martin's Patent Bars 
— Henderson's Patent Door and Bars — Crude Petroleum Residuals — Stop Valve — 
Safety Valve — Size of Safety Valve — Internal Pipes — Feed Valves — Automatic Feed 
Valves — Blow-off Cock — Scum Cock — Water Gauge — Steam Gauge — Sentinel Valve — 
Air Valve — Weir's Hydrokineter — Steam Whistles — Separator — Boiler Clothing — 
Cameron's Patent Lagging. 

CHAPTER XXVI, 

Fitting in ol Machinery, Starting and Reversing of Engines, etc., pp. 691-717 

Fitting Machinery into a Ship — Boring the Stern Post — Engine Seatings — Thrust 
Block Seating — Pedestals for Tunnel Shafting — Boiler Seatings or Bearers — Fitting 
Machinery on Board the Ship — Holding-down Bolts — Staying Engines — Ramming Chocks 
and Stays — Boiler Seats — Copper Pipes — Starting and Reversing Engines — Brown's 
Patent Reversing Gear — Steam Gear for Reversing of Propellers driven by Turbines — 
Blenkinsop's Arrangement for Manoeuvring — Ships with Four Screws driven by Turbines 
— Regulation and Stop Valves — Steam Turning Gear — Ash Hoists — Governors — Dunlop's 
Governor — Steam Governor — Durham & Churchill's Velometer — Coutt's and Adamson's 
Governor — Westinghouse Governor — Gauges — Lubricators and Impermeators — Cadman's 
Patent Lubricators — Centrifugal Lubricators — Sight Feed Lubricator — Mechanical 
Impermeator — Drain Pipes — Jacket Drains — Feed Heaters — Weir's Feed Heater — 
Evaporators — Ladders — Gratings and Platforms — Feed Filters — Stokehole Ventilators. 



CHAPTER XXVII. 

Weight and other Particulars of Machinery relating thereto, . pp. 718-744 

Weight of Engines and Boilers — Warships— Torpedo Boat and Destroyer — Battle- 
ships and Cruisers — First, Second, and Third Class Cruisers — Piston Speeds — Boiler and 
Mean Pressure — Materials — Design — Effect of Loads and Stresses — Parts subject to 
Intermittent Stresses — To Alternating Stresses— Shearing and Torsion — Elastic Limit- 
Safe Working Stresses — Stretch under Tension — Paddle Engines — Naval, Express, and 
Cargo Ship Engines— High-speed Craft — Boilers — Cost — Auxiliaries and Appurtenances 
— Spare Gear — Fuel — Weight of Modern Machinery — Relation of Weight to Tonnage — 
X. Atlantic Service — Steamers on Routes to India, Cape, and Australia — Battleships and 
Cruisers — Scouts and Destroyers — Cross-Channel Steamers — Shallow Water Steamers — 
Paddle-wheel Steamers — Turbine Steamers. 



COXTK.NTR. 



XV 



CHAPTER XXVIII. 

Effect of Weight— Inertia and Momentum— Balancing the Same, . pp. 715-801 

Fixed Parts — Moving Parts — Momentum — Balancing : — Preliminary Definitions — 
Harmonic Motion — Inertia of Parts of an Engine — Of Connecting-rod — Of Valve Gear — 
Of Air Pomp — Single-crank and Valve Gear balanced by Rotating Weights only — Do. 
by Bob Weights and Rotating Weights combined — To Balance an Engine of any Number 
of Cranks — Curves of Free Force and Couples for a Multi-crank Engine — Four-crank 
Engine — Yarrow-Schlick-Tweedy System — Modified Principle of Balancing — Effect of 
Inertia ami Weight — Stresses due to Inertia — Design of Balance Weights. 



CHAPTER XXIX. 

Materials used by the Marine Engineer, . . pp. 802-835 

Cast Iron — Different Kinds of Pig — Cold- blast Iron — Mixtures — Specific Gravity of 
Cast lion — Strength — Wrought Iron — Rolled Bar — Merchant, Best, etc., Bars — Weight 
of Bars — Forgings — Steel — Bessemer, Siemens-Martin — Steel for Boiler Construction — 
Admiralty, Board of Trade, Lloyd's, British Corporation Rules — Sizes of Bars and Plates 
— Price — Steel Forgings — Castings — Nickel Steel — Chrome-Vanadium Steel — Soft Steel 
for Solid- drawn Tubes — Manganese Steel — Copper — Tin — Zinc or Spelter — Lead — 
Aluminium — Duralumin — Antimony, Alloys, Brass, Muntz Metal — Naval Brass — Tube 
Metal — Gun-metal — Admiralty and Phosphor Bronzes — Crotarite — Immadium — Tur- 
b.ulium — Stone's Bronze — Bull's Metal — Melloid — Delta Metal — Parsons' White Bronze 
— Babbit's White Metal — Admiralty White Metal — Plumtine — Magnolia — Fenton's 
White Metal — Stone's White Bronze — German Silver — Richard's Plastic Metal — Tables 
on Strength, Composition, Properties, and Prices of Metals, etc. — Effect of Temperature 
on Strength of Metals. 

CHAPTER XXX. 

Oil and Lubricants — Engine Friction, . . pp. 836-844 

Consumption of Oil — Kinds of Oil — Animal Oils — Vegetable Oils — Mineral Oils — 
Density — Viseosity — Boiling, Setting, and Flash Points — Compound Oils — Soaps — 
Adulteration — Solid Lubricants — Tallow — Grease — Coefficients of Friction. 



CHAPTER XXXI. 

Tests and Trials : Their Objects and Methods, . . pp. 845-858 

General Considerations — Factors or Margin of Safety — Admiralty Rules for Testing 
Machinery — Italian Government — British Corporation — Lloyd's — Board of Trade — 
Germanischer Lloyds — Trials under Steam — Admiralty Ship Trials — Merchant Ship 
Trials — Modern Steamship Trials — Progressive Trials — Do. at Sea — Endurance Trials — 
Mercantile Marine. 



APPENDIX A. 

The Diesel Oil Engine, PP- 859-875 

Efficiency — Rate of Revolution — Two-cycle Engines — Double-acting Oil Engines — 
Junker Engine — Reversing Oil Engines — Manoeuvring — Twin-screw Oil Ship " Fiona " 
— Doxford's Double-piston Engine — Examples of Oil-engined Ships — Semi-Diesel 
Engines. 



iVi CONTENTS. 

APPENDIX B. 
Lloyd's Register Rules for Internal Combustion Engines, . pp. 876-884 

APPENDIX C. 
Bureau Veritas Rules for Internal Combustion Engines, . „ . pp. 885-886 

APPENDIX D. 
Turbines on Shipboard, . . . . . . . pp. 887-896 

APPENDIX E. 
Superheated Steam and Superheaters, . . . . . pp. 897-903 

APPENDIX F. 

Board of Trade Rules for Boilers, ..... pp. 904-911 

Surveyor's Duty and Responsibility — Working Pressure — Girders for Flat Surfaces 
— Compression Stress on Steel Tube Plates — Cylindrical Boilers — Rules. 

APPENDIX G. 

Lloyd's Rules for determining the Working Pressure to be allowed in New 

Boilers, ........ pp. 912-916 

APPENDIX H. 

Rules of the British Corporation for the Working Pressure, Thickness of 

Plates, and Sizes of Stays for Steel Boilers. . . . pp. 917-921 

APPENDIX I. 

Bureau Veritas Rules for Boiler Shells, Working Pressure or Thickness of 

Plates, and for Sizes of Stays, . . . . -pp. 922-930 

APPENDIX J. 
Admiralty Rules for Steel Boilers and Materials and their Tests, . . pp. 931-935 

APPENDIX K. 
German Government Rules tor Boilers. . . . . .pp. 936-944 

APPENDIX L 
Lloyd's Rules lor Electric Light Installations on Board Vessels, . . pp. 945-948 



CONTENTS. XVli 

APPENDIX M. 
Refrigerating Machinery and Appliances, Lloyd's Registers' Rules, . . pp. 940-954 

APPENDIX N. 
Rules for Vessels Fitted for Burning and Carrying Liquid Fuel, . • .p. 955 

APPENDIX 0. 
Board of Trade Rules tor Safety Valves, . . . . . pp. 956-960 

APPENDIX P. 
Bureau Veritas Rules for Boiler Safety Valves and Metal Tests, etc., . . p 961 

APPENDIX Q. 
Admiralty Rules for Testing Materials for Machinery, . . . pp. 962-964 

APPENDIX R. 

Recommendations for Main Boilers of the British Marine Engineering Design 

and Construction Committee, . . ... • • PP- 965-967 

Index, ......... P- 969 



LIST OF TABLES. 



TABLE 

I. Showing Progress of Naval Engineering — Screw Propulsion, . 
II. „ „ Engineering in the Mercantile Marine, . 

III. John's Coefficients for Computing Effective Horse-power, . 

IV. Coefficients for Computing I.H.P. by the old Admiralty Formula, 
V. Coefficients (Prismatic) of Displacement suitable for Various Speeds, 

VI. Relation of Power to Speed (Sir W. H. White), 

VII. Results of Trials of Ships at Speeds from 9 to 12 Knots, . 
VIII. „ „ „ „ „ 12 to 15 „ 

IX. „ ,, ,, „ 15 to 18 -„ . 

X. „ „ „ „ 18 to 21 ,, 

XI. „ „ „ „ „ 21 and upwards, . 

XII. „ „ „ fitted with Turbines, 

XIII. „ „ „ „ Paddle-wheels, 

XIV. Relation of Power and Displacements, .... 
XlVa. Comparative Trials of Ships fitted with Turbines and Reciprocators, 

XV. Water Consumption of Engines of H.M.S. " Amethyst " and " Topaz' 
Compared, ........ 

XVI. Results of Measured Mile Trials of s.s. " Otaki," 
XVII. Steam used Expansively, ...... 

XVIII. Factors for Estimatimg Mean Pressure of Steam in Practice, 
XIX. Results of Trials of Naval Twin-screw Ships, . 
XX. „ „ Merchant Steamships, . . . 

XXI. „ „ Two-stage Compound Engines, . . 

XXII. „ „ Three-stage Three-crank Engines, . 

XXIII. „ „ „ Four-crank Engines, . 

XXIV. „ „ Four-stage Expansion Engine, , . 
XXV. Efficiency of some German Engines, . . . . 

XXVI. Yarrow's Experiments on Efficiency, .... 
XXVII. Tyacke's Experiments on Efficiency, ..... 159, 
XXVIII. Work done Theoretically by One Pound of Steam, . . 
XXIX. Examples of Typical Engines, ..... 

XXIXa. Rates of Revolution of Screw Propellers, . . . 

XXIX6. Strokes of Engines as in Common Practice, 
XXX. Trials of a Typical Cargo Steamer, .... 

XXXI. Factors for obtaining Sizes of Steam, Exhaust Pipes, etc., 
XXXII. Weight of Steam passed through Pipes with a Drop of 1 Lb., 

XXXIII. Stuffing-boxes and Glands, Sizes of, ... . 

XXXIV. Ratio of Maximum to Mean Twisting Moments at Various Cut-offs, 
XXXV. Board of Trade Factors for Shafts, .... 

XXXVI. Lloyd's Factors for Shafts of Steam Reciprocators, . 
XXXVII. and XXXVIIa. Lloyd's Factors for Shafts of Oil Engines, . 
XXXVIII. British Corporation Factors for Shafts, .... 

XXXIX. Factors for Shafts per Author's Formula, 
XXXIXa. „ Paddle-shafts, per Author's Formula, 

XXXIX6. „ Crank-shafts of Screw Engines, per Author's Formula, 

XL. Effect of Vacuum on Steam Consumption, 

XLI. Temperature, Latent Heat, and Volume of Steam at very Low Pressure, 
XLII. Number of Tubes in a Condenser per Square Foot of Tube Plate, 
XLIII. Trials of Battleship " Ibuki "—"Curtis Turbines, 
XLIV. Ratio of Cooling Water to Steam Condensed, . 



PACK 

17 

18 
26 
27 
28 
41 
42 
43 
44 
45 
46 
47 
48 
49 
92 

93 
97 
111 
125 
130 
131 
132 
133 
134 
135 
156 
159 
, 160 
161 
176 
208 
209 
232 
247 
248 
260 
298 
321 
322 
323 
324 
329 
331 
332 
362 
363 
367 
368 
370 



XX 



LIST OF TABLES. 



TABLE 

XLV. Ratio of L.P. Cylinder to Circulating Pump Capacities, 
XLVI. Centrifugal Pumps for Circulating Cooling Water, 
XLVII. Particulars of Screw Propellers, Solid and Loose-bladed, 
XLVIII. Thickness of Copper Pipes for Various Purposes, 
XLIX. Steam and Fuel Consumption of R.M.S. " Lusitania," 

L. Particulars of Admiralty Type of Fans for Air Circulation, . 
LI. Fuels of all Kinds (Solid), ...... 

Llcr. Fuels (Liquid), ....... 

LII. Comparison of Various Designs of Boilers, .... 

LIII. Allowance of Total Heating Surface in Practice, 
LIIIo. Capacity of Funnels, ...... 

LIV. Particulars of Marine Boilers made under Old Rules, 
LV. „ „ „ „ „ New Rules, 

LVI. Basin and Sea-going Trials of H. M.S. " Sheldrake," 
LVII. Particulars of Surface, Weight, etc., of Express Boilers, 
LVIII. Special Trials of Belleville Boilers in H.M.S. " Sharpshooter," 
LIX. Results of Steam Trials with Niclausse Boilers, 
LX. Trials of the Miyabara Boiler, ..... 

LXI. Water Consumption of Main and Auxiliary Machinery — H.M.S. 
"Hermes," . . . . 

LXII. Water Consumption of Main and Auxiliary Machinery — H.M.S. 

"Diana," . . . 

LXIII. Results of Trials of Various Modern Marine Boilers in Lbs. per Hour, . 
LXV. Trials of Various Naval Ships — Water Tube compared with Tank 
Boilers, ........ 

LXV1. Trials of Various Marine Boilers, ..... 

LXVII. Tests of Boiler Material prescribed by Various Authorities, . 
LXVIII. Strengths of Various Kinds of Riveted Joints, 
LXIX. Single-riveted Lap Joints, ...... 

LXX. Particulars of Double-riveted Lap Joints of Plates, . . 

LXXI. ,, Treble-riveted Lap Joints of Plates, . . 

LXXII. „ Treble Special Riveted Lap Joints of Plates, . 

LXXIII. „ Double -riveted Butt Joints of Plates, . 

LXXIV. „ Double Special Riveted Butt Joints of Plates, 

LXXV. „ Treble Ordinary Riveted Butt Joints of Plates, 

LXXVI. „ Treble Special Riveted Butt Joints of Plates, . 

LXXVI1. Relative Thickness of Shell Plates for Different Tensile Strengths, . 
LXXVIII. Values of Factor F in Formula for Shell Plates, 
LXXIX. Boiler Tubes, Standard Sizes and Thicknesses, 
LXXX. Particulars and Scantlings of Modern Cylindrical Boilers, 
LXXXI. Details of Riveting for Funnels, Casings, etc., . . 

LXXXII. All-round Reversing Gears, Particulars of, . 
LXXXIII. Limits of Safe Working Stresses on Various Metals, . 
LXXXIV. Conditions obtaining with Machinery of Modern Ships, 
LXXXV. Particulars of Machinery of Various Naval Ships, 
LXXXVI. ,, ,, ,, Ocean Express Steamers, 

LXXXVII. ,, ,, „ Passenger Cargo Steamers, 

LXXXVIII. ,, ,, „ Small Passenger Steamers, 

LXXXIX. „ „ „ Paddle-wheel Steamers, 

XC. ,, „ „ Turbine Screw Steamers 

XCI. Weights of Moving Parts of Balancing Engines, 
XCII. Admiralty Tensile Tests for Boiler Steel, 
X ( ' 1 1 1 . Board of Trade Tensile Tests for Boiler Steel, 
XCIV. Composition of Metal and their Alloys, 
XCV. ,, Various White (Bearing) Metals. 

XCVI. Phvsical Properties of Various Metals, 
XCVII. Current and Past Prices of Various Materials (1912). 
XCVI 1 1. Standard Tests for Steel adopted in U.S. America, . 
XCIX. Constituents of Solid Matter in River and Sea Water, 
C. „ ,, „ deposited in Boilers, . 

CI. Effect of Temperature on Metals, 
CII. Specific Gravity of Various Oils at "80° F., 



PAGE 

397 
410 
478 
509 
513 
522 
544 
545 
575 
579 
583 
584 
585 
596 
612 
616 
619 
623 

627 

628 
628 



630 

. 631 

. 635 

. 638 

. 639 

. 644 

. 644 

. 645 

. 647 

. 647 

. 647 

. 648 

. 650 

. 652 

. 661 

. 667 

. 671 

. 699 

. 724 

. 735 

. 740 

. 740 

. 742 

. 744 

. 7446 

. 744c 

. 779 

. 809 

. 810 

823, 824 

. 825 

826, 827 

. 828 

829, 830 

. 832 

. 832 

. 834 

. 838. 



LIST OF TABLES. 



XXI 



TATttK PAQB 

(III Density of Various Oils at Different Temperatures, . . . s:{'.» 

CIV. Viscosity of Various Oils at Different Temperatures, . . . 839 

CV. Boiling, Setting, and Flash Points of Various Oils, . . . siu 

CVI. Prices of Various Oils, . . . . . .sin 

('VII. Coefficients of Friction of Various Materials, . . . 84.'! sll 

CVIII. Examples of Oil Engine-driven Ships, .... 879 

CIX. Methods of Transmission from Turbine to Screw, . . . 895 

CX. Steam Consumed per S.H. P. -hour (various conditions). . . 898 

CXI. Superheated Steam through Pipes, Lbs. per Minute, . . . 903 

CXII. Board of Trade Factors of Safety for Boilers, . . . 907 

(XIII. Constants for Joints of Boiler Plates with Different Kinds of Riveting, 911 

CXIV. Values of 2 R in the Formula of the Bureau Veritas for Boiler Shells, 923 

CXV. Value of Bureau Veritas Multiplier Factor for Flat Plates, . . 927 

CXVI. Board of Trade Allowances for Safety Valve Area*;, . , . 957 



LIST OF ILLUSTRATIONS. 



FIO. 



DESCRIPTION OF ILLUSTRATIONS. 



1. Thornycroft's Stern for Shallow-draught Screw Ships, . 

2. Yarrow's Drop-flap Shallow-draught Screw Ships, . 

3. Paddle Steamer " Charlotte Dundas," 1802, . 
Machinery and Wheels of Paddle Steamer " Comet," 1812, 
Modern American Lake Paddle Steamer, 
Turbine Steamer " Duchess of Argyll," 
Engines and Paddle Wheels of a Stern Wheeler, 
p.s. " Empress Queen," 11,000 I.H.P., 
Early Screw Propellers of Smith and Ericsson, . 
Kirk's Analysis — Block Model, 
Curve of I.H.P. for Varying Speeds, 
Curves of Power, etc., of s.s. " Lusitania " on Trials, . 
Trials of H.M.S. " Cossack " in Water of Different Depths, 
Harold Yarrow's Experiments in Water of Different Depths, 



4. 

5. 

6. 

7. 

8. 

9. 
10. 
11. 
12. 
13. 
14. 
15. 
16. 
17. 
18. 
19. 
20. 
21. 
22. 
23. 
24. 



N. D. Lloyds Co.'s „ „ 

Thornycroft Four-Cylinder Engine using Light Oils, 
Belliss and Morcom Heavy Oil Engine, 
Compound Diagonal Paddle Engine — Two Cranks, 

M „ „ — Cranks Coupled, 

Three-crank Compound Paddle Engine, 

„ Triple-compound Paddle Engine, 
American Steamer Beam Engine, 
Engine of the " Comet," 1811-12, 

25. Side-lever Engine, 

26. Engines of p.s. " Regent," 1816, 
27 and 27a. Oscillating Engine, . 

28. Steeple Engine, 

29. Twin-cylinder Engine, 

30. Diagonal Compound Paddle Engines, U.S.A., 

31. Trunk Engines, 
31a. Return Connecting-rod Engine, 

32. Three-crank Triple-expansion Engine (Naval), 

33. Vertical Quadruple-expansion Engine (Express), 
33a. „ „ „ (Mercantile Ordinary), 

34. Engines of H.M.S. " Salmon " (Destroyer), 

35. High-speed Cruiser Engines, .... 

36. Triple-compound Engines for Fishing Craft, . 

37. Five-cylinder Quadruple-expansion Engines of s.s. " Inchdune," 

38. Eight-cylinder „ „ „ <( " K.W.," ;> 

39. General Arrangement of Turbine, etc., of R.M.S. " Lusitania," 

40. Diagram showing Results of Trials of H.M.S. " Amethyst " and " 

41. Combination of a Central Turbine with Twin Reciprocators. . 

42. „ Twin Turbine's „ „ 

43. Longitudinal Section through Engine-room of R.M.S. " Olympic,' 

44. Plan of Engine-room of R.M.S. " Olympic," . 

45. The Wheel Gearing of s.s. " Vespasian," 

46. Combination of Twin Turbines with a Single Reciprocator, 

47. Oil Engine, Single-acting, Marine Type, Six Cylinders, 

48. „ Double-acting „ „ 

49. Mirrlees-Diesel Single-acting Marine Engine, . 

50. Marine Oil Engine, Two-cycle Single-acting, Diesel System, . 
50a. Section of „ 



Topaz," etc. 



PA OK 

2 

3 

6 

7 

8- 

9 

11 

14 

IS 

32 

38 

60 

51 

51 

53 

54 

58 

59 

61 

62 

63 

64 

65 

66 

67 

67 

68, 69 

71 

72 

73 

74 

74 

75 

76 

78 

79 

80 

81 

84 

87 

90 

91 

95 

96 

98 

99 

100 

101 

102 

104 

105 

106 

107 



XXIV 



LIST OF ILLUSTRATIONS. 



JIG. DESCRIPTION OF ILLUSTRATIONS. PAGE 

51. Two-cycle Oil Engine, showing Three Fundamental Stages, . . .108 

52. Turbo-motor, Single-stage, on De Laval System, . . . .110 
52a. ,, Two-stage or Compound, De Laval System, . . .110 

53. Rankine's Graphic Method for determining Mean Pressure, . . .112 

54. Theoretical Indicator Diagram showing compression, . . .114 

55. ,, ,, for Triple-expansion Engine, . . . 129 
55a. Diagrams as in Practice from Theoretical Diagrams, . . . 129 

56. Reaction Turbine, Equivalent Nozzles, . . . . .136 

57. Impulse „ „ „ ..... 137 
57a. Diagram showing Stages of Pressure Reduction, .... 139 
576. ,, ,, Variation in Velocity due to Pressure Reduction, . . 139 

58. Ljungstrom Turbine for Electric Propulsion, ..... 140 

59. Curtis Turbine as fitted in a Japanese Battleship, . . . . 141 
59a and 596. Zoelly Turbine for Marine Work, . . . . .142 

60. Fottinger Torsion Meter, . . . . . • .144 
60a. Diagrams from a Fottinger Torsion Meter, . . . . .144 

61. Hopkinson and Turing's Torsion Meter mounted on Shaft, . . . 146 
61a. „ „ „ Diagram of, . . . . 147 

62. Collie's Mechanical Torsion Meter, ...... 149 

63. Crank Effort Diagram from Torsion Meter of G. Hamilton Gibson, . . 151 

64. Friction Trials with Belliss and Morison's Engines, .... 158 
64a. Diagram showing Effect of Superheat on Consumption, . . .171 

' 65. Indicator Diagrams and their Combination of some Quadruple Engines, . 193 

66. Diagram showing the Various Stages in the Use of Steam Expansively, . 194 

67. Indicator Diagram, . . . . . . . .199 

68. Weir's Condenser as fitted to a Quadruple-expansion Engine, . . 213 

69. Single-crank Compound Engine — Two Cylinders, . . . .217 

70. Single-crank Triple-expansion Engine, ..... 218 

71. Triple-expansion Engine with Special Valve Gears, . . . . 219 

72. „ „ with Four Cranks, Balanced, . . . 220 

73. Various Arrangements of Cylinders of Triple-expansion Engines, . 236, 237 

74. „ „ ,, Quadruple-expansion Engines . . 239 

75. Cylinders of a Two-crank Quadruple-expansion Engine, . . . 240 

76. „ Three-crank Triple-expansion Engine, .... 242 

77. H.P. and L.P. Cylinders of a Naval Engine, . . . 244, 245 

78. Section through a Cylinder, ....... 250 

78a. Admiralty Method of fitting Liners, ...... 250 

79. Shortened Cylinder with Port in Cover, ..... 257 

80. Combination Metallic Packing, ...... 259 

80a, 806. Stuffing-boxes and Glands, ...... 259 

81. 81a. Piston with Ramsbottom Rings, ...... 264 

82. Piston with Common Packing and Junk Rings, .... 264 

83. „ Mather & Piatt's Packing Ring and Springs, . . . 264 

84. ,, Buckley's Packing Ring and Springs, .... 264 

85. „ Quarter & Hall's Packing Ring and Springs, . . . 264 
' 86. „ Cameron's Wave Spring and Common Rings, . . . 264 

87. Methods of Jointing Ends of Spring Rings of Pistons, . . . 264 

88. Packing Rings of Piston restrained on Admiraltv Method, . . . 266 

89. Forged Steel Piston, . . . . . . .267 

90. Cast Steel Piston, . . . . . . . .207 

91. Piston-rod with T End and Double Slipper Guide Shoes, . . - 270 

92. „ „ „ Single Slipper Guide Shoes, . . . 270 

93. „ formed with Head and Slipper Foot and Loose Gudgeon, . . 270 

94. „ Crosshead with Gudgeons in One Piece, .... 274 

95. „ „ „ „ Steel Casting, . . . 275 

96. Diagram showing Thrust of Connecting-rod, ..... 275 

97. Crosshead and Guide Block for Double Piston-rods, .... 278 

98. Connecting-rod with Solid Ends — Gudgeon shrunk in Place, . . . 279 

99. „ „ Fitted Brasses at each End, .... 2T'.t 

100. „ „ „ „ „ ConicalJaws, . . 284 

101. Indicator Diagram of a Diesel Oil Engine (Two-stroke Cycle), . . 287 



LIST OF ILLUSTRATIONS. 



XXV 



FIO. DESCRIPTION' OF ILLUSTRATIONS. 

L02. Diagram showing Turning Moment at various Angles of the Crank, 

103. Diagram of the Curve of Twisting Moments of a Single Crank, 
103a. Diagram of Combined Twisting Moments of Two Cranks, 

104. Crank Effort Diagram of a Three-crank Engine — Triple-compound, 

105. The Cranks of a Paddle-wheel Engine, 

106. Built-up Crank-shaft, 

107. Naval Crank-shaft. . 

108. Ccdervall's Screw Shaft Fittings, 

109. Michel's Thrust Block, 

110. Thrust Block with Adjustable Collars, 
110a. Overhung Crank (Bureau Veritas). 
1106. Double Ciank (Bureau Veritas), 

111. Crank-shaft Bearing, . 

112. Improved Form of Crank -shaft Bearing, 

113. Solid Steel Main Bearing Frame — Naval Engines, 

114. Cast Steel Main Bearing Frame — Naval Engines, 

115. Weir Uniflux Condenser for a Turbine— 11,000 S.H.P. 

116. Cylindrical Condenser on Morison's System, . 

117. Ordinary Marine Engine with Contraflo Condenser 
117a. Diagram showing Flow of Steam in the Contraflo System, 

118. Contraflo Condenser with Feed Temperature Regulator, 
118a. Morison's Diagram for Air saturated with Water Vapour, 
1186. Indicator Diagram showing Effect of High Vacuum, 

119. Condenser Tube fitted with Wooden Ferrules, 

120. „ ,, „ Screwed Glands and Tape Packings, 

121. „ ,, „ Safety-screwed Glands, 

122. Air Pump of Ordinary Type, . 

123. Edwards' Air Pump, .... 

124. Weir's Dual Air Pumps, 
124a. ,, ,, Diagrammatic Section of, 

125. Indicator Diagram of Air Pumps, 

126. Air Pumps, Motor-driven, 

127. Condenser with Kinetic Air Pumps, . 
127a. Parsons' Vacuum Augmentor, 
128-132. Air-pump Valves of Kinds. 

133. Impeller or Wheel of a Centrifugal Pump, 

134. 17-inch Circulating Pump, 
134a. Circulating Pumps by Allen for a Battleship and s.s. " Mauretania," 

135. Common Locomotive Slide Valve, 

136. The Trick Valve, .... 

137. Common Double-ported Flat Valve, . 

138. Piston Slide Valve, .... 

139. Self-balanced Dual Valve, 

140. Thorn's Patent Piston Valve, . 

141. Dawe & Holt's Patent Relief Frame for Flat Valves, 

142. Common Naval Relief Frame for Flat Valves, 

143. Spring and Packing Rings, Relief Frame for Flat Valves, 

144. Martin & Andrews' Valve Relief Frame, 

145. Church's Patent Valve Relief Frame, . 

146. Valve with Exhaust through its Back, 

147. Joy's Assistant Cylinder for Slide Valves, 
1476. ,, „ Indicator Diagrams of, 

148. Valve-rod Guide, 

149. Proportions of Locomotive Slide Valves, 

150. Slot Link and Rod Ends, 

151. Double-bar Link with Rods inside, 

152. „ „ outside, . 

153. Marshall's Patent Valve Gear, 

154. Joy's Patent Valve Gear, 

155. Effect of Notching-up on Indicator Diagrams, 

156. Diagram of the Piston Path, . 



PAGE 

295 

296 

296 

297 

299 

307 

308 

315 

319 

319 

326 

328 

336 

336 

338 

338 

351 

352 

354 

354 

356. 

357 

358 

36£ 

365 

365 

375 

376 

377 

378 

380 

383 

387 

388 

394 

399 

400 

403 

413 

414 

414 

415 

417 

418 

418 

419 

41!) 

420 

421 

423 

424 

424 

426 

427 

430 

432 

432 

435- 

430 

441 

444 



XXVI 



LIST OP ILLUSTRATIONS. 



fio. 

157. 
158. 
159. 
160. 
161. 
J62. 



DESCRIPTION OF ILLUSTRATIONS. 



Zeuner's Diagram for the Motion of a Slide Valve, 

Valve Diagram showing Effect of " Notching-up," 

Link Motion " Notched-up," .... 

Diagram of the Motion of an Independent Expansion Valve 

„ „ of an Expansion Valve on the Back of the Main Valve, 

„ ,, of another Expansion Valve on the Back of the Main 

Valve, .... 

163. Adjustable Expansion Valve on Back of the Main Valve, 

164. Effect of Sequence of Cranks on the Indicator Diagrams, 

165. Paddle-wheel with Feathering Floats, 

166. Locus of the Flat of a Feathering Wheel, 

167. Feathering Paddle-wheel with Wooden Floats, 

168. „ „ „ Steel-plate Floats, 

169. „ ,, ,, • Gear on Inner Side of Wheel, . 

170. Smith's Screw Propeller as first fitted in the Navy, 

171. Diagram showing Curve of Indicated Thrust, . 

172. Diagram showing Blade Pressures of Screw Propellers, 

173. Solid Cast-iron Screw Propeller, 

174. Bronze Screw Propeller with Loose Blades, Ordinary Surface Ratio, 

175. „ „ „ „ Large Surface Ratio, 

176. General Form of Modern Mercantile Marine Screw, 
176a. Typical Sections of Screw Propeller Blades, . 

177. Hirsch's Patent Screw of 60°, 

178. Solid Parsons' Bronze Screw of R.M.S. " Mauretania,' 

179. Damaged Bronze Screws, 

180. Bevis Patent Feathering Screw, 

181. Kingston Valve — Typical Design, 

182. Details of Inlet Valves on Steel Ships, 

183. Double Ram Flywheel Auxiliary Pump, 

184. Section of Worthington Duplex Pump with Externally-packed Rams, 

185. Weir's Automatic Pump, ..... 

186. ,, ,, with Float, Tank, and Automatic Gear, 

187. Compound Engine for Electric Generating — Enclosed Self -lubricating, 

188. „ „ „ „ — Single-crank, 

189. „ „ „ „ — Two-crank, Naval, 
189a. „ „ „ „ „ Section of, 

190. Turbo Electric Installation for Ship Lighting, etc., 

191. Air Compressors for High-pressure Naval Work, 
191a. „ „ „ „ Section of, 

192. Steam Steering Gear with Hand Gear (vertical), 

193. „ . „ „ „ (horizontal), 

194. „ „ as fitted in Warships, 
'195. „ „ Brown's Patent, 

196. Refrigerating Apparatus for Warships, 

197. ,, ,, Merchant Ships, . 

198. Section of Hall Refrigerating Apparatus, 

199. Auxiliary Condenser for Merchant Ships, 

200. „ „ Warships, 

201. Holden's Apparatus for Burning Oil Fuel, . " 
201a. Lassac Lovekin Apparatus for Burning Oil Fuel, 

202. Furnace fitted with Apparatus for using Oil Fuel. 

215. Goldsworthy Gurney's Water-tube Boiler of 1827, 

216. Ward's High-pressure Coil Boiler, 

217. Babcock & Wilcox Boiler, Large Tube, Naval Type, 
"218. ,. „ Headers for Small Tubes, 
218a. „ „ „ „ Large 

219. „ „ Small Tube, Naval Type, 

220, 221. „ „ Mixed Sizes of Tubes, Naval Type, . 698, 

222. Normand Water-tube Boiler, . 

223. Fleming & Ferguson's Water-tube Boiler, 
■224, 224a. Seaton's Water-tube Boiler, 



PAGE 

445 
450 
450 
451 
453 

454 

454 

455 

462 

463 

468 

468 

470 

474 

477 

479 

489 

491 

493 

494 

494 

495 

496 

498 

500 

503 

504 

517 

518 

519 

520 

523 

524 

524 

525 

525 

526 

527 

528 

528 

529 

531 

534 

534 

535 

536 

537 

540 

540 

542 

587 

593 

594 

595 

596 

597 

599 

600 

602 

603 



LIST OF ILLUSTRATIONS. 



xxvn 



pia. DESCRIPTION OF ILLUSTRATIONS. 

225. Yarrow's Large Tube Water-tube Boiler, 

226. „ Small Tube Water-tube Boiler, 

227. „ Boiler as made in Japan, 

228. Blechynden's Water-tube Boiler, 

229. Thornycroft's Water-tube Boiler, 

230. Mumford's Large Type of Water-tube Boiler. 

231. Reed's Water-tube Boiler for all Classes of Ships, 

232. Modern Belleville Boiler, with Economiser, . 
232a. Tube and Headers of a Belleville Boiler, 

233. Modem Niclausse Marine Boiler, 

234. Stirling Marine Boiler, .... 

235. Hohenstein Boiler, . " . . 

236. Miyabara Boiler, ..... 

237. Thornycroft-Marshall Boiler, 

237a. Special 9-rivet Quadruple Joint, . . 

2376. „ 11 -rivet „ „ . 

238-240. Methods of Flanging Boiler Ends, 

241. Furnace Stiffened with a Bowling Hoop, 

with an Adamson Joint, . 
by Differential Diameter, 
by Corrugation (Fox's Plan), 
by Ridges (Purves' Plan), 
by Curved Grooves (Morison's Plan), 
by Corrugation (Holmes' Plan), . 
„ (Deighton's Plan), 

Removable Furnace (Ashlin's Design), 



Methods of Connecting the Furnace to End Plate, 



242. 
243. 
244. 
245. 
246. 
247. 
248. 
249. 
250. 
250a. „ „ „ Tube Plate 

251. „ „ „ End Plate, 
251a. „ „ „ Tube Plate, . 

252. Admiralty Ferrule for Fire Ends of Tubes, . 

253. Funnel and Casing for Naval Ships, Yachts, etc., 

254. „ „ Merchant Ships (Ordinary), 

255. Naval Bulkhead Self-closing Stop Valve, . , 

256. Spring Safety Valves as fitted in Merchant Ships, 

257. Adams' Form of Valve and Seat of Safety Valves, 

258. Cockburn's Form of Valve and Seat of Safety Valves, 

259. Safety Valves as fitted in Naval Ships (Short Type), . 

260. „ „ ,, (Naval), Combined with 

261. Feed Check Valve (Improved Form) as fitted to Boilers, 

262. „ and Stop Valve Combined, 
262a. ,, Valve, Automatic, on Mumford's Plan, . 

263. Weir's Hydrokineter, or Water Circulator, 

264. Steam Syren and Organ Whistle Combined, . 
264a. Steam Syren — Admiralty Type, 

265. Steam Reversing Gear, Brown's Original Pattern, 
265a. „ „ ,, Improved Self-contained, 
2656. „ ,, „ and Hand Gear Combined, 

266. Reversing Gear for Twin-screw Turbines, 

267. Steam Manoeuvring Gear for very Large Turbines, 

268. Engine Stop and Regulating Valve for Small Engines, 

269. „ „ ,, Balance for Large Engines 

270. „ ,, ,, Equilibrium (Cockburn's) 

271. Ash Ejector, Steam-worked (See's), . 

272. Combination Governor (Murdoch's), . 

273. Centrifugal Lubricator for Crank Pins, 

274. Automatic Drainer, .... 

275. Feed-water Heater and Automatic Regulator (Weir's 
275a. „ ,, (Live Steam), 

276. 277. Fresh Water Evaporator and Distiller, . 

278. Fresh Water Supply Installation as in Battleships, 

279. Grease Extractor and Feed-water Filter, 



Stop V 



alve, 



713, 



PACK 

(ill I 

605 

606 

607 

608 

610 

till 

6H 

615 

617 

620 

621 

622 

624 

649 

649 

653 

655 

655 

655 

656 

656 

656 

656 

656 

658 

659 

659 

659 

659 

662 

670 

670 

676 

677 

677 

677 

678 

679 

681 

683 

684 

687 

688 

688 

698 

698 

698 

700 

702 

703 

703 

703 

704 

708 

709 

710 

712 

713 

714 

716 

716 



XXVIU 



LIST OF ILLUSTRATIONS. 



no. 

281. 

282. 
283. 
284. 
285. 

286. 

287. 



DESCRIPTION OF ILLUSTRATIONS. 

Indicator Diagram of a High-pressure Cylinder, 

» Weight, 
,, ,> »> >> >> »> inertia, 

Quadruple Engine, Two Cranks, Balanced (M' Alpine System), 
Cranks and Curves of Forces for Quadruple Engine (Yarrow-Schlick-Tweedy), 
Triple-compound Engine with Two Cranks (Wigzell's System), 



in Balancing 



Diagrammatic Single-crank Engine showing Forces, . 
288-319. Diagrams illustrating the Problems involved 
Engines, ..... 

320. Heavy Oil Engine, Single-acting Two-cycle (by M. A. N. Co.) 

321. Cylinder Section of the above Engine, 

322. Heavy Oil Engine — Diesel System, Italian. . 

323. „ „ — Junker's Design, 

324. „ „ — Westgarth-Carel System, s.s. " Eveston,' 

325. Emrines (Diesel) of Twin s.s. " Fiona," 

326. Section of a Cylinder of a Westgarth-Carel Oil Engine, 

327. Cylinders of Oil Engines, 

328. Doxford's Double-piston Oil Engine, 
328a. Semi-Diesel Engine (Kromhout), 

329. Geared Turbine, 

330. ,, s.s. " Normannia," 

331. Double Reduction Gear for Turbines, 

332. „ ,, Floating Frames 

333. Fottinger's Turbine Transformer, 

334. Superheaters in a Three-furnace Boiler, 

335. Robinson's Superheater Header, 

336. Treble-riveted Butt Joints with Half Number of Rivets in Outer Rows that 

are in the Inner Ones, .... 

337. Treble -riveted Butt Joint with Inner Strap taken by Inner 

only, ...... 

338. Hlustration of Irregular Staying of a Flat Surface, 

339. Gauge Cock required by German Government for Testing, 



PAGE 

747 

747 

747 

7c 3 

754 

755 

. 756 

. 757 

Marine 

777-8C0 
. 861 
. 862 
. 863 
. 864 
. 869 
. 870 
. 871 
. 872 
. 873 
. 874 
. 888 
. 889 
. 890 
. 891 
. 894 
. 9C0 
. 902 



Rows of Rivets 



924 

924 
926 
939 



I 



A MANUAL 



OF 



MAEINE ENGINEERING 



CHAPTER I. 

GENERAL INTRODUCTION. 



Fundamental Principles. — The first object aimed at by the marine engineer, 
is to propel a floating body through the water at a certain speed ; the second, 
so to construct the propelling apparatus that the motion may readily be 
reversed ; and the third, to adopt such an arrangement of propeller and 
engine as shall be convenient for the floating body and the service on which 
it is employed. 

The principle on which nearly all marine propellers work is the pro- 
jection of a mass of water in the direction opposite to that of the required 
motion. The only exception to this rule is the case of ferry steamers and 
some river craft, where a chain or rope lying in the bed of the river passes 
over a wheel or barrel in the ship itself. 

The water, in modern practice, is projected by — (1) One or more screws 
at the end of the ship (as will be described under the heading of propellers) ; 
(2) by one or more paddle-wheels outside of the ship ; or (3), by a form 
of wheel in the inside of the ship, which is generally spoken of as a /^-pro- 
peller, as the water issues in jets from orifices in the ship's side. 

The latter, however, is seldom used, although it has some features that 
make it attractive in particular cases. The paddle-wheel, too, is slowly 
dying out ; and although there is reason to believe that for river service, 
especially in the tropics, and certain special duties, it will survive, it is, 
nevertheless, gradually being displaced by the screw in some of these, while 
in all other services, even in the shallow water ones which at one time seemed 
reserved for its use, it is practically gone. 

The paddle-wheel with feathering floats has certain qualities of its own, 
which render it more serviceable than the screw in particular cases ; for 
example : — 

In tug boats quick manoeuvring is of extreme importance. The sudden 
stopping, and equally sudden and certain starting, of the boat is most desir- 
able ; this cannot be done with the screw, nor can the turning round in 

1 



2 -';.'■ '. .' MANUAL OF MARINE ENGINEERING. 

confined spaces be- accomplished even with twin screws so dexterously as 
wit^' .disconnected paddle- wheels. 

In river steamers running in very shallow water the paddle-wheel, especially 
when fitted at the stern, has distinct advantages over the screw, even when 
the latter is placed in a well or enclosure, as done by Mr. Yarrow and Sir 
John Thoinycroft, inasmuch as it can free itself or be easily freed of weeds, 
.s less liable to injury, and when injured is easily repaired. 

In steamers making frequent and short calls at piers and wharfs, the 
paddle-wheel permits of a more rapid service on rough and windy days than 
can screws, especially the small screws of the turbine-driven ship. 

On the other hand, the paddle-wheel is a heavy and somewhat clumsy 
instrument exposed to wind and sea, and so liable to damage, even when 
protected with boxes and guarded with sponsons, fenders, etc. ; it can, 
however, be repaired by simple means, and even when badly damaged can 





Fig. 1. — Thornycroft's Stern for Shallow Draught Screw Ships. 

be generally sufficiently repaired by the ship's staff to permit of proceeding 
on the voyage. 

Its position in the ship and the space occupied by the paddle engine 
interferes with the general economy ; the machinery is heavier and more 
cumbersome than that driving a screw developing equal power, inasmuch 
as the speed of revolution is necessarily restricted ; the wheel exerts a thrust 
on a part of the ship less calculated to take pressure than that to which 
the screw applies ; but with the modern steel ship, however, that can be 
remedied, and provided for in a way which was not so easy with the wooden 
ship of the past. 

The paddle-wheel was the propeller of the first steamers put to practical 
use, and to-day its efficiency is little short of that of the best screws. 

The modern paddle engines, with their long stroke and choice design, 
have brought this about to a very great extent. 

The screw is much smaller than even the small high-revolution paddle- 



[NTRODUCTION — THE SCREW. 3 

•wheel of to-day ; it is wholly immersed and largely protected by the quarters 
of the ship. It is generally of so small a diameter as to be wholly below 
the water line, but in shallow draft ships this is not always possible ; it is, 
however, totally immersed when working by surrounding it with portions 
of the body of the ship, so that it revolves in a channel, as done by Sir John 
Thornycroft (tig. 1), or by dropping a flap behind it, as done by Mr. Yarrow 
(tig. 2). The thrust from the screw is applied lower down, and nearer the 
centre line of the ship's resistance, than is that of a paddle-wheel, so that 
the tipping moment is quite small, and the thrust block is attached to a part 
of the ship's structure eminently calculated to take the force. 



V 




60,000 



% 40.000 



Z SO, 000 

? 20.000 
I 
<fc 10,000 



i£- 



*&' 



-^°& 



&[.<m 



He 



fio? 



_^1 






\\t 



itf9 



** 



V 



-A 



7 a a w 

Speed ( Statute Miles per Hour) 



II 



50,000 



1 10,000 




£783 
Speed [Statute Miles perHourJ 



Fig. 2. — Yarrow's Drop Flap for Shallow Draught Screw Ships. 



There being no restriction to the number of revolutions of the screw, 
the engines can thereby have a high piston speed, and be comparatively 
light, cheap, and made to occupy less space, and moreover can be placed 
in positions more convenient to the general arrangements of the ship. It 
used to be urged that the screw caused much vibration and consequent 
discomfort to those in the ship ; that it was inherent in the screw, however, 
was erroneous, and it is known now that vibration was as much due to the 
momentum of the moving parts of the engines as to the screw ; and. while 
it is admitted that even a good screw may cause horizontal vibration at 
the stern, owing to the difference in pressure of water on the lower and upper 



4 MANUAL OF MARINE ENGINEERING. 

blades, it is only with the two-bladed variety that it is pronounced, and even 
then only with those of large diameter and small submersion is it excessive. 
With well designed three- and four-bladed screws driven at fairly high revolu- 
tion by engines carefully balanced and well preserved, the vibration is virtually 
nil. That the screw is liable to foul itself with ropes, nets, etc., which only 
a diver can remove, and that it may damage itself by striking dock walls, 
semi-submerged wreckage, etc., so as to necessitate dry docking, or tipping 
must be admitted ; but this seldom happens. 

Hydromotors. — There is, however, another method by which ships have 
been, and may be, propelled distinguished from all others by the absence 
of any propelling instrument. In the past many proposals have been made 
and patented for ejecting a stream of water from the stern of a ship which 
has been taken in at the bow or through the bottom. 

As far back at 1729, John Allen proposed this method, and in 1793 James 
Ramsey constructed and tried on the Thames a boat fitted with a pump 
and pipes to do this : the speed attained, however, was only 4 knots per 
hour. It was not, however, till 1849 that John Ruthven patented what 
has since been known as the jet propeller, and associated with his name, 
although he was not the first inventor of the system. In this case, however, 
the impeller of the centrifugal pump is really the propeller. In the case 
of the s.s. " Hydromotor," built and equipped in Holland in 1876, the 
stream was ejected from a pair of cylindrical receivers by steam pressing 
on the surface of the water which had been caused to flow into them by 
reason of the vacuum formed by condensing the steam of the previous dis- 
charge ; in fact, they were a pair of pulsometers. The enterprise was a 
failure, and no one is likely ever again to attempt to propel a sea-going ship 
by such extravagant means — extravagant in steam consumption and in 
space occupied both by receivers and water passages. 

Motive Power. — Both the screw and the paddle-wheel revolve around 
their axes, consequently the engine employed to move them must have 
circular motion, and inasmuch as both instruments are reversible, and can 
thereby reverse the motion of the ship, the engines should be capable of 
reversing. The engine must also be able to work efficiently at all speeds 
varying from the maximum to the slowest ; for although most mercantile 
ships usually run at " full speed " — that is, nearly approaching the maximum 
— naval ships, cruising yachts, and some other vessels travel at varying 
speeds, and seldom for a long period at full speed. Such engines are by 
preference connected direct to the propeller shaft, but they may be again, as 
they have been in the past, geared to them by means of spur wheels and 
pinions, in order that there may be no compulsion to drive the propeller 
at the same revolutions as the engines, or vice versa. The first screw steamers 
had engines similar to those in paddle-wheel ships, and moving at the same 
revolutions, and it is a fact that in the well-known trials of the s.s. " Rattler " 
versus the p.s. " Alecto," the engines in these ships were of the same size, 
design, and construction, but the screw of the '"Rattler" revolved four 
times to one of the engines. ::: 

Such engines as are suitable for the purposes of marine propulsion may 
be worked by means of steam, or by gas generated from coal or oil. Hitherto 
steam has been exclusively employed on sea-going craft, and oil vapour on. 

* Vide "The Screw Propeller," p. 202 



INTRODUCTION — MARINE STEAM ENGINES, 6 

small craft in home waters only. The success of the oil engine in such craft. 
and the fact that engines using only crude and non-dangerous oils can now 
he obtained, will no doubt lead to their use on large ships, and a more extended 
service now they are made reversible, so that the propellers can be reversed 
without wheel or other objectionable gearing, which when of small size do 
not develop defects so rapidly as in the case when larger power is put through 
them. The Diesel engine, which can be reversed by means of compressed 
air. and uses heavy oil, has been fitted to several ships, and is now being 
supplied to both the mercantile and naval marine in quite large sizes, by 
multiplying the number of cylinders. The suction gas plant, which on 
shore is used with fairly satisfactory results, may not be found so attractive 
on a service where the working day is twenty-four hours, and the working 
week seven days. It is, moreover, heavier than an oil engine installation. 

The steam-worked machinery is more flexible than the oil and gas-worked, 
is less liable to derangement from shocks, and more easily governed in a 
sea way. But perhaps the most important feature to remember in making 
a comparison is the fact that steam is water vapour produced by heat gene- 
rated by the combustion of anything that will burn, so that for the boiler 
of a steam engine fuel of sorts can be found in all parts of the world, whereas 
for the internal combustion engine only certain oils and certain coals can be 
used, and if they are not obtainable the engine is useless. There is, moreover, 
another feature that cannot be altogether overlooked, and that is, while 
the steam engine works with a back pressure of only 1 or 2 lbs. per square 
inch, the gas engine has over 15 lbs. 

Marine steam engines are of two kinds, the one works by means of the 
elastic pressure of steam during expansion, the other by the kinetic energy 
of the steam developed on expansion ; in the one the action is static, in the 
other it is dynamic ; and just as in hydraulics, there is the common wheel 
with its buckets filled with water, whose weight causes it to turn ; likewise 
the ordinary ram which is forced outwards against a load, or as the recipro- 
cating hydraulic engine when work is done by the pressure due to head, while 
the Pelton wheel is moved by a jet of water impinging on its vanes with 
velocity generated by the fall of the water, so there is the ordinary recipro- 
cating expansive engine, and the turbine in which the steam acquires a high 
velocity by expanding into and through a nozzle, which directs its flow on 
to the blades of a rotor, where it gives up its energy by producing motion 
in it against resistance. The efficiency of a reciprocating engine is little 
affected by the velocity of flow of steam, be its piston speed high or low ; 
the revolution for maximum efficiency of the turbine must be such that the 
peripheral speed is half that of the flowing steam. This means that the revolu- 
tions of a simple turbine must be exceedingly high or the diameter of the 
rotor very large ; and for large power both. By methods adopted by Mr. 
Parsons and others, the rate of revolution has been brought down to reason- 
able limits for high-speed ships, while preserving a good efficiency ; but it 
is still too high for such ships as are engaged in cargo carrying if they are 
to have screws of decent efficiency and effectiveness. Mr. (now Sir Charles) 
Parsons has met this difficulty, therefore, by fitting pinions to the shafts of a 
pair of turbines, and geared them to a spur-wheel on the screw shafting. 

The employment of a quick-revolution engine to drive the paddle shaft 
by means of a wheel gearing was practised by the Butterley Company so far 



G MANUAL OF MARINE ENGINEERING. 

back as 1823, but it was not followed till almost the end of the nineteenth 
century, when it was revived by one or two firms for river steamers, in order 
to use the quick-running triple-compound three-crank engine. 

Wheel-gearing has found little or no favour in the eyes of the marine 
engineer since the geared screw engines of the middle of the nineteenth 
century were given up, and no doubt there will be much prejudice against 
it now, although it may be fairly urged that to-day the teeth are double- 
helical machine-cut, made of more suitable material, having no backlash, 
and the pinion driving instead of being driven (v. fig. 45). Experience- 
condemned the system in the past, and it is the successful experience of Mr. 
Parsons which will revive it to-day. 

With oil engines there seems to be, for practical reasons, a limit to the 
diameter of the cylinder (20 inches at present), and all increases of power 
is obtained by an increase in the number of cylinders. Moreover, since 
they are generally single-acting, and act at most once in two, and generally 
in four, strokes, their size for the power developed is large compared with 
that of the steam engine. The first oil engine was exhibited by its inventor 
at Cambridge 90 years ago. In 1817 Neepce proposed to use volatile oils 




Fig. 3.— Paddle Steamer "Charlotte Dundas," 1802. 

in his explosive engine for propelling ships, and an internal combustion 
engine was used so far back as 1825 to drive a screw propeller in a boat on 
the Thames. The system is even now only emerging from its infancy, and 
the room for developments is extensive. In the submarine ship there have 
been the greatest developments, as these engines are almost the only ones 
possible. 

Steam used expansively. — The earliest marine engines were naturally 
near akin to those on shore, but it is very interesting and noteworthy that 
the one in the " Charlotte Dundas " of 1802 was a horizontal double-acting 
engine having a connecting-rod from the piston-rod end to the crank-pin, 
and, therefore, much in advance of and differing from entirely the engine 
of James Watt. The honour of designing the engine (see fig. 3) is due to- 
William Symington, who patented the fitting of the connecting-rod, 1787, 
Pickard having taken out his patent for the crank and connecting-rod of 
the beam engine in 1780. 

The " Comet," which was the first steamship in this country to earn 
money by conveying passengers and goods between Glasgow and Greenock 



INTKODI'CTION — .MARINE STEAM ENGINES. 



m 



fl A 




8 



MANUAL OF MA1UNE ENGINEERING. 



in 1812, had a modification of the beam engine of the type known later on 
as a Side Lever (v. fig. 4). These ships, as the ones that followed them, had 
engines using steam at or a little above the atmospheric pressure, and ex- 
hausting into a jet condenser with an air pump, etc. Later on, after the screw 




X 



X 






S 



3» 
© § 

CO ""* 



■-: 



03 — I 

£ - 

c 03 

® s 

60 . 

£ — 

to a 

CO "H 

^~ 

: 3 



o 



03 



>> 

-^ 

3 



x> 



03 
03 



60 



propeller had been adopted, some small ships, and even a few large ones, in 
the Royal Navy^had non-condensing engines using steam at 50 to 65 lbs. 
pressure generated in cylindrical tubular boilers. Fig. 6 is a modern steamer 
as now built for service on the Clyde estuary, and doing the work that the 



INTRODUCTION — MARINE STEAM ENGINES. 



■E 



A 



ts 






Y~ 



b 

X 

o 
re 

X 

o 

'-"5 



SO 






3 
- 



si) 



L 



10 MANUAL OF MARINE ENGINEERING. 

"Comet" established; she is, however, driven by three screws and turbines 
instead of four paddles and a one-cylinder engine, and carries more passengers 
in a trip than the " Comet" in a couple of months. 

In the early marine engines, as in those on land, the cycle was a simple 
one. Steam was admitted to one or more cylinders direct from the boiler 
during about 70 to 85 per cent, of the stroke, when working at full power ; 
at or near the end of the stroke it was allowed to escape to the condenser ; 
the cylinder on that side of the piston remained in open connection with 
the condenser during about 85 per cent, of the return stroke, and conse- 
quently its surface was exposed to the cooling action of the comparatively 
cold vapour remaining in it. The earliest'screw ships in the Navy had their 
boiler safety valves loaded to 5 lbs. per square inch ; by 1851 the load had 
been increased to 14 lbs., and two years later 20 lbs. was taken as the standard. 
In 1855 several special ships were fitted with non-condensing engines of con- 
siderable size supplied with steam from 50 to 65 lbs. pressure generated in 
cylindrical boilers of the so-called Scotch type, but the ordinary ships with 
condensing engines continued to work with steam generated in box boilers 
at 20 lbs. pressure. In 1861, the new ironclads had boilers with safety 
valves loaded to 25 lbs. per square inch, and four years after a further increase 
was made to 30 lbs., which was the usual or standard pressure till the box 
boiler ceased to be made. 

The cut-off in the cylinders of these ships when running at full speed 
was never less than 60 per cent, of the stroke, so that the maximum rate of 
expansion was no more than 1*67 under these circumstances ; but with 
the increase in pressure from 25 to 30 lbs. came the supply of expansion 
valves, whereby a much earlier cut-off could be obtained with a corre- 
sponding economy in fuel consumption. With such valves a rate of expan- 
sion of 4*0 could be obtained, so that the terminal pressure at which the steam 
exhausted to the condenser was under 10 lbs. absolute. In the mercantile 
marine about 1870 the boiler pressure was increased to 50 lbs. for expansive 
simple engines, and the cut-off at full speed was about 30 per cent, of the 
stroke, giving a rate of expansion of 3*33 ; a few ships with these engines 
had boilers loaded to 60 lbs., but in a general way 50 lbs. was the highest 
pressure with simple engines. The pressure at exhaust was then about 
15 lbs. absolute at full speed. 

The compound engine in which the steam, after doing duty in the first 
cylinder, exhausts into another, instead of delivering to the condenser, was 
the invention of Hornblower, a Cornish mining engineer, in 1771, improved 
by Wolff in 1804, and first used on shipboard by Randolph and Elder in the 
screw steamer " Brandon " in 1854 with a boiler pressure of only 22 lbs. 
This same enterprising firm supplied the first compound engines for H.M. 
Navy in 1863, and fitted them in H.M.S. " Constance." It is worthy of note 
that their designer was Professor Rankine, and they worked with a boiler 
pressure of 32*5 lbs. per square inch, and that the rate of expansion at full 
speed was about 5, and the referred mean pressure was 11*3 lbs. per square 
inch, which is not bad, considering that the theoretical would be not more 
than 22 lbs., or 5T5 per cent, efficiency. 

It was not, however, till about 1870 that the compound engine was 
accepted as suitable and desirable for marine purposes, but after experience 
had proved its economy in consumption of steam, and that the expansive 



INTRODUCTION — MAU1NE STKAM ENGINES. 



11 



economy 
engineers 



engine using high-pressure steam developed mechanical troubles from which 
it was free, it soon superseded all other engines, and was the accepted type 
until it was found that, with the increase of boiler pressure from (><> lbs. to 
100 lbs., their 



gam in 
was not what 
had reason to 
expect from such increases. 
In 1881 the late Dr. 
Kirk, whose name is as- 
sociated with that of John 
Elder, of the Randolph and 
Elder firm, and himself a 
most able and enteif>rising 
engineer, developed the idea 
of multiple cylinders, or, as 
we would now say, of ex-> 
panding by stages ; he built 
a three-stage engine, where- 
by the steam exhausted 
from the first engine into 
the second, and from the 
second to the third engine. 
Dr. Kirk had, however, in 
1874 fitted an engine of this 
kind in s.s. " Propontis," 
with cylinders 23 to 41 
inches, and 62 inches dia- 
meter by 42-inch stroke. 
Unfortunately, the water- 
tube boilers of this ship 
proved dangerous, and 
caused the whole installa- 
tion to be doubted. Looking 
at the subject another way, 
it may be said that he in- 
troduced a third cylinder 
between the ordinary high- 
pressure and low - pressure 
cylinders, so as to avoid the 
big drop in pressure between 
the high and low. But the 
principle involved was one 
of stage, whereby there was 
a decrease in difference 
between the temperature 
during admission to, and 




that at emission of steam from each cylinder. There was, however, a 
mechanical gain as well, due to the decrease in initial and general loads 
on the pistons of the compound systems as compared with those obtaining 
in the expansive engines. The pressures of steam adopted by Dr. Kirk 



12 MANUAL OF MARINE ENGINEERING. 

and by the author himself in the early triple-expansion engines was low 
compared with those now used ; in fact, the pressure was little beyond that 
with which the compound engine was then working on shipboard. Since 
that time working pressures have gone on increasing till 180 lbs. was a common 
practice for triples in the mercantile marine, and on the introduction of the 
water-tube boiler into the Navy, steam of 250 lbs. pressure was and is used 
by the triple- compound engine. In the mercantile marine now steam up 
to 225 lbs. is generated in the cylindrical or tank boiler for use by quadruple 
expansion engines with advantage ; these engines were introduced by two 
or three of our leading manufacturers in 1885. The rate of expansion had 
gone on increasing, till now, with quadruple-expansion engines it is common 
practice to work at 16, while with the triple engine working with a boiler 
pressure of 180 lbs. the rate of expansion will be 12 to 13 in the mercantile 
marine, and about 11 to 12 in H.M. Navy at full power. 

Until quite recently all attempts at economy in consumption of fuel 
were in the direction of increase of pressure and increasing the number of 
cylinders. The upper end of the steam expansion diagram was the field in 
which further gains were looked for, and the dictum of the old chief engineers 
that there was no economic gain in working with a vacuum over 24 inches 
was accepted without a close enough examination of the foundations for 
such a statement, hence few, if any, engineers turned their attention to the 
possibilities of the gain to be got at the lower end (v. fig. 66). The adoption 
of the turbine for marine propulsion by Mr. Parsons opened his eyes to this 
by the great increase in efficiency of his machines when working with the high 
vacuum so cheaply obtained with an unlimited supply of cooling water 
and a really good air pump on shipboard ; and further, the general use 
of feed heaters on shipboard has quite removed the bugbear of cold-feed 
water for the boilers. Before 1865, with the old common jet condenser, 
the vacuum was seldom more than 24 inches, whereas with the surface con- 
denser, which soon became the common practice after that date, 28 inches 
was not uncommon, and 27 inches easily maintained ; the Admiralty require- 
ment was that the vacuum on trial should be within 3 inches of the barometer. 
In passing, however, it may be mentioned that the pressure in the con- 
denser is in no way subservient to that of the atmosphere, although those 
registered by the common vacuum gauges, of course, are, for they really 
only exhibit the difference between the pressure of the air and that in the 
condenser. To-day gauges can be obtained which indicate the exact pressure 
in a condenser, without regard to the atmospheric pressure, and should be 
always used. ^ 

It was not an unknown thing forty years ago to maintain, in a surface 
condenser, a vacuum of 29 inches with high barometer (30'5 inches), but 
it was only with one of exceptionally good design, and it, the pumps, and 
everything in the best of order that so good a vacuum was got. To-day, 
with the better design of both condenser and pumps, and the relay system 
or other means for obtaining high efficiency of air pumps, 29 inches can 
be maintained with comparative ease, consequently a rate of steam expansion 
that was useless formerly may now be worked with advantage and with 
considerable gains both in power and economy, especially with turbines. 

The reciprocating engine has its limitations, one of which is the size and 
nature of the valves for admitting and releasing the steam. It is practically 



INTRODUCTION MARINE STEAM ENGINES. 13 

impossible to provide means in this direction to ensure high efficiency to the 
low-pressure cylinder of a compound system ; wire drawing and clearance 
losses are great, and the difference in pressure between the cylinder and 
the condenser necessarily great compared with those of a turbine. Hence 
the last addition to a marine engine, whereby steam efficiency is considerably 
enhanced, is the low-pressure turbine taking its steam from the exhaust 
pipe of the low-pressure cylinder, and expanding it from about 10 lbs. to 
1 lb. absolute, or even less. It is claimed by Messrs. Denny that the increase 
of power and an economy of fuel has been found as high as 15 per cent. 

The compound turbine so much in use to-day in H.M. Navy and in express 
steamships (v. fig. 6) of high speed is capable of a rate of expansion far beyond 
that of a reciprocating engine ; in fact, there is no practical limit to it. But 
in the early stages of the compound turbine the thermal efficiency is not so 
high as is the triple compound engine ; it follows then that the maximum 
efficiency will be obtained by the combined arrangement of a reciprocator 
with a low-pressure turbine, taking the steam from the low-pressure cylinder 
at something like 15 lbs. pressure absolute, which is about the terminal pres- 
sure of such an engine working at 200 lbs. boiler pressure. 

Propellers. — The " Charlotte Dundas " (fig. 3), the first ship to be pro- 
pelled by steam in a practical way, had one paddle-wheel at the stern. The 
" stern wheeler," as she is now called, remains as the surviving representative 
of the paddle ship in the construction of steam craft for river purposes, 
especially those working in the tropics, where she is the favourite, and appears 
likely to continue as such. The first steamer in America, the " Claremont," 
of 1807, and the " Comet " in this country in 1811, as the first British steamer 
to be put to commercial purposes, had a pair of side wheels — that is, a wheel 
on each side with an axle or shaft common to the two. It should, however, 
be noted that the " Comet " had two pair of wheels (v. fig. 4) when first 
tried, a system followed by Sir Edward Reed in 1874 for special reasons 
when he designed the swing saloon steamship " Bessemer." The pair of side 
wheels continued to be the practice for general purpose down to recent times, 
and is still followed when such ships are built for the special services already 
alluded to. The last of the ocean-going paddle ships was the " Scotia," 
6,871 tons displacement, 362*5 feet long, having engines of 4,950 I.H.P., 
driving a pair of wheels 40 feet diameter, weighing 156 tons. The largest 
modern American paddle steamship is the " Priscilla," 424 feet long and 
5,200 tons displacement; she has engines developing 9,345 I.H.P., which 
drive her at a speed of over 19 knots per hour. Fig. 5 shows the paddle 
steamer as now constructed in America for lake and river service to convey 
passengers and parcels expeditiously. The largest modern British paddle 
steamship (v. fig. 8) is the " Empress Queen," 360 feet long, 2,900 tons 
displacement, attaining a speed on service of 2T7 knots with 11,440 I.H.P. 
The fastest British paddle steamship is " La Marguerite," 330 feet long, 
1,868 tons displacement, and having a speed of 22 - 3 knots. 

Screw propellers, brought into practical use by Bennet, Woodcroft, 
Francis P. Smith, and John Ericsson, in 1836 (v. fig. 9), are now the most 
important instruments of propulsion, and, although much time, genius, 
and money have been expended in exploiting screws of the most varied 
forms, engineers have to-day settled down to the almost universal use of the 
screw, such as used by Smith and his friends in 1840 on the s.s. " Archimedes " 



14 



MANl'AL OF MAHINK ENGINEERING. 



M 

o 




■ 


LI 


1 


© 




CO 




© 




1^ 




f-H 




X 


. 1 


lo 




<N 




-f 




X 




^, 




© 




CO 


( 


a 


. 1 


ca 




s> 












■~-} 




GO 
0) 










j 


a 




a 


' 




■ 


*■ 




0D 




I 




GO 


* 


. 




to 



OJ 

CD 

CO 

CO 

Ol 

CD 



Q 



X 



o 
o 
© 



CD 

o 
"Si 

DO 



INTRODUCTION — MARINE STEAM ENGINES. 



15 



— that is, in the essentials — viz., making it a portion of a true helix and 
of moderate diameter and acting surface. So far as shape is concerned, 
the practice now is very like that of Robert Griffiths of 1805 — that is, the 
blade is narrow at the tips compared with the breadth at middle, and the boss 
spherical of considerable diameter. The prevailing form for express and 
naval ships is, of course, a modification of the Griffiths, inasmuch as their 
tips are somewhat fuller and rounded, so that the blade is the shape of the 
longitudinal section of a domestic hen's egg ; and in some cases where the 
diameter is small for the power, the blade is even circular. In fact, all of 
them are such as would be formed by taking the Griffiths, and reduce the 
diameter by rounding the tips until the maximum breadth is nearer the top. 
The number of screws has increased from the one situated in a gap in the 
deadwood of the ship to the four of the modern high-speed warship driven 




Francis Pettit Smith, 1836. 





Francis Pettit Smith, 1838. 



John Ericsson, 1836. 



Fig. 9. 



by turbines. The twin screw, an obvious development, especially if required 
to compete with the paddle-wheel in shallow waters, was first introduced 
by the Rennies in 1854 for service on the Nile ; since then they have replaced 
the single screw in all express and ocean-going high-speed steamers. With 
such twin screws, not only is total immersion attained and the " feed " to them 
unobstructed, but the safety of the ship considerably insured, inasmuch as 
the liability to total breakdown is reduced by a-half, and a means of steering 
is provided, should the rudder or its gear be disabled. Two screws, one 
at the bow and another at the stern with a line of shafting common to them 
have been used frequently for tugs and ferry boats with advantage, and 
four screws similarly arranged have been used for the same services. 

Three screws, one in the deadwood and one on each side, somewhat 
ahead of the middle one, have been used in France, Germany, and Italy, and 
the United States, but the system did not find favour in this country till 



16 MANUAL OF MARINE ENGINEERING. » 

the turbine wa3 used as the prime mover. The development is a natura 
one, where large power is required in a comparatively shallow vessel, and 
also with the view of keeping the engine within moderate dimensions. Italy 
was the first to give the system a practical trial in 1886 on the cruiser 
" Tripoli " ; France followed on a large scale with the " Dupuy de Lome " 
in 1890. In 1892 the United States of America adopted the system for 
the " Columbia," of 7,375 tons displacement, 22'8 knots speed. In the same 
year the German Government fitted the cruiser " Kaiserin Augusta," of 
6,330 tons, and 22"5 knots, with three screws, while in 1896 Russia adopted 
the system for the " Rossia," a large cruiser of 12,130 tons and 14,500 I.H.P. 
In our own Navy, the " Amethyst," cruiser of 3,000 tons, having Parsons 
turbines operating on three screws, was the first attempt, if the experiment 
with H.M.S. " Meteor " in 1855 is excepted. To-day the rule in the British 
Navy is to fit four screws* to all but the very small ships ; the newest cruisers 
of 75,000 H.P. are so fitted, as are the new class of " Destroyers." In the 
mercantile marine our largest and fastest mail steamers have four screws ; 
the smaller ones, three when driven by turbines or combined reciprocators 
and turbines. The newest and largest ship, s.s. " Olympic," has thus three 
screws (v. figs. 43 and 44). With reciprocators only the twin screw is now 
almost the universal rule for all express steamers, and for cargo steamers of 
the largest size. 

Multiple Screws. — In some ships of special design, or for special service, 
a larger number of screws have been fitted, and their arrangement varied ; 
for example, the saucer-shaped Russian warships " Popofiskas " had six 
screws, and the Russian ice-breaker, " Ermack," has three screws at the 
stern and one at the bow. 

Mr. Parsons has also in some few cases fitted more than one screw on 
each shaft, and more recently the German Naval Authorities have tried, 
on the cruiser " Lubeck," eight screws, two on each of her four shafts ; the 
results of this experiment are by no means satisfactory, being much inferior 
to those of sister ships having twin screws driven by reciprocating engines. 

Table I. contains examples of naval ships' engines typical of the 
periods since the introduction of the screw propeller, and shows the progress 
made in the use of steam expansively. 

Progress in the use of steam expansively in the mercantile marine during 
the past 50 years, as shown by the typical examples in Table II. 

* Quite recently, however, there is a tendency in both the British and Foreign Navies 
to revert to the twin-screws for the smaller class of ship, and even in second-class cruisers, 
with a complete turbine for each propeller, and since fcae introduction of gearing the 
tendency is to twin-screws for all kinds of ships. 



INTRODUCTION — MARINK STKAM KNGINKS. 



17 



o 

►J 

ft. 
O 

« 

H 
at 
o 
OQ 

o 

K 
K 
K 

I— I 

o 
to 

W 

■j 
•< 

► 
< 

b 
O 

m 

K 
« 

o 
o 
a 

Pi 








i 


In 


2 rs !S 4) 1 




£ -2 <* ?> £ Sa B £ c 


■ — 


a - i > o o o o o 
^><3K,u •.•.•ii-. , jj ^AP.-ii ? -o : : : : : : : : : 


9 

03 


Q 


d * -g jg c ^i2a^^ C ^=JS c S fe * 




K 






atical 
ure. 


„; . ,09 p p p .pocpoooooooooiftitS'Oiraooirao 


fc*a 


-^ : : : oi ■** o »m r- :c»Tj<ci-H , t^?bc<r)t^.t^3i — mwn- — -r — ^r » 

1-1 MlO JIMtl f)M«OKM«MM(NnC0'*e»»l^l^»« 


a" 9 2 


lal 
ure. 


^^c(jHosi«t-oo«!ooiaotio)«offl^wccHii»oi-ooffl 


Act 

Me 

Press 


•^M-H'M-io'Mi'ooi-inciffloombo-ociinbHiOfflioiOH' 

^--H'HlM^-Slw-irtiNIM-HOlrtfMC-INflOllMUHiM^^I'^'* 


Approxi- 
mate Rate 
Expansion. 


K5«iO?;0»nOMOO!00(MO!B'i''* 


CICC^IMUMWOi'OiQOOOOOOOOOOOOtOWOOiOOOKJO 


H-H-HrtlHrtrtfl-l(3rtH^I-H}l'Hrtrt!0t-.!0«Oh'00S10lON 


.— t ^-1 ^ 


S = 

.— to 


.OOOOOOOOiCOOOOOMO'-SOO ooo 


Bibboooooitifubi'oNOHb-'Oooi'oooionoo^h 


=5 ■» 
ftH 


t-5 «-MO(NIM01!MMINHMMMnMniO»tOffl-MiOOca^l 


—l — — CO — i — cn 


— 03 


ClOOiOCMCSH'iOOin "JiOMIOOMh CO - ' H ° © CO l~- ■<* '.O 


> 3 


O'M«Mh-t0i0 0!l^H l -00OC<:«Hi«H , c;0)^iOO'0O - OOi-0-H 


°"S 


5IO-*ifiiQCHQl0^ir5iOtO>Ot-t>tOl>.(Ot-OJ"»00(MO>Hi-(N«-i 


d — 


— — i — i— ■— Ol 


-3 


(»OXX5)Oi0iCt-«l>.MOOOHt)O»O , t'i l -«Wi0-iOOO 


CS <* *• 


ClT«MfflCli'0-OOi'MOOO'tffl--IMI10i8©ntOOOO 

Tioo>H i »ooji"OMoom»or'«aio>OH | c-rtH"*a)inifii- 


— ™ * 


"2 ^Pn 


« — -H<0^C<5IM«l«>CO«O^t^.t-^ira(M fflC©MOl«00MH" 


H* 


CH p-H <-* ~* C^ *— ' 




S75-^;a3C0050COC'l , H , OS'*C^03CO-H"OOTtl05000C'10C05'-<05COOO'MO 




^o , H"iCMcocce«j-t-H , Tficoiooco'*io-»*<ir5torc — * ■* coiowii^^M 




"""■g xxxxxxxxxxxxxxxxxxxxxxxxxxxxx 




Tjl -H< 




_:-©:"? IN •H'OO-tS'l CO 51 :0-*Oi -«-* — c**©©©V5'H<t---H , CC-f-H<r--t— t— 


«j 


-s-*^iOffl(M©~. C0Xl^OO0JOfl>O(N00OH'ON»at'!B(0t^W 


V 
« 


•O O 115 MOHhhh 


II HP II =£ II II jgdaJ8«?.8to-H"«op«7.io 


— 


00 — t-- h. M«000)*i 


« 


r-- u5 • t^ TfMiotowo 


o 


. S§ n 8 co JJ !* go jg"!? 




a. a, c p- -p r s « 




o o ? o '.n c ' 








■t H 1 H"M Tl ;i Jl M 51 to O tl ■* !1 CQ#1 <M ■* <M <M H 


Ship. 


i 

,2 . C 


JjCr-l.. . « , 


»4 




--^ © *a .a o C'S , .&, &. a. 




0) 

£ 

a 

S5 


»5-»os S?'S S5 ox »> u'' 2 i'S-S » « S^S ^S^" 
J.§J23^|l|§-a t | : 33c3l|§*Sc? , J 2^2 "5 S3 




HCOiowi'OH(it(;iflii5ia!SNCOO--oiO"oooo«)o 


«5 


^i«coin'"t0 3:ooo»3tot;i-i^i.i-t^!»M»eio)30oo 




»*XXX»M*X00X»00XMXa0K»X»»»a000C15>O0» 





18 



MANUAL OF MARINE ENGINEERING. 



5 
< 

a 

■j 

55 

■< 
o 

a 



a 

a 



as 

a 
a 

SB 

I-H 

O 

w 

a 
o 

m 

tn 

a 
a 
o 
o 
a 
Ah 



W 
cq 



1 


u 










to 




CO 


6 


0> 


6 






CO 


o 




^ 




o 








4a 










4a 




Tl 


*a 


Tl 


-ta 






Tl 


— 




4a 


CO 


4a 








C 










C 




rrt 


C 


m 


C 




91 


efl 


c 




a 


T) 


C 






• 












at 




u 


« 


i. 


cS 




CO 


u 


es 


■ 


at 


1* 


crt 






o 

1-4 




d 


00 
co 


CO 




— 
< 


6 


"3 


■ja 

< 


H 

"3 


-*a 
< 


d 

ft 


TJ 

G 
— 




-4a 

< 


c8 
CO 


4a 

< 


03 


4a 


d 


d 
P 


n 
00 


J5 




J2 

-*a 


co 


-ta 




cu 




0> 


X 

-4-* 




4a 
in 

CO 


CO 


-ta 


— 
4a 


4a 


4S 
4a 














>-. 


. 1 


f-i 




C 


hi 


c 


(a 




c 


ki 


'-I 


Sh 


IS 

e 

i— i 










O 




cS 


o 


at 


O 




V 


O 


CO 


o 




CO 


O 


O 


O 


O 








5 




PL, 


S5 


PQ 


fc 




o 


S5 


o 


55 




C5 


& 


^5 


iz; 


55 








™ O 


■H* 


r- 












m 


•n 


m 


r-» 


m 


© 




m 




© 


© 




«M 


°aa 




O 
CO 


o 

CM 


CM 


<M 


CO 
(M 


CO 
CM 


CM 


© 


© 


CO 


© 
— ^ 


© 


© 








■n 


■* 




CO 


a 

o CD 

'■SIS 


«9 


o 


o 


W 


4 


o 


© 


© 


© 


© 


© 


© 


© 


t- 


© 


© 


© 


© 


© 






t. a> <° 


S3 


IM 


~H 


^H 


CM 


o 


<M 


© 


CM 


(M 


© 


m 


00 


CO 


CN 


© 


00 


© 


m 


on 




4)-= g 

£ a 


CO 


CM 


CO 


CO 


CO 


■* 


m 


■* 


•* 


o 


■* 


m 


-# 


© 


m 


m 


m 


m 


m 












































co 


H 










































»fH 


, 










































45 


ee c S 


.o 


o 


CO 


CO 


CM 


oo 


r- 


© 


© 


m 


© 


CM 


© 


co 


00 


© 


© 


© 


00 


© 


3 


s « 5 


£ 00 


o 


o 


■n 


CD 


o 


00 


© 


t» 


■* 


CO 


!>• 


© 


© 


l~- 


^ 


© 


CO 


in 


m 


H 


a 


a-* 


CM 




CM 


i-H 


CM 


CM 


CO 


CM 


CM 


CO 


CM 


CO 


CM 


co 


CO 


-* 


CO 


CO 


CO 


-3 
S 


B 












































1 § 










































CO 

O 


.E ""■» 


■> 


CO 


o 


o 


w 


o 






























s, 


5^3 


CO 


CO 


«5 


00 


cm 


iS 


© 


k>- 


© 


© 


© 


© 


m 


m 


o 


© 


■n 


m 


© 


© 


- — A 


^H 


^H 


in 


-# 


CO 


ifi 


m 


t^ 


k> 


l> 


l> 


<M 


oo 


CO 


© 


m 


© 


CO 


CO 


m 


CO 


Q4 H 
























t ~ l 


















Dh 


< 












































hg 

2 s 




o 


o 


00 


■n 


CO 


© 


in 


© 


© 


© 


© 


m 


© 


© 


© 


© 


8 


(M 


m 


m 


"R 2 


■C CM 


CM 


CO 


■<* 


CO 


CO 


r^» 


—. 


© 


© 


— 


in 


■<* 


© 


00 


© 


© 


© 




^H 


ng 


a 














"* 
















CN 


CM 


CM 




CM 


CM 


a 












































5« 

o c 


o 


o 


o 


o 


o 


o 




m 








CM 


© 


o 


© 


© 


© 








© 


m 


CO 


^H 


m 


r- 


© 


© 


t- 


. 


l> 


. 


00 


■* 


00 


00 


00 


r» 


_ 


© 


© 




5S 


<* 


-* 


CN 


CO 


m 


© 


m 


m 




m 




in 


© 


-# 


t-- 


© 


00 


t- 


r» 


00 


t~ 


a 










































t^ 


■o 












































0) . ^ 


o 


o 


O 


o 


r- 


© 


© 


i-H 


© 


r» 


© 


00 


© 


(N 


© 


m 


© 


© 


8 


© 


© 


iSo«> 


CM 


o 


iO 


t^ 


o 


00 


© 


-1" 


© 


■* 


© 


© 


in 


00 


© 


© 


© 


CO 


© 


© 


U hi £ 


eo 


■O 


CO 


CO 


■> 


r-* 


■* 


■M 


CO 


CO 


CO 


"t 


in 


© 


© 


Tfl 


CO 


rti 


© 


© 


© 














































^H 


CM 


f-H 






CM 


m 


^H 


© 


CM 


T 


CM 


m 




k> 




00 


CM 


00 


© 


CO 






















^H 
















CM 


-# 


m 




<DCM 


IM 


o 


CO 


CO 


00 


© 


© 


© 


© 


CI 


l~- 


© 


(M 


© 


"* 


■* 


on 


CM 


,_ 


in 




HH O 


■<** 


CO 


CO 


CO 


■* 


© 


© 


© 


© 


l> 


m 


© 


■* 


© 


CN 


m 


■* 


t- 


l> 


i-» 




X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 


X 










IO 


























m 












c S>0 


p 


CO 


r-~ 


r^ 


© 


CO 


© 


© 


tH 


m 


CM 


© 


l~» 


© 


© 


•n 


© 


00 


CM 


t^ 




Miea 


© 


© 


CO 


CO 


© 


00 


o 


© 


00 


© 


00 


© 


m 


i-H 


CO 


r- 


k> 


© 


^H 


© 


(A 


•o 




^ 




<** 


°tt 


=w 


<& 


<M 


CM 


i-H 


© 


CM 


© 


r—t 


t^ 


co 


© 


00 


^ 


t~- 




















© 


■* 


^H 


m 


© 


■**» 


00 


(M 


•n 


■* 


© 


IO 


© 


•a 






on 




Hei 


(M 


on 


© 


> 




r~» 






• 


© 


• 












a 










CO 


in 


•»tt 


fM 


© 






^H 


CM 


00 




uro 


<n 


r» 


© 




■* 












CO 








© 




m 

© 


CO 


•* 


<M 

© 
CM 


CO 


00 


© 

CO 

in 
in 

CM 


CM 


in 


© 

CO 


00 

o 




CM 


CI 


-# 


CM 


CM 


<M 


■* 


CM 


co 


CM 


CO 


CO 


CO 


Tj< 


© 


^ 


00 


CO 


00 


© 


X 


<u 


• 


„ 












: 






cS 






_ 






fl 






: 




Nam 

of 

Ship 


4a 
< 


a 

c 

co 

o 


a 

o 

Pm 


t-l 

co 


6 

Tj 


00 


o 

■3 

!-i 

PQ 


H 

t-H 

^5 


C8 

S 
N 


d 

o 


© 

a 


id 


o 

CO 

£ 
o 


->a 

C 

o 


CO 
-*a 
CO 

a" 


d 


CI 

CO 

c 


d 

-a 

S 


6 

a 
o 
O 


a 


o 
d 

PQ 




© 


© 


o 


o 


o 


m 


m 


© 


© 


© 


IO 


m 


m 


m 


g 


© 


in 


in 


p 


© 


"* 


S 


co 


CO 


CD 


r» 


1— 


r— 


t— 


on 


00 


00 


on 


00 


00 


00 


C5 


© 


© 


o 


© 


i-H 


rt 


00 


oo 


00 


f/> 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


00 


on 


00 


on 


© 


© 


© 


O 













































VALUE OF TRIAL TRIPS. 19 



CHAPTER II. 

RESISTANCE OF SHIPS AND INDICATED HORSE-POWER 
NECESSARY FOR SPEED. 

Although, strictly speaking, it is not the province of the engineer to deter- 
mine the power necessary to drive a ship at a certain speed, but rather that 
of the naval architect, still it is a point of great importance to the engineer, 
and one with the investigations of which he should be fully acquainted. 
Circumstances sometimes require, indeed, that the engineer shall name the 
power, as the naval architect may submit that, inasmuch as he is unaware of 
the efficiency of the particular engine to be supplied, he cannot say what 
indicated horse-power will be necessary, but only what effective horse-power. 
Moreover, the subject is one possessing great interest at all times, and some- 
times of the utmost importance to the engineer, as the deficiency of speed 
obtained at the measured mile from that anticipated may be attributed to 
the inefficiency of the engine and propeller. This charge may be, and often 
has been, proved to be true ; but, on the other hand, it may be without 
foundation, the blame really belonging to the designer, who has given the 
ship lines unsuited to the speed. 

Value of Trial Trips. — Trial trips are now conducted, both in the mer- 
cantile marine and the Royal Navy, with more care and interest than obtained 
formerly ; and it is not sufficient to prove at the measured mile merely 
that the ship has done the speed expected, or that the engines have developed 
the power for which they were designed. Both engineers and naval archi- 
tects desire to determine whether the speed has been obtained with the 
minimum of power, and the engineer can satisfy himself on a most important 
point — viz., the efficiency of the propeller, and, to some extent, the efficiency 
of the machinery, while the owner, if it be a private ship, is enabled to judge 
whether he is paying for what he calls " big horses " or " little horses." 
Another point (and one most important to the owner) which, to some extent, 
is determined on a trial trip is — at what expenditure of fuel a ton of dis- 
placement is carried over a mile. It is not an unknown thing to find that 
the engine which burns least fuel per I.H.P., does not compare so favourably 
with others when measured by this latter standard. The apparent con- 
tradiction here is not very difficult to understand when fully looked into ; 
it may be, perhaps, best comprehended by taking extreme cases. Suppose 
the blades of the screw are set so as to have no pitch ; the engine will work, 
develop a certain power necessary to overcome its own resistance and that 
of the screw, but it will not drive the ship an inch ; the coal consumption 
per I.H.P. will probably be somewhat heavier than that of the same engine 
when working with half its load, but still may be light. Now place the 
blades fore and aft, so that the pitch is infinity, and although there may be 



20 MANUAL OF MARINE ENGINEERING. 

now a large development of power, there will be no appreciable speed — 
theoretically, none at all. In both these extreme cases the consumption 
per I.H.P. may be very satisfactory, but the satisfaction would not be 
experienced by the owner. It is manifest, then, that between these two- 
extreme limits of pitch there is some value and one position of blade which 
will give the best result, so far as economy of fuel for load propelled is con- 
cerned. Not only is the pitch of propeller an important function in all 
calculations relating to the speed of ships, but the diameter has a very im- 
portant bearing also on the subject, and more than was generally thought 
previous to the remarkable trials of H.M.S. " Iris." 

The Resistance of a Ship passing through water is not easily determined- 
beforehand, as it may vary from more than one cause, and in a way often 
unanticipated, as has been seen during the trials of the very fast torpedo- 
boats and destroyers. The investigations of the late Dr. Froude on this 
subject have shown that the older theories were sometimes erroneous, and 
the old-established formulae unreliable ; and perhaps the best source of. 
information on the intricacies of this somewhat complex subject is to be 
found in the many able papers read by him, and others since his day, before 
the Institution of Naval Architects and other learned societies. 

When the screw or paddle first commences to revolve, the ship makes no 
headway, and it is only after some seconds have elapsed that motion is 
observable. The engine power has, during that period, been employed in 
overcoming the resistance to motion which all heavy bodies possess, and 
which is called the vis inertia. When the engine is stopped at the end of the 
voyage, the ship will continue to move, and come gradually to rest, unless 
otherwise, retarded by the reversal of the engine or by check ropes. The 
ship is then said to have " way on her," a phrase which, in scientific language, 
means that she possesses stored-up energy, called momentum, which is given 
out, when the engine stops, in overcoming the resistance of the water to 
the passage of the ship through it. This energy was stored up at starting 
of overcoming the inertia, and remains stored until there is any retardation 
in velocity. In this way the weight of the ship helps to preserve a uniformity 
of motion, as that of a flywheel does to an engine, and, therefore, it is im- 
portant that tug-boats should have weight as well as power, to prevent 
towing in the jerky fashion so often observable. When the vis inertia has. 
been overcome, the power of the engine is directed to overcoming the resist- 
ance of the water, and wind if there be any, and in accelerating the velocity 
of the ship ; as the speed increases, the resistance much more increases, 
until the surplus power available for acceleration becomes nil, and the whole 
engine power is absorbed in overcoming the internal resistances, or those- 
belonging to the engine itself and the propeller, and the external, or that 
of the ship. 

The Resistance of the Water is Twofold. — First, the ship in moving 
forward has to displace a certain mass of tvater of the same weight as itself, 
and the water has to fill in the void which would otherwise be left by the 
ship. The work done here is measurable by the amount of water, and since 
it is equal to the displacement of the ship, displacement becomes a factor in 
the calculations of resistance. But to effect this displacing and leplacing of 
water with the least amount of energy, it is necessary to do it gently — to 
set the particles of water gradually in motion at the bow, and let them come 



KESIDUAL RESISTANCE. 21 

gradually to rest at the stern. If it is not clone gently, and the water is 
rudely separated, a wave is formed on either side, showing that energy has 
been spent in raising the water of this wave above its normal level. Although 
every ship, however well designed to suit the intended speed, causes these 
waves of displacement, it is the object of the naval architect to reduce their 
magnitude as much as possible. 

The chief cause of resistance to the passage of a ship through the water 
is, however, the friction between the surface of the immersed portion and 
the water. Resistance from this cause is generally spoken of as skin resistance, 
and is in well-formed ships much greater than the resistance due to other 
causes. However fine a ship may be, there is, of necessity, a certain area 
of skin exposed to the water, and though the displacement be very small 
indeed, and the section transverse to the direction of motion reduced to a 
minimum, it is found that a considerable amount of power is required to 
propel the ship through the water, and that, roughly, the power is propor- 
tional to the wetted surface at the same speeds. It is from this cause that 
the older rules for speed, involving only displacement, or area of midship 
section, together with speed as variables, are found to be so misleading. 

Residual Resistance is the term generally used to express the sum of 
all other resistances to be overcome in propelling a ship through the water, 
and includes that due to wave-making, eddy-making, etc. In tank experi- 
ments this is differentiated from skin resistance, and ascertained with accuracy, 
but it may be calculated with a very fair approximation to it by a formula 
arrived at by Mr. D. W. Taylor, U.S.A., and published by him there. It 
is as follows, viz. : — 

Residuary resistance™ lb, -™->™ 

where b is the block coefficient ; 

D is the displacement in tons ; 

V is the speed in knots ; 

L is the length on the water-line in feet. 

V 2 
The formula is applicable only to speeds for which y- is less than 1'2. As 

will be seen, it conforms to the law of comparison, and, compared with other 
formulae, gives extremely good results for all classes and sizes of ships. It 
has been applied to a large number of different ships, and the general results 
of the application are satisfactory. 

Most of the examples taken for comparison have been warships, where 
it was found in applying the formula that a slight modification gave better 
results. This modification consists in taking the length as the extreme 
length of immersed vessel in place of that on the water-line, and in calculating 
the block coefficient on the extreme length instead of that between per- 
pendiculars, as usually taken. For merchant ships the two lengths are 
generally the same, but for warships the difference in the two is appreciable. 
Mr. A. W. Johns found that for vessels in which the block coefficient varies 
from *6 to -65, the calculated results are generally correct for a speed such 

V 2 

that y - i s about '85. Below this speed, the results are generally about 



22 MANUAL OF MAUINE ENGINEERING. 

7h per cent, larger, whilst above this speed the results are somewhat smaller 
than the experimental results. 

For vessels in which the block coefficient varies from - 5 to "55, the calcu- 

V 2 

lated results are generally correct for a speed given by-y- equal to 1. Above 

this speed, the results are slightly smaller, whilst below they are generally 
about 10 per cent, greater than the experimental results. 

The resistance of the ordinary ship roughly varies as the square of the 
speed, so long as the highest speed does not exceed that for which the ship 
is suited, and so long as the wave formation is not so great as to cause 
considerable variation in the trim of the ship. When such conditions prevail, 
as they do largely in modern very high-speed ships of small size as are 
Destroyers and Scouts, the variation in resistance does not follow so simple 
a rule, and, moreover, the fluctuation is apparently capricious, as may be 
seen by examining the figures given by Sir William White for destroyers 
(q.v.). Assuming, however, as we may in the case of ordinary ships, that 
E varies as S 2 , then to complete the expression, which will give a definite 
value to the resistance of a given ship, it was necessary to multiply the 
product of the above two variables by a quantity found from practice ; and 
if the law were absolutely correct, this quantity should have a fixed value, 
whatever the size and form of the ship, and would be a " constant " multiplier 
for all cases in choosing values for C and K. Actual values for them can 
be found in the tables of performances of ships on trial trips. 

Coefficient of Fineness. — To determine the form of a ship, as to whether 
it is " fine," " fairly fine," or " bluff," it is usual to compare the displace- 
ment in cubic feet with the capacity of a box of the same length and breadth, 
and of depth equal to the draught of water ; the coefficient by which the 
capacity of such a box must be multiplied to give the displacement being 
called the coefficient of fineness. Thus 

Coefficient of fineness = T =; rr- T , 

L x B x W 

D being the displacement in tons of 35 cubic feet of sea-water to the ton ; 
L the length between perpendiculars in feet ; B the extreme breadth of 
beam in feet ; and W the mean draught of water in feet, less the depth of 
the keel. Strictly speaking, the length should be measured from the stem 
to aft part of body-post on the water-line, instead of to aft part of rudder- 
post ; but as this dimension is not easy to ascertain without referring to 
the plans, and the calculation is made for the sake of comparison, rather 
than as an accurate computation, no inconvenience will arise from this, so 
long as all the ships under comparison are measured in the same way. 

It will be easily seen that the above coefficient only expresses a relation 
between the cubic contents of the immersed portion of the ship and a box 
of the same dimension, and gives no certain clue to the fineness of the water- 
lines, which is really what is wanted for consideration iu dealing with the 
question of power for speed. 

Two ships may have the same dimensions and the same displacement, 
and, consequently, the same coefficient of fineness, and yet one may have 
bluff lines and the other fine — the difference arising from the latter having 
a flat floor, and the former a hi^h rise of floor. To take an extreme case,. 



THR WKTTED SKIN OF A SHIP. 23 

the fine ship might have a rectangular midship section, and the bluff one a 
triangular one ; and if the " coefficient of fineness " was 0*5, the bluff ship 
would have rectangular water-planes, while those of the fine ship would be 
two triangles base to base. 

Now, if a coefficient be obtained by comparing the displacement with 
the volume of a prism, whose base is the midship section, and height the 
length of the ship, it will indicate the general fineness of water-lines, and 
form a guide in the choice of the constants for speed calculations. 

D x 35 

The Prismatic or Coefficient of water-lines = 



area of immersed mid-section x L 



The Skin Resistance is, in all classes of ships, the most serious, although 
in destroyers at full speed (say 30 knots) it only amounts to 45 per cent, 
of the total, as compared with 80 per cent, at 12 knots. With cruisers it 
is as much as 80 per cent, at 20 knots, and more than 70 per cent, at the full 
speed of 23 knots, while at 12 knots it is 90 per cent. 

From experiments made by Dr. Froude with varnished surfaces, such as 
given by the modern spirit mixed anti-fouling compositions applied to ship's 
bottoms to discover «, the index of V, the following was deduced : — 

Resistance = j X A X V n . 

j is a factor which varies with the length of the ship. 
A the area exposed to water rubbing. 
V the velocity in feet per second. 

n had a value 2 close to the bow, T85 at 20 feet from it, and 1*83 at 
50 feet. The average value throughout may be taken at T83. 

At a speed of 10 feet per second the mean resistance was found to amount 
to 0*25 pound per square foot. 

Taking as an example an express steamer, 400 feet long, at a speed of 
20 knots, what will be the resistance per 100 square feet of wetted skin ? 
Here j = '00886, and V = 34. 

R = 0-00886 x 100 X 34 1 8S = 563 lbs. 

The power required to draw this surface through the water at the velocity 
will be 

Power = 563 x 34 X 60 = 1,148,520 foot-lbs., or 34*8 H.P. 

This, of course, is the net horse-power, and not that developed by the engine 
driving the propeller. If, however, the efficiency of machinery, propeller, 
etc., were, say, 0'7, then 

The gross I.H.P. = 34'8 -^ 0'7 = 49-7. 

The Wetted Skin of a Ship should be accurately measured, but this is 
a somewhat long and troublesome business, and, moreover, necessitates 
having the full lines of the ship, which, as a rule, are not available in the 
initial stages of design, and certainly not accessible to the engineer as a rule. 
There are, however, methods of obtaining the area with sufficient accuracy 
for the purpose of estimating the indicated horse-power necessary to drive 
a ship at a required speed, and certainly is accuracy sufficiently close if the 



24 MANUAL OF MARINE ENGINEERING. 

allowances have been obtained from practice with wetted skins calculated 
by the same methods. 

Kirk's Analysis is a system introduced by the late Dr. Alexander Kirk, 
whereby the qualities of ships can be compared and incidentally the fitness 
of a proposed ship for the speed required. By the means he provided the 
wetted skin is found, and for ships with a high rise of floor, as were the rule 
at the time the area so calculated was within 3 per cent, of the actual, and 
often almost identical with it. With modern ships the error is often as much 
as 5 per cent., and even with ships having deep bilge keels and large fins 
for the side screws the error is 3 per cent. 

Mumford's method of calculating wetted skin gives fairly accurate results 
with ships of normal form and proportions, but with shallow draft ships 
and flat bottoms the actual surface is somewhat in excess of that given by 
his rule, which is as follows : — 

Wetted skin = L (1-7 d + b B). 

L is the length between perpendiculars, d is the depth of immersed midship 
section, B is the greatest beam, and b is the block coefficient of displacement. 
Seaton's Modification of Mumford's method is as follows : — 

Wetted skin = (c X d X L) -\ . 

a 

Where c is a coefficient = 2 X area immersed mid-section -fBxrf. For 
flat bottom shallow-draft ships c is 2*0, while for ships with a high rise of 
floor, as usual in yachts and fast-sailing ships, c is 1*6. For ordinary ships 
with a draught of water not less than one-quarter the beam c is 1*8.* 

D is the displacement in tons, and d the mean moulded draft of water. 

Seaton's Method for sea-going ships, whose draft of water is more than 
one-quarter the beam permits of a ready computation of wetted skin with 
the same in formation ; here L is the length, B is the beam, and d the 
moulded draft. 

(1) K = L -=- (055 B+d). 

(2) F = 42 4/K. 

(3) Wetted skin = F X D s — that is = 42 ^K X Di 

The following are the values of F for variations in K : — 



When K is 4, the value of F is 59-4 
,, o, ,, „ o2*9 

„ 6, „ „ 65-7 

,, 7, „ „ 68-3 



When K is 8, the value of F is 70-6 
Q 72-8 

10, „ „ 74-7 

11, „ „ 76-4 



The form suitable for a required speed is gauged by the coefficient of 
fineness of water lines, called usually the prismatic coefficient, as already 
stated; it also may be obtained from the block coefficient by dividing it by 
the coefficient of mid-ship section, when knowing the area of midship section. 

_ . . „ . Displacement X 35 

Prismatic coefficient = -j ^r- — — — -. -r • 

Area mid-section X lengtn 

This may be called the criterion of form. 

* To ships with multiple screws c is increased by 3i per cent., so that the multiplier 
is 207 instead of 20. 



DETERMINING THE POWER. 25 

The author has devised a rule for guidance in designing a ship, as also 
to provide a means whereby engineers may avoid trying to do impossible, 
or, at least, non-economic things in the way of forcing a ship beyond her 
capacity for speed. 

Seaton's Rule for Limitation of Speed is as follows : — 

Suitable prism coefficient F = 04 £JL -»- £/S. 

When L is the length of ship in feet, and S the speed in knots, being the 
highest at which the ship can be driven without an excessive expenditure 
of power. That is, with a length of ship L, 

The maximum economic speed S = (04 £/L -=- F)i 

In a general way the resistance will vary very closely with the square of 
the speed with ordinary ships, which are not driven at a speed higher than 
that given by this rule. 

The least length of ship for a given speed and coefficient of fineness can 
also be ascertained, thus : — 

Minimum length L = ( — j • 

Table V. gives the prism coefficient suitable to the length of a particular 
ship for a given speed ; that is to say, the actual displacement of the ship 
of a given length and for a certain speed, should not be greater than that 
given by multiplying the displacement of a prism of the same section as 
the midship section of the ship and the same length by the factor given. 

For example, — A ship 300 feet long, for a speed of 15 knots, should have 
a displacement not greater than *674 of the prism displacement — that is to 
say, she must have a coefficient of fineness of waterlines of - 674. 

To determine the power necessary to drive a ship at the required speed, 
the facts already stated must be ascertained — \jz., general dimensions, dis- 
placement, area of midship section, and the wetted skin. The first investi- 
gation must be to discover the highest speed possible under these conditions, 
and that is not less than that required. 

Having done this, a calculation should be made of the maximum total 
resistance R in pounds. The efficiency of the machinery and propeller 
must also be known, and if the general efficiency is E, and S the speed in 
feet per minute — 

Gross indicated horse-power = =;XSv 33,000. 

The total resistance is made up of three components : — 

(1) That due to the skin friction, called frictional resistance. 

(2) That arising from the making of waves and eddies, called residual 
resistance. 

(3) That due to the action of the propeller on the hull, called the aug- 
mented resistance. 

In the case of a paddle steamer the velocity given to the water by the 
wheels is higher than that of the water flowing past it, and increases the skin 
friction both before the wheels when the water is flowing into, and said to 
be feeding the race, and abaft the wheels in the race. But with a screw 



26 



MANUAL OF MARINE ENGINEERING. 



steamer there is also the increased velocity caused by the feed, but a greater 
loss is due to the decrease in pressure at the stern, owing to the action of the 
screw pushing the water away. So great is this in bluff ships that the water 
flows into the space behind the stern on each side of the race, and so causes 
an eddy stream to follow the ship. 

The following table has been calculated by Mr. Johns, of R.C.N. Con- 
structors, as applicable to all modern ships with clean, fresh-painted bottoms, 
and may be used for estimating the net horse-power necessary for over- 
coming the skin resistance. The horse-power to overcome the residual 
resistance can be calculated by means of Taylor's formula (p. 21). The 
two results added together will give the total net horse-power called E.H.P. 
If the propulsive efficiency is 06, then — 

I.H.P. = E.H.P. - 0-6. 



TABLE III. — Coefficients for Computing Effective Horse-power 

REQUIRED TO OVERCOME SKIN FRICTION BASED ON Mr. FrOUDE's CON- 
STANTS. AS GIVEN BY M.R. A. W. JOHNS. 

If S is the wetted surface in square feet, then 

E.H.P. = / . S, where / has the values given below. 











Length of Ship in Feet. 








Speed in 
Knots. 








































100 


150 


200 


250 


300 


350 


400 


450 


500 


25, . 


•2516 


•2477 


•2458 


•2444 


•2434 


•2415 


•2415 


•2407 


•2399 


24, . 


•2242 


•2207 


•2190 


•2178 


•2169 


•2160 


•2152 


•2145 


•2138 


23, . 


•1988 


•1957 


•1942 


•1931 


•1923 


•1916 


•1908 


•1902 


•1895 


22, . 


•1753 


•1726 


•1713 


•1703 


•1696 


•1690 


•1683 


•1677 


•1672 


21, . 


•1537 


•1514 


•1502 


•1494 


•1487 


•1481 


•1476 


•1471 


•1466 


20, . 


•1340 


•1319 


•1308 


•1301 


•1296 


•1291 


•1286 


•1281 


•1277 


19, . 


•1159 


•1141 


•1132 


•1126 


•1121 


•1171 


•1112 


•1108 


•1105 


18, . 


•0995 


•0979 


•0972 


.0966 


•0962 


•0958 


•0955 


•0951 


•0948 


17, . 


•0846 


•0833 


•0827 


•0822 


•0819 


•0815 


•0812 


•0810 


•0807 


16, . 


•0713 


•0702 


•0697 


•0693 


•0690 


•0687 


•0685 


•0682 


•0680 


15, . 


•0594 


•0585 


•0580 


•0577 


•0575 


•0573 


•0570 


•0568 


•0567 


14, . 


•0489 


•0481 


•0478 


•0475 


•0473 


•0471 


•0469 


•0468 


•0466 


13, . 


•0397 


•0390 


•0387 


•0385 


•0384 


•0382 


•0381 


•0379 


•0378 


12, . 

; 


•0315 


•0312 


•0309 


•0308 


•0307 


•0305 


•0304 


•0303 


•0302 



In the above table skin friction is taken as varying as V 1 ' 825 . 
In Table IV. are the values of C in the old formula I.H.P. = 



D« 



C 



for ships whose length varies from 100 feet to 900 feet, and the proposed 
speed from 10 knots to 28 knots, and designed with a form suitable for the 
speed — that is, the fineness of the water-lines of the ship is such that the 
prismatic coefficient of displacement will be not greater than given by the 
formula 0-4 4/L- VS. 

If the ship is finer than determined by this criterion, the value of C may 
be increased somewhat ; also in the case of a ship having engines of high 
olficiency, such as possessed by most turbines, and high-class reciprocators 



VALUKS OF COEFFICIENT. 



27 



a 

H 
< 



o 

H 
S5 



El, 

o 
o 

& 
a 

w 



CO 



xbW|!$Q 



ii 

p-i 
W 



< 

D 

s 
03 
o 



a 



a 
a 
a 
o 

o 

a 

o 

CO 

a 
P 



i 



CO 



6 

o 

K 
H 

03 

a 

H 

< 
a 

03 

H 
O 

CO 
fH 

CO 

a 



! 

00 






o 


l"» 


© 


CO 




00 


■■* 


© 


1(0 


© 


■f 


00 


CO 


r^ — « 






00 


© 


© 


H 


IM 


<M 


CO 


CO 


-f 


-f 


if5 


>o 


© 


© r- 












<M 


<M 


<M 


<M 


(M 


CM 


(M 


CM 


c< 


CM 


CM 


IM CM 








00 


© 


00 


CO 


Tt< 


~H 


r^ 


CM 


© 


CO 


00 




© 


© -f 








00 


© 


o 


F-H 


CM 


CO 


CO 


"+ 


f 


1Q 


>o 


© 


© 


t^ r- 








i-H 




(M 


IM 


IM 


IM 


IM 


(M 


<M 


CM 


CM 


<M 


(M 


CM CM 










CM 




© 


r-- 


-* 


© 


in 


© 


© 


~H 


m 


© 


-* oo 


CM 






© 


© 


i-H 


t—l 


IM 


CO 


-H 


-f 


■o 


l.O 


© 


© 


1^ 


t- t--- 








(M 


(M 


<N 


<M 


<M 


CM 


(M 


CM 


CM 


CM 


CM 


CM 


C-l CM 






o 


CO 


m 


■H* 


(M 


© 


r- 


CO 


© 


Tt< 


© 


o 


© 


CO 


00 Ol 






GO 


© 


o 




cm 


CO 


CO 


-+l 


•V 


«0 


© 


© 


© 


r~ 


t- 00 






F-H 


fH 


CM 


<M 


IM 


IN 


(M 


CM 


(M 


CM 


CM 


CM 


IM 


CM 


Oi CM 






(M 


m 


r— 


CO 


m 


IM 


© 


© 


CM 


r- 


CO 


00 


CM 


l~- 


f-h m 


CI 




on 


© 


o 


H 


IM 


CO 


-H 


-H 


W 


"O 


© 


© 


t^ 


r- 


00 00 




f»h 


FH 


CM 


(M 


IM 


CM 


IM 


CM 


<N 


CM 


CM 


CM 


CM 


CM 


<N <N 






m 


00 


© 


© 


00 


© 


CO 


© 


© 


F-H 


1— 


CM 


© 


© 


m © 






00 


© 




CM 


IM 


CO 


-* 


B5 


m 


© 


© 


r- 


I-- 


00 


00 00 






I-H 


i-H 


CM 


(M 


<M 


<M 


<N 


CM 


CM 


CM 


<N 


<M 


IM 


CM 


CM CM 






Of) 


-H 


CO 


CO 


IM 


© 


r- 


CO 


00 


-+l 


-H 


© 


© 


m 


© ■* 


CM 




on 


o 




CM 


CO 


"* 


-* 


m 


lO 


© 


r- 


t-- 


00 


00 


OO © 




^ H 


(M 


CM 


IM 


(M 


<M 


(M 


CM 


CM 


CM 


(M 


(M 


CM 


CM 


CM IM 






© 


•<f 


© 


CO 


W 


CO 


© 


I— 


CO 


00 


■* 


© 


•<* 


© 


CO 00 


CM 




© 


o 


iH 


IM 


CO 


■* 


m 


iO 


© 


© 


r- 


00 


00 


00 


© © 






i-H 


CM 


cm 


<N 


iM 


(M 


(M 


IM 


CM 


CM 


CM 


M 


CM 


CM 


<M CM 


c 




ec 


00 


© 


© 


© 


00 


w 


CM 


00 


"* 


© 


■n< 


© 


-* 


© CO 




ffi 


o 


i-H 


CO 


Hh 


rh 


m 


© 


© 


I— 


r- 


00 


00 


© 


©. © 






~ 


CM 


CM 


<M 


<N 


(M 


IM 


iM 


IM 


CM 


CM 


IM 


CM 


CM 


CM CO 




f-h 


F-H 


o 


00 


00 


t- 


© 


CO 




1— 


(M 


© 


tC 


© 


Tt< 


00 CM 




00 


© 




IM 


CO 


■* 





© 


r-- 


r^ 


00 


00 


© 


© 


© 


© —I 




r-H 


64 


CM 


<N 


IM 


(M 


(M 


<N 


(M 


CM 


c>» 


CM 


CM 


CM 


CO 


CO CO 


CO 


oo 


© 


Tf 


to 


r^ 


r~ 


© 


■^ 




00 


tH 


w* 


© 




r- 


— 1 © 


00 


o 


CM 


CO 


-+ 


>o 


© 


r— 


00 


00 


© 


© 


© 


F-H 


p-H 


<N CM 




^ H 


CM 


IM 


IM 


(M 


(M 


IM 


<M 


CM 


0-J 


CM 


CO 


CO 


CO 


CO 


CO CO 




CD 


r- 


CO 


r- 


00 


© 


l~- 


© 


CO 




t^ 


Tf 


© 


m 


© 






© 


fH 


CO 


■* 


W 


© 


r— 


00 


© 


© 


© 


F-H 


CM 


<M 


CO 






f-h 


(N 


IN 


<M 


<N 


<M 


(M 


(M 


IM 


CO 


CO 


CO 


CO 


CO 


CO 


1 


IN 


«o 


© 


ifl 


© 


iM 


(M 


(M 


F-H 


00 


© 


CO 


© 










o 


CM 


■*> 


eo 


r- 


00 


© 


© 


© 


^H 


(M 


CO 












CM 


CM 


CM 


<M 


<M 


CM 


<M 


CO 


CO 


CO 


CO 


CO 












© 


Tt< 




r- 


© 


i-H 


--H 


-H 


© 




















-f 


CO 


r- 


00 


© 


rH 


IM 


CO 


















CM 


CM 


CM 


(M 


<M 


CO 


CO 


CO 


CO 


















do 
































c 


c 


































o 




































































































S 


43 


































CO 
































Pi 


O 
































V. 


<— 


































o 


3 


O 


O 


^ 


© 


© 


© 


© 


© 


© 


© 


8 


© 


© 


© © 




c 


W 


© 


lO 


© 


© 


ITS 


c 


35 


© 


l.O 


•n 


© 


m © 








IN 


(M 


CO 


CO 


■* 


t 


W 


to 


5 


© 


I- 


t^ 


30 


CO © 



28 



MANUAL OF MARINE ENGINEERING. 



w 

X 

O 

Z 



z 

<! 

72 
Q 

a 
a 
Pu 

Jl 

A 
O 
ft 

a 

H 

t— I 

«j 

OO 
Ph 

as 

&. 
o 

z 
.a 



ft 
a 
a 
o 

a 

CO 

t— I 

D3 



> 

PQ 

_ 



1 


9 ! 
do 

(M 


ISe«OI)9OOOOHOt<O4ll>C03OHOQ 

• • • • c: o ci co -# lo r— cc © O ci co lo cc t- x o — ■ ci co "# 

• ' ' ■ — • lo »o lo lo io o o o — — — cc cc — — t— r- t— r- t— 




1 








p 


i-— « « O W O S t- t" '-S ^ H O C « 13 t> 5 5 - C CC 

• • • oo © — i«f iost-aoaH«*c hx r. i'mcC'Cl'j 

• ■ • t o o o to lo lo lo lo lo co cc cc — cc cc cc t— t- t- t- i~- 




p 
to 


NXO(''OC0^4l>0^<NQ^09)>Ot'01AOC O 
• • • si © Ol CO LO CC t— X © O O -* 'O t— 00 © © — < CI •**> >~. — 










^. • • • o — i ci -* lo t- oc © o — co -t — t- 05 © — ci -+ ic — t- 




p 


• •MONeoia»t'CSO-*w* 1 -'5fcoo-HiMr5L'jer-x 

• ■ ^ 13 LO Q lij LI O 13 ffi CJ a C5 CS CE CS I- t» r> t> t» t» t" h- 










O 
5-1 


cm cr. m o « c; x o o © x s ci t^ - -)i a o - ?! ci « 

• • O -CI ■* •- > X S O -i CI -* '- X S -■ CI ?5 L3 CI t- x a 
■ ' Tf L3 L3 L3 L3 L3 L3 L3 C 3 S S S S '- 1^ > > t^ > t' t^ t" 














OOL3MXffl^!0Xt»XC5t*Mt»O'tMOi- l M»L3 
• , ON«i3C3t"CiO^CIWL3t-OOW!<!^C5t»XaO 














p 
S3 


-tfX-*N3XCi5L3 1>3t- t— t— CI t- CI LO X >— CI CC LO © 

•xoo'*et^xo-'CiM'*coxc-<M-*L3f"Xao- 

■ •* LO L3 L3 L3 L3 L3 3 3 S 3 » C S > > r> t^ h- t' > r^ X X 




o nOCBWHIfJOrtOOOOOOOt'MCSMONiOt-OC- ' 

§ • ■* o o o o o « o » o s t a •- r-r-t-t-t-t-t-xxx 


3 
O 

w 


p 


— LOLO-*©LO©CO©X©©©X©C1X©COCO©-*OJCC 
•OIM'*fflXffi-HN«'<i | L0 3 00»l'*i3f«0'-Ne2S 

• o lo lo lo lo lo 3 co » co co » 3 > r- t^ i^- r- i^ cp op x x op 


P. 

ai 


O 

00 

— 


CO a ■* ■* ^ O t- O^XCO-i — ( CI — iKCSh IO © <N •>* t- X © 

oo©coLOi^©©oico^©t-x©cjcOLOi-x©;— cico^co 
■* lo lo lo lo lo co © © co © cc co t- t- t- t- t- t- x x x x x x 




q 1 ,— iO-*©'*<NX<Nt-©LO-t | CC©LOTt<©©©^X©CO-*© 

J5 ^ e 3 o o s s s a a ? ? 9 > t^ t^ i^ r- oc x x oo cc cc cc 




o 

— 


Hiiiflt"CiaoooN^t»OJHOxiSrtso*t>o-|2 

OC0L0fffl-H«'i , fflt»X»OC0L0 3XO-M'*i0r'XO 

iJ0LOLOLOL0©cc©©©©cct-t--t--t-r-xxxxxxx© 




3 lOioiontefflotooot't-fr-t-t'Xxooxxeoxca 




p 

l-H 


>o-*©eO'* | co©-*©co©©cO'*Tt<LO--HXLO©co©©-H'* 

(NOXOMDHOt-aOClM'tSXONMOt-XO-CJM 

mono©©©©©©!— t-t-t^t-t^xxxxxx©©©© 




p 

CO 


aOLOXOXLONt^oaxOrt^M^OHONiSOOrli 
CO©© — ^tLOt-©©C»CO'*©X©ClTj<©t-©© : -<CO^'LO 

LoiOLO©©©©©t-t-t— r-t— t-xxxxxx©©©o© 




el 


iocdco©x©lo©©-— x©-— coio^co—h© — !— — ict— 
SSoS^Sooo-HW^LOixa- »iot>»o-irt*|o 
ioioco©o©©t— r— r— i— r-t— t— xxxxx©©©ro© 


• 




p 

r-l 


Sx-h co lo t- © — c*-*lo©x©cw©x©-hcwjo ; 
LOioocsost-f 1 r-t— r-t— xxxxxx©©©© 








— 


©(N©"Ot-t-(M©t-<M©©CN©t-Xt-©©lO--CD 
0©C1t*<©X©C1COLO©X©— COLOt— © — iCI^LO ■ 

lo lo © co © © r- r— t— r-t— r—r— xxxxx©ss©© 






p 
— 
— 


00— i©-"#X©-*-h©CJ© — -too©©— <©©c>© 
CC©CllOt-©-HC0-^<©t-©©C)L0t-CC©Cl-*LO 

»o©co©©©r— t— t— r-t— t— xxxxx©©©© 








U3 

o 

— 


x— <0'"*t-©»oci©co©co©©eoLO"'*eo©io 
t^-H^?«x©ci-*!cor-©©-Hrt<©x©ci-f"0 ; ; ' 
io©©©©r-t— t— t— t-t— xxxxx©©©© 








p 

6 


O«Hf.»000010C)'*Cl'0»M»e:00MM 

lO©©©©t— t— t— t— t— xxxxxx©©© 




Length 

in 
feet. 


©LO©>0©0©>0©13©10©©0©0©©©©C)©©© 
S CN l« r- © CI LO r- O CI LO 1— © m © »o © LO © LO © m © LO © 

3rtrtrtNNNN«WMM^'*L3"OCBCBt't»MM»aO 



DETERMINING THE POWER. 29 

having all the pumps disconnected from the main engines, and with efficient 
propellers C, may be somewhat higher. In the latter case the increase will 
be generally about 5 to 7 k per cent. If, on the other hand, the efficiency 
of the machinery is from any cause low, as it used to be with the horizontal 
engines, and even with some of the vertical ones with air, circulating and 
feed pumps driven by the main engine ; and if the efficiency of the screw 
is for some reason low, then the values given in the table are rather too 
high. 

Further, it may not be assumed that such coefficients will be developed 
from the trials of such ships designed for and fitted with engines to drive 
them at high speeds when running at low ; for example, it is a common 
experience to find the highest value at a speed 10 to 20 per cent, below the 
highest trial speed, and a decrease in value with the decrease of speed below ; 
this is due to the fall in efficiency of the engines and propeller, both being 
too large for the power developed ; as a matter of fact, the value of C is 
a measure of the general efficiency of a ship. 

However, the calculations of resistance and net horse-power is scarcely 
the province of the engineer, and even the naval architect has found that 
experiments with models of ships in a tank are the most reliable way of 
ascertaining resistance and power. The Admiralty have employed a tank 
and trained staff of experimenters for more than 30 years, first under the 
guidance of the late Dr. W. Froude, since under that of his gifted son, Dr. 
R. E. Froude. Until recently tanks were a luxury enjoyed only by a few 
large wealthy shipbuilding companies ; thanks, however, to the munificence 
of the eminent engineer, Sir A. F. Yarrow, there is now, at Bushey, a tank 
equipped with the very best apparatus open to all who desire to have 
experiments made with the models of proposed ships. 

Tank Experiments with models of ships are very interesting, and of great 
importance to the builders of vessels out of the common order of things 
as to form and speed, as it is only by such means that their exact resistance 
under varying conditions can be computed with such accuracy and reliance 
as to permit of an exact provision being made of the power for the propulsion 
of the ships such models represent. In this way the designers of cruisers, 
scouts, destroyers, etc., whose form is uncommon and speed high, can deter- 
mine the horse-power necessary for them, as also that for the very high-speed 
express steamers now required for service on channels and oceans. 

Dr. William Froude's experiments with H.M.S. " Greyhound " and her 
model led to his establishing the laws which govern the true relation between 
ships generally and their models, as also those between one ship and another 
ship whose forms are similar but their dimensions different. Dr. R. E. 
Froude has for many years followed on with the work begun by his father, 
and from time to time has published his investigations and their results by 
reading papers at the Annual Meetings of the Institution of Naval Architects, 
in whose transactions they may be found recorded, and read with advantage. 
The fuller consideration of the subject, however, is one outside the scope of 
this work, except to say that Dr. R. E. Froude in this country, and Mr. 
Taylor in America, have developed methods whereby screw propellers may be 
examined and tested by their models, and their efficiency measured, not only 
per se, but when working at the stern of a model of the ship for which the screw 
itself is intended. By such experiments the effect on the ship of the screw 



30 MANUAL OF MARINE ENGINEERING. 

working astern of it is also ascertained — that is, the augmented resistance due 
to the screw. 

The tank is a canal wide enough, generally about 20 feet, and deep enough 
for such models as are used to pass through it without abnormal resistance ; 
the model itself is made of paraffin wax to a suitable scale, and towed by 
mechanical means at a speed given by the following formula, when L and S are 
the length and speed of the proposed ship, and I and s that of the model : — 



Then speed of model s = a/^ X S. 



If the model is made to the scale of a quarter of an inch to the foot, it 
will be one forty-eighth of the length of the ship, and consequently the speed 
practically is one-seventh that of the ship itself. The towing apparatus is on 
a travelling platform athwart the tank, which is caused to move at the speed 
required by electrical driving gear. The tension on the tow rope is carefully 
gauged and registered automatically, and gives the resistance of the model 
when free from the screw. The screw is then fixed to an apparatus on the 
same platform in rear of the model, and submerged so as to come into the 
exact position relative to the model that the real screw would be to the 
real ship. It is caused to revolve at the rate of revolution due to the speed 
of the ship and designed slip, but without propelling or even touching the 
model ; its thrust is carefully measured, and the torque or power necessary 
to turn it also noted. The tension on the tow rope under these new conditions 
is also recorded, and compared with the preliminary records. The model is 
also tried in the same way at lower speeds, so that it has progressive trials 
similar to those the ship will or may have. By comparing the thrust with 
the torque at each speed, the screw's efficiency is ascertained, and can be 
plotted on a curve ; by doing the same by the tension of the tow line and 
torque, the general efficiency of the ship can be compared in the same way, 
and by referring to the tests without the screw the augment of resistance 
due to the screw can be determined. In the case of the tension and thrust 
the speed in feet per second or minute is used as a multiple, while in that 
of the torque 2 ir X revolutions in the same time is the multiplier to give 
the power usefully employed and that developed. The thrust multiplied 
by the speed is the measure of the screw as a " pusher " ; the tow rope tension 
multiplied by the same speed is the useful work done, and, therefore, the 
measure of the screw as a propeller. Let T r be the tow rope tension, T the 
screw thrust, and t the torque, R the revolutions, and s the speed in feet 
for time unit. Then 

Efficiency of screw 

Efficiency of propulsion 

For engineers' use there are several rules, which may be employed with 
advantage, and which will give the indicated horse-power under normal 
conditions with a fair amount of accuracy. 

(1) Professor Rankine's Rule may be mentioned, although its employment 
is restricted ; it is as follows : — 



T X* 


6-28 R x t 


X ■)* /\ S _L y 

T^7 or T 



PROFESSOR RANKINE'S RULE. 31 

Rule I. — Given the intended speed of a ship in knots ; to find the least 
length of the after-body necessary, in order that the resistance may not 
increase faster than the square of the speed : take three-eighths of the square 
of the speed in knots for the length in feet. To fulfil the same condition, 
the fore-body should not be shorter than the length of the after-body given 
by the preceding rule, and may with advantage be one and a half times 
as long. 

Rule II. — To find the greatest speed in knots suited to a given length 
of after-body in feet, take the square root of two and two-third times that 
length. 

Rule III. — When the speed does not exceed the limit given by Rule II., 
to find the probable resistance in lbs. : measure the mean immersed girth 
of the ship on her body plan ; multiply it by her length on the water-line ; 
then multiply by 1 + 4 (mean square of sines of angles of obliquity of stream 
lines). The product is called the augmented surface. Then multiply the 
augmented surface in square feet by the square of the speed in knots, and 
bv a constant coefficient ; the product will be the probable resistance in 
lbs. 

Coefficient for clean painted iron vessels, . . 0*01 

,, ., coppered vessels, . . 0*009 to 0*008 

,, moderately rough iron vessels, . 0*01 1 and upwards. 

Rule Ilia. — For an approximate value of the resistance in well-designed 
steamers, with .lean painted bottoms, multiply the square of the speed 
in knots by the square of the cube root of the displacement in tons. For 
different types of steamers the resistance ranges from 0*8 to 1*5 of that 
given by the preceding calculation. 

Rule IV. — To estimate the net or effective horse-power expended in pro- 
pelling the vessel, multiply the resistance by the speed in knots, and divide 
by 326. 

Rule IVa. — To estimate the gross or indicated horse-power required, 
divide the same product by 326, and by the combined efficiency of engine 
and propeller In ordinary cases that efficiency is from 0*6 to 0'625 (Rankine, 
Rules and Tables). Marine engines to-day have a combined efficiency of 
0*65 to 0*70, and some even higher. 

Although the method here proposed has been found to give much more 
accurate and reliable results than those obtained by the older plans, it is 
open in practice to two very strong objections. First, it is necessary to 
have an accurate plan of the ship from which to measure the dimensions 
required ; and second, it is difficult in actual practice to measure accurately 
the angles of obliquity of stream lines, and the calculation requires more 
time than can generally be devoted to the purpose. Often the horse-power 
requisite to drive a ship at a certain speed must be calculated at the time 
the lines are being got out, and it would be too late to wait for a plan of 
the ship before getting some idea of the power. Again, the size and fineness 
of a ship cannot be finally decided upon until the weight of machinery is 
roughly known ; and as this will depend on the power, it is necessary to 
approximate to it on very rough and ready information, for which rough 
and ready rules are more suitable than the more refined ones. Hence, the 
rules based on immersed midship section and displacement could be con- 



32 



MANUAL OF MARINE ENGINEERING. 



veniently used to obtain that approximation, and the power calculated 
accurately from the augmented surface afterwards. 

Dr. Kirk's Analysis. — A method of analysing the forms of ships, and 
calculating the Indicated Horse-Power, was devised by the late Dr. A. C. 
Kirk, of Glasgow, and met with much favour on all sides. It is often 
used by shipbuilders on the Clyde and elsewhere for comparing the results 
obtained from steamers with those obtained from others, and likewise to 
judge of the form and dimensions of a proposed steamer for a certain speed 
and power. 

The general idea proposed by him is to reduce all ships to so definite 
and simple a form that they may be easily compared ; and the magnitude of 
certain features of this form shall determine the suitability of the ship for 
speed, etc. As rectangles and triangles are the simplest forms of figure, and 
more easily compared than surfaces enclosed by curves, so the form chosen 
by him is bounded by triangles and rectangles. 

The form consists of a middle-body, which is a rectangular parallelepiped, 
and the fore-body and after-body prisms having isosceles triangle for bases ; 
in other words, it is a vessel having a rectangular midship section, parallel 
middle body, and wedge-shaped ends, as shown in fig. 10. 

This he called a block model, and is such that its length is equal to that of 




Fig. 10.— Kirk's Analysis. 

the ship, the depth is equal to the mean draught of water, the capacity equal 
to the displacement, and its area of section equal to the area of immersed 
midship section of the ship. The dimensions of the block model may be 
obtained by the following methods : — 

Since A G is supposed equal to H B, and D F equals E K, the triangle 
A D F equals the triangle E B K, and they together will equal the rectangle 
whose base is D F and height A G. Therefore, the area A D E B K F 
equals E K X A H. The volume of the figure is this area multiplied by the 
height K L. Then the volume of the block is equal toKLxEKxAH. 
But K L X E K is equal to the area of mid section, which is by supposition 
equal to the area of immersed midship section of the ship, and the volume 
of the block is equal to the volume displaced by the ship. Hence, 



Or, 
Now 



Displacement X 35 = immersed midship section X A H ; 
A H = displacement X 35 -r- immersed midship section. 
HB = AB— AH, and A B = the length of the ship. 



Therefore, the length of fore-body of block model is equal to the length of the 
ship, less the value of A H as found above. 



DR. KIRKS ANALYSIS. 33 

Again, the area of section K L x E K is equal to the area of immersed 
midship section, and K L is equal to the mean draught of water. Therefore, 

E K = immersed midship section -r- mean draught of water. 

Dr. Kirk also found that the wetted surface of this block model is very 
nearly equal to that of the ship ; and as its area is easily calculated from 
the model, it is a very convenient and simple way of obtaining the wetted 
skin. In actual practice, the skin of the model is from 2 to 5 per cent, in 
excess of that of the wetted skin of the ship ; for all purposes of comparison 
and general calculation, it is sufficient to take the surface of the model. 

The area of bottom of this model = E K X A H. 

The area of sides = 2xFKxKL = 2(AB— 2HB)xKL = 2 

(Length of ship — 2 length of fore-body) x mean draught of water. 

The ar ea of sides of ends =4xKBxKL=4 V H B 2 + H K 2 X 
K L = 4 >/Length fore-body 2 -f half breadth of model 2 x mean draught 
of water. 

The angle of entrance is E B L ; E B H is half that angle ; and the 
tangent EBH = EH-HB. 

Or, tangent of half the angle of entrance = half the breadth of model 
-T- length of fore-body. 

From this, by means of a table of natural tangents, the angle of entrance 
may be obtained. 

The block model for ocean-going merchant steamers, whose speed is from 
15 knots upwards, has an angle of entrance from 24 to 15 degrees, and a 
length of fore-body from 0*3 to 036 of the length. 

For that of ocean-going steamers, whose speed is from 12 to 15 knots, the 
angle of entrance is from 30 to 24 degrees, and fore-body from 0*26 to 0*3. 

Rule. — For angle of entrance of " block model " — 

Angle in degrees = 70 —=— 

L is the length of ship in feet, S is the speed in knots. 

Dr. Kirk measured the length from the fore-side of stem to the aft-side of 
body-post on the water-line. This is an unnecessary refinement when screw 
steamers alone are being compared, as then the length may be taken as that 
" between perpendiculars." However, when small or moderate size screw 
steamers are being compared with paddle-wheel steamers, it may be neces- 
sary to measure in this way. 

(2) The old Admiralty rules are as follows* : — 

(a) Indicated horse-power = D« X S 3 -r C. 

(b) „ ,, = area immersed with section X S 2 ■*■ K. 

* If D x be the displacement in pounds, 8 t the speed in feet per minute, R the resist- 
ance in foot-pounds per minute, A the constant, then 

R = D^ X Si* X A. 

Multiply both sides of this equation by S v then 

R x S, = D,' x Si* x A. 

Now R ,< S x is the work done in overcoming the resistance R, through a distance S u 
and is, theuefore, the power required to propel D x at a speed S lf and if B is the efficiency 

o 



34 MANUAL OF MARINE ENGINEERING. 

D is the displacement iu tons, S the speed in knots, C and K are coefficients 
determined from previous practice. Their value varies with the size of the 
ship and the speed — that is, C and K will be less for a long ship than a short 
ship, the speed being the same — and if the length is the same, the value will 
be greater at the slower speed than at the higher. 

The above two rules were, for many years, the only ones used by ship- 
builders in determining the necessary power for a given speed. Their 
partial accuracy depended on the fact that the wetted skin varies very nearly 
with the displacement in ships of somewhat similar form,* and that the 
proportions of steamships were such that the wetted skin varied nearly with 
the area of immersed section. Their usefulness depended on the information 
in the hands of the user, and on his discretion in choosing values for C and 
K. These rules are, in experienced hands, a good check on the newer methods, 
and can be used by themselves with fewer data than are required when rules 
based on wetted skin, etc., are employed. Actual values for C are given 
in Tables vii., viii., ix., etc., on pages 41 to 48, deduced from the per- 
formances of ships on trial trips made with every care ; but in choosing 
values discretion must be exercised that the ship for which a calculation 
is to be made is somewhat similar in form, size, and speed, to the one whose 
constants are selected. The values on Table iv. are made to suit all conditions. 

The value of C may be taken as approximately = 140 X \/L -5- \^S v. 
Table iv. gives them calculated in this way. 

(3) Horse-power by calculation from wetted skin. 

This is a simple and efficacious method, and one giving very satisfactory 
results in practice. It is based on the assumption that so long as a ship is 

* Note. — Let L be the length of edge of*a cube just immersed, whose displacement 
is D and wetted surface W. Then 

D = L 3 or L = 3 Jl), 

and 

W = 5xL» = 5x( \fD)K 
That is, W varies as D. 

of the machinery and propeller combined, so that B X I.H.P. is the effective horse- power 
employed in propelling, then 

33,000 (B X I.H.P.) = D, 3 x S^ X A 
I.H.P. = (D x 3 x S^) X 



33,000 B 

Now, it is more convenient to express the displacement in tons and the speed in knots ; 
so that if D and S be substituted for D x and Sj, D being equal to D l 4- 2240, and S = 
(Si X 60) ~- 6080 s= S x — 101-33, it involves the introduction of other constant quanti- 
ties, which do not, therefore, alter the expression, so that the whole of these constants 
may be replaced by a single constant, C, which will express them. Therefore 

I.H.P. = D§ * S '. 

D being the displacement in tons ; S the speed in knots ; and C the coefficient. 

It was also supposed that the resistance would bear a direct relation to the area of 
section transverse to the direction of motion, as this would be the measure of the 
channel swept out by a ship ; hence the following rule : — 

T area of immersed midship section X S 3 
T.H.P. = ^- , 

K being also a so-called constant, but really only a coefficient. 



DR. KIRKS ANALYSIS. 35 

not over-driven — that is, the speed does not exceed that appropriate to her 
form — the power will vary as the cube of the speed with machinery whose 
efficiency is not less than 0'9. and propellers suitable to the conditions, both 
as to diameter and area of blade surface ; an allowance of 5 I.H.P. for each 
100 feet of wetted skin at 10 knots is a fair basis for calculation, and easily to 
be remembered. It is obtained by supposing that the resistance per square 
foot of clean painted bottom at that speed should not exceed 1 lb. Dr. 
Froude found, with the " Alert," having a coppered-bottom, it was about 
1£ lbs., but a modern ship with a smooth steel surface coated with varnish 
paints will not cause so much resistance as old copper sheathing on a wooden 
ship. The author has come to the conclusion, from a careful examination 
of the trial results of a large number of ships, that it is frequently less than 
1 lb., and that 1 lb. is a fair all-round allowance. The resistance, then, is 
100 lbs. for 100 square feet, at the velocity per minute of 101*3 feet. 

Net horse-power = 100 X 1013-3 -=- 33,000 = 3*07. 

Taking the net horse-power ae 62 per cent, of the gross, the 

I.H.P. = 5-0 nearly. 

If the efficiency, as is the case nowadays with high-class reciprocating engines, 
is 70 per cent., then the 

I.H.P. = 4-386 per 100 square feet at 10 knots. 

The same remarks apply in this case as to the former, as to the variation 
in value assignable due to the influence of length on speed ; hence a suitable 
value to each case may be calculated as follows : — 

Rate of I.H.P. per 100 feet wetted skin at 10 knots = 8*5 >/S -=- \/L. 

If, then, the allowance for 10- knot basis is Q, then 

/S\ 3 
Gross I.H.P. = ( — J x Q for a speed of S knots. 

For Example 1. — Find the I.H.P. required to drive a twin-screw steamer 
500 feet long at a speed of 23 knots, whose wetted skin is 40,000 square feet. 

Here 85 v/23 -=- VoOO = 5' 10. 

Allowance for 23 knots = ( ^q) X 5*1 = 62*05. 

Total I.H.P. = 62-05 X 400 = 24,820. 

Example 2. — How much I.H.P. will be necessary to propel a steamer 
300 feet long at a speed oi 21 knots, the wetted skin being 13,500 square 
feet, and the displacement 1,950 tons. 

In this case Q = 8"5 V2T - V300 = 5-63. 

Allowance for 21 knots = (~) X 5-63 = 52' 2. 

(a) Total I.H.P. = 52"2 X 135 = 7,042. 

By Admiralty methods, C = 140 X x/300 - 1/21 = 212. 

D« = 156. S« = 9,261. 

rr * , TUD 156 >< 9 > 261 *Q1A 

(b) Total I.H.P. = 212" = 6 ' 814 



36 MANUAL OF MARINE ENGINEERING. 

Example 3. — A cargo steamer 400 feet long lias a displacement of 7,500 
tons, a wetted skin of 29,500 square feet, what power ia required to propel 
her at 12 knots ? 

Here Q = 8'5 #12 -s- V400 = 4-38. 

C = 140 X ^/400 -^ V 12 = 274. 

Ds = 383. S 3 = 1,728. 

/12\ 3 
Allowance for 12 knots = ( ^J X 4'38 = 7'56. 

(a) Total I.H.P. = 295 X 7'56 = 2,235. 
By Admiralty method — 

(b) Total I.H.P. = 383 **' 728 = 2,397. 

Example 4. — A yacht 230 feet long is required to steam 15 knots, her 
displacement is 1,100 tons, and the wetted skin 8,800 square feet, what 
I.H.P. should she develop ? 

Here Q = 8*5 #15 + J/2S6 = 5:37. 

C = 140 x v ; 230 -=- VTB = 221. 
D§ = 107. S 3 = 3,375. 

(15\ 3 
~j X 85 = 1,594. 

(h\ 107 X 3,375 

(o) » = 22l = ' 

It must not be forgotten, however, that these rules for speed and horse- 
power apply only to ships working under trial trip conditions, and the results- 
shown in the following schedules are from trials made under the same cir- 
cumstances — viz., in practically smooth water, free from eddies and cross 
currents, and when the wind and weather are moderate ; further, the ship 
herself is in the best trim and condition, her bottom clean and fresh-painted, 
and, finally, the water is deep enough to preclude the influence of the sea 
bed seriously affecting her progress. The latter condition is an important 
one, and has had only in late years the consideration it demands, although 
it was patent to every one having to do with steamships long ago that in 
shoal water there was a marked diminution in speed. Now, however, we 
know, from the careful observations of many nautical authorities, that 
the influence continues, though in a lesser degree, when the ship is in deeper 
water than was formerly thought necessary for successful trials. 

These things being so, the following rule should be observed if it i& 
intended that the ship shall be able to attain the legend speed under some- 
what unfavourable conditions of wind, weather, and condition of wetted skin, 
there must be provided a margin of power beyond that sufficient under trial 
trip conditions. This margin should be one of power, and not one of speed, 
for by the latter method the conditions are by no means satisfactory. It 
was, and is still, a common practice for the owners of express steamers to 



PROGRESSIVE TRIALS. 



37 



specify a trial trip speed considerably in excess of the speed desired on service ; 
the hull is, as a consequence, of much finer form, with the corresponding lack 
of capacity for carrying deadweight. Moreover, the power and consequent 
weight of machinery is far in excess of what is necessary to fulfil the real 
requirements of the service. 

For example, suppose a cross-channel steamer is required to perform 
her service in fair average weather at 20 knots. The owners specify that 
the trial speed is to be 22 knots, their real intention being to have a 10 per 



cent, margin. 



Now, as a concrete example, suppose she is 350 feet long X 40 feet beam X 
12 feet draught water, having a displacement of 2,644 tons, the following 
table of comparison of her with what she might have been had a 20 per cent, 
margin of power been provided for contingencies. It will be seen that the 
latter ship has 100 tons greater displacement, and her machinery, with 
the 20 per cent, excess power, is 120 tons lighter, and the cost will be about 
£7.000 less. 







s.s. A, in accord- 


s.s. B, for same 






ance with Spectfled 


Service and 






Requirements. 


Scheduled Speed. 


Length, . . . . . 


. ft., 


350-0 


350-0 


Beam, • 


. ft., 


40-0 


40-0 


Mean draft water, . . 


. ft., 


12-0 


120 


Prismatic coefficient, . . . 


• 


0-601 


0-623 


Displacement, . . . . 


. tons, 


2,644 


2,744 


Area immersed mid section, 


. sq. ft., 


440 


440 


Maximum I.H.P., 


• • 


9,000 


7,800 


I.H.P. for 20 knots, . 


. a 


6,322 


6,488 


Maximum speed — Maximum I.H.P 


•» • 


22-36 


21-2 


Weight of machinery, . 


. tons 


900 


780 


Difference in cargo capacity, 


. 


. • 


220 


Increased consumption of coal per 


100 miles, 




13 cwt. 



It will be observed in examining the schedules that, whereas the merchant 
ships are generally somewhat over-driven on trial for their forms, the naval 
ships are often finer than demanded by the legend or even actual speed, 
consequently they will attain their legend full speed under somewhat un- 
favourable circumstances. 

The high speed of torpedo boats and destroyers depends almost wholly on 
their lightness of both hull and machinery, which enables them to do with so 
small a displacement that they literally skim the water, and the resistance 
per square foot of wetted skin is consequently comparatively small. Unless 
small boats are made to float at a very light draught they cannot be driven 
at high speeds, and all experiments with fast river steamers on the Clyde 
and elsewhere have shown the decided advantage of light draught. The 
effect on such ships of the depth of water in which they move is also consid- 
erable, and very interesting data have been given already showing this. 

Progressive Trials. — In modern ship-trials information is usually sought 
both as to the power required for the highest speeds and (what is equally 
important) for lower speeds, as such knowledge gives the means of gauging 



38 



MANUAL OF MARINE ENGINEERING. 



the efficiency of ship and engines, jointly and separately, and is useful in 
ship designing. 

The system of examination is as follows : — Let P x P 2 P 3 be the power 
developed in obtaining the speeds $! S 2 S 3 in knots with R x R 2 R 3 revolutions 
per minute. Take a line AN as a base line (fig. 11) ; on it take points B, C, 
and D, so that A B, AC, AD are proportional to S t S 2 S 3 ; at the points 
B, C, D erect ordinates, B b, C c, D d, so that they are proportional to P x 
P 2 P 3 . Through the points b, c, d draw a curve, which is called the curve 
of power, or curve of I.H.P., and it is such that if an ordinate be drawn through 
any other point, X, on the line A N, the part X x intercepted will measure 
the power corresponding to the speed measured by A X. If the curve ia 
accurately drawn, it will be found that it does not pass through the point 
A, but above A, at a distance A a ; this would signify that when the engine 
was indicating the power measured by A a the ship would not move, and so 









./ 


/ 










2 
o 




/ 

/i 






e 




/ 1 

>^ r\-- 


> 


J 


' 




..-~ 


\ 


\ \ 


t 




Soeed in knots. B ) 


I t 


! i 


) H 



Fig. 11. 



A a is the amount of power required to overcome the resistance of the 
machinery and propeller at starting, or rather, when not propelling the ship, 
and hence A a is said to represent the initial friction of the machinery. 

Curve of Revolutions. — A curve of revolutions is constructed in a similar 
way, by taking points r x r 2 r 3 on the ordinates, so that Br x , Cr 2 , Dr 3 are 
proportional to R x R 2 R 3 . When the slip is constant the curve of revolutions 
becomes a straight line. 

Curve of Slip. — The slip may be shown by a curve whose ordinates are 
proportional to the percentage of slip at the speeds S t S 2 S 3 (v. fig. 12). 

Examination of Curves will show — (1) what indicated horse-power revo- 
lutions and slip correspond to any speed intermediate to those observed ; 
(2) the efficiency of the engine at its lowest possible speed; and from it an 
idea may be formed of its general efficiency, and a comparison made with 
other engines ; (3) the efficiency of the ship, as tested by the rate of increase 



SEA PERFORMANCE OF STEAMERS. 39 

of power for speed, which is seen by the form of the curve towards the higher 
speeds — if it begins to mount upwards suddenly it is certain that the resist- 
ance has there begun to increase abnormally ; (4) that, if the curve is one 
fairly following the law of resistance increasing as the square of the speed, 
an estimate may be made from it of the power requisite to drive a similar 
ship at speeds higher than the highest observed curves, or lower than the 
lowest ; (5) any sudden rise in the slip alone indicates the propeller to be 
defective in either diameter or surface, or both. 

Another method of expressing the results of progressive trials is by 
setting out A B, A C, A D proportional to S x 3 S 2 3 S 3 3 , and erecting ordinates, 
etc., as before. If the indicated horse-power throughout varies as the cube 
of the speed, the " power curve," or line drawn through the points b, c, d, 
will be a straight line ; and if the power increases at a higher rate than the 
cube of the speed at any point, the line will again assume the curved form. 
The advantages of this plan over the one before described lie in the fact 
that a straight line is more easily drawn than any curve, that any deviation 
from a straight line is more easily detected than that of one curve from 
another, and that the production of a straight line is less liable to error 
than a curve, so that the interception of Aa is less open to error than by 
the previous method. Of course, the curves of slip and revolutions cannot 
be examined so well by this latter method as by the former, and it is only 
the " power curve " that should be analysed in this way. 

The values of the different " constants," rates, etc., for ships found from 
calculations made from the results of carefully conducted trial trips, are 
more reliable than those got by taking averages when employed in calcula- 
tions for proposed ships. Tables vii. to xiii. give the values of constants, etc., 
as obtained from the performances of some well-known ships of various 
types and sizes. 

Sea Performance of Steamers. — That the engines may work economically, 
both in consumption of coal and stores, as well as in wear and tear, it is 
advisable to run them at such a speed that they develop about 80 to 90 per 
cent, of the maximum power. Steamship owners do not always care to 
pay for 20 per cent, more power than is requisite to drive their ships at the 
speed intended, but there can be little doubt that it is true economy in the 
end to do so. For short voyages there is not the necessity for this reserve 
of power, and very fast steamers could not afford to carry the weight entailed 
by such an excess of power beyond the actual requirements ; but ordinary 
sea-going steamers making long runs can, as a rule, easily do this without 
much sacrifice, and the wisdom of such a course would be shown by the 
saving in working expenses at the year's end. 

Although, as a rule, trial trips are made honestly, and what the engine 
has done on the day of trial it can easily be made to do again, still, with the 
limited staff available for continuous service when the ship is at sea, there 
cannot be that attention devoted to the working parts which was bestowed 
by the staff of the manufacturer ; and the application of water to the bear- 
ings and brasses to prevent heating, which is often deemed absolutely neces- 
sary by sea-going engineers when the engine is running at full speed, cannot 
but affect those parts prejudicially, and is a poor substitute for an attendant. 
Unless, then, the engine is run at a power somewhat below that of the trial 
trip, either a larger staff of engineers and attendants must be employed, or 



40 MANUAL OF MARINE ENGINEERING. 

the wear and tear may be appreciable. It is true, on the other hand, that 
some engines will develop more power after a voyage or two than was obtained 
on' their trial trip, due to the polishing of the rough surfaces of the guides 
and cylinder walls, and to the general " smoothing " of all the rubbing sur- 
faces ; but it is also true that even such engines should not be run for length- 
ened periods at their maximum power. The improvements in design and 
the employment of better materials for guides and bearings, however, admit 
of modern engines being worked at high speeds with less risk than was 
the case formerly. The better balancing of the engine with three cranks, 
and the almost perfect balancing of the four-crank engine, together with 
the extended use of superior white metal in guides and bearings, metallic 
packing in the principal stuffing boxes, have permitted of a high rate of 
revolution with less risk than obtained formerly with the slow-running 
mercantile engine ; but, notwithstanding all this, a reserve of power for bad 
weather and emergencies is in every ship highly desirable. 

Such remarks are, however, scarcely applicable to the turbine, inasmuch 
as it is cased in, and every provision made for high revolutions. Moreover, 
it is at high speeds that it is most economical, and in the mercantile marine 
is fitted to express steamers whose service demands full power. Moreover, 
it is claimed as one of the advantages of this form of motor that, in spite 
of high revolution, a smaller engine-room staff can be employed. 



RELATION OF POWERS AND SPEEDS. 



41 



ft 

= 



QQ 



CO 

a 

K 
fa 

a. 
02 

O 

<! 

CO 

a 
- 

o 
Ph 

fa 
o 

z 

o 

— 

H 

-< 
J 
fa 



ft 







to 


CO 


CO 


a 


CO 


CD 


CO 


P-H 






O 


IO 


00 


10 


OJ 


■>■ 


t- 


CO 




<9 

«3 


-" 


~ * 


™" 


—• 


CN 


CN 


CN 




O 

M 




































o 


fa 


O 
CO 


s 


8 


O 
O 


O 

8 






O 
O 






w 


— 








O 


10 


IO 
























fa 


~^ 


CO 





o> 


— 


CN 


■^1 










•-H 




^H 


. — 1 


>—i 






ti 


t— 


§ 




I- 


ira 


CN 


tr^ 


«0 




V 


■fl 


CO 


IM 


05 


00 


a> 


a 


DO 


O 






-* 


ex 


CN 


CM 


CN 


w 


o 


















*3 

i 


3 


































6 

o 
O 


00 










O 





O 


O 


O 




fa 


■>• 





O 





O 


O 


O 




W 


00 


m 


■"* 





ItO 


O 


O 


4a 
























CM 


CO 


co 


t>- 


os 


O 


73 


















~H 


5 

1 








































(a 


Ol 


CO 


a 





■>■ 


00 


T« 




5 


CO 





10 


UO 


-* 


Oi 


O 


T3 


a 
45 





(N 


CN 


CN 


CN 


CO 


OJ 


co 


■ 


O 


















• 


2 


































fa 


■* 


fa] 


O 


O 


O 














W 


r—* 




CD 


O 


O 








8 







a 


CN 


^H 


•-H 


-* 





CO 


HH 




fa 




' H 


0» 


CM 


CO 


■* 


Tj< 






w 


O 




■* 


OS 








10 






s 


O 


00 


CO 


l~- 


CO 


01 


«5 




<ia 

o 







f^\ 


PH 


<N 


Ol 


co 


CN 


CN 


























































o 


fa 


O 





O 


O 





O 


O 




■H 




^H 


10 


O 


O 





O 


O 






W 


pH 


^* 


f- 


CO 





IQ 


O 






1-4 












p-H 


CM 




i a 




co 


Iffl 


O 








O 


O 









£? 


O 


CO 


05 


O 


10 




&9 s 




' — 


fr» 


CO 


CO 


CO 


1— ( 


lO 




.2SE- 








of 


CO 


t-T 


oT 


^H 


















" - 




*3 . 

si 5 
S53 




M 


CO 
CO 


CO 
CO 


CM 

CO 


co 


OS 


CO 




Q~ 






r-H 




CN 


~l 


CN 




A 




















40 




















"a It 




t 


t- 


i-H 


co 


O 


m 


^0 




• 


•■-*. 


CN 


t*I 


•*< 


co 


CO 


CO 




■i f= 


1 


















M 




















A 




















teat 

C •••• c 


1 


iO 





U0 








•0 


»o 




1 


co 


CO 


CO 





CD 


t-- 


(M 




3 * 


1 


pH 


(M 


CN 


co 


CO 


co 


10 










~jv — 


^_ 


_<-, 


_^ 





_*_ 








, 


•."» 1 


„. , 




- ■ 


- 1 


t_, 










*3 " 


h 


u 


u 


u 


CD 

00 










eS t* 


CD 





a) 


<D 










V 


m 


to 


to 


to 






e . 




■ 


5 ° 


'3 ■ 


sV 




'3 


a 








u 


h 


U R 


to 




C CO 




c3 


3 g 
C3 m 


Os_ 


2 


O 


°s 


CO 




... eg 

CO 
CO 

p 




O 
W 

O 
^) 

O 

pi 


orpedo 1 
" Sharp 
Class, 


ct3 
CO 3 


Class 
Terpsic 


Class 
Edgar,' 


J2 g 


mtio P 
tearaer, 











■H •* 


T3 S 

a 

CN 


*s - 


*a - 


g 00 


1 






EH 


H 


M 

CO 


to 

P-H 


1—1 


< 



42 



MANUAL OF MARINE ENGINEERING. 



03H 



Ift CO © CO J^ 

©cn©uo>ocsio£J 

CN tfi ON — ■ — r- © co 
-* — co — CN 



o 

CO 



£8 



no ■* 



00 



CO 

o 
■* CO 



, 8 

tf GO 



^ f 



OOOiOOlOn 
OOOffl-OOO™ CN 

• • • oo eo eo • w " cs 

CX 'O ■"*)< . . .Co j^. 
^ i-* ^ CO 



cs 
x 



TT CO 
— CO 



5" p 

GO o 



CO 
CM 






03^ 



OOOIMMOh 
OOl-OCIMWOo) ° 

CO uo (M co — i cM — ' 'ft • 
•<* — i -* -h CN 



CO 
IQ 



' na O '■"' 

■ 55 os ip to 

ONJO *"" "* ON 

/m CO ^ 



o 1 



© 10 

g « 

o 



co 

H 
O 

M 

CM 



o 

H 

OS 

H 
< 

GO 

P- 

ts 

GO 

a 

63 
E-i 
GO 

6- 
O 

CO 

< 
H 

O 

CO 

H 

k3 

& 
CO 
63 



!> 

PQ 
Eh 



co f 



o too* 

OOOOOMaoO O — ' 

ocoiSg-.ifeg S3 

cn tj< p^ co — ' r- 

^ CO 



CO 



o 

.'CO t- 

rH O 

-h CN 



00 



r-» "* 



CO 

CN 



© uo 

g g 

o 



til 



_ o © _ ^ no© 

PP^c5cOcN P " 

cn o> —i ?i r4 °1 co o 

CS CO CN l« CO -h CO 
CO CN — —i 



CO 
CO 



CO 
OS 

CO 



8: 



°2 

CN t> CO 



o 



co 

CO 



o 

CM 

CO 



_ 'ft 

eo co 

^ © 



G0TJ 

PQ 



OOeogiS^-o R £ 

CN-^rt^rHt-CO. ^ 

CO CN CN 



'CO 



t- 

•* 



• t~- CO 



— I CO 
•0 CN 



o 

CO 



no 

CO 



co'63 

«! f 



rt< 



_ © © '.«..,» 'ft © 00 OS 

2co-co' n c-.c?°co t- * ! 

CO ■— I ^H 



o 

CN 



CO 

os 



co co 



CO 
CO 
CO 






CO" 

«j I" 

CO > 

o 



©©eo©©cs©co 

- O OS o ' 



o t— 1-» tc m 
j-hco — co 
co 



01 



CO OS 



CN 



o 

CO 



<?>!o - 



CN 

co ■* 

rt!C O » ^ 



eo co 
50 6 



03 w 

* I' 



o o ^ » 

OiOCHOLIooO lO 

lb 4t< OS^ CN ^.OL, g 

m eo — Ooo3 • 
CN —i °' 



o 

o 



o 

Mm (S W 

H™ O ip 

-h °° do ib 



eo 

CN 
CN 



o 



U5 



co">! 
w f 



coO 

co r 



oocoOinoooS co 

OCNCNt^CO -OSo m 



lOMHH 

CN 



CO 
CO 



^ CN 



i— i CO *P 

p-h 35 cb 



9 »o 

Id ON 



^ — -« SSS3 on • -' io co 

0<-*-*TCOOS -Ooi^. r^. 

CO •* ON co >o t- o • ■ 

eo on co 



ipcN 

o co 






•^ o 

tc eo 



o 



o § 

00 CN 



00 < 

50 Pi 



© O © rS 

© ° - 1 S "* X ° CO "* "5 

J^ ^. Z, °0 fM ON • ' J CO OS 

©OCNcnSJcO'-'o co CO 

© co -i X3 eo 3g tc co » cp 
ci o) 



00 rt 1C 

— . CO ^ 



9 ON 
CO 0M 



r- — 

50 6 



coS 

H 63 



O 

EH 

■< 
cs 

°3 

u 

Q 



© © © ~ 

©© — CN©OOo©CO © 
•••O0l«©- o -h lO 



T'cn w o 

OS OS ON t(< 



-* 

t— 





CJ 



e3 
5-0 .T3 



■3 a 



i £ 

C CD 



cr o 

to ^ 










© 

CO 00 

O CO 



S-s^ c ' eo 
e " ° *« k/5 ' 

O-O,!- o tt' .p 



+= .s 



I S ® g o 'I- S 







eo ■ - 

©ft 



-2 J 


CO 

HP 


§ % 


o 

a 


■.-*■! 


■ M . 


- -tJ~ 


a 
o 


o cc 


a 

0) 




. o . x 

■^ o 


d 

CO 

o 


CO 4> © 


wrj 


rt 


x mid 

se-pow 
per 10 


eo ~ « 
JUCV2 


CO 

" 1 


. to 

til's 

h-i .1. 


Length 
Speed, - 
Ind. hot 
I.H.P. 


X 

H to 
P 


W 2 

• s 
i— i 


a; 
w 

t-H 



RESULTS OF TRIALS OF STEAMSHITS. 



43 



»: i" 

00 i 



oo'« 

Si 



00 
oo\ 



aj 



oo-nffl co © © S£ co co 

■ >o *i^ w a 

1 * <**2« ™ * 

^H f^ 



- o 

SCO CO 10 

co 



CO H 



; 00 © 

fH CM — 



10 

t"- Tf (^J 



o o co 10 — <-> N 

©© iO © CM »?- 

© © CO •. •» 

CO lO <M ^< -H 
iO — 



©^CM 



© 



10 



«^2 ~ 2 






co — 1 



o 
!* co 



00, 






000 

O O 01 O CO 

.U " ,i. *o oS "* © 

lO — CO OS fH «Zi o 

NiOaOl — © ^ CO 

■* CO " CM 



t- 






! £ £ 



S co 
■* 10 CM 



CQ 



© 
© 
CM 



CM © 

CO f-. 



O 

r» co 

CO ^* 

© ~ 



00 

CO 



CM 

IO 



00 

O 

W 

10 



o 



CI 



H 

<< 

00 

- 

— 

a 

00 

53 
>< 
w 

go 

fa 
o 

OQ 

►J 

< 

5 

fa 
o 

00 

E- 
.J 

P 
ao 

fa 

« 



00 >J 

ao' 

00, 



u 



00 o 

00 t 



o o 

O CM CO <M t— 

O iO » 00 t^ 
O Tj" — XQ 



O CO — < 00 
-CM 00 t^ CO 



1^52 co t o 
■*Jr -* ■* cm 



_ © © f-. —. °° "J" © 
©O©© S «• «1 ■* 
CM CM ^- t~- • co co 

50 ©*2© «P <° 

SI i-h 



o — t- ^ 

CO "~ ^ 



■^ CO "* 

— 1 O) — ' 



co 00 



CM 

CO 

CM 



IO 

IO •»»< 
■* CD 

50 6 



CO 
CO 
CO 



o 
o 

1 - 



oriH 



o o 

00 3 o o 

~ • • fc« CO 

OO—O © 



>o o CO _„ 

O • CO t^ 

••CO » t- t^- 

CO aj t-- • 
CO OT CI 



09 



CO 2 fH ■* 

FH W fH 



o 

-* CO 

o 

CO 



o 
o co 
r«- ■* 
eo ft 



S Q — 
99^0 

35 co-h(n 

CN 



<5 oi co o 01 
;2 © • o t 



"5 -* © 

ffl M"J M ifl 5) t£> 



ao' > 
» )' 



© co _ 
O©co © 

■ ■ 00 

^•*Oo 

o co ~< — . 



© rSSn 

© CN "? " CN 

CO © 00 ?« © 



© CO © CN w 



© 

CN 



10 

00 

© 



CM 

© 



00' fc; 

Si 



©© © 

OOt-CN 

© © © l - 

© O CM CM 



CI 



co © So 

10 ©jy © 

»©o I-, 

•* -< CM 



S ©^ 2 £ © 

© • 2 ' w co 

^H CO © ^ ^* 



00 

© 

© © 



co 



00.0 



00 a 

■3 



© 01 © 

© © © CO 

— -U -U W 

© CO © ^ 
10 © CM CM 
U0 CM 



© © © © 
r^ © © B J © 

. ••© o qo 

FH © „ F< • 

© CO 



© So © © 

© . © © CO 

<?> EJ5: co co 



CO 



© © 
©©© !2 cM 

• 00 F-H 

© CO CO © t-- 
CO •«* H tf) 
CO 



© r^> ^ CO 
CM -^ CM 



-f © 

• -M 

CM CO 
— CM 



CO 

co 



© 

00 



© ■* 



© 
© 

CM 



co 

00 F- 

© CM 



CO ■* 

10 6 



.„-» 



u 



© — ^s 

CM CO CO © 
© -# — 1 lO 



■<*©» 

© 10 © 

10*22 

CM — ^ 






„ 00 _ 
00 CM » ■ 
© "?" © fH 

t^ CM '-i • 
hCN « 



r^ — « 



© 

CM 



© 
© 



CO 

© 



pq 

< 



CO'lJ 

00 L 



© © 
© © CM OJ F-. 

• " © © 
© CM fH 10 ■<*< 
© •* — 1 CM 
CO 



C*2 o *l 

-J" *- ' i,i *-» It- cm >o 

— -1 






CO 

CO 



© 



© 10 



CO 
CM 

CM 



■* © 

00 © 

^ © 



7. 

o 

— 

■< 

2 

- 



00 

u 
o 

c 

CO 



COrj-J 

9 ® ' 

O ffl 



=rc^>>8 
a 



03 
-P 

o 

c 



© 



s 

c 

■ 

-c""3 

■** T3 
S m 

f^m 



2 




Q 

■* d a 

■« i '- 

S " 

s "H-f2 

fiQS 



fW 



IO 

CO 



' I eux 
— ? ^"a 



_ J3 

03 I 



'■3 

a> 

43 
43 

a> 

FT* 

o 

^^ 

ft 

03 = 

E 

o M 

*%* 



CD 
-►3 

o 

(3 



;- c 









CO 
X 

o 

s 

00 

.'a 
* • s 

1.2 hh a 

o fS 



o 

CO 

o 

3 

£ co 



43 

a 

CD 

a 

s> 
o 
1 

I— i 

a" 

fH ' 

ft 
a 



44 



MANUAL OF MARINE ENGINEERING. 



eoH 



co" 






CO 



pH ITS 



CI 



~> "0"$ t^PH 

cn • o • 

^H ^ I— ( 



CN 

CO 
CO 



o *r 

Jo "O 



©bb~°-^» 

COt^C0©<MCo!_(CN 
to CO 1-- « 



„ O — lO 

co » **h • « 

r-» co _ f to co 



05 

CO 



■■M ™ 



00 

o 
W 

00 



o 

E-i 



10 



Eh 

-4 

CO 

a. 
n 

CO 

S3 
< 

K 

H 

02 

fa 

o 

CO 

►J 
Htj 

-—I 

03 

H 

fa 
o 

CO 

ij 

D 

co 

H 



1—1 



CO 1-1 

» I' 



to^ 



co J, 



« l" 
HpS, 



coO 

CO 
COr 



op!, 



© -h o O t^o - 
OOOCO — I^-OCD <N 
" ■ • CO CO © CO o CO CO 
OCOCOCOi-hcO-h^j • 



to 



CN 



CO 



hShQ 

9§t9 
co -r 1 ® ■* 



CN 

CO 



o 






/_ ,i, ,i- CM QO »o co " -h CM 
CO >« CN CO 00 i_ iO • • 



CN 



CO -+i 
CN © 



CO CO CN -H« 

1— 1 CO (M 



ao 

CN 



to °° 
© *- 

co 



OO © <-> ^ 

OOfflWlOcO • O CO CN 

cS^J-TS^efSS! * <° 

CO ©1 ^ 



00 J5 00 tp 



© 

CO 
CN 



to 50 

(M 9 



822 



o o 

• s, x. co o ■* , 

CM CN CO 05 tf 

CN 1— 1 



o 



CO 

■a 



Ow-om 

>p g T*< lO 

b * b •* 

— . pH CN 



CO 
CN 



to _ 



o^§ocoS?2 « 
10 ■* « 00 w S f • 



«o, 



s 9 

*• 2*PM 



. -* CN 



O T* 



O© 
Id© i« ■* <-q 



CO 



co-< 

M* f 



CO'fc 
CO I 



co ■• 



■Pi 
co" 

«i [ 



«S f 



z, 
o 

EH 

V. 

a 

M 

cn 
W 
P 



a>, 



oS^cocoSPg -* 
o co co o^ co *e<i o t^* 



•,-> O ^ O CO 

I-- iO a CO >* cq 

* i-H ■<* 1— I 



co jn 

Hi o 



•a » >-<■ §1 »o S ;* • 

CO fh p-* 



Jg COS t^O "5 
co to S ■* •* c^ 



h(Mh 



00 2 

to a 



ogSo.oi5?2 3 co 

O "^r CO CN ^ ~cci to to 

O CO h c* ■* S "3 • • 

CO f* f- 



■<*< „ CO to 

10 co eo •>* 
ihCNh 



OS 
CN 



CO ^ 



o§§ocog«« co 

O00h m ^ .._, o tO 

CDCOHN M"^ • 
CN ^^ ^ 



23 p gcp 00 
b fi b tj< 



o 
to 



«0 
CN 



O CO 



Ot-HCOOiCO «tO° CO 
COCOph^h CNoO 1 ^* 

CN pH W pH 



J2 o f5 to ■># co 

to W COCO U5 pq 



- 1 ?J 

o « 



' to 



43 *3 



to CO 

E 

3 



CO 

O 

5 



t-. 



3 









•r— 


(li 


c 

CP 


S 
I- 


(- 


X 







n 


d 


-S 


.0 




1 


C 





^ 


U 

pq 



, pa 

CD 
09 

■ e 
a 



■a s 

(0 cS 

' c 



rco 



~9 _ 

C » 0«h CO ^ •" 

.2*3 ja °-~^> a 1*3 

slffl>.|.i E 



bc_5« jg 
OPS 



- a,? 



o 

CO 

m 



i ' 
j< - 

■ 43 

pfl O 

co a 

^ ? pH 

o • o 

Ph4h- 

^H S cj 

■a &£™ 

CO 



gti^^^^ 2x0- 

J-giBfinf «g la |p3wfi.2H 



o !pJco.- 





■ 


M " 


no 


a 




CO 


X 


a 





CO 




m 


03 




■ Cm 


T3 


Ph.2 


B 


w-f 


Tl 


\A T 


cq 


PM 


s 


W 


1— 1 


p-i 



RESULTS OF TRIALS OF STEAMSHIPS. 



46 



H 
O 

W 

CM 

c 

H 
00 

H 
«! 

03 
Ph 
I— t 

B 
m 

% 
< 
W 
H 
02 

O 
03 

<! 

os 
H 

En 

o 

03 

-H 

i-3 

D 

03 

Ed 

P3 



X! 





O -« O «M 00 J ly 




© 00 








OS ^ 


OOfflrtNOW* 


IS 


© oo © r- © 


r^ 


CM 


C>1 


Si 


.?3obci s i v e ij;o 

0C©CMi3-iCM^'t2 


CM 


^r cm-xco 


© 

CM 


IS 

© 


© 


© CM, I"- ~* 




CM 








ndrf 


o o © o cm 9 © 
o o 9 ;o 5 j « ■* 




© © t- 








© 


SO CM O — •>* 


© 


© 


00 


« f 


© 00 © .'",,,- o 


I- 

© 


S ©^lis.s 


CO 
CM 


© 


© 


H r \ 


t— 13 CM © — • IS 00 


. CM © •># 


l-H 


CJ 


■** co — 




■"• 








W 


© © © © is 2 © 




© © CM 








°"ai 


© © © © is -Ji 00 is 


© 


-< © ■* © IS 
■? Cl © •* 


00 


IS 


H* 


^: _" 


OOO-^.'^o 


3 


00 
CM 


»— 1 

© 


t^ 


- * ^5 


»SN3«iO IS 


• 


^H 


~ a 


so cm — 




■" ' 








.Ph 


© © is is co ? cm 




13 CM 








ccaj 


© IS © SO ** 00 ■<# -* 


© 


© 00 t-- © t- 


CM 


-H 


© 






© 

13 


JS ©*i-.3 

.- C-J t- IS 




IS 


CM 

SO 


a 


CO -H -H 












jM 


ONoeov « 




"# © t^ 








99 


©©©©— it~«M 


© 


© SO CM CM CM 


© 


© 


I— 




© ob — p "t "* ~L ~ o 
© so i—i cm so i— 


© 
i-S 


g 6 x .^ is 

T CNJ IS ■* 


CO 
CM 


© 


CM 


Ph 


CM r* r* 












rfW 


O O D h M V C 




© © 








©©ISIS— iiSOOim 


© 


CO SO 13 © © 


t-- 


© 


oc 


a 


r- is — < — < © •* 


CO 
i-S 


- f— SO SO 


© 

CM 


© 
13 


1^- 

CM 


o 


CM •—!** 












• ^ 


is ^ 

© © © CM 00 . © 




© © t^-* 








os ^" 


IS © IS 00 © CM CM -Jt 


r- 


•«# r- © © oo 


© 


— 


IS 


S£ 


^^ >J ^^ *. #. ^— O 

-IM-- CM CM 


© 


? 2 5"^^ 


CM 


© 


CM 
CS1 




es ■— i ■-! 












~*fc 


oooonI'n 




t-~ — t 13 








CO 


© © LS © © t- © — i 


SO 


-# IS -* CM © 


© 


CO 


00 


DO f 


<n a> ss ■*. ■* ■*. z; o 


© 


9 ob H © o 


c\ 


© 

13 


-H 


— — 


© SO — CM ■* © 




T — L3 SO 


CM 


^ Q 


SO i— i i— i 














c _ 












. — 


© 00 © IS SO ? 00 




© X CM 00 








X 


© IS t» © CM t)< © — 


r- 


SO CM © CO — i 


CM 


CO 


© 


QQ f 


©•* » e l'* c l2o 


s 


2 CO^-HIS 


CO 
CM 


00 


© 


— — 


© SO i— i CM -^ IS 


. 


. — 't s-o 


CI 


" Q 


so — i — 














so 












-P 


3Ot»530,i;. 

© © r~ so © r- :=; *s 
©cox™~< ~ 2°, 

© SO —i © 5° 




— 1 ■* 








GO 


so 


IS © so ■* © 


© 


© 


CM 


— o 


© 

13 


§ ob*iso4< 

. — i CO CO 


CM 


IS 

»s 


© 


Oh 


CM - 1 










CM 


















'— **-* &*- l-l —*•-*-« 














5 cr „ 




2 












. 


• .5 • • • 


• 


• 


• 




T3 © 














o c 














s — - 


























• • • -i-ssl 


• 


" ■ .5 


• 


• 


. 




« -^ O O 




' ^ ■/ 








| 


"3 *** 




It 








•< 
ss 


■3 a © - C 
S a +j a oo 

§ fi g *f § '« 

© © o X © * 

ft " S a S 




eo 


© ^i 
"a fe* S rt 

1 £! 2 

o . c c 

© P^^J -^ 

« i - 4 

a • o — s 

C -P . -3 

X -3 © u 

© &i 

MS .© -! S 




• 


a 
© 

E 


H 

a 


Iqq 

i 


X 

43 

a 
© 

© 
© 

* 


n 

X PL,' 


n 

S. 
X 


© 

09 




© 
© 

ft 

m 

+ 

ft 




§2 g.s2 .a 


© 

so 

6 


Q 


c © -a bd — 
©. ft a ^ 

P rd nh — 


3 


•c 

5 




w 



46 



MANUAL OF MARINE ENGINEERING. 



So 


cm 


p 
-* 


•* 


O 

o 
o 


CO 


o 
o 

CO 




eo 

00 
CO 


o 
o 


© 

© 

CO 


t^ 


© 


CO 

CO 


00 


1*2! 


IO 


© 


CO 


CO 


eo 


o 








•># 


OS 


•* 


CM 


00 . 


rf« 


CO 






IO 




i-H 


© 


o 


CM 


U5 


•«*< 






CO 


CO'q 
00 h) 


© 

eb 


O 


© 

OS 


' © 
CN 

CD 


CO 
i© 


O 
i— 1 


o 


»-H 


O 
ITS 


© 
© 
© 


© 




© 


© 


eo 

CO 


CO 


CM 


CO 
CM 


-H 


<M 

CO 


o 


o 


eo 

CM 


CD 

eo 


o 


■>* 


CM 


■<4I 


oo'O 

00-1 


© 

do 


o 

CM 


OS 


o 
o 


s 

OS 


© 


CM 

F- 1 


eo 

»-H 


O 


© 
© 
© 


© 

OS 


CM 

© 


© 


eb 


HM 


so 


t> 


CM 


cc 

CM 


i-H 


i© 

CO 


o 


© 


CO 
CM 


© 

CO 


U5 


-*i 


CM 


-* 


ir. ;. 


© 
© 


CM 

IO 


p 

OS 


© 

o 

I© 


© 
CM 
I— 


o 

CO 


•<* 

o 


OS 
(M 


OS 

© 


SO 

eo 
© 


00 


CO 

© 


00 


CO 
00 


H« 


o 

CO 


CO 


<N 


I-H 
CM 


i-H 




o 


O 


CM 

CM 


© 
CM 


IO 


«o 


CM 


CO 




p 

o 


p 

i-H 


© 

CO 


O 

© 
— 


CM 

CO 


o 
o 


CO 


o 

i-H 
CO 


i-H 


© 
© 
CM 


© 


© 

IO 


CM 

© 


CO 


g 


t~ 


CM 


-* 

-h 


i— t 


o 


o 


© 


-* 

CM 


i-H 

co 


t- 


IO 


CM 


IO 


d J 


o 


o 


>© 


O 
O 


CO 
CM 


© 
o 
t- 


to 

CO 




© 
OO 


© 
© 


© 


i-H 

CO 


CM 


© 




3 


CO 


"# 


CO 


-*■ 


CO 


>o 




i-H 


1^- 


© 


CO 


CM 


OS 


i-H 


I-H 
CO 


o 


© 


CO 
CM 


CM 

CM 


t- 


o 


CM 


-* 


2* 


p 
© 


p 
© 


i© 
CM 

-tf 


-*• 

OS 


o 
o 
1© 


o 
eo 


OS 
IO 


CM 


© 
CM 


© 
© 
00 


eo 

© 

p-H 


© 
-* 


<M 

© 


CM 
CM 


CO 

eo 


^ 


i—t 


CM 




CO 


© 


o 


CM 


CD 

i-H 


© 


CM 


X 




p 
© 


ob 




© 

OS 


o 
I-- 


O 

r- 

CM 


OS 

OS 


eo 

OS 


CO 


CO 

I"- 

i-H 


CD 

>o 

© 


© 


IO 

© 


CO 
CO 


CO 


eo 


i-H 


CM 




CO 


o 


o 


CM 


i— i 


© 


-H 


00 


00*. • 


O 


i© 

CM 

CO 


CM 


© 
CC 


p 


o 

CO 


W5 
OS 
U5 


© 

>-H 

CO 


oo 

00 


© 


eb 
© 


00 

© 


© 
© 


<* 

■* 


H* 




eo 


1—1 


CM 


-H 




o 


© 


in 

CM 


© 


>o 


CM 


00 


«5 . 


o 


t- 


co 


CO 


© 


OS 

eo 


eo 

00 


CO 


CM 


© 
© 


e© 

CO 

i— i 


© 

00 


© 


© 


s* 


© 


-H 


■© 


© 


o 


CM 


■rj< 


o 


p-H 


eo 


i-H 


CO 


I-H 

94 


CM 




CO 


OS 


-* 


© 


© 


CO 


© 


■* 


CM 


*-H 




p 

O 


•O 
© 




© 

CM 


© 


eo 

CM 

eo 


00 


CO 
CO 

MS 


© 


CM 

eo 

eo 


© 
© 

eo 

-H 


© 

00 


© 
CO 


-H 


3 W 


i-H 
CN 


CM 




co 


OS 


■>* 


© 


© 


© 
eo 


O 


-* 


CM 


~ 


05 . 


O 


«o 


CO 


CO 


© 


© 
OS 


00 

OS 


© 
o 


cc 


© 

00 


CO 




r- 


<N 


S 55 


Q 


as 


■© 


CO 


t~ 


00 


■<+l 


CD 


t- 


io 


eo 


eo 


o 


.oo 

w 


O 
CM 


|-H 




CM 


f- 


eo 


© 


o 


CM 


eo 


© 


•* 


CM 


f-H 












, — ■ — ,' 












, — •— * 










»4H 

• 


• 


MH 

• 


of 

a 
o 

4* 


43 

<u a 1 


00 


• 




m 

43 

o 




s 

43 


0? 

43 

o 

a 




• 
• 
























© 
















73 












o 


p-H 


$4 




*5 


of 


• 


a 


. 


a 

E 
f) 


• 


* 


. 


• 


i~" 


■8 • 


o 

43 


w 


. 


© 


-3 
"3 




S 




a 
a 






D 




o 


o 

*4H 

OS 


*t2 

3 


t— 1 

•1- 


•in 


z 
2 

M 
Q 


c 
s 

& 

o 

f 


0> 

a 

o 
E 
4» 
X 
4> 

J3 

+» 

eS 


s 

43 
el 
& 

•— 
O 

4» 

bC 
3 


■i 

a 
® 

3 

a> 
o 

Jh 


a 
_o 
'-3 

■ 
3 


• 

a 

n 
V 

43 


• 

Irf 

CO? 

+ 


"5 

a? 

o 
o 

o 
"■i» 

a 

a 

.23 

P>H 


• 

o 
a 
6 

a 

CO 


i 
m 
E 
o 

-3 

T) 
AS 
43 

ES 

o 


It . 
d 

in 

h ' 

CO 

Oh 


-a 


X 

a 

o 

"Bh 


a 

S3 

•i- 

Ch 




a 
1 




c8 


.23 

ft 


'U 

• ** 

S 




o 


S 
Ph 

cc 


T3 
13 

l-H 




» 


.23 


i-i 



D 

Ph 
P 

Q 

H«J 

H 
O 

M 

— H 

CO 

Ch 

.5 

a 

w 

H 

« 
o 

&H 

o 

co 

►J 

rH 

fa 
o 

co 

s 

O 
so 

fa 



PQ 

< 



RESULTS OF TRIALS OF TURBINE SCREW STrAMSHIPS. 



47 



TB.3.8. 

LSTA. 1 

1 


1 

g 


00 


tN 

eo 


o 
■** 

co" 
eo 


© 
© 

of 


© 
o 
eo 

of 

00 


— 

© 


© 
© 


© 

CN 


© 
© 
CO- 

© 


ob 


© 
>* 


00 
CM 


CO 
LO 


■* 




o 
© 


CO 


p 
© 


o 
o 

CM 


CO 

© 

CO 


© 
eo 


CO 

© 


© 

CO 

© 


© 

CO 


© 
© 

CO 


© 
© 


OO 

CO 


© 
00 


00 

© 


»* 


eo 


r- 


CN 


t^ 




t-» 






r- 


t«. 


~m 


•* 


CM 




B3 M 


io 






-— 




»# 


© 


© 


CN 


■* 










CO 

to 

CD 


LO 

05 


p 
CO 


o 

5 

CO 
CM 


X 

CJ5 
CO 

CN 


© 
a 
©_ 

© 


LO 

© 
© 

© 


LO 

© 


eo 
do 

CN 


00 

eo 
t> 

oT 
oc 


eo 


© 

CO 


© 
© 
CN 


p 

CO 

eo 


-t 


,_; eft 

S3 


© 
© 


© 


CN 


© 

o 

CO 




© 

LO 

CN 


CO 

© 


LO 
CN 

© 


© 
CO 




>* 


CM 


CO 

© 


CN 

© 
CO 


■«* 


B 3 


CO 


■>* 


i—i 


Tj« 


© 


CN 
CN 


© 


© 


CO 
CN 


LO 
CM 




© 


CN 




«!4 


© 
uo 


uo 


to 

CO 


o 

CO 
CO 


CO 

t— 


© 
© 


LO 

CO 
LO 


© 
© 


oo 


© 
© 
© 


CJ5 


CO 

© 


r- 

© 


o 


^ 


K' s 


eo 


■"# 




ct 


Tf 


00 


6 


o 


CN 


00 

1 — 1 


© 


T»< 


«N 


CO 






o 

•O 


CM 


lO 

6 


© 
© 

CO 


© 


© 
© 
© 


if5 


© 

© 

CO 


© 
CO 


© 
© 
© 


© 

LO 


LO 
CO 


© 
CO 


CO 

© 


«* 


B« 


eo 


eo 


— 


— 


eo 


CN 


© 


© 


CO 


© 
eo 


CN 


o 


CN 


CM 




2 m 


© 


r» 


CO 


© 


i-- 


© 
CO 


CO 


«N 


© 


© 

LO 


© 


l^- 




LO 




«aj 


o 


lo 


CO 


CO 


00 


>o 


lO 


© 


eo 




CO 




LO 


lO 


CO 


K o 


uo 

CM 


CN 




CO 


— 




© 


© 


CO 


— 


<-H 


•>* 


CM 






CCC-i 
i_'oq 

*:S 


© 
© 

CO 
CO 


© 

© 


LO 

— < 


© 

8 

CO 


00 

© 

lO 


© 

© 
© 

CO 


© 
© 

© 


© 


CO 

© 

CO 
CN 


© 
© 
CN 


ob 

CO 


© 


CM 

eb 


ep 

ob 
© 


eo 


""-si 


o 


p 
© 


eo 


CO 

LO 

CO 


© 

C5 


© 

CO 
CO 


© 

© 

© 


© 

CO 
LO 


CN 

— 


8 


© 
ob 


LO 


CN 

© 


p 


eo 


g rt 


CO 


•* 


*H 


CO 


>o 


CO 


© 


© 


CN 




r- 


LO 




© 




CO . 


© 
© 


p 
© 


6 


© 

CO 

© 


LO 


© 
© 

CN 


CN 

© 
© 


LO 
CN 

© 


CO 


© 
a 

© 


© 


© 


© 


© 


eo 


3 ** 


eo 


■*1< 




OI 


co 


CN 


© 


© 


C) 


00 


© 


© 


CN 


■* 






© 

6 


© 


© 


© 
© 


© 

© 


© 


CO 

LO 


© 

© 


LO 


CM 

© 


eo 


© 
CO 


© 

LO 


LO 


CM 


E^ 


^^ 


CM 




— 








© 


CO 


CN 




t— * 


~—i 




CO 






^* 




00 


© 


© 


CN 


^-1 












CO 


© 

© 


© 
© 


O 
CO 


g 


© 


© 
© 
•<* 


co 

CO 


eo 


00 


© 
© 




eo 


eo 
© 




CO 


lo 


eo 




r- 


i-h 








© 






© 


^^ 






H 


<N 










© 


© 


© 


CN 


CO 


































__ 












r 


r 


r 


a> 


■ 


« 


• 


• 






-o • 




• 


• 


• 




<4-t 




43 


a 
o 


43 

EM 


43 






4-» 
O 




© 

43 


X 

43 










• 


• 


• 


43 


CO 


D 


• 


• 


a 

.1* 


• 


© 

is • 

43 


O 

a 

.3i 

© 


• 


• 


• 


. 




a 


. 


. 


. 


• 


• 


• 


• 


. 


© 
© 


O 

43 


bT 


• 


• 


as 
o 

M 


tn 

— 




s 




© 






. 






© 

B 

cS 

a 


© 
o 




<*> 




^ 


3 




• 


t* 


• 


• 


^ 


• 




a 4 " 


a 




43 




n 
3 


-3 
a 

© 
a, 
t. 
9 
&, 

.c 
a 


a 

© 


b 

CO 

C3 




© 

a 
s 






.S 




fc-t 

© 
o 


© 
© 


13 
© 

B 


■1- 
02 


E 


00 

© 
fi 


o 


B 

43 

H 




43 

e 
© 


r" 


a 

■ — 


© 
o 
© 


n 

c 


- 

© 


hi 


J 


X 


gg 


o 

CO 


_ . 


© 

-»3 

cS 
C9 


o 

43 

a 

c3 
B 

Q 


a 
© 
© 

X 

5 


o 

43 
© 
© 
35 

T3 


© 

43 
43 

© 


+ 


# © 

43 

ee 

s 

.2 

04 


» 

a 

© 
© 

a, 


03 

E 
O 

A 

43 

02 


© 

• a. 

H-i -^ 


s 


43 

s 

_© 

"E. 

s 


•1- 

a 

02 


O 

h 

© 

a 

9 

55 



48 



MANUAL OF MARINE ENGINEERING. 



■r. 

ft, 

B 
oo 

S 
< 

« 
H 
EC 

►5 
K 
H 
K 
S* 

►4 
O 

o 
<: 
Pm 

&. 
o 

CO 

- 

< 

I— 

H 

fc 
o 



►J 
& 
CO. 

K 

P3 



X! 



a?^ 


* 5 


O 




CO 


O 
co 




03 


X 
OS 







OS 




CM 


CO 

O 

CO 


CO 
CO 


eo 


IS 




Pi J 


<=a 


10 

•* 





~ 




CO 


X 


6 


6 


X 


l> 


CM 


■* 


CM 


CM 


cCy 


« E 


O 


CM 



L ? 

CM 


g 

o« 


8 


O 
O 


© 

X 
CO 





OS 


eo 


f4 


S 


01 


Ol 


Pn=n 


«d 


CM 


O 


" 


L0> 


CO 












CS 


eo 





01 


CO 


.55 


£s 


O 


CM 


p 
5^ 







CO 

CO 


O 
CS 


CO 


eo 

CM 

CO 


r» 


<M 


p 


CO 
OS 


eo 

X 


X 


2- "?. 


«S 


CO 

eo 


■** 




Ol 


•<* 


co 





6 


CM 


F-H 


t^ 


CO 




10 


«2 


ti 2* 


O 


O 


MS 


00 

CO 


<N 


CM 

X 


<* 


«5 


CO 


8 

to 


CO 


CO 


eo 


CO 


.2 S 


O 


6 


00 


X 


S>1 





CO 


CM 


l> 


T 1 


CM 


CS 


-^ 


OS 


eo 
eo 


•** 




- 


eo 


eo 


6 


6 


CM 





>o 





CM 


-* 


eog 


ti& 





© 


LO 


01 


10 


CO 


■* 





m 


8 


CS 


»o 


X 





« s 


O 


CM 


>~ 





t> 




CO 




OS 


CO 


LO 




(-H 




*« 


5 ° 


CO 


CO 









CS 








~ * 




CO 


■<* 


CM 


CO 




■e » 


cp 


O 


CO 


LO 


t^ 


8 

X 

0" 


CS 


8 

p 

6 


LO 





l> 





r- 


CO 




eo 


eo 

CO 







X 


p 



OS 





eo 


«o 


eo 

CM 


eo 


£| 


• o5 





1^ 











Ol 





X 


1^ 




CO 


>^3 


^ 


eo 


t- 


CM 


CO 


do 


•>* 


OS 


t^ 





CO 


f-h 





OS 




O 


OS 


«H 


eo 


CO 




- 1 


6NJ 


1— < 








CM 


l> 


ic 


«o 


CM 


"<J< 






O 
i-O 
01 


00 

CO 


© 
p 


LO 

1— 1 
00_ 


s 

eo 


1 


CO 

p 


CM 

p 


CM 


co 


X 

6 

CO 


CO 


CM 


X 

eo 


v5Z 


«3 


O 


cjs 

CM 





p 


p 


CO 
CS- 

co' 


-*- 

OS 

6 


CO 

p 
O 


CM 


O 

CO 

oi' 


p 

OS 

CO 



p 


8 

CM 


eo 










i-0 
























00 Si 


— 


O 
oi 


,1, 


CM 

CO 





01 

01 


(M 

CS 
X 


'CH 


CO 


X 


CM 

X 


CM 

oj 


OS 


O 





fc H 


>c 


O 


CM 




-t 




■* 


O 


O 


4" 


X 




10 


CM 


r—t 


•/z 


■*s zz 


p 


00 




00 





05 





X 


CO 





X 


t^ 


_*. 


CM 


CO 




33 j! 


CM 


ob 


CO 


X 

01 


OS 




co' 







6 


>o 


CS 


CM 
CM 


CO 


CS 


X 

l-H 


»5 


sir* 

09 - 




CM 


^ 

c^ 




CO 
CM 






Ol 


CO 
CO 

p 






i> 




p 

CM 


CO 


X 


CO 
X 






CM 




•— v— 


„ 


* 


„ 


^ 


m 


fc 










-c3 • 














43 


43 


_, 


43 


43 










.2 


— 




















^-1 





































O 




43 


43 


















cr* 


a* 






C 







c 






" 






• 


• 




. CO 


03 


• 


• 


^4 




> • 

43 
OS 


-5 



1— t 


• 


• 


















• 


• 


• 


• 


t^- 





Pu 


• 


S5 


CO 

W 
S5 


CD 

s- 

s 


• 


c 

03 

a 


• 




CO 

-— 



• 


• 


4^ 

d 

us 


• 


09 

O 

P4 




cS 

3 

CT 1 * 
CO 


43 

OS 
CS 

"5 


n 

1— « 

•i- 


43' 

£ 




en 


!5 


■~cj 

e 

OS 
Ph 
fc» 
O 
Ph 


O 


9 

cS 




s 
s 




\rn 







i 

CO 






OS 


CO 


a 


W 


a 



it 


c 
01 


c 


c 


^f 


111 
c 



C 
e3 





r— ( 


• 


X 


fl 








O 
43 







43 


CO 


•1- 





4) 




OS 




43 


•1- 






J5 
C 

3 




JZ 





rj 


TS 


i-3 


cj 




43 


n 




s 








s 

eg 


to 

C3 
d 
B 

P 


CB 

5 


CO 

s 


3 

43 

as 
5: 


6 


s 

CO 


T3 

i 

— 

CO 


O 

e 
1— 1 








CO 

5 


CL, 

1— i 



RELATION OF POWERS AND DISPLACEMENTS. 



49 



TABLE XIV. — Relation op Powers and Displacements.* 





No. L 


No. 2. 


No. 3. 


No. 4. 


- t 

No. 5. 


Length in feet, 


280 


300 


360 


435 


600 


Breadth in feet, 


35 


43 


60 


69 


71 


Mean draught in feet, 


13 


16} 


23| 


24* 


26£ 


Displacement in tons, 


1,800 


3,400 


7,400 


11,000 


14,200 


LHP. for 20 knots, 


6,000 


9,000 


11,000 


14,000 


15,500 


I. H.P. per ton of displace- 












ment, • 


3 3 


2-65* 


1-48 


1-27 


109 



The following horse-powers were required to drive cruisers Nos. 4 and 
5 in the above table at the speeds named : — 





No. 4. 


No. 5. 


10 knots, 


1,500 I. H.P. 


1,800 LHP. 


12 „ 


2,500 ,, 


3,100 „ 


14 „ 


4,000 „ 


5,000 „ 


16 „ 


6,000 ,, 


7,500 ,, 


18 „ 


9,000 „ 


11,000 „ 


20 „ 


14,000 „ 


15,500 „ 


22 „ 


23,000 ,, 


23,000 „ 



The frictional resistance of clean painted surfaces varies about as the 
1*83 power of the speed, but resistance due to wave making may vary very 
widely, since it is dependent on form. The total resistance of " Destroyers" 
has been found to vary as follows : * — 

nearly as speed " 
„ speed 8 
„ speed 3 ' 3 
„ speed 27 
,, speed 2 
practically as speed 183 

and the resistances other than frictional vary as follows : — 

Up to 1 1 knots, - - - as speed 2 

At 12} to 13 knots, ,, speed 3 

,, 14^ knots, * - „ speed 4 

„ 18 „ - - - ,, speed (more thau 5th P° wer > 

„ 24 „ - - - „ speed 2 

and at higher speeds as still lower powers of the speeds. 

The relation of the frictional to the total resistance is * : — 



Up 


to 11 knots, 


At 


16 




i) 


18-20 




» 


22 


ft) 


»> 
>> 


25 
25-30 


9 9 





" Destroyer." 


Cruiser. 


At 12 knots, 
,. 16 ,, 

20 
»> 23 ,, 
„ 30 „ 


80 per cent. 
70 „ 
nearly 50 ,, 

45 „ 


90 per cent. 
85 
nearly 80 , , 
over 70 ,, 



- If the coefficient of friction be doubled (as it might easily be with a foul 
bottom), the maximum speed of the " Destroyer " would fall fully 5 knots, 
and that of the cruiser would be reduced to 19 knots. See Tables vii. to xiv. 

* Sir We White, British Association Address, 1899. 



50 



MANUAL OF MARINE ENGINEERING. 



Progressive Trials should be made with all ships when at the measured 
mile, and it should be remembered that for practical purposes it is more 
important to know the power, revolutions, slip of propeller, etc., at speeds 

i 







tarn 


I3U43 


IMS 


nsos 


i %!09 


*09 


x 


ov 


Stat 


tfOl 


«0l 





1 

5 






>. 






Sd3i13dOic JO 


dllS 


«3H 




%ISl 


XjOt 


%IS 

1 





s 






< 


**© 












/. 










3 

Ik 
















• 




^}y> 








/ 






\> 




f 








10 












■ 


\ 






| 










i 






















H 


\ 


V 


/ 






~v 










* 


















\ 












1 


>1 




































\ * 




if 

s 








M 




























1 






























J 










J 




1 








(M 


















n| 










°l 




v* 


























L 














\l 












i 

i 












■ V 
o 

z 

Id 






































IT 
tap 
u 










~\ ^ 




1 


:> 






o 


















u 

> 

5 








































3 
0. 
O 










\ 


v 










u> 


>. 
















\ 










s 

hi 

_j 
■j 
wj _ 


\ 


l\ 


1 
i 

i 














. 






\ 










HI 
0.1 






» 






«1 




















i 








el 

a J 




\ 


"c>~"^ 


































Z/ 




\ 


1 






K 




























5| 




V 


II 






































\ 


1 






g 




i 




























|^9 












































m 


umoc 


3SH0 


1 ooo 


re 


ooo 


71 ] OOO 

.1 


39 


OOO 


OS 


ooo 


0» 


OOO'OS 


000 

1 


at 


ooo 


M 



















S»31 


1949 


U" u 


no* 4 


Sh 


pij-m 


9A3S. 


NV3H 


00 


z _ 


Si 


i ' 


■o 





o 

g 

o 

-a 

■3 



> 



o 

CM 



75 

%3 



3 
- 1 



05 



so 



less than the maximum than those at the utmost speed ; it is especially 
important that the ship shall be tried at as low a speed as possible consistent 



EFFECT OF DEPTH OF WATER. 



61 



MAPLIN 7 4- rATHOMS — — 

SKELMORLIZ 40 •• 




Fig. 13. — Effect of Depth of Water on Performance. 

Speed Trials of H.M. Torpedo-boat Destroyer " Cossack " at Maplin and Skelmorlie. 
270'x 26'x 93' draught Displacement, »36 tons. 14,000 S.H.P. 



52 



MANUAL OF MARINE ENGINEERING. 



with obtaining accurate observations; and that between it and full speed 
there should be one or more trials at intermediate speeds. There should 
be three consecutive runs made on the mile for each rate of speed, with 
the steam, vacuum, and revolutions kept as steady as possible during the 
whole time occupied in doing them. The mean speed should be calculated 
in the usual way — that is, if the speeds observed are at the rate of x, y, and 
z knots per hour — 

m , x -f 2y -f- z 
irue mean speed = f 



The revolutions should be taken from a 


counter, 


ES 


the number i 


DQr 


mile, 


4500 










































/ 










































C 


> 






4000 










































*-8| 


4-000 
3500 






































/ 


f I //J 










































/ 






3500 


























■*" 












/ 


/ 
o 








































/ 






"3 






3000 






"EC7 






























/ 


' 4 


» 








2500 


EF 


IVE. 
























// 
















£500 


H0R. 


>EFC 


WER 






















.■■'/, 




1 

















2000 




















A 


5FE 


ET 


-^// 




i 
















2000 


















M)F 

A 


£ET 






/ 


■9- 


















I50& 
















10 FE 

B 


ET j 




y 


- , 
1 




3^ 


ClIRVEA 












4500 




trials of 400 ton Destroyer 
on a depth of 40 feet assum- 


















1 




i J 


f 


•V- 




1000^ 














/ 
/ 






V 


'** 






be 62 per cent of the Indica- 
ted Power. 
- —curvtb is deduced from the results 
of the Model Experiments 


1 000 












/ 


/ 


<\i 


i 










500- 










/ 


















ATA DEPTH OF 30FEET 

CURVEC is simTlar to Curve B but 


500 










It 


' V 




















ATA DEPTH OF 


4ir 


EET. 




1 


O 


-f 




1 












' 




, 


















1 



Fig. 14. — Harold Yarrow's Experiment. 

Curves of Comparison between Model Experiments of 400 tons at depths of 30 and 45 feet, and 
Actual Trials with a Destroyer of 400 tons at a depth of water of 40 feet. 

this, when divided by the time taken on it, will be the true rate of revolu- 
tion per minute. The results of horse-power, revolutions, slip of screws, 

])§ x S 3 
and value of C = T „ p as a measure of efficiency for each speed should 

be set up as ordinates, with the speed in knots per hour as abscissce : a 
curve drawn through their upper ends of each set will indicate all that is 
desired to know of the ships' performance. If, however, the ship's resistance 
has been ascertained by model experiments or otherwise, a curve of E.H.P. 
can be inscribed, and the curve of propulsive efficiency will then be obtained 



EFFECT OF DEPTH OF WATER. 



53 



rup 

by inscribing the values of t ' tt *t>* as ordinates for it. 



Fig. 12 is such a 



I.H.P. 

diagram, as it is highly desirable to have for all important ships, showing, as 
it does, clearly the performance of the " Lusitania " on the measured mile. 
Being a turbine-driven ship, the power is practically Brake Horse-Power, 
being that obtained by observing the torque on the propeller shafting. It 
is usual to speak of this as Shaft Horse-Power. 

- The Effect of Depth of Water on the speed of steamships is somewhat 

erratic, as may be seen by carefully examining the curves of performance 

of several ships and models at trials made from time to time for this purpose. 

Sir Philip Watts exhibited those of H.M.S. " Cossack," obtained as the 




<M0TS 



Fig 15. — Curve of Effective Horse-power and Speed with Various Depths pf Water. 

Model Experiments by Harold V arrow, I.N. A. 

results of her trials at the Maplin Sands, where the water is comparatively 
shallow (7*4 fathoms), and those carried out at Skelmorlie, where the watei- 
is deep — viz., 40 fathoms — and the influence of the bottom only felt by 
the largest ships at high speeds. It will be seen that, at what may be called 
the critical speeds of this ship, there were changes of trim corresponding 
to the changes in revolution and torque at about 18 knots; these were 
of a violent nature at about 20 knots, while at 26*5 knots things became 
normal again ; at speeds above this the power required was actually less 
1*1 the shallow water than in the deep {v. fig. 13). 

Figs. 14 and 15 are equally interesting, as being the results of special 



54 



MANUAL OF MARINE ENGINEERING. 




S1V09 03ZIS linj JO U3M0d 3SH0H 3AU33JJ3 



EFFECT OF DEPTH OF WATER. 55 

trials made by Mr. Harold Yarrow for the same purpose of finding the effect 
of depth of water on fast ships. Fig. 14 shows the curves of comparison 
between the model experiments and those made with the actual ship. The 
ship herself acted in much the same way as did the " Cossack " in 7-4 fathoms ; 
her critical point was 18 knots, but became normal at about 23 knots. 

The North German Lloyd Company, of Bremen, had an interesting 
series of experiments made by Herr Popper, and fig. 16 shows the results 
of two sets of them, each being made with the boat and her model ; the 
effects in both cases are even more striking than the former ones, inasmuch 
as the changes are more emphatic and pronounced. 

The full accounts of all these trials are given in the Transactions of the 
Institution of Naval Architects, and may be studied there with advantage. 

Dr. D. W. Taylor's formula for ascertaining the least depth of water in 

which a ship should undergo her speed trials for a satisfactory performance 

10 x d x s 
is as follows : — Minimum depth of water in fathoms = j , where 

d is the draught of water of the ship, I its length between perpendiculars, 
and s the speed in knots. 

Example. — The minimum depth of water for the trials of a cruiser whose 
draught is 26 feet, the length 500 feet, and the speed 25 knots. 

n ., 10 X 26 X 25 1Q . ,. 

Depth = — ^r = 13 fathoms. 

500 



56 MANUAL OF MARINE ENGINEERING. 



CHAPTER III. 

MARINE ENGINES '. THEIR TYPES AND VARIATIONS OF DESIGN. 

The marine engineer, when dealing with design and construction, is faced 
with, and has to solve, problems much more complex and involved than 
those corresponding ones familiar to builders of land engines. Moreover, 
he is hampered by circumstances and limitations quite unknown to the latter. 

The space occupied by, and the weight of, the machinery of the ship 
is limited at all times ; in the case of the cargo ship each ton of weight means 
a ton less cargo on which freight is payable, and in that of the express steamer 
and warship where the power is larger in proportion to the size of the ship, 
both weight and space are of great consequence, and generally quite extremely 
limited. The design, therefore, must be such to allow of inclusion in the 
machinery space allotted, while leaving sufficient room to permit of accessi- 
bility to all parts as required for working and overhauling ; the weight is 
as strictly limited to that share of the displacement provided by the. naval 
architect in his design. In bygone years more than one good ship failed 
to comply with the conditions prescribed for her by her designers from a 
miscalculation of the weight of machinery, or the adding to it by the engine 
builders without regard to the consequences. To the fast paddle steamer it 
was fatal, for, in addition to the extra displacement and wetted skin, there 
was an increased immersion of float, which reduced the efficiency of the 
propeller immensely. The large engines found in central electric light and 
power stations on shore are secured on massive concrete foundations which 
never move, whereas those on board ship, while being strong, are not massive, 
and may and do move about in a way most trying to them and their bed- 
plates. Not only does the engine move forward with the ship in the line 
of the course she pursues, but it is liable to inertia stresses due to the acceler- 
ation and retardation of the velocity of the ship from various causes, and 
when pitching and rolling the angular motion is often considerable; so that 
the inertia stresses thus set up are even more serious. It has not often 
happened that a marine engine was torn away from its bed, but all of them 
are liable to such an accident if ample provision is not made for such a 
contingency. 

In warships very special provision was formerly always made for the 
shocks that would be set up when ramming the enemy ; this, however, is 
a nautical manoeuvre no longer contemplated by naval commanders, as 
it was apt to be more fatal to the rammer than the rammed, so that for this 
and other reasons ramming is no longer spoken of. It remains, however, 
as a contingency common to all ships, for both naval and mercantile ships 
are liable to run on a rock or other massive obstruction, and even to ram 
another ship accidentally. It is not desirable that the displacement of the 
engines shall follow such a catastrophe. 



VARIOUS TYPES AND DESIGNS OF ENGINES. 57 

The marine engineer has also to produce an engine that may be depended 
on to keep running without a stoppage for an indefinitely long time, and 
at a uniform speed, as it is highly desirable that there should be no slowing 
down of the engine except when so desired by those navigating the ship. 
Slowing down or stopping the engines at a critical moment might mean 
the loss of the ship ; in the case of a naval ship, it might mean her capture, 
or even the loss of a battle, which would decide the fate of a kingdom. It 
is, therefore, of the very first and highest importance that the marine engine 
shall be free from every extraneous fitting which might cause temporary 
derangement, and itself should be so carefully designed, manufactured, 
fitted, and cared for as to preclude the possibility of compulsory stoppages. 

Further, in both naval and mercantile ships, the whole of the machinery 
must be practically noiseless when in motion, and the engines free from 
vibration, which would spoil the gunnery of the one and the comfort of the 
passengers in the other ; even the auxiliary machiner)' must comply with 
these conditions of absence of noise and vibration. Finally, since there is 
a limitation to the weight of fuel which can be carried, and it is desirable 
that it shall last over as long a voyage as possible, it is necessary on that 
ground, as well as for the sake of economy in cost, that the consumption 
of it be as low as possible consistent with a satisfactory compliance with the 
conditions already insisted on as essential. 

The marine engineer enjoys one advantage over his brother on shore ; 
he has an unlimited and cheap supply of cold water, whereby he may con- 
dense as much steam as he desires, and work with as high a vacuum as his 
apparatus can produce and maintain. And if it be an advantage, and no 
doubt it is in many ways, his engine varies in velocity as its load varies, 
instead of running at constant velocity whatever be the load. 

Various Types and Designs of Engine, of which in the early days of steam- 
ships there was a very large variety employed, and although some of them 
were developments of those common and successful for land service, the 
greater number were evolved to satisfy the conditions imposed on the engineers 
of the day, many of them showing considerable originality in form as well 
as ability in design ; some of them display features which indicate that their 
originators possessed a technical knowledge, for which they have not been 
always accorded credit. On the other hand, not a few had inherent defects, 
which, while not being apparent in the model state or in engines of small 
power, were very evident in the larger engines, and soon caused their early 
dismissal to the scrap heap ; such defects were generally due to a want of 
technical knowledge, or to the misapplications of the empiric formula? of 
the day, which were the rough and ready pilots of these earlier times. 

The engineer of to-day is not required to waste time and energy in differ- 
entiating the claims of many types before coming to a conclusion as to what 
will suit the circumstances of his particular case, for time and experience 
have decided much of this for him, and so now for a paddle steamer it is 
almost the universal rule to choose the direct-acting compound engine, either 
horizontal or nearly so, and for the screw the inverted form of the same. 
He may, however, have to debate with himself or his advisors whether his 
is a case for turbines pure and simple or for reciprocators ; he may also 
compromise the matter by having reciprocators at the boiler end of the 
installation, with low-pressure turbines at the condenser end, as is the case 



58 



MANUAL OF MARINE ENGINEERING. 




VARIOUS TYPES AND DESIGNS OF ENGINE. 



5$ 




& 

o 



o 
01 
u 



s 

-1-5 

o 
d 

_i 
"He 
g 

>> 

o 
o 



o 
o 

t- 
o 



23 



g 



> 
d 

H 

u 
© 

-3 

o 
X 

33 



00 



€0 MANUAL OF MARINE ENGINEERING. 

now in many large ocean-going ships of high speed {v. fig. 44). When the 
power is not very great a Diesel oil engine may be adopted, and with smaller 
power even a semi-Diesel type will prove attractive. For small craft of low 
or moderate speed the paraffin engine is popular and convenient. Petrol, 
however, should be avoided on shipboard except in very special cases. 

Although there is very little variety in the type of engine now employed 
for paddle-wheel ships and screw ships, it is well to know why particulai 
ones have survived, while others have died out, and in order to appreciate 
the selection, some knowledge of the virtues and vices of those others now 
discarded is desirable as well as instructive. 

Paddle-wheels were the first practical form of propeller employed foi 
steam navigation, and, as has already been stated, the earliest ship had a 
horizontal direct double-acting engine. To-day this form of engine survives 
as the fittest for the purpose of both side wheel and stern wheel paddle-ships 
of all sizes ; in the case of the former it is, however, inclined to the horizontal 
position, and called the diagonal or inclined engine. In this form it was 
patented by Marc Isambard Brunei, the famous engineer, in 1822, who 
claimed as his invention two inclined cylinders, the cranks at right angles, 
the piston-rod ends fitted with roller guides, and the weight of the pistons 
relieved by spring supporters on the extremities of the beam head ; the engines 
were to be governed by means of a stream of water pumped through an 
orifice. The condenser was to be formed of an assemblage of pipes, which 
collectively formed a spacious chamber ; they were to be connected together 
with a set of smaller pipes, and the whole placed in an iron reservoir. This 
must be admitted to be a very ingenious and comprehensive invention, con- 
sidering it included roller guides, balanced pistons, a cataract governor, 
and last, but not least, a surface condenser. The roller guide never took 
on for large engines, and a cataract governor seems almost to have met 
with the same fate, otherwise Brunei's invention may be said to have sur- 
vived as the fittest for the purpose of marine propulsion with paddle-wheels, 
just as the vertical direct-acting inverted first adopted by the Thomsons, 
of Glasgow, has for screw propeller work. 

The inclined direct-acting engine (figs. 19 to 22) is the engine of to-day, 
as being the one complying best with the conditions ruling on board such 
ships as are still propelled by side paddle-wheels for the following reasons : — 

(1) The design is simple and without complications or make-shifts of 
any kind, hence less liable to derangement and breaks down. 

(2) It is as light and cheap in construction as any other, and may be 
arranged so that its framing helps to stiffen a lightly built ship rather than 
to over-strain her. 

(3) Great flexibility of design is possible ; the engine may have as many 
cylinders as are desired within the limits of beam, and may be compound 
or simple without any difficulties or objections. 

(4) The stroke of piston may be long ; in fact, there is practically no 
reasonable limit to it. 

(5) The steam pressure may be as high as that in screw ships without 
any disadvantages. 

(6) The shaft may be at any desired height from the ship's floor, and 
so permit of a small diameter of wheel with a corresponding increase in number 
of revolutions if desired. 



PADDLK-WHEKI.S. 



61 



On the other hand, it has the defects of all horizontal engines, sucli as 
heavy pistons running on the cylinder liners or bodies either unsupported or 
inadequately so, whereby they are worn barrel shape in course of time ; there 
is also the friction due to this which retards the movement. Further, the 
momentum of the pistons and rods in connection therewith causes a pulsa- 
tion fore and aft-ways in the ship, which in light river craft is sometimes 




Fig. 19. — Diagonal Compound Paddle Engines (J. Brown & Co., Clydebank). 

by no means agreeable to the passengers ; this is especially the case in some 
of the older river steamers having only one cylinder, the more modern steamers 
having three cylinders (fig. 22) are practically free from this defect, and in 
those with two cranks at right angles it is not so serious, but it is yet observ- 
able, whereas with vertical cylinders it was absent, although they had a 



€2 



MANUAL OF MARfNE ENGINEERING. 



vibration of theif own, and the steeple engines had the objectionable pulsa- 
tion also. But, as a rule, the paddle steamer in old days was a more com- 
fortable one in which to sleep than were the screws. 

Of the inclined types, there may be one or more cylinders side by side, 
as shown in fig. 19 ; or, when the ship is narrow, and from this or other cause 
& limitation in athwartships space, the design of fig. 20 is a good one, and 





Fig. 20. — Diagonal Compound Paddle Engines. 

works quite satisfactorily. This latter permits of more deck room amid- 
ships, and space available for passengers, etc., and generally gives a clear 
deck fore and aft in even small ships. 

The fore and aft space taken up by the diagonal engine is of course very 
considerable, and formerly in some classes of ships would be of great disad- 
vantage, but in modern paddle-steamers this is of no consequence, and is 



PADDLE- YVHEKLS. 



63 



more than compensated for by the advantages in other ways, not the least 
of which is the ease with which a paddle-wheel of small diameter can be 
adopted, whereby high revolutions are rendered possible. This, together 




with the long stroke of piston which can be permitted with it, allows of a 
very high rate of piston speed, amounting in some large steamers to as much 
as 750 feet per minute with a low-pressure piston 108 inches diameter, the 



64 



MANUAL OF MARINE ENGINEERING. 




BEAM ENGINES. 



65 



.revolutions in this ease being 52 ; this and other examples of these engines 
may be studied on Table xiii. 

On the American rivers the stem-wheel steamers of shallow draught have 
the horizontal engine in general use ; a remarkable example of it is found in 
the tugboat "Sprague," of 2,200 tons displacement ; her two engines have 
cylinders 28 inches and 63 inches diameter, and a stroke of piston of 12 feet 
driving a wheel 40 feet diameter and 40 feet long. At 16 revolutions the 
I.H.P. is about 2,500. 

Beam Engines, by their family likeness to those on shore, show then- 
descent from them ; and the survival of type locally is illustrated in their 




Fig. 23. — American Steamer Beam Engine. 

•case as it is in that of the direct-acting variety. The second practicable 
steamboat constructed was Fulton's p.s. " Claremont," built by him in 1807, 
and used for service on the River Hudson ; she was fitted with an ordinary 
overhead beam engine supplied by Boulton & Watt, from their Soho Foundry 
at Birmingham, England. Probably, as a consequence, this type of over- 
head beam engine soon came into general use in America, and continued to 
be the favourite one till quite modern times, and may be still seen on service 



66 



MANUAL OF MARINE ENGINEERING. 



to-day with cylinders of enormous size with " drop " valves and wooden 
framework, etc. (fig. 23). Indeed, it is to this latter fact that it continued 
30 long to hold its own, for wooden structures and wooden connecting-reds 
properly braced and bolted were permissible with such engines, and were, 
and are, the peculiar product of the people inhabiting a new country where 
wood is more plentiful and more easily manipulated than iron. Never- 
theless, the machinery of a large Fall River steamer constructed on this 
method was a splendid example of engineering genius, displaying as it did, 
the adaptation of what was found to hand to the needs of the problems. 

Side Lever Engines are a form of beam engine, and was the type developed 
in this country from the beam idea. The third practicable steamboat was 
the P.s. " Comet," already alluded to and shown in fig. 4. This ship was 




Fig. 24.— Engine of the "Comet," 1811-12. 



installed by Bell, her owner, with a single-cylinder engine having a beam 
at each side, as shown in fig. 24, connected to the piston-rod cross-head and 
the-crank pin by rods. A modification of this design was adopted many 
years ago by the builders of tug boats, especially by the Tyneside builders, 
and called the Grasshopper engine. Such engines are still in use on tugs,. 
and are exceedingly suitable for them. They are very cheap in construction, 
have a very long stroke of piston for such shallow ships, the racking action 
when in motion is consequently comparatively slight, and is taken by the 
keelsons, the stiff est part of the hull ; when only one cylinder is employed 
there is in practice no " dead " point — that is, the crank can be moved by 
the pressure on the piston from any position in which it may have stopped. 
This latter quality is due to the position of the connecting-rod with respect 




*7 



Fig. 25. — " Side Lever " Engine. 




1 
































— \ 


k.-».-l 


^ ^ 


^ 


.0^ 


&N 


^ 


sa 


^ 


^ 


£9 


^ 




— . — ^_ — 




5n3 ' 
SSI 


^ 


5 

£: 


































-n 



Fig. 26. — Engines of Thames Steamer " Regent," 1816 (Maudslay). 



68 



MANUAL OF MARINE ENGINEERING. 



to the levers when the piston is at the end of its stroke, and aided by the 
slight amount of " play " in the brasses. This class of engine is capable of 
working satisfactorily when in such a state of disrepair that would in any 
other form of engine prove dangerous ; it is also not exacting in the matter 
of attention from the attendants. 

The most important development of the side lever engine was, however, to 




Fig. 27. — Oscillating Engine — Section through Trunnions. 

be found in the sea-going steamers of the early half of last century, where it 
might have been seen in large varieties both of size and design. It was the 
means whereby the Cunard and some other important companies established 
the reputations of their ships by regular passages and freedom from breaks 
down. Fig. 25 is a good example of such an engine, and shows all its salient 
features. It continued to be the most popular type to the end for sea-going 



OSCILLATING ENOINKS. 



69 



ships, and the " Scotia," the last of the Cunard paddle ships, had such 
engines with two cylinders, each 100 inches diameter and 12 feet stroke, 
and developing 4,950 I.H.P. Fig. 26 shows a modified form, inasmuch as 
the lever is of the bell crank type. 

Oscillating Engines were first suggested by Trevithick ; they were used 
originally by Dr. Goldsworthy Gurney about 1827 for driving his steam 
motor cars. In the same year Joseph Maudslay patented the use of such 




Fig. 21a. — Oscillating Engine — Section through Valve-Boxes. 



cylinders, and in 1830 William Church took out a patent for their applica- 
tion to driving paddle-wheel shafts. The firm of Maudslay, Sons & Field, 
however, developed the oscillating marine engine, and until they quarrelled 
with the Admiralty over some repairs to one, had the sole supply of them. 
The patronage of the Admiralty was transferred to John Penn & Son foi 
their further supply, and since then this type ha6 been always associated 



70 MANUAL OF MARINE ENGINEERING. 

with this firm. It was, moreover, brought by them to a very high state of 
efficiency, and employed largely on river and cross-channel services, as well 
as in all kinds of ships in the Navy. 

" The Great Eastern " steamship had four oscillating cylinders, 84 inches 
diameter and 145 feet stroke ; but perhaps the largest and most interesting 
example was that of the p.s. " Ireland," 1,952 tons, built by Messrs. Laird 
for the City of Dublin S.P. Company, for service between Holyhead and 
Kingston, which, with two cylinders 102 inehes diameter and 8'5 feet stroke, 
7,000 I.H.P. was developed, and a speed of 23 knots was attained. 

The oscillating engine has survived to the present time, as they may 
still be found in common use on dockyard tugs, and on river craft throughout 
the world ; and although there have been compound engines of this kind 
with cylinders of considerable size for steam pressure of 60 lbs., the type 
is not so good for the higher pressures or for triple engines as the diagonal. 
Figs. 27 and 27a show them to be simple in design, having only one rod 
from piston to crank-pin, requiring no guides, and generally are free from 
complications, except that the valve-gearing may be considered somewhat 
complex. They are very light ; occupy little space, and permit of a fairly 
long stroke, even in somewhat shallow ships. In the case of the " Great 
Eastern," they were placed diagonally two to each crank-pin, and this design 
has been followed down to modern times for the British Dockyard Tugboats, 
where each wheel can be worked independently by a pair of cylinders. 

It is not necessary to dwell on the other types of paddle engine, as they 
have had their day and disappeared, except to say that the Steeple Engine, 
as shown in fig. 28, was a favourite one with some Clyde engineers, and it 
was the type adopted by Messrs. Laird when refitting the p.s. "Violet" 
with triple-compound engines of 4,070 I.H.P. , in which the L.P. cylinder 
was 108 inches diameter and 6 - 5 feet stroke. The space occupied by this 
form of engine was small, and generally permitted of a fairly long stroke, 
but the thrust of the connecting-rod caused a tilting action on the engine 
framing, which, unless well provided for by bracing, severely strained the 
entablature, etc. 

Vertical Direct-acting Engines were used by some makers, and with the 
old sea-going ships with their good depth of hold, and a radial wheel of large 
diameter, a fairly long stroke could be obtained ; but it was always some- 
what short compared with that of other types of engine. 

Twin-cylinder Engine of Maudslay & Field (fig. 29) is now chiefly inter- 
esting as being the one fitted in the first screw ship of the Navy, H.M.S. 
" Rattler." It was geared to the screw shaft, so that the latter made four 
revolutions to one of the engine, whose cylinders, four in number, were 
40 inches diameter, and the piston stroke 4 feet M 

Screw Engines were at first, as in the case of H.M.S. " Rattler," of the 
same type as used for paddle-wheels (v. fig. 29), with spur and pinion gearing 
connecting them to the screw shaft, and what, perhaps, was of more import- 
ance, the necessary revolutions for the screw were got thereby without sub- 
mitting the engines to such speeds as they were then not fit to bear. Gearing 
was eventually discarded, and direct-driving adopted, both for safety and 
economy. The higher speed of engine permitted of the use of smaller ones, 
which were lighter and cheaper ; and as tooth-gearing involved considerable 
friction as well as frequent repairs, further economies were affected by 



SCREW ENGINES. 



71 



abolishing it. To-day, however, we are witnessing a revival of wheel-gearing, 
to permit of the use of turbines in cargo steamers where a small high revolu- 
tion screw is not permissible, and in ships generally when turbines of high 
revolution and consequent economy may be employed. It is, of course, under 




Fig. 28. — Steeple Engine. 

much more favourable circumstances that this experiment is being carried 
out, for in this case the torque of the engine shaft is absolutely uniform, instead 
of being highly variable ; the pinion is driving instead of being driven ; and, 
moreover, there are two, one on each side of the spur-wheel ; finally, the teeth 



72 



MANUAL OF MARINE ENGINEERING. 



arifflfr. 



of the cog-wheels are to-day double helical machined to exact shape and size, 
instead of being common toothed of rough cast metal or ol wood that, while 
tough, was not particularly hard (v. fig. 45). The efficiency of the helical 
wheel connection is 98" 5 per cent., which is very high, and so far satisfactory, 
as to permit of high-speed reciprocators, as well as turbines for the drivers. 

The Horizontal Direct-acting Engine was the product of Humphrys & 
Tennant latterly, and in an older form of Boulton & Watt for ships of the 
Navy ; they and one or two other engine builders adopted this type also for 
the mercantile marine. Although itself now superseded, it displaced and sur- 
vived the other kinds of horizontal screw engine once so popular on shipboard, 
and itself only succumbed when it was found that protection against shot 
could be afforded to a vertical engine in a warship without inconvenience 
or serious cost. Till then some horizontal type, which was well below the 
water-line, was considered absolutely necessary, and as the direct-acting 
kind of engine suited the higher steam pressures of modern time better than 
the others, it prevailed ; moreover, with twin screws a longer stroke of piston 

was permissible than was the case 
when only half the breadth of the 
ship was available for the cylinder 
and rods. Very early on Boulton & 
Watt made a horizontal four-cylinder 
oscillating engine that was fairly 
successful, and admitted of a longer 
stroke than the ordinary kind. Ex- 
perience with the "Simoom" does- 
not seem, however, to have inspired 
sufficient confidence in them to cause 
many repeats. 

Such direct-acting engines possess 
the same good features as the direct- 
acting paddle engine had when placed 
horizontally or nearly so ; they like- 
wise have the same objectionable 
features in the way of heavy pistons 
rubbing on cylinder sides, effect of momentum of the moving parts to cause 
stresses on the hull, and general vibration notwithstanding balance weights 
on the cranks ; and, finally, the general want of accessibility to some parts, 
when running, and to others, such as the pistons, etc., when requiring 
overhaul. 

Trunk Engines, invented and introduced by John Penn & Son into the 
Navy in 1849 on board H.M.S. " Arrogant," are of the horizontal type and 
direct-acting, inasmuch as the rod connects the piston direct to the crank 
pin, as is the case in the oscillating engine only. Fig. 31' shows such an engine 
and how the connecting-rod is enclosed in a hollow cylinder or trunk attached 
to the piston, and with a similar trunk in rear whereby the rod end and 
gudgeon could be fitted and examined ; it also served to ventilate these 
important parts, and prevent them from overheating. Both in the mer- 
cantile marine and Navy they were, in their day, favourite engines ; they 
were splendidly made and finished, and worked even at comparatively high 
revolutions, and latterly with steam of 60 lbs. pressure quite satisfactorily. 




Fig. 29.— Twin-Cylinder Engine. 



TIU'NK ENCINKS. 



73 




Jet Condenser 
Y\a. 30.— Engines of P.S. " C. D.," 6,370 tons. Lake Service. U.S.A. 
Cylinders 92 ;, -62"-92" diam. X 102" stroke. 7,000 I.H.P. at 28 revs. 



74 



MANUAL OF MARINE ENGINEERING. 



There was, however, considerable heat loss by radiation from the trunks, 
and condensation in the cylinders by their cooling action ; the friction at 
the stuffing-boxes was also somewhat of a drawback, being often considerable, 
and could be easily made so great as to slow down the engines. 



rfes 




Fig. 31.— Trunk Engines. 




Fig. 31a. — Return Connecting-Rod Engineo. 



Their screw was generally " left-handed," so that the thrust of the con- 
necting-rod when going ahead was upward ; but when at full speed it was 
considerably in excess of the weight of the piston, so that the rubbing 
was sufficient to " barrel " the upper side, and when in stern gear the 
thrust and weight were combined in doing the same thing to the lower 
side 



RETURN CONNECTING-ROD ENGINE. 



75 



The Return Connecting-rod Engine is shown in fig. 31a, by which it will 
be seen that it is a steeple engine laid horizontally with the two piston-rods 
prolonged to the gudgeon cross-head instead of into a banjo-frame. This 




Fig. 32. — Three-Crank Triple-Expansion Engine (Naval). 



type was introduced by Maudslay & Field to the Navy, and used by them 
till the vertical engine displaced them. Several other builders also adopted 
this type, inasmuch as therebv a much longer stroke was possible with ample 



79 



MANUAL OF MARINE ENGINEERING. 



room behind the cylinders for examination and repair, and all the working 
parts were quite accessible when moving. An attempt was made latterly 




d 

eg 

d 
o 

3 



-a 

d 

eS 



es 



o 






* l 



CO Q 

g d ® 

CO t» 

••ti o b 



s o S 
O S 

00 2 fl 

a5<2 "" 

h.a s 

I eel *o 

en " .3 

d « ' e» 
60, 



H 



M -2 



life 

es a 
W §3 



a 
-d 

eS 

d 



es 
o 

4) 

t'. 

SO 

tab 

s 



to support the pistons by fitting a substantial tall rod with a slipper support 
and guide at the back ; this arrangement was only a palliation, and did not 
prevent rubbing of the cylinder walls at bottom. With the vertical engiue 



VERTICAL DIRECT-ACTING ENGINES. 77 

there is no absolute need of such a fitment, and although the tail rod through 
their cylinder covers tends, no doubt, to steady the piston, and when the 
ship is rolling to prevent excessive side play. 

The Vertical Direct-acting Engines. — Figs. 32 and 33 are now the universal 
type of screw engine employed throughout the world for both war and mer- 
chant ships ; its advantage over others are so obvious as to need no further 
■comment, after having pointed out the defects of those others whose com- 
petition lasted. This type was first employed by the Thomsons, who founded 
the business now so extensive and successfully carried on by John Brown & 
Co. at Clydebank. In the mercantile marine it quickly found favour, and was 
soon adopted by most builders of marine engines for it, and is still largely 
used for generating electricity on shore. For naval purposes it was deemed 
unsafe formerly, as having the vital parts of the machinery above the water- 
line, therefore much exposed to shot and shell in a way not obtaining with 
the horizontal engine. This objection was, of course, from the military point 
of view, and formed a bar to their use long after the introduction of armour 
plates had provided a means for their protection if naval designers had so 
desired, and as was eventually adopted, and still prevails. 

In the mercantile marine, not only is the foundation or bed-plate of 
■cast iron, but the columns supporting the cylinders, both front and back, 
and forming guides for the piston-rod ends, are often of cast iron. In the 
Navy, for lightness and strength, and in a measure to resist concussion, the 
columns are of forged or cast steel, and the foundations themselves also of 
steel. Figs. 32 and 33 show two different makes of engines, the one for the 
Navy, the other for an express steamer. Fig. 32 is that of an engine with 
■cast-steel foundation and front column with cast-iron back columns. In fig. 
33 the columns on both sides are of cast iron, as also are the foundations. 
In fig. 34 is seen an example of a wrought-steel structure of extreme lightness, 
and yet so rigid and strong as to permit of 380 revolutions per minute being 
run quite satisfactorily. The cost of the latter, and such structures as seen 
in fig. 35, compared with either of the others, but more especially with the 
engines of fig. 33, is very high, but, notwithstanding this, the saving in 
weight, combined with the perfect rigidity, warrants the use of similar struc- 
tures for the support of the cylinders of express steamers of quite large size. 
Some builders of mercantile engines prefer these wrought-iron or steel columns 
for the front of the engines of all kinds (v. fig. 36), as they permit of freer 
access to the moving parts ; moreover, when secured in the foundation by a 
single nut and to the cylinders in the same way, so that these columns can 
be turned down from a plain rolled bar, they are even cheaper than cast- 
iron ones. 

In comparing the engine design and practice of to-day with that of past 
generations, it must not be forgotten that the earliest engineers had to make 
almost every part of cast iron ; for there were no steam hammers, and the 
forgings made by the tilt hammer and the rolled-iron bars of that period 
were of very limited size and form. Steel could be had only in very small 
pieces, and was very dear. Modern engineers have an unlimited supply of 
splendid mild steel, and now even reliable high tensile steel can be obtained 
at quite moderate costs. Forgings, rolled bars, and plates of huge size can 
be obtained cheaply from various sources of supply, and quickly compared 
with the time required twenty-five years ago. Our predecessors knew not 



78 



MANUAL OF MARINE ENGINEERING. 



the advantages of " dumping," or of steel castings, nor did they enjoy such 
white metal and stroug zinc bronzes as we know them. 

Arrangement of Cylinders in the modern marine engine is a much more 
important problem than was the case when there were two cylinders only. 




o 

=3 



to 



£ a* 

'5b 5* 

a ti- 

to 

% - 

— ?i 

"d v 

3 % 

0} 2+ 



o 

S-l 



o 



I — 

o 

'. a 

3 — 
So "> 

X 

cib «» 

.Si a> 



13 



CJ 



To-day, with a multiplicity of cylinders, their arrangement is governed by 
circumstances not obtaining and never dreamed of formerly. We have 
now to consider the question of balancing the momentum of the moving 
parts, as well as to determine what is the best sequence for the flow of steam 



ARRANGEMENT OF CYLINDERS. 



79 



through the system from the boilers to the condenser to attain the greatest 
steam efficiency, together with the minimum amount of "piping for that 
purpose. N These and other practical considerations impose themselves od 




u 
O 
J= 
♦J 
3 

< 






<u 
a 
so 

Q 



a 



o 
o 
o 



u 

■*> 

in 



a 
o 

a 

n 
02 



a 



c 

'5b 

c 
- 

I 



the designer, especially when the engines are of such great power as to require 
excessively large cylinders if their number is to be as limited as they may be, 
and are generally just as found in engines of smaller powers. Large cylinders 



80 



MANUAL OF MARINE ENGINEERING. 



can be, of course, manufactured, and large pistons, being now of solid cast 
steel, are no longer so heavy and cumbersome as to be almost impossible 




of handling with the engine-room appliances . Moreover, all engines to-day- 
are of the vertical type ; there is, consequently, no longer any fear of damage 



ARRANGEMENT OF CYLINDERS, 



81 







2 

o 

a! 



•3 
C 
* 
50 

.3 









=3 
Ah < 



5 -3 



"* 


"3 


o 


3 




■ 


m 


3 


0) 


a 




09 


60 


< 


S 




W 








"3 


— ■ 


a 

3 

o 


3 


a, 


V 


a 


•z. 


o 




O 


g 


in 


u> 




I s 

. CO 

CO © 






a 



s 



82 MANUAL OF MARINE ENGINEERING. 

to cylinders, liners, and walls, or difficulties due to the position of the cylinders 
whereby the use of mechanical means for getting them in and out of place 
was precluded, besides rendering them difficult to examine. But the risk 
of casting large cylinders is still great ; the cost of replacing such an one, 
if damaged, great, and perhaps, what is more important in these days of high 
piston speed, there is always the impossibility of designing the low-pressure 
ones with ports of sufficient size for a flow of steam slow enough to be con- 
sistent with high efficiency. The area of port possible for a cylinder with 
slide valves varies practically as the diameter, inasmuch as the length of port 
axially does not increase rapidly with increase in the diameter. The volume 
of steam passing through it varies as the square of the diameter. When 
piston speeds seldom exceeded 500 feet per minute, ports with a sufficient 
area for very large cylinders could be designed easily enough ; but with a 
speed of piston of 1,000 to 1,200 feet it is impossible to provide them large 
enough for moderate flows and with passages small enough for moderate 
" clearance." 

With the Diesel and other oil engines, where the initial pressures are very 
high, while the mean pressures are quite low, the latter condition requires 
a large total cylinder capacity and the former small units ; hence such engines 
have as many as twelve cylinders per engine when of the large horse-power 
required on shipboard for good speed of ship. 

Even in the steam engine of to-day the boiler pressure, and consequently 
the initial pressure in the H.P. cylinder at starting is as much as 230 lbs., 
while the mean pressure of the compound cylinder system is not very much 
more than that obtaining when it was a fifth. In the Navy, with water-tube 
boilers, the pressure is often even greater still, but the referred mean pressure 
in such ships is much higher than in the mercantile marine when running 
full speed. 

With a Compound Engine the least number of cylinders is of course two, 
which may be in line axially with their pistons attached to a common rod, 
and operating on one crank by one connecting-rod (v. fig. 75). Such engines 
were at one time used in certain cargo steamers, and thought well of at the 
time, in large measure owing to the small space occupied by them and their 
comparative cheapness. More commonly the two cylinder compound engine 
had them side by side, each operating on a separate crank, the one opposite 
the other when the steam expanded direct from the H.P. to the L.P. cylinder, 
or set at right angles with a receiver between the cylinders into which the 
H.P. exhausted, and from which the L.P. took steam. In some few cases 
the cranks were set at a different angle, generally about 105°, so as to favour 
the timing of exhaust from H.P. and flow to the L.P., and to reduce thereby 
the variation in pressure in the receiver. These latter engines were, how- 
ever, somewhat unhandy in starting and reversing, and eventually were 
given up in favour of the cranks at 90°, and having a somewhat larger 
receiver between the cylinders. 

The Three-cylinder Compound Engine, each cylinder with a separate 
crank and connections, was deservedly a favourite one for large powers ; 
for not only was there with it the advantage of having the L.P. cylinders 
of moderate size, but a much higher initial pressure could be thereby main- 
tained in them, consequently with the minimum amount of drop between 
H.P. and L.P. cylinders ; a fair division of the work between the three cranks 



TWO-CRANK SCREW ENGINES. 83 

was made, and they were usually set at angles of 120 3 . This engine was 
a favourite for electric generation on shore stations with large, units. 

Four-cylinder Compound Engines enabled the same subdivision of the 
L.P. pressure member of the system without resorting to an additional 
crank by dividing the H.P. member, and placing one H.P. cylinder tandem 
with one L.P., each pair side by side operating its own crank, which was at 
90° with the other. This type of engine was adopted by Messrs. Maudslay, 
Sons & Field in the celebrated White Star steamers, " Britannic," etc. ; 
moreover, it was the method used by many engineers as a cheap and ready 
way of compounding the old expansive engines after the superiority of the 
compound system was assured about 1870. 

Six-cylinder Compounds, each pair of H.P. and L.P. being tandem and 
side by side, each operating on a separate crank, as with the four-cylinder 
system, of which it is an extension ; this design was adopted for the " City of 
Rome," as a further subdivision of cylinder due to her then large power. 

With a Triple-expansion Engine the least number of cylinders is, of course, 
three, as with a quadruple expansion it is four. It does not, however, follow 
that each must have its own separate crank, as is the rule with the single- 
acting oil engine ; and, although it is common practice now to do so with 
the marine steam engines, it was a few years ago not an unusual thing to find 
a triple-expansion engine with two cranks and the cylinders as in fig. 73, 
Nos. 1, 2, and 3. Moreover, the quadruple engine when first placed on the 
market as a competitor with the triple had only two cranks (fig. 73, No. 1) ; 
this, as a matter of fact, was at that time placed to its credit as a means 
whereby it occupied no more space than a common two-cylinder compound, 
and less than a three-crank triple or compound engine. When, however, 
the demand for a non-vibrating engine was insisted on by the Admiralty, 
and much desired by all interested in the passenger service, the Yarrow- 
Schlick-Tweedy system of balancing the four-crank quadruple engine gave 
it such an advantage over the three-crank triple that makers of the triple 
engine had to adopt the four-crank arrangement, and with two L.P. 
cylinders, either as Nos. 3 or 4 in fig. 73, to obtain like advantages. 

A Single-crank Engine is seldom or never seen to-day, except in quite 
small and cheap launches, or harbour service boats. Even single-crank 
tandem compound engines, which at one time were much in favour with 
one or two shipowners and some few engineers, are no longer made. Single- 
crank paddle engines are also a thing of the past. 

Two-crank Engines still survive in paddle steamers of moderate power 
with compound cylinders (v. fig. 19) ; with the larger power compound 
engines three cranks are favoured (v. fig. 21), as each of the two L.P. cylinders 
are of moderate size, and the ratio of maximum to mean torque is lower than 
with two cranks, consequently the movement of the wheels is less jerky, 
and a smaller shaft possible. Triple-expansion paddle engine; are invariably 
of the three-crank type, as shown in fig. 22. 

Two-crank Screw Engines are still made with compound cylinders for 
quite small powers, as in tug boats, steam launches, and other small craft ; 
they may be adopted with advantage still in cases where prime cost, weight, 
and space occupied are of more importance than fuel consumption. With 
triple-compound cylinders such engines are now very seldom, if ever, made, 
but with quadruple-compound cylinders something like those shown in fig. 75. 



84 



MANUAL OF MARINE ENGINEERING. 



the two-crank arrangement is quite a good one for smaJl craft, especially 
those having water-tube boilers for the &ake of quick raising of steam, as in 
yachts or their launches. 

Three-crank Screw Engines (v. figs. 32 and 36) still continue to be the 
common and, to a great extent, the favourite form in the mercantile marine, 
when the power is not great, and the service either cargo-carrying or com- 
bined with a passenger service of not high class, inasmuch as it is cheaper, 
occupies smaller space than the four-crank, and can be balanced sufficiently 




Fi^ 37. — Eive-CMiiKler Quadruple-Expansion Engines 01 S.s. 

Marine Engine Works, West Hartlepool). 



Inchdune " (Central 



well, if thought necessary, by simple and comparatively inexpensive means, 
so as to cause little or no inconvenience from vibration. 

The Four-crank Engine (figs. 33 and 33a) is, however, more easy to balance, 
and the balancing, when done, is more perfect ; the quadruple system permits 
of the use of higher boiler pressures with higher efficiency and, therefore, 
with advantage ; on this triple-expansion system, however, there are two 
L.P. cylinders, with such advantages as arise from their smaller size. 

It may be taken, then, that as a rule for express passenger ships and 



HEAVY OR CRUDE OIL ENGINE. 85 

warships, the four-crank engine is a better one to adopt ; for very small ships, 
or where the engines are small, the three-crank engine has advantages which 
may outweigh those which would favour the four-crank in a general problem. 

Five-crank Engines (fig. 37) have been made to a limited extent with 
cylinders on the quadruple system, thus having two L.P. cylinders. It has 
been claimed for them that there is a superiority in the matter of balancing 
together with a steam consumption as low as any other. Such an engine 
may be employed with advantage when, owing to size, it is desirable to have 
two L.P. cylinders ; but for small power it seems an unnecessary expense 
with considerable complication ; moreover, the mechanical efficiency cannot 
be so good as that of a three-crank engine of the same power. 

Six-crank Engines for steam have been used to a very limited extent 
by marine engineers. Fig. 38 is a good example of such an engine of very 
large power. Oil engines have now come into use on ships of considerable 
power, so that six cranks are common ; they are generally in two pieces, 
and so coupled that No. 1 and No. 6 cranks are in line, as are also Nos. 2 and 
5 and Nos. 3 and 4, each pair being at angle of 120° with the others. With the 
two-stroke cycle they may be in sequence at 60° angles. 

Eight-crank Engines will also be used largely, as they are already for 
internal combustion systems, when the power is very great, since there is 
a decided limit to the size of the cylinders of such engines. It may be that 
steam engines will also be made with these numerous cranks to compete 
with such engines and turbines. 

Of Oil Engines there are three kinds used on shipboard — viz., the petrol, 
the paraffin, and the crude or heavy oil engine. 

(1) The Petrol Engine, requiring as the fuel supply the light volatile oil 
of that name, having a flash point from 80° to 100° F., is not admissible on 
board ordinary passenger ships, from the danger attending the carrying and 
storing such a highly inflammable liquid. It is, however, used extensively 
on launches and other small craft for coasting work or on rivers and small 
lakes, and is as efficient and convenient in them as in motor vehicles on shore. 
The ease of starting a cold engine is always a strong recommendation for 
this oil. 

(2) The Paraffin Engine using a refined light oil obtainable almost every- 
where, and safer to carry, use, and store than petrol, inasmuch as it is not 
nearly so volatile, and its flash point is considerably higher — viz., 120° to 
150° F. — is employed for many purposes now, as it is almost as easy to start 
as the former and runs quite as well. Both it and the petrol engine work 
on the usual Otto or four-stage cycle, and require ignition by an electric 
spark, which may be produced by a secondary battery or by a small dynamo 
worked by the engine itself, called a magneto. 

(3) The Heavy or Crude Oil Engine, which is the desideratum for ship- 
board, on account of its comparative safety, uses oil as fuel whose flash point 
is over 200° F., and not volatile enough to be used in the same way as the 
light oil engines. Moreover, as the oil is unrefined, or else refuse, special 
treatment has to be accorded to it for the different varieties of fuel used. 
Texas, Batoum, and similar oils practically free from bitumen may be used 
with comparative ease with suitable carburetters, etc., and with suitable 
engines and care will run sufficiently well as to be used for driving dynamos 
for power purposes, and most engines of this kind can be trusted, therefore, 
to drive propellers of ships quite well for considerable periods if there is no 



86 



MANUAL OF MARINE ENGINEERING. 




§ 1 
© -* 

: 3 



00 






-f IN 

a j, 
.S 

>5 t- OS 



a 
I- 

(3 

c? 



oo 



60 



>» 

a 



stoppage at any time sufficiently long to permit of the fooling of the cylinders, 
etc. "With a large number of the oils obtainable at other parts of the world 
the crude oil contains varying amounts of bituminous matter, which not only 



THE SEMI-DIESEL ENGINE. 87 

■causes excessive deposits of tarry matter, but what is v/orse, a formation of 
coke, hard and refractory, which, if deposited in the cylinders or among the 
valves, will cause serious damage to them. 

The Griffin and some other engines were designed to run with the gas ox 
-vapour distilled from crude heavy oils and their residuals by the high tem- 
perature of the exhaust gas ; this vapour, mixed with air, was drawn into the 
cylinders as in the ordinary gas engine and exploded there by a spark. The 
tarry matter and pitch were in this way thrown down in the distiller and 
•excluded from the internal parts of the engines, so that they remained clean 
and free from grit. The early promise of these engines was unfortunately 
not maintained as fully as was desirable, and after the Diesel engine had 
proved so successful it displaced them. The cylinders, etc., were kept suffi- 
•ciently cool by the usual water- jacketting. 

The Diesel Engine,* which is now largely used for ship propulsion, differs 
fundamentally from other oil engines, inasmuch as it draws in a charge of 
air only and compresses it highly, generally to about 500 lbs. per square 
inch, when it becomes so hot as to ignite with certainty the spray of oil injected 
into it at the commencement of the active stroke ; moreover, the ignition 
is gradual instead of instantaneous, and the pressure is practically that at 
which compression ceased, so that there is no shock due to explosion ; ex- 
pansion commences before combustion is complete, and continues to the 
end of the stroke, when the products escape through the exhaust valve, being 
driven out by the piston of the four-stroke cycle and by a blast of fresh air 
in the two-stroke cycle. Almost any oil will do, but generally heavy crude 
oils freed from highly volatile constituents, or residuals of oils from which 
petrol, paraffin, and lubricating products have been extracted, are used ; in this 
country tar oil and shale oil residues are home products which can be used 
with satisfactory results, their flash point is about 220° F. 

The Semi-Diesel Engine, which also uses similar fuel, works with less 
compression, about 150 lbs. or 60 per cent, of the initial pressure on ignition, 
and consequently requires a hot plate or bulb on which the spray impinges 
to effect it. It works on the two-stroke cycle, and is scavenged by air pumped 
into a receiver by the imderside of the piston. It is fairly efficient, and much 
used on small ships with cylinders up to 16-5 inches diameter. 

The Diesel engine, like the other oil engines, was single-acting, working 
on the four-stage cycle ; air only is admitted on the first descent of the 
piston, on its return it is compressed, sometimes to the extent of 40 atmospheres, 
so that even with means for cooling the cylinder the temperature is very 
high. Just as the piston is about to make a second descent the necessary 
supply of oil is sprayed into the cylinder top by means of a jet of air com- 
pressed higher than that in it ; the finely pulverised oil at once ignites, and 
burns during the early part of the stroke, and so maintains the pressure 
attained by compression (v. fig. 101) ; at the end of the stroke the exhaust 
valve is opened, and the piston on returning to the top scavenges the cylinder 
— that is, drives out the products of combustion. It again descends, drawing 
in air alone as before. 

To compete with the steam engine, especially on shipboard where 
weight and space are of importance, Dr. Diesel, and those acting with him 
at Nuremburg, devised the double-acting two-stage cycle engine, whereby 
an explosion takes place at each end at every revolution, so that its activities, 
so to speak, are equal to those of the steam engine. With the two-stage cycle, 

* Vide Appendix A 



88 MANUAL OF MARINE ENGINEERING. 

whether the engine be single or double-acting, the modus operandi is the same, 
and as follows [v. fig. 51) : — The cylinder is charged with air alone, as before, 
but now it is forced in under quite moderate pressure by a special pump 
worked by the engine, much in the same way that the air pump of a steam 
engine is worked [v. fig. 68). The piston compresses, and the oil is injected 
quite as before, so that when it descends for the first time it is filled with 
products of combustion. An exhaust port opens at the end of the stroke, 
and by another the admission of air is made in sufficient abundance to 
thoroughly scavenge it, and leave it filled with pure air. In the case of 
the double-acting engine the exhaust port is at the middle of the cylinder, 
and the piston sufficiently deep to act as the valve for each end to close it 
before compression, and keep it closed during explosion and expansion. 
This, of course, means a very long cylinder, and a very hot one, too, when 
there is an explosion every revolution at each end. Both it, the piston, and 
piston-rod will require to be water-cooled somehow. 

The oil consumption of the Diesel engine when in good working order 
is very low, under half-a-pound per horse-power developed, and in some 
cases it has been as low as 0-38. This latter is equal to 0475 of best Welsh 
or 0-51 Newcastle coal, or less than half that of a turbine or quadruple- 
expansion reciprocator, but much more lubricant is expended. 

It is estimated that of the total heat of combustion in the cylinder of the 
Diesel engine 40 per cent, is usefully employed, iO per cent, passes away with 
the gases at exhaust, and 20 per cent, is absorbed by the cooling water. 
The cycle of operations in its cylinders may be followed in fig. 51, Nos. 1, 2, 3, 
and the engine itself with its various pumps in fig. 50a, and in Appendix A. 

The Reversal of Oil Engines is accomplished by using one or more of the 
cylinders as an air engine, supplying it with compressed air carried for the 
purpose in small craft, or accumulated by the oil engine when running, and 
stored for the purpose as it is in larger ones (o. fig. 50a). 

Reversal of the Propeller, when driven by oil engines, can be accomplished 
by means of wheel-gearing, as in a motor car, or, better still, by twisting 
the blades sufficiently to bring the base or part of them near the boss to 
a transverse position of no pitch, when the outer part and tips will have 
reverse pitch sufficient for navigation purposes. This was done in various 
ways, and one similar to the method adopted by the late Mr. Bevis for modi- 
fying the pitch is quite a successful one. In this case, instead of a nut on 
the shaft, there is a sleeve, which is free to slide on the shaft, and is carried 
round on it ; it is grooved as is a sliding clutch, and the lever operates in the 
same way as in a clutch gear. In head gear the two blades of the propeller 
are of true pitch and the usual form to be highly efficient when at normal 
work. Since in stern gear it is only the outer portion which is effective, 
it is desirable that the tips be made fairly broad. 

Turbine Machinery in its various forms is now common in express steamers 
of all sizes and services, as well as in the ships of the British and some foreign 
navies, where it has quite displaced the reciprocating engine. The Parsons 
turbine has long been a favourite, and the success of this instrument, both 
on land and shipboard, is due so very largely to the genius of that gentleman 
that it will ever be associated with his name. The Curtis, Zoelly, Rateau, 
and other turbines are fitted to ships with results equalling, if not excelling, 
those already recorded by the Parsons, the Brown-Curtis being very efficient. 

The turbine must have, of necessity for high efficiency, a great peripheral 



TURBINE MACHINERY. 89 

speed ; the diameter of rotoi or the revolutions must, therefore, be great, 
but for very large power direct-driven the latter cannot be. In the case of 
the s.s. " Lusitania " (fig. 39), the L.P. rotor is 140 inches diameter, with 
the last set of blades 22 inches long, and running on bearings 33 inches 
diameter at 194 revolutions per minute on trial, and about 185 on 
service, and weighing 120 tons. For smaller powers the rate of revolution 
is much higher, and in case of a warship of 5,000 II. P. per shaft the rate of 
revolution is as much as 500, and higher, even up to 700 revolutions, in 
smaller ships of 4,750 H.P. per shaft. Such revolutions necessitate not only 
a screw of small diameter, but one of such very small pitch that the pitch 
ratio is so low that the efficiency of the propeller is lower than in competing 
ships. The slip ratio, however, is wonderfully small in the turbine-driven 
ship, taking all these things into account, being only 15-3 per cent, in the 
" Lusitania," and seldom over 25 per cent, in others where revolution speed 
is not very excessive. Formerly Sir C. Parsons fitted more than one screw 
to each shaft, but without the success he anticipated ; the German Admiralty 
also tried the same method for improving the propeller efficiency of the 
turbine-driven ship, " Lubeck," with the same disappointment ; conse- 
quently for such a ship a single screw of moderately small diameter and large 
disc ratio is the rule. That is, the screw is somewhat larger than would be 
fitted, if the regard of the designer were limited to turbine efficiency only. 
As a matter of fact, in all cases of this kind the choice of screw and all that 
pertains to it is governed by combining the efficiency of propeller and gene- 
rator, as will be seen later on.* The turbine, when of a power exceeding 
1,000 S.H.P., is superior to the best reciprocator in steam consumption 
per unit of power ; the turbine of any size has a higher mechanical efficiency 
than a reciprocator of equal power ; it occupies about the same amount 
of floor space as the ordinary triple and quadruple engine, but is of less 
height, so that much of it can go under deck, consequently the engine hatches 
can be very much smaller for them than the reciprocator. The consumption 
of lubricants is less and fewer attendants are required in the engine-room 
when on service. The difference in weight is trifling, but the prime cost 
and the repair and wear and tear account of the modern make of turbine 
compares favourably with that of the average reciprocator. On the other 
hand, the steam efficiency of the turbine falls off on reduction of load, and 
since the marine turbine suffers reduction in velocity when the speed 
of the ship is reduced, the fall in efficiency is considerable ; so much so, 
indeed, that in the case of the " Lusitania," where the consumption of steam 
per S.H.P. per hour of the turbines alone was only 12-77 lbs. at the full speed 
of 25-4 knots per hour, it was as high as 21-23 lbs. at 15-77 knots. The 
corresponding coal consumptions, which, however, of course included that 
due to the requirements of the auxiliary machinery, were 1-46 lbs. and 2-76 
lbs. respectively per S.H.P per hour. With the reciprocator there is no such 
rapid increase in coal consumption at the lower powers ; on the contrary, 
the rate is lower at moderately reduced speeds than at full, and at quite 
low speeds is not very high. For example, on the trials of H.M.S. " Achilles " 
the steam consumption at full speed of 23'275 knots, with the engines 
developing 24,000 I.H.P., was 19-9 lbs. per I. H.P. .per hour, and the coal 
2-03 lbs., while at 14-6 knots it was only 16-95 lbs. of steam and 1-88 lbs. 
of coal; moreover, at 21-58 knots and 16,000 I.H.P. the coal consumption 

* The employment of gearing has solved the problem of screw and turbine efficiency 
{vide Appendix A). 



90 



MANUAL OF MARINE ENGINEERING. 




. ® «K 




rasafflflt 



Ck 

a 

CO 



to 

f 



a 



CO 

2 



S3 



CO 

S3 



IS 

3 



= 
© 

s 

s 

S3 
c3 



O 
S 

C5 



o 



b£ 



ADMIRALTY TURBINE TEST. 



91 



was 1'85 lbs. only. Her engines were four-cylinder four-crank triple- 
expansion reciprocators. 

The Admiralty, to test the Turbine, caused to be carried out some exhaustive 
comparative trials with H.M.S. " Amethyst," fitted with Parsons turbines, 
and her sister ship, the " Topaze," having the four-cylinder triple-compound 
reciprocators. These ships are each 360 feet long, 40 feet beam, and 14*5 feet 
mean draught of water ; their displacement is 3,000 tons, and wetted skin 
about 16,100 square feet, with block coefficient of fineness of 0*503. Their 
boilers were similar in all respects, and their trials were conducted on exactly 



13,000 




$>I2 2,000 
10 1, 000 



20 21 22 23 



16 17 16 19 
Scale of Speed. 

Fig. 40. — Trials of "Amethyst " (Turbines) and Sister Ships (Reciprocators) 



25 Knots 



was exerting 



the same lines, with the result that, whereas when the " Topaze 
the maximum power of her reciprocating engines, 9,868 I.H.P., the speed 
attained was 22'103 knots, the turbines of the "Amethyst" developed a 
S.H.P. equivalent to about 14,200 I.H.P., which drove her at a speed of 
23*63 knots. Fig. 40 gives the curves of power consumption, etc., of the 
two ships and their sister ships fitted with reciprocators. 

The Experiments made in the U.S. of America with the cruiser " Bir- 
mingham," fitted with triple-compound reciprocators, and the " Salem," 
having Curtis turbines, is equally interesting. These ships are otherwise 



92 



MANUAL OF MARINE ENGINEERING. 



oo 

w 
z 

Cs 

S5 

H 

<J 

C 
O 

« 

o 

o 

-« 

S3 

n 
« 
s 
H 

a 



H 

E-i 
H 



00 

ft, 
a 

02 

O 
oo 

Hi 

H 

K 

> 

i— i 
H 
««) 

« 
<! 

S 
o 
o 



XI 
PQ 



s 




g^ _-05 90 CM -g 


. as 
<•£? 


»o o o 


o o i> o o © 


o o 


<M • A t> O O 

«M 

1 


5 =. 


u > 


s 


-co o „ o, H 
:= 90 . ;* cm © . a •- 

^CM W- 1 <N -< fig 






• 90 © rr\ ~~* 
O JL 90 . . -+l . • . — 






. . 


o o o 


gSrJ M M C.22 u 


«< S 


O O h> o o o 


n "S P 


OB 5 


CM 


■ b a r ° 




. o o _ P O « 
-©© .oo© .5oH-^ 






s ia °l • ^ '-"0 . • 






f—t 






.oo •* t-» © 






&©.»o©oot--© 00 


»:s 


oo« 




© O © © © © 
© • • © — no 

" ■* m eo » o 


J 0) o 


a=2 


>.-=; sr 


o .a"tL 
. © oo oo o _L «h ~ 


(-H 


S^h ©©OO-^©^^* 3 © 






3^ ,»^ NA™ O 






.© © © © 






£© . lO © I— O — < co 90 


43 

* <n 
CO >> 

as 


© © 90 




© © © © © © 


-3 o g 


" -* — , 90 © © 


.90 O IN ■* r, a H 
S"-C .©©cOTtlt^^PH © 










1-1 






. © © «5 lO 00 






sE©.COr~cM©00 in 


CO «> 


© © t- 


= -« -*^ co»^ 2 

02^- -*00"rt .JO" -1 


© lO © © © © 


fcj <B O 


»3 


^ r« CM 90 © © 




t^ & ■*■ ^~~~ ~* 


.t~- Ot*< —90'V'SH 


— 90 












CM 






(e F- Q . © cm <^ ?- 


rt 


© © ©OS 


T ^ /^ — 90 CM ^' Si 1 


03 S 


i lO >o ^ © © 


3 o s © i 


CO — 


53 2 o o 

a - JS .O 00 05^H(L 
s no Q0 . © c^ v • d- 
fe CM oo ^ 0^ 2 


Hi 


t- 00 90 © CM © 


90 00 










© 


■ 


* *• *• «r r . 


-*-N 




43 -+3 43 J- £ 
*«-, ^ *+H q 


ccT ai co 




43 J © 


• 


43 — 


• O — ' — i 




Q 


c 


* 


. . . . M . 


■a S 


1 


• ••••• 


oj • • • . c • • •• 

a *< -e 






co co p^ 


PS 


* *« 


f 6 a 




CO CD 


C c» P-i r co g 


to 


2 a 


H js i PL, fl.2 




£ - ® 
■2 *fl 


O » • O -g 




c3 K K C 

C CO CO (/J q> CO 


o£.2g x "3- "I, °-2 

» <S -O > A 43 e3 SP ^ 53 




.sassls 


e-i-s c ,g « o o a o o 



COMBINATION OF TURBINES WITH RECIPROCATORS. 



93 



alike, and 420 feet long, 47 feet beam, 16*75 mean draft, the displacement 
4,700 tons, and the wetted skin about 22,000 square feet. They are of 
very fine form, the block coefficient being 0*50. The Curtis turbines drove 
the " Salem " at a speed of 25-947 knots, with 19,200 S.H.P., and a coal 
consumption of 201 lbs. per S.H.P. per hour (equivalent to 181 per I.H.P. 
of a reciprocator). The reciprocators drove the "Birmingham" at 24*325 
knots speed, with 15,540 I.H.P., consuming 1*92 lbs. of coal ; at 12*23 knots 
with 1,600 I.H.P., the consumption of coal of the " Birmingham " was 
2-89 lbs., against the 2-68 lbs. of the "Salem" at 11*93 knots with 
1,360 S.H.P. 

Experiments were made by the German Government with the cruiser 
'* Lubeck," of 3,170 tons displacement, 341 feet long, 43 feet beam, and 
16" 4 feet draft of water, driven by turbines operating on four shafts, each 
shaft having two screws at a speed of 23 knots, and the twin-screw sister 
ship, " Hamburg," having reciprocating engines ; but in this case the superi- 
ority of the turbine was not demonstrated, inasmuch as it took 14,158 H.P. 
to attain a speed of 23*16 knots with the "Lubeck," and only 11,582 for 
23*17 knots with the reciprocators. Doubtless, however, this was in no 
small measure due to the screw arrangement, which was a bad one, and 
probably the propeller efficiency was very low. Since those experiments 
were made, the German Admiralty have followed the lead of the British, 
and now are having turbines of kinds fitted in their warships. 

Table XlVa. summarises the above. 

The following table is interesting and instructive, showing, as it does, 
the comparative steam consumptions of the " Amethyst " and " Topaze " 
at speeds varying from 10 knots to 22, including the steam used by the aux- 
iliaries. It must be borne in mind, when considering the same, that the 
engines of the " Topaze " were not designed primarily for economy, conse- 
quently their consumption of steam per I.H.P. is very high compared with 
that of the quadruple-compound engines of the mercantile marine, such as 
fitted in s.s. " Saxonia," constructed by John Brown & Co., and tested by 
the Admiralty Boiler Committee, where it was found to be 14*5 lbs. per I.H.P. 
when on service. At the same time, the thermal efficiency of the " Topaze's " 
engines was by no means bad considering the conditions under which they 
worked, for her steam consumption in main engines only was but 16*91 at 
a speed of 20 knots and 15'45 at 18 knots : — 



TABLE XV.- 



Water Consumption per I.H.P. per Hour of H.M.S. 

on Progressive Trials. 



" Amethyst " and " Topaze 



Speed in Knots. '. 10 


11 


12 


13 


14 


15 


16 


17 


18 


19 


20 


21 


22 


1 
H.M.S. " Amethyst" (Turbs.), 29-3 

H.M.S. "Topaze" (Recipros.), 238 


260 
22-0 


237 22 
20-6 19-6 


20-4 
18-8 


19-0 
18-4 


179 
18-3 


168 
184 


159 
187 


1.5-2 
19-2 


14-7 
198 


14-3 
20-5 


140 

21-4 



Combination of Turbines with Reciprocators seems to be the most likely 
development of marine steam machinery in the future, and it has already 



94 MANUAL OF MARINE ENGINEERING. 

been adopted successfully by Denny & Co., also by Harland & Wolff 
in the Transatlantic steamships recently built by them for the White Star 
Company and others. Since the triple-expansion engine is more economic 
than is the boiler end half of a turbine, while the other or condenser end of 
the latter is much more economic than the reciprocator, and can make good 
use of a high vacuum, such a combination of the two is obviously a fitting 
one, and a desirable thing. Further, since at slow and cruising speeds the 
turbine is not economical, while the triple-compound engine is very fairly 
so, these latter engines can be employed by themselves to propel without 
using the turbines by exhausting direct to the condenser. Fig. 41 shows 
an arrangement proposed by Parsons for dealing with a three-screw 
ship ; here each wing screw is driven by a four-cylinder triple-expansion 
engine arranged to exhaust either to its own condenser or to a low-pressure 
turbine operating a central screw and exhausting to the same condenser. 
Fig. 42 is an example by the same gentleman for a four-screw arrangement, 
in which each inner screw has its own triple-compound engine exhausting 
direct to its own condenser, or to a low-pressure turbine operating a wing 
screw, and exhausting to the condenser. That is, in each case a turbine 
is interposed between the L.P. cylinder and the condenser, in order that 
the steam from it at 10 lbs. pressure absolute may be usefully employed 
in expanding down to the pound or even less pressure of the 
condenser. 

Messrs. Denny Brothers, of Dumbarton, fitted the s.s. " Otaki " with 
triples and low-pressure turbines, and demonstrated that the gain over the 
arrangement with triple engines in the sister ship, s.s. " Orari," was as much 
as 17 per cent. Careful experiments with triple-expansion engines at electric 
power stations on shore show that fully 15 per cent, more power is obtained 
with the same consumption of steam if a low-pressure turbine is interposed 
in this way between it and the condenser. 

The following are the figures given by Com. Wisnom, of Denny's, from 
the trials of the above two steamers : — 

The Performance of s.s. " Otaki," having two sets of ordinary triple- 
expansion engines, each driving a wing propeller as in a twin-screw ship, 
and both exhausting to a low-pressure turbine driving a propeller on the 
middle line as in a single-screw ship; built by Denny Bros., of Dumbarton, 
for the New Zealand S. Co., and sister ship to the twin-screw steamer " Orari," 
which was, however, 4*5 feet shorter. 

The " Otaki " has a displacement of 9,900 tons on 27'5 mean draught, 
and her principal dimensions — 

Length between perpendiculars, „ . . 464*5 feet. 

Beam moulded, 60*0 „ 

Depth ,, 34 - „ 

24"5"-39"-58" 
Each of her reciprocating engines has cylinders - ~-^ Q „~ , while each 

24'5"-4r3"-69' y 
of those of the sister ships, " Opawa " and " Orari," are j^„ . The 

turbines of the " Otaki " has a rotor 90 inches diameter. 



COMBINATION OF TURBINES WITH KECIPKOCATOKS. 



»5 




n 

a 
o 

CO 



n 

n 

s 

a 
o 

o 

— 
o 

.3 






■A 






a- 

a 

a 

e 

o 
O 






96 



MANUAL OF MARINE ENGINEERING. 




a 

3 



(4 

O 

o 

o 
t* 

.Oh 

o 

o 

3 






m 

4) 

g 

u 

a 

H 






.2 

{3 

.9 

a 

o 
O 






PERFORMANCE OF S.S. " OTAKI.' 



97 



The following is a comparative summary of results of trials at 146 knots 
speed on the measured mile : — 



Name of Ship. 


E.H.P. 


I.H.P. 


Propulsive 
Coefficient 


Water Consumption 


per Hour. 


Total. 


Per E.H.P. 


Pei I.H.P. 


3-screw s.s. " Otaki " (turbo- 

recipro.), 
2-screw s.s. " Orari " (reci- 

pros.), .... 


3,350 
3,210 


5,880 
5,360 


Per cent. 
57 
60 


73,300 
88,300 


21-9 
27-5 


13-7 
16-5 


Gain per cent, in " Otaki," 


•• 


•• 




17 


20 


17 



TABLE XVI.— Measured Mile Trials of s.s. " Otaki," 
October 31, 1908, on 20 Feet Mean Draught. 





Mean of 


Mean of 


Mean of 


Mean of 




A Runs. 


B Runs. 


C Runs. 


D Runs. 


Total horse-power, being I.H.P. (recipros.) -f 










S.H.P. (turb.) 


6,857 


5,348 


4,704 


3,282 


Mean speed, .... knots, 


1502 


14-28 


13-83 


12-52 


Revolutions, recipros., .... 


103-5 


97-9 


93-5 


83-4 


„ turbine, . . t 


224-5 


209-7 


197-2 


1721 


Total water consumption per hour, . lbs., 


82,000 


67,300 


60,200 


44,600 


„ „ „ per H.P. „ 


11-95 


12-6 


12-8 


13-6 


Mean absolute pressure at H.P. cylinder, „ 


193 


178 


166 


135 


„ „ turbine inlet, „ 


9-5 


7-62 


6-76 


5-0 


Vacuum at exhaust end of turbine, 


28-1 


28-2 


28-4 


28-5 


,, on condenser gauge, 


28-2 


28-4 


28-3 


28-5 


Temperature of sea water, . . . F.° 


56 


56 


56 


56 


„ circulating discharge, . 


70 


67 


70 


70 


„ hot well, .... 


72 


70 


73 


74 


Steam consumption based on the I.H.P. of 










s.s. " Orari " by tanks, .... 


13-66 


13-7 


13-8 


13-07 


As measured by pumps per I.H.P. per hour, . 


14-12 


14-1 


14-3 


15-2 



Figs. 43 and 44 show the general arrangement of the machinery of the 
s.s. " Olympic," built by Harland & Wolff for express service between 
England and New York. She is one of the largest ships, and has worked 
with satisfactory results ; she is 860 ft. long, 92-75 ft. beam, and 32-5 draft 
of water, displacement 50,000 tons, and I.H.P. 54,800 ; propelled by three 
screws, the two wing ones worked by reciprocating triple-compound engines, 
each having four cylinders, 54, 84, 97, and 97 inches diameter, and 63 inches 
stroke, and the middle by low-pressure Curtis turbine, taking the steam 
supply from the L.P cylinders of the reciprocators (v. fig. 44). 

A Development with existing Single-screw ShiDS could be made by fitting 
two low-pressure turbines abaft the old triple engine, each operating a wing 
screw, and both exhausting to the old condenser ; this would probably 

7 



98 



MANUAL OF MARINE ENGINEERING. 




ENGINE-ROOM OF R.M.S. "OLYMPIC." 



99 







a 

o 

.2 

'35 

5 

Oh 
M 

w 



H 

o 
is 
H 



43 
C 

o 

3 © 
~S 

O 3 

oH 
o . 

- P* 

3, <c 
- i» 

? s 

-j CO 



t&'P-yl 



- t i i i i i i i rp i inii i-i j 1 1 ° ! i 



<Z5 

P5 



o 



a 

o 
o 

I 

.3 
g 



e 

S 



- 



I T I I 



be quite as good for cargo and " mixed " steamers as pulling out the old 
engines and fitting two turbines geared to the original screw shafting (tig. 45), 



TOO 



MANUAL OF MARINE ENGINEERING. 



as done by Sir C. Parsons in the s.s. " Vespasian." Fig. 46 shows such an 
arrangement proposed by this gentleman ; it is one quite easily carried out. 
and although it may be questionable if it is worth doing to an ordinary cargo 
boat, it certainly would be quite a good thing for very many of the combined 
cargo and passenger steamers designed . for long voyages at speeds from 
12 to 15 knots with single screws. The gain of 13-7 per cent, in power shown 
by him (in the schedule later on) is very material, and the alternative 15 per 
cent, of fuel important in all places, but very much so where coal is 20s. to 
30s. per ton delivered on board the ship in ordinary times. 

Oil Engines of moderate power are usually of the well-known single-acting 
type (figs. 17 and 49), having trunk pistons connected direct to the crank-pins 




Fig. 45. — Wheel-Gearing of s.s. " Vespasian " (Parsons). 

by ordinary connecting-rods and valve-gear wheel driven ; they seem to work 
very well at steady loads, and are handled by compressed air. Increases 
of power are obtained by increasing the numbers of cylinders, so that it is 
not at all uncommon to find eight cylinders in line operating on one line of 
shafting with a screw at the outer end. There is, of course, in this way 
no limit to the number of cylinders, and probably the cost of increase by 
this method is no greater than if the greater power were obtained with 
cylinders of large size*, sometimes cylinders are placed in tandem axially — 
that is, in rear of or above each cylinder is another operating with it on 
the same connecting-rod. 



OIL ENGINES. 



101 



Trouble, however, is experienced 
sometimes with these trunk pistons 
when of large size, so that recourse is 
had now to the piston-rod type when 
the cylinders increase in diameter ; at 
present 30 inches diameter* is con- 
sidered about as large as should be 
made for marine engines on the Diesel 
system, where the initial pressure is 
exceedingly high, especially if com- 
pared with the mean. It is sometimes 
as high as 40, but is generally 35 
atmospheres in these engines at 
commencement of the stroke, but, 
owing to the high compression 
effected, there is little or no shock 
at commencement of stroke. This 
engine, however, notwithstanding such 
initial pressures, has become popular 
in this country, as it had been for 
some time on the Continent, where 
it has proved successful when work- 
ing with that comparatively safe 
fuel heavy oil with a high flash point 
or residuals. 

The Design of Oil Engines was at 
first very similar to that followed for 
land engines. Now, however, the 
tendency is to conform to the marine 
practice found to be the best for 
steam engines ; their builders also 
adopt the enclosed type with forced 
lubrication, and so obtain good and 
safe running at high rates, of revolu- 
tion. Fig. 17 is a good example of a 
Thornycroft 100 H.P. marine petrol 
engine, the reversing of which is made 
by means of a clutch gear, which 
does well enough in small craft, 
being similar to the method by 
which motor-driven vehicles are re- 
versed. 

The Diesel Engine is generally 
designed to work on the usual 
four-cycle system, and its modus 
operandi is as already described 
('■. figs. 50 and 50a), but at- 
tached to it is a three-stage air 
compressor, which not only supplies 
* Vide Appendix A 




102 



MAM7AL OF MARINE ENGINEERING. 




5 

I—* 

X 

5 






3 



a 

o 

fl 

v 

.» 
d 

SO 

.E 

*3 

<! 

J2 

"So 

H 

CM 

«? 
a 
'5) 



I 

■i 



TWO-CYCLE DIESEL ENGINE. 10 



9 



the air for injecting the oil, but provides a means of working the engine and 
reversing it when required by using the oil cylinders as in an air engine, 
when, by special means, it is put out of action as an oil engine. The engines 
on the two-cycle system have also one or more low-pressure pumps (v. fig. 
50a) to supply the air for scavenging and filling the cylinders. In some 
marine designs these pumps are worked by levers, and like the ordinary 
air pumps of a steam engine. The high rate of compression of the 
air prevents shock on exploding the mixture of air and oil vapour, as would 
be the case if the load came on suddenly, as it does in the ordinary engine 
when there is no such compression. The ratio of maximum to mean 
pressure (4-31) is exceedingly high in all such engines, and consequently 
the rods, framing, crank shafts, etc., must be large by comparison with 
those of steam engines, it follows that the weight of these engines per horse- 
power is very great, and not much less than that of a steam installation 
including the boilers ; on the other hand, the consumption of oil fuel is only 
about 40 per cent, that in an oil-fired steamship. 

The double-acting cylinder in which explosives take place on both sides 
of the piston have yet to be proved equal to continuous service ; doubtless 
the pistons must be in that case water-cooled, and the stuffing-boxes most 
carefully made and maintained to be satisfactory. To compete with large 
powers the double-acting engine is desirable, but it still remains for engineering 
skill and resource to get over these practical difficulties. In time, too, the 
reversing may not be effected by quite such clumsy and indirect means as 
prevail at present with gearing or subsidiary compressed air arrangements. 
Fig. 49 is a sectional view of a Diesel engine as made by Mirrlees 
Bickerton Company for small power on the four-cycle system, and 
non-reversible. 

Double-acting Diesel two-cycle engines of considerable power have been 
made in large sizes on the Continent. Here the cycle with its compression 
and combustion follows its course as in the single-acting engine, but in 
this case on both sides of the piston, as in fig. 48. The power developed 
is thus practically doubled in a given size of cylinder ; or the same power 
obtained with about half the cylinder capacity. There are the usual practical 
objections to this system of overheating both pistons and cylinders, 
although they are water-cooled, and the difficulty with stuffing-boxes exposed 
to such high temperatures. Moreover, to obtain such high compression 
the clearance must be very small, for with 35 atmospheres at each end it 
must be less than 3 per cent, of the cylinder capacity, and consequently 
the stroke clearance about 2 per cent. — that is, with an engine of 10 inches 
stroke it is not more than ^u lnca at eacn en( J> an d must not vary materially 
from this at any time. 

The stuffing-box difficulty may be overcome by having hollow rods, 
through which water is passed to the pistons, and kept in circulation as 
Dr. Kirk did with steam for heating the L.P. pistons of the early compound 
engines made by J. Elder & Co. for the Navy. It is, however, very 
doubtful if the two-cycle double-acting engine will be satisfactory for even 
intermittent running, and there is reason to think it is unlikely to be so for 
continuous work on shipboard in large sizes. 

The Two-cycle Diesel Engine differs from the four inasmuch as the 
fresh air is admitted above the piston when it is at the bottom of the stroke 



104 



MANUAL OF MARINE ENGINEERING. 




a 






o 

u 

s 



"3 

a 



-a 
'3 



50 



3 
3 




after combustion, the products having begun to escape at six-sevenths the 
stroke are by it ejected and replaced, the scavenging orifice closing again 



THE FUEL CONSUMPTION. 



105 



one-seventh from the bottom ; the piston continuing its stroke compresses 
it to the end when the oil is sprayed in as before. In this way an impulse is 
made at every revolution, instead of at every other one as in the " Otto " 
cycle. This still further increases the power developed by a unit of cylinder 
capacity, but it likewise increases the heat production and the difficulties 
that arise from high temperatures. 

The Fuel Consumption of these engines is generally less than 04 pound 
of oil per horse-power, and in very favourable cases as low as 0-348 lb. 
Taking 10 lbs. of steam as a very good performance for a turbine, and 16 lbs. 

ft 




Fig. 49. — Single- Acting " Mirrlees-Diesel " Marine Oil Engine, 
of steam per pound of oil when burned to be produced in a good boiler, then 
Consumption of oil per hour of a turbine = 10 -f- 16, or 0*625 lb. 

If the consumption of the " Lusitania " be taken as 12*77 lbs. in the turbines 
alone, and 14*46 lbs. the total for all purposes, the oil fuel consumption 
will be 0*798 and 0*903 lb. per S.H.P., equal to 0*766 and 0*867 per I.H.P. 
respectively. 

Taking, however, the consumption of the " Otaki " with the combined 
turbo-reciprocators as 12 lbs. of steam per hour, the oil fuel for her would 



106 



MANUAL OF MARINE ENGINEERING. 



be 0*75 lb. Although this consumption of oil fuel is considerably greater 
than that of the oil engine, the amount of lubricating oil in the Diesel is 




a 

— 

CO 



C 
O 

60 

,C 

'-3 
o 

<! 

J2 

"so 

.s 

t/2 

_© 

"3 

& 

6 

is 
H 



« 

o 
-tfl 

.3 
to 



o 

.a 






very much higher ; in fact, excessive compared with that of a reciprocating 
steam engine of equal power. 

When oil engines are used on board sea-going ships, it is necessary for 



THE FUEL CONSUMPTION. 



107 



them to have power for steering and other purposes generated by an inde- 
pendent oil engine and electrically distributed. The whistle may be blown 




Hi 

CS 

ac 

■a 

CO 

u 
CD 
> 
CD 

W 

c 

e3 



s 





a 

a 

o 

o 

.a 
<! 

tao 

.5 

BE 



- 

X 

« 

o 

a? 

So 

a 

>» 

O 

6 

is 
H 



> 
a 

CD 

CD 

— 



3 
60 



with stored compressed air from the main engine compressor, so that there 
are no insuperable difficulties now these engines can be relied on to stop, start, 



108 



MANUAL OF MARINE ENGINEERING. 



and reverse quickly, and to run for a week on end without stopping to clean 
carburetters, etc., and certainly without stopping unexpectedly at awkward 
times from this latter cause. Mr. Westgarth thought it prudent to provide 
a small steam generator for such purposes above-named in the ship fitted 
by him with oil engines, so that there might not be so many novelties at 
one time. 

Gas Engines on board ship were pretty much of the same general design 
as the oil engine, but they require a gas producer, etc., to supply them with 
fuel, which adds to their weight and the space occupied by them. More- 
over, as Prof. Vivian Lewis very properly pointed out, " the suction plant 
suffers from the limitation that before it can achieve commercial success 
afloat a form of generator and scrubber, occupying small space, must be 
devised, in which bituminous coal can be used as the fuel to be gasefied, 
J and the gas supplied freed from all tar vapour. ... I am not aware 






1. 

Charged with Compressed 
Air, and Fuel being ad- 
mitted. 



Expansion completed and 
Piston about to open 
Exhaust. 



End of Down Stroke. 
Scavenge Valve open. 
Clearing out. 



Fig. 51. — Two«Cycle Oil Engine (Diesel System), showing the Three Stages. 



that it has yet been successfully done. The mechanical troubles of caking 
and arching of the fuel in the generator can be overcome, but many years' 
experience of efforts to decompose or get rid of tar vapour has impressed 
me with a great respect for the difficulties of the problem and a perfectly clean 
gas, absolutely free from tar vapour, is the first essential for success with the 
gas engine." 

All this is very true, though it must be a matter of regret, seeing that 
this country abounds in coal, but has very little oil. For this reason alone 
the adoption of the oil engine in British waters is scarcely politic, either in 
warships or cargo-boats. 



STEAM TURBINES. 109 



CHAPTEK IV. 

STEAM USED EXPANSIVELY. 

In the Reciprocating engine work is done by the elastic force of the steam 
acting on the pistons, and pressing them forward on their strokes against 
the back pressure behind them during its expansion from the time it enters 
the cylinder to the time it is allowed to escape to the free atmosphere in the 
case of a non-condensing or to the condenser of a condensing engine. 

From the time of entering to the time of cut-off expansion is taking place, 
though it is slight and often not appreciable ; after cut-off it is real, considerable, 
and quick ; it is continuous to the end, and the rate is expressed by the ratio 
of the capacity of the L.P. cylinder to that portion of the H.P. which was 
filled with steam at cut-off. This is nominally the rate, and would be really 
so if the admission valve were large and wide open till it closed and cut-off 
made suddenly, as is the case with the Corliss or " drop " valves. In practice, 
however, at the exact point of cut-off with the ordinary slide valves the 
steam is wire-drawn down considerably below the pressure in the valve 
chamber, which latter may be taken as the initial pressure. 

With Steam Turbines steam is admitted continuously instead of inter- 
mittently as in reciprocators ; the expansion commences immediately it 
enters the nozzles or their equivalents, and may be wholly carried out in one 
nozzle, and the whole velocity due to the fall from boiler pressure to exhaust 
pressure generated at once, and the kinetic energy expended on the blades 
of a single rotor, as shown in fig. 52. The well-known De Laval turbine 
is on this principle, and consequently runs at a very high rate of revolution, 
since the velocity acquired by steam in falling from 160 lbs. to 3 lbs. absolute 
(vacuum 24 inches) is 3,660 feet per second, and the peripheral velocity of 
rotor for this flow, if the efficiency is to be good, must be not much less than 
1,800 feet ; for a turbine of this kind with a rotor 38 inches diameter, the 
rate of revolution should be 180 per second, or 10,800 per minute ; as a matter 
of fact, a De Laval turbine of 300 H.P. has a rotor 30 inches diameter running 
at 10,600 revolutions per minute. 

The Expansion may be in Stages, however, and fig. 52a shows how this 
may be accomplished in a rudimentary way by causing the first rotor to run 
in a chamber where the pressure is below that at entry, but is somewhat 
above that at the final exhaust to condenser. The velocity of flow will then 
be much less than that stated, as the drop will be less (say to 25 lbs. abso- 
lute instead of 3). The steam from that chamber will pass into and through 
auother nozzle, where further expansion takes place with a renewal of velocity 
to the steam, thereby giving it further kinetic energy to be expended on 
the blades of a second rotor. In all cases of turbines, the rate of expansion 
is expressed by the ratio of the initial pressure to that at exhaust — that is, 
Pi + Vo = r. 

In the above example r — 160 -s- 3, or 533. 



110 



MANUAL OF MARINE ENGINEERING. 



Moderately Moist Steam expands in accordance with Boyle and Marriott's 
law — viz., whereby the pressure varies inversely as the volume — that is, 
pr = c. 

Then, 

mi 1 + hyp. log r 
lhe mean pressure = p x — ■ — — — . 

The hyperbolic logarithm of a number may be found by multiplying the 
common logarithm of that number by 2 "302585. 

There is a simplicity in this rule that commends it to the practical mind, 
and as steam in marine engines is usually fairly moist, it may generally 
be used to solve with sufficient accuracy the every-day problems connected 




~ . . ;:,y$L 





Fig. 52. -Turbo-Mofcoe 
(i>e Laval system). 



Steam from 
boiler 

Fir. 52a. — Compound Turbo-Motot 
(Do Laval system). 



with the marine engine. Such an expansion at constant temperature is called 
isothermal. 



The terminal pressure = 



_ Pi 



Therefore, 



YD 

(1) Ratio of mean pressure to terminal pressure = -— . 

.„ - Pi 



(2) Ratio of terminal pressure to mean pressure = 



rp m 



(3) Ratio of maximum pressure to mean pressure = — . 

(4) Dry steam is assumed to expand pv" = constant. 



STEAM TURBINES. 



Ill 



The following table will be found useful, and contains the multipliers 
for ascertaining the mean pressure of steam when expanding on Boyle's 
jaw, as well as when expanding adiabatically. 



TABLE XVII. — Steam used Expansively. 





l 


rp m 


_£l 


El 


Pm 


£»* 


/' 


r 


P\ 


rp n 


Vm 


Pi 


20 


0-050 


4-00 


0-250 


5-00 


0-1998 


0-186 


18 


0-055 


3-89 


0-256 


4-03 


0-2161 


0-200 


16 


0-062 


3-77 


0-265 


4-24 


0-2358 


0-220 


15 


0-066 


3-71 


0-269 


4-05 


0-2472 


0-230 


14 


0-071 


3-64 


0-275 


3-85 


0-2599 


0-242 


13-33 


0-075 


3-59 


0-279 


3-72 


0-2690 


0-254 


13 


0-077 


3-56 


0-280 


3-65 


0-2742 


0-258 


12 


0-083 


3-48 


0-287 


3-44 


0-2904 


0-271 


11 


0-091 


3-40 


0-294 


3-24 


0-3089 


0-292 


10 


0-100 


3-30 


0-303 


3-03 


0-3303 


0-314 


9 


0-111 


3-20 


0-312 


2-81 


0-3552 


0-340 


8 


0-125 


3-08 


0-321 


2-60 


0-3849 


0-370 


7 


0-143 


2-95 


0-339 


2-37 


0-4210 


0-408 


6-66 


0-150 


2-90 


0-345 


2-30 


0-4347 


0-417 


6-00 


0-166 


2-79 


0-360 


2-15 


0-4653 


0-450 


5-71 


0175 


2-74 


0-364 


2-08 


0-4807 


0-466 


500 


0-200 


2-61 


0-383 


1-92 


0-5218 


0-506 


4-44 


0-225 


2-50 


0-400 


1-78 


0-5608 


0-540 


4-00 


0-250 


2-39 


0-419 


1-68 


0-5965 


0-582 


3-63 


0-275 


2-29 


0-437 


1-58 


0-6308 


0-616 


3-33 


0-300 


2-20 


0-454 


1-51 


0-6615 


0-648 


3-00 


0-333 


2-10 


0-476 


1-43 


0-6993 


0-670 


2-86 


0-350 


2-05 


0-488 


1-39 


0-7171 


0-707 


2-66 


0-375 


1-98 


0-505 


1-34 


0-7440 


0-733 


2-50 


0-400 


1-01 


0-523 


1-31 


0-7664 


0-756 


2-22 


0-450 


1-80 


0-556 


1-24 


0-8095 


0-800 


2-00 


0-500 


1-69 


0-591 


1-18 


0-8465 


0-840 


1-82 


0-550 


1-60 


0-626 


114 


0-8786 


0-874 


1-66 


0-600 


1-51 


0-662 


1-10 


0-9066 


0-900 


1-60 


0-625 


1-47 


0-680 


1-09 


0-9187 


0-913 


1-54 


0-650 


1-43 


0-699 


1-07 


0-929:. 


0-926 


1-48 


0-675 


1-39 


0-718 


1-06 


0-9405 


0-940 



When steam expands in accordance with the law p v = constant, the 
curve drawn through the extremities of ordinates representing the pressure 
at any position of the piston is a hyperbola. The mean height of such a 
system of ordinates is found by the formula given above ; this mean height 
will represent the mean pressure. 

The mean pressure may be obtained without the aid of logarithms, by 
resorting to arithmetical calculation of the ordinates, and finding the mean 
by the method usually followed with indicator diagrams. 

Example. — Initial pressure 80 lbs., rate of expansion 5. Suppose the 
length of stroke divided into ten equal parts by points 1, 2, 3, . . . 9, 
10. The cut-off is !. or two-tenths the stroke. 



112 



MANUAL OF MARINE ENGINEERING. 



The pressure at commence of stroke is 

1 tenth 
2 
3 
4 
5 
6 
7 
8 
9 
10 



>> 
»» 
»> 

ii 
ii 



80 lbs. 

80 „ 
80 .. 



f of 80 or 53-33 
4000 
32-00 
26-66 
22-86 
20-00 
17-78 
16 00 



2. 

* 
2 

T 
i! 

ii 



5- 
2 
J 



)> 
)) 
»> 
)1 
II 
>> 
)) 



J>» - 



_ 1 



( — *- + 80 + 80 + 53-33 + 40 + 32 + 26-66 + 22-86 + 20 + 17-78) 



= 42-063 lbs. 



By reference to Table xvii., with 5 for the rate of expansion. ^~ = - 5218, 

and for 80 lbs. pressure p m = 41*744 lbs., or about 0*75 per cent, less than 
that given by the summation above, the excess of which is clue to the small 
number of points of observed pressure. 

Graphic Method. — Professor Rankine proposed * a geometric method of 
ascertaining the mean pressure, which is of interest and value, and which 
first appeared in the Engineer. Draw a straight line, A B, of definite length, 

A B 

produce it to A C, so that A C = —r-. Through 

A draw A D at right angles to CAB. With C as 
centre, and C B as radius, draw the arc of a circle, 

D A 

cutting AD at D. Then if =r-g; is the rate oi 

expansion — = -r-^z. 
Pl AB 

To suit this diagram for actual use, A B should 

be taken of such a length as is convenient for scale 

Fig. 53. measurement, say 4 inches ; A D should be divided 

into ten parts and subdivided into quarters ; through 

the divisions faint lines should be drawn parallel to A B. Scales should be 

constructed 4 inches long, suitable for the usual pressures coming under 

consideration. 

The object of columns 3, 4, in Table xvii., is that mean pressure and 
initial pressure may be easily determined from terminal pressures, when 
the rate of expansion is known ; or when initial and mean pressures are 
known, the rate of expansion may be found. Column 5 is given to show 
the relation between maximum and mean pressures at the various rates of 
expansion. 

Adiabatic Expansion. — When steam is dried by slight superheating, so 
that it is surcharged with heat, and is capable of very considerable expan- 
sion without liquefaction taking place, it expands according to the law of 
perfect gases, and then 




♦Rankine, Rules and Tables p. 291. 



STFAM TURBINES. 113 

r " may be found by extracting the square root of — four times. 



r-*-VVVVF. 



Column 6 gives the value of — as calculated from the above formula 

T\ . 
It will be seen that, except at very high rates of expansion, there is no very 
great difference between the ratio as calculated by this method and by the 
method for moderately moist steam. 

Clearance. — In practice the mean pressure in the cylinder is very materi- 
ally affected by what is called clearance. However accurately the engine is 
constructed, there is always at the commencement of the stroke a space 
between the piston and cut-off valve, made up of the part of the cylinder 
between the piston and the cover or cylinder end, and the passage between 
valve face and cylinder ; this is called the clearance. Supposing this space is 
equal to one-tenth of the capacity of the cylinder, and the cut-off is at two- 
tenths the stroke, the effective cut-off is not two-tenths, but something more, 
due to the fact that the expansion of a volume of steam equaling three-tenths 
the capacity of cylinder is being effected, instead of that of a volume of two- 
tenths. This practically amounts to making the cylinder 10 per cent, longer, 
and cutting off at three-tenths the stroke without clearance. It is, there- 
fore, customary to speak of the clearance as equal to a certain fraction 
of the stroke. This, however, must be distinguished from the lineal clearance 
or distance of the piston from the cylinder ends while at the extreme 
limits of its stroke. This should be expressed always as a definite length, 
and not as a fraction of the stroke ; it may be, and often is, called stroke 
clearance, while the space is volume clearance. 

To allow for the effect which the clearance will have when steam expands 
in a cylinder, let r be the nominal rate of expansion as before, and r 1 be the 
actual rate allowing for clearance, c the clearance as a fraction of the cylinder 
capacity. Then 

1 

1 r J l + • /AN 

- = , — and r, = r ■ ... (A) 

7-,1-f-c * 1+cr y 

- + c being the volume of steam at cut-off between the piston and the cut-off 

valve, and which expands to the volume 1 + c at the end of the stroke. If 
there is no cushioning of the steam before admission, then the whole of the 

space - + c must be filled at each stroke with fresh steam. 

Then the real mean absolute pressure will be 

V'm ~ C iP\ ~ Pm) • - ' • ( B ) 

p' is the mean pressure obtained by means of Table xvii., the actual rate of 

8 



114 



MANUAL OF MARINE ENGINEERING. 



expansion being taken, and p 1 is the absolute initial pressure. If, however, 
there is sufficient cushioning to fill the clearance space with steam at the 
initial pressure, the volume of steam used at each stroke will be only that 

swept by the piston at cut-off and equal to — . 

Compression or Cushioning. — There will be an increase of back pressure 
caused by this cushioning, and its effect on the mean pressure is as follows : — 

Let p' m be the mean absolute pressure due to the effective cut-off — ; p the 

absolute back pressure ; c the clearance ; and p l the absolute initial pressure 
as before. 




Fig. 54. 

Fig. 54 is the indicator diagram of an engine working under these condi- 
tions. A B is the stroke, A C the clearance, E F the nominal cut-off. and 

C B 
D F the effective. The actual rate of expansion is therefore -^-= C D 

D F 

represents the initial pressure and H K the back pressure. Cushioning com- 
mences at K with pressure p , and at A the pressure is p v 

The figure C D E H K is the diagram due to the expansion of steam of a 
volume equalling c at a pressure p, to a pressure p M so that the rate of its 
CK 



expansion is 



DE. 



Now p x x D E = p x C K, and, therefore, 
ft CK 



Po DE" 
Since p x and p are both known the rate of expansion is known, and by 
referring to Table xvii.the mean pressure p" m due to this rate of expansion is 
found. 

Ihe area H E F G = area CDFGB - (area CDEHK + area H K B) 
CDFGB = p' m x (AB + AC) = p' m {\ - c). 

HKB =» (AB-AK) = Po (l-( c - Pl -4 

\ Po ' 

„ CDEHK = P \(CK) =p'J&c). 


„ H E F G = p m x 1, or the effective mean pressure. 



HJ * 



STEAM TURBINES. 115 

Therefore, the effective mean pressure 

-/.<i*«-{i-.(*-i)}*-*.(£-.) 

= P« X « (P'n. ~ Po) + C Pl (l - P y\ 

= fa'*. " Po) (1 + C) + Oft (l - £=). - - - (C) 

General Effect of Clearance and Cushioning. — Let p', the absolute initial 
pressure, be represented in Fig. 54 by C D, p the back pressure by B L, A B 
the length of stroke, A C the clearance c, A K the compression a;,EF the 
nominal cut-off, r the nominal rate of expansion, r x the real rate of expan- 
sion, &c., (fee, as before. p' m the mean pressure due to an initial pressure ;/, 
and a rate of expansion r r ; p m the real mean pressure of N E F G L H. As 
before 

1 + c 1 + c /1X 

r, = = r- . (1) 

1 1 1 + cr 

- + c 
r 

Since the steam at point K is shut up in a space x + c, and is compressed 

OS "4" C 

into a space c, the rate of compression is ; and the pressure after com- 

OC "I - C 

pression at N is p e = p , and represented by ANorCM; let p c m be 

c 

the mean pressure of the figure MNHKC, which is that due to a pressure 

x + c x + c 
■ p , and a rate of expansion -. 

The area NEFGLH = CDFGB - DENM - MNHKC - KHLB. 

NEFGLH = Pm x 1. 
CDFGB = p' m (1 + c). 

DENM = (p' - Po ^-p) c = (p'~ Po ) c - p x. 

MNHKC = p' m (x + c). 
KHLB = p (1 - x). 

Therefore, 

P m = P' m (1 + c) - {(p' - p ) c - p x} - p° m (x + c) - p (l-x) 
= p' n (1 + c) - p'c - p (1 - 2x - c) - pi (x + c). - (2) 

Example I. — To find the effective mean pressure in a cylinder having a 
clearance space equal to one-seventh its capacity, the initial pressure 80 lbs. 
absolute, the back pressure 10 lbs. absolute, and the nominal cut-off -g- the 
stroke : 

CD If no compression 

r = 5 ^-±-i = 5 x » = 3-33. 
1+5 ia 

By reference to Table xvii.— - = 0-6615 for a rate of expansion = 333 



H6 MANUAL OF MARINE ENGINEERING. 

Then 

p' m = 80 x 0-6615 = 52-92 lbs., 
and 

The effective mean pressure = 52-92 - 10 = 42-92 lb* 

(2) If full compression to 80 lbs. : Here rate of compression 

r ■ «&•« 

° Po 
Therefore, 

Pi r p e m 



80 8 



= 3-08 (Table xvii.) 
Po Pi 
p' m = 52-92 lbs. as before. 

Then the effective mean pressure by formula (C), p. 115, 

= (52-92 - 10) (1 + |) + so (l - 3-08). 

= f x 42-92 - ^j^ = 25-28 lbs. 

If there was no clearance, the effective mean pressure would be 41-74 - 10, 
or 31-74 lbs. 

The s'eam used in the case (2) is the same as if there had been no 
clearance, and as the effective mean pressure was only 25-28 lbs., there is a 
loss due to clearance of 6-46 lbs., or 20 per cent. In case (1) the quantity of 
steam used is ^| the volume of the cylinder per stroke, or one-seventh of the 
volume in excess of the quantity with no clearance, so that with this increase 
of steam, if there was no clearance and the rate of expansion 5, there should 
be an increase in the work done, and that increased work will be to the work 
done by the smaller quantity of steam as 12 is to 7. 

The equivalent mean effective pressure is then X T 2 of 31-74, or 54-41, as- 
against 42*92 lbs. which was obtained, showing a loss of 11-49 lbs., or 21 
per cent. 

The example given is a very extreme case, and such as would be rarelv 
found in practice. The effect of clearance in the high-pressure cylinder of a 
compound engine may be seen in the following : — 

Example II. — The nominal rate of expansion is 2, the initial absolute 
pressure 90 lbs., and the absolute back pressure 22\ lbs., the clearance being, 
one-ninth the capacity of the cylinder. 

(1) No compression : 

1 + i 
r = 2^4 = 1-82. 

1 + f 

By reference to Table ix., — = -8786 for a rate of expansion of 1-82. 

Then ,) m = 90 x 0-876 = 79 lbs. 

Mean effective pressure = 79 - 22-5, or 56-5 lbs. 

The equivalent mean pressure when \ + \ or \\ of the volume of the 
cylinder of steam is used will be y- of 5368 lbs., or 65-61 lbs., showing a 
loss by clearance of 13-88 per cent. 

(2) If full compression to 90 lbs. : 

Here - 1 = 4, which is the rate of compression ; so that 

?JL«^J?= 2-39. • - - Table xvii. 
P Pi 



MEAN PRESSURE IN A COMPOUND ENGINE. 117 

The effective mean pressure by formula (C), p. 115. 

= (79 - 22-5) (l + ») + ^ (l _ 2-39) 
= 02-77 — 13-9 = 48-87 lbs. 

Thus showing a loss of 5*81 lbs., or 10*8 per cent. only. The loss from the 
clearance in a compound is not so serious as in the expansive engine, and 
in the H.P. and M.P. cylinders of a triple or quadruple engine is of no conse- 
quence, as the steam in the former (which has passed from the high-pressure 
cylinder without giving out its full work), will do more work in the medium- 
pressure and low-pressure cylinders ; whereas, with the expansive engine, the 
■exhaust steam passes direct to the condenser at a higher pressure than if there 
is no clearance. Further, since the cut-off in an expansive engine is much 
earlier than in a compound, and the clearance from practical considerations is 
very much the same, the ratio of clearance to volume at cut-off will be much 
higher in the former than in the latter. Considerable loss is, however, ex- 
perienced in compound, triple, and quadruple engines if the clearance in the 
low-pressure cylinder is large, therefore in that it should be as small as possible. 

The beneficial effect of cushioning is seen in both the preceding examples, 
but its value is greater still when the cut-off in the high-pressure cylinder is 
somewhat earlier, as may be seen by the following : — 

Example III. — The cut-off in the cylinder of example (2) is altered to 
^ the stroke, so that the nominal rate of expansion is 4. 

(1) No compression : 

14-1 
r — \ ~ 9 — 3 

Then p m = 62*1 lbs. 

and the effective mean pressure = 62T — 22*5 = 29'6 lbs. 

The equivalent mean pressure due to the amount of steam used is now 
" of 31 '185, or 45 lbs., thus showing a loss of 12 per cent. 

(2) If full compression to 90 lbs. 

The effective mean pressure by formula (C) 

= (62T - 22-5) *■£- + ^ (1 - 2-39) 
= 44 - 13-9 = 30-1 lbs. 

Thus showing a loss of T085 lbs., or 3*4 per cent. only. 

The economy effected by working with a considerable amount of cushioning 
is. therefore, very appreciable, and experience has proved the correctness of this. 

In actual practice, however, it only happens that so much cushioning can 
be effected as to fill the clearance space with steam of pressure equal to that 
entering in the H.P. cylinder of a triple or quadruple engine which exhausts 
steam of high pressure ; but still even what is conveniently obtained materially 
adds to the economic working of the engine. It must not, however, be over 
looked that the effective mean pressure is considerably reduced by cushioning. 

Mean Pressure in a Compound Engine. — If the effective mean pressure in 
the high-pressure cylinder of a compound engine be divided by the ratio of 
capacity of low-pressure to that of the high-pressure cylinder, the quotient 
represents the mean pressure necessary to do the same work in the low- 
pressure cylinder as is effected in the high-pressure cylinder. If this be 
added to the effective mean pressure in the low-pressure cylinder, the sum 
will be the mean pressure necessary to obtain from the low-pressure cylinder 



118 MANUAL OF MARINE ENGINEERING. 

alone, the whole work done by both cylinders, and may be called the equiva- 
lent mean pressure. If there be no loss of mean pressure, owing to drop in 
the receiver, or other cause, this equivalent mean pressure will be the same 
as the effective mean pressure obtained by the steam expanding in one 
cylinder at the same rate as the total expansion effected in both cylinders of 
the compound engine. In the two-cylinder receiver form of compound 
engine, there is sometimes a considerable fall in pressure from the release 
point to the exhaust, owing to the low pressure maintained in the receiver, 
or to the late cut-off in the H.P. cylinder. 

il) Two-cylinder receiver compound engine. 
.<et p i be the initial pressure, p Q the back pressure in the low-pressure 
cylinder, p r the pressure in the receiver and back pressure in the high- 
pressure cylinder ; R the ratio of cylinder capacities; r the total rate of 
expansion, r : the rate of expansion in the high-pressure cylinder, and r„ that 
in the low-pressure cylinder ; p' m the mean pressure due to an initial 
pressure, p 1 , and a rate of expansion, r x : p" m , the mean pressure due to an 
initial pressure p r , and a rate of expansion r„. P m is the mean pressure due 
to a rate of expansion r, and an initial pressure p l : 

The effective mean pressure in the high-pressure cylinder is then 
(p'm - Pr) \ and that in the low-pressure cylinder is (p" m - p ). 
Also _ 1 + hyp. log. r 

*- ~ Pi ~ ' 



Pm=Pl 



Pm = P, 



1 + hyp, log. r T 
» 

1 + hyp. log. r.. 



r 2 



Since the work performed in the engine is supposed to be equally divided 
between the two cylinders, 

Pm ~ Pr= R(j>"m ~ Pdh " " " M 

But if there be no loss due to " drop," and the mean pressure in the high- 
pressure be referred to the low-pressure cylinder, then 

^i* + (/. - Po) = P- " Po- 

By suostituting the value of (p' m - p r ).ot (1) in the above 

P"m - PO - <*. ~ P0)h \ 

and R } - - (2) 

P'm ~ Pr " ( P m - Po)-2 I 

Let x be the efficiency of the system, so that (1 - x) is the proportion of 
lo.^s due to drop. Then 

P'm - Po = x ( p ™ - Po)h | 
inc R > • - - (3) 

P'm -Pr = x(?m " P ) 3 ) 

To find the actual mean pressures when there is loss due to " drop," the 
value of x must be determined ; this may be done by substituting the value 
of p' m and p" ms found by the preceding formulae ; but an approximate value 
may be found by determining the value of p r in equation (3) ; from the 
"j,lue thus found calculate p" m , and refer the mean pressures of both 



F m - Po = 12-36 + ~ 



MEAN PRESSURE IN A COMPOUND ENGINE. 119 

cylinders to the low-pressure cylinder. If (P' m - p Q ) be the equivalent mean 
pressure thus found, then, approximately, 

Pm ~ Pq . 

P m " Po " W 

Example. — To find the mean pressure in a compound engine using steam 
of 90 lbs. absolute pressure, the total rate of expansion being 7, the ratio of 
the cylinder capacities 3-5, and the back pressure 4 lbs. 

P m = 90 x 0-421 = 37-89 lbs. - - Table xvii. 

r = 7 + 3-5 = 2. 
Then 

p' m - 90 x 0-8465 - 76-18 lbs. - - Table xvii. 

Pr = 76-18 - ^ (37-89 - 4) = 16-88 lbs. 

_p r r_ 16-88 x 7 _ 
r * ~ p, ~ 90 " l 313 ' 

jT m = 16-88 1 + hyP- log. 1-313 = ^ 

1 O I o 

That is, the effective mean pressure in high-pressure cylinder is 
76-18 - 16-88, or 59-3 lbs., and that in low-pressure cylinder is 16-36 - 4, 
or 12*36. Referred to low-pressure cylinder alone, 

59-3 
7 5~ 

P m _ p Q = 37-89 - 4 = 33-89 lbs. 
Therefore, 

--ana- °' 865 - 

Then, actual effective mean pressure I _ . 865 (37 . s9 _ „, _ u , 65 lbs 
in low-pressure cylinder J x '•* 

And the actual pressure in receiver) _ 7fi , fi ?jL/9Q-3\ 24-88 lb* 
is then ....-( 2 ^ 

(2) 2'^e three- cylinder receiver compound engine, having two low-pressure 
cylinders. Ratio of each low-pressure cylinder to the high-pressure 
. R 

In this case only one-third of the work is done in each cylinder. Then 

R 

Pm ~ Pr = -S (Pm ~ P )> " (1/ 

and as 

" " R ^ + {j> " m ~ Po) = Pm " * 
Then 

P"m - 7>0 = I ( P ™ ' Po) ) 

and /9\ 

V m " Vr = -o (Pm " 7Po)J 



120 MANUAL OF MARINE ENGINEERING. 

Also the actual values 

P"~ ~ Po = |( p «- Po) x \ 

R >--•(») 

p' m - p r =j(P n - p )x\ 

Example. — To find the mean pressures in a three-cylinder compound 
engine (having two low-pressure cylinders), using steam of 90 lbs. absolute 
pressure, the total rate of expansion being seven, the ratio of the combined 
capacity of the low-pressure cylinder to that of the high-pressure being 3-5, 
and the back pressure 4 lbs. 

P m = 90 x 0-421 = 37-89 lbs., 
r, = ^ = 7 -*- 3-5 = 2, 

p' m = 90 x 0-8465 = 76-18 lbs., 

p r = 76-18 - ~ (37-89 - 4) - 36-64 lbs., 



]> T r 36-64 x 7 



r 2 =^=^7^ =2-85, 



f n= 36-64 1 + teft 2 ' 85 = 26-4 

Then the mean effective pressure in the high I -,. , D o^» r> , nn * . ,, 

i- j f = 7o*lo - dbo4 = 39*54 bs. 

pressure cylinder J 

Then the mean effective pressure in the low ( a „ . n ., . .. 

1- i > = *o*4 - 4 = 22-4 lbs. 

pressure cylinder J 

Then 

P'*, - Po = 22-4 + i*||* = 32-7 lbs. 

Then 

x = 32-7 -*- 33-89 = 0965. 

Then actual effective mean ) or 

pressure in high-pressure > = -5- (37*89 - 4) x 0*965 = 38-15 lbs. 
cylinder- - - - ) 

Then actual effective mean ] 9 

pressure in each low- V = ^(37-89 - 4) x 0-965 = 21-8 lbs. 
pressure cylinder - - j 

(3) The three-cylinder compound continuous expansion engine, having one 
high-pressure, one low-pressure, and one medium-pressure cylinder, generally 
called triple-compound. 

K is the ratio of low-pressure to high-pressure cylinder ; R : the ratio of 
low-pressure to mean-pressure cylinder; p the initial pressure, &c., as before. 

p' m the mean pressure due to expansion, r v and pressure p', 

P m it ■>■> >» **2> >> P > 

P m J> » |] P3, ,, P . 

p" is the pressure in the receiver between high-pressure and mean- 
pressure cylinders, p"' that in the receiver between mean-pressure and 
low-pressure cylinders. 



MEAN PRESSURE IN A COMPOUND ENGINE. 



121 



Then effective mean pressure in high-pressure cylinder = p' m - p", 
„ „ mean-pressure „ = p" m - p", 

low-pressure „ = p'" m - p 



But 



Then, if there is no loss due to drop, 

p' m ~ P = R (P'" m - P°) and p" m - p" = R x (p'" m - p»). 



(i) 



Therefore 



Ri 



R 



P - p° 
P' m -P°=^^- 

p m -p"'=^(P n -p°) 

Pm-p-^Pn-p*) 



r-) 



This is true when there is no loss from " drop," but, as in practice there 
is generally some loss from this cause, an approximation must be found in a 
similar way to that for the two-cylinder compound engine. 

The cut-off in the high-pressure cylinder will be, as before, 

1 R 



The cut-off in mean-pressure cylinder in order to maintain a pressure, />", 

in the receiver between it and the high-pressure cylinder can be found in the 

same way as before. Since R is the ratio of low-pressure to high-pressure 

. R . 

cylinder, and Rj that of low-pressure to mean-pressure cylinder, will be 

the ratio of the mean-pressure to high-pressure cylinder, then 

R 



and 



p tx p 

r* X R^ ~ K 



± = h«K. 



R 



1 

X — • 



p 



1 \ p 

Substituting the value of — . Then, — = Rj -V- ■ 

r \ r 2 P r 

The cut-off in low-pressure cylinder to maintain a pressure, p"\ in the 
receiver between it and the mean-pressure cylinder, 

1 1 



Ri p" 



l 

x — • 



1 1 p' 

Substituting the value of — . Then, — = -^-. 

r 2 r 3 p r 

But since the terminal pressure in the low-pressure cylinder will be 
that due to an initial pressure, p, and a rate of expansion, r. Then 



E. 
r 



P 



or — = 



p r 



122 MANUAL OF MARINE ENGINEERING. 

Therefore. 

R 

Cut-off in high-pressure cylinder = — 

T 

„ mean-pressure „ = R x ^~ \ - - (3) 

p r 

„ low-pressure „ = -4tt— 

To avoid any lengthy or elaborate calculations, a result sufficiently 
accurate for practical purposes may be obtained by assuming a value 
for x, and using it only in the first step of the calculation. This value 
will vary from 1*0 to 0-9 in well-proportioned engines of this class, when 
the steam pressure is not less than 120 lbs. absolute, and the rate of 
expansion not less than 10 times. 

Example.*— To find the mean pressures in the three-cylinder continuous 
expansion engine, using steam of 120 lbs. absolute pressure, and expanding 
12 times. The ratio of low-pressure to high-pressure cylinder being 6, and 
of low-pressure to mean-pressure cylinder, 2 "5 ; the back pressure in low- 
pressure cylinder being 4 lbs. : 

Assume x = 0*9. 

P m = 120 x 0-2904 = 34-85 lbs. ) T „ ,. 
p' m = 120 x 0-8465 = 101-58 lbs. f laDle 
p' m - p " = • (34-85 - 4) x 0-9 = 55-53 lbs. 



Therefore, 
Now, 

Then, 



p" = 101-58 - 55-53 = 46-05 lbs. 

- = 2 " 5 x ^nl 20 to = ' 5U > orr 2= 1-838. 
r 2 46-05 x 12 

„ 1 + hyp. log. 1-838 ' , 

v . - v rsss = 38 lbs - 

If the work performed in the second cylinder is to equal that done in the 
first, then 

p" m - p'" = ^!x 55-53 = -^ x 55-53 = 2314 lbs. 

Then, 

p" = 38 - 23-14 = 14-86 lbs. 

7 3 = 14-86 2Q x 12 i or r,» 1-486. 

P'" m = p" L^Z2_^I^ = 13 . 96 lb, * 

p'" m - p = 13-96 - 4 = 9-96 lbs. 

Therefore, the mean pressures are 55-53 lbs., 23-14 lbs., and 9-96 lba. 
Referred to the low-pressure cylinder, 

F m - Po = 9-96 + *£g + 55 ^ = 28-471, 
P m - p Q = 30-85. 



WIREDRAWING OF STEAM. 12$ 

Therefore, 

x _ 28H1 _ . 923 
" " 30-85 ° 9i6m 

So that if the work is exactly equally divided between the cylinders, then — 
Mean pressure in low-pressure cylinder 

= ^-^> 0923 = ^ x 0-923 = 949 lbs. 

Mean pressure in mean pressure cylinder 

= 5i (P m - p ) 0-923 = 2 3 5 x 30-85 x 0923 = 23 72 lbs. 

Mean pressure in high-pressure cylinder 

R 

= -3 (P» - i> ) 0-923 = 4 x 30 " 85 * 0923 = 5694 lbs. 

Actual Mean Pressure in Practice. — In the preceding pages, the mean 
pressure spoken of is only such as would be obtained from a perfect engine 
in which steam is dry and expanding at a constant temperature, and as such 
is what may be called the theoretical mean pressure. In an actual engine, 
however carefully designed, manufactured, and worked, there will be certain 
causes of loss of pressure, so that the actual indicator-diagram will show a 
mean pressure considerably less than that due to the initial pressure and 
the rate of expansion, allowing that during expansion work has been done. 

The following are the principal causes of loss of pressure in the cylinder 
of a marine engine : — 

(1) Frictional Resistance in the Stop-valves on the Boiler and Engine, 
and in the Pipes connecting these. — If the initial pressure is taken as that 
in the valve case, of course this particular loss does not affect the indicator- 
diagram at all. If the stop-valves are opened to the extent of one-quarter 
of their diameter, and the steam-pipe is fairly straight and short, and of 
sufficient diameter, so that the flow of steam at any point does not exceed a 
velocity of 8,000 feet per minute : the loss of pressure at the valve-case will 
be very slight, and not exceed 2h per cent. If the capacity of the valve-case 
is nearly equal to that portion of the cylinder filled at the cut-off point, the 
loss will be still less, as the case then acts as a reservoir in which steam is 
stored between the cut-off and admission periods. 

(2) Friction or Wire-drawing of the Steam during admission and cut-off.— 
This is one principal cause of loss of pressure in most marine engines, and is 
generally due to defective motion of the valve-gear, combined with small 
steam ports and passages. If the opening to steam during admission is 
small at the most, and the valve closes slowly, large passages and ports are 
of no avail ; and, on the other hand, if the passages are too contracted, there 
will be considerable loss of pressure in the cylinder, however efficient the 
valve-gearing may be. But the slow and limited motion of the ordinary 
slide-valve itself is the most serious obstacle to the obtaining of good diagrams. 
The slow opening of the valve causes no loss, as the piston speed is low at 
that period. A perfect valve should open wide enough to allow the steam 
to pass at a velocity of 8,000 feet per minute, and remain open until cut-off, 
which should take place quickly ; the valve should remain closed until very 
nearly the end of the stroke, when it should open quickly and wide to exhaust ; 



124 MANUAL OF MARINE ENGINEERING. 

the slow closing to exhaust, and sjow opening to lead, are of no consequence, 
and cause no practical loss. The loss of pressure from these causes with 
engines having common slide-valves, and the ordinary link-motion, is con- 
siderable ; especially is this the case when steam is cut-off early in the stroke, 
by setting the main valves with very little lead, and having only single ports. 
As has already been stated, however, the steam becomes superheated by the 
friction ; it is, therefore, a little more efficient during expansion than it 
would otherwise be. 

When cut-off is effected by means of special valve-gear, or by a separate 
cut-off valve, the pressure at cut-off is very little below that in the valve-casing, 
and sometimes very nearly equal to it ; when effected by the ordinary single- 
ported slide-valve and link motion, the pressure at cut-off is sometimes as much 
as 15 per cent., and seldom less than 7^ per cent, below that in the valve-case. 

(3) Liquefaction during Expansion, due partly to the cooling action of the 
walls of the cylinder and the passages, is a frequent source of loss of pressure, 
and this was especially so in expansive engines working with moist steam 
in unjacketed cylinders ; it is observable also, though in a smaller degree, 
in all compound engines working under similar circumstances. 

(4) Exhausting before the Piston has reached the end of its stroke, although 
conducive to the good working of a fast-running engine, will show a loss of 
pressure in the indicator-diagram. The loss from this cause is, however, more 
imaginary than real ; but it must not be forgotten that the I.H.P. will be there- 
by less, which is important when the I.H.P. is deemed essential in the contract. 

(5) Compression and Back Pressure due to " Lead " also tend to reduce 
the mean pressure of the diagram when compared with the theoretical mean 
pressure. But these are both essential to the good working of an engine, 
and (as has been shown in a previous part of this chapter) compression tends 
to balance the loss due to clearance. 

(6) Friction in the Ports, Passages, and Pipes, between cylinder and. 
cylinder and condenser, produces a loss of pressure, and, although not large 
when the velocity through them does not exceed 6,000 feet per minute, 
sometimes amounts to 2 or 3 lbs. in badly designed engines. 

(7) Clearance has been shown to serve to increase the mean pressure 
'beyond that due to the nominal rate of expansion, and therefore cannot be 
reckoned as a source of loss, unless the equivalent cut-off is taken to obtain 
the rate of expansion. 

It will be seen, then, that the actual mean pressure expected to be 
deduced from the indicator-diagram of an engine depends very much on 
the proportion and arrangement of the cylinders and their valves, etc., and 
in calculating the expected mean pressure from the theoretical mean pressure, 
due allowance must be made in each individual case. 

If the theoretical mean pressures be calculated by the methods laid 
down in this chapter, and the necessary corrections made for clearance and 
compression, the expected mean pressure may be found by multiplying the 
results by the factor in the following Table xviii. 

If no correction be made for the effects of clearance and compression, and 
the engine is in accordance with general modern practice, the clearance and 
compression being proportionate, then the Theoretical Mean Pressure may 
be multiplied by 0"96, and the product again multiplied by the proper factor 
in Table xviii., the result being the expected mean pressure. 



CLEARANCE. 

TABLE XVIII. 



125 



Particulars of Enoink. 



(1) 

(2) 
(3) 

(4) 

(5) 
(6) 



Expansive engine, special valve-gear, or with a separate cut-off 
valve, cylinders jacketed, ....... 

Expansive engine having large ports, etc., and good ordinary 
valves, cylinders jacketed, ....... 

Compound engines, with ordinary slide valves, cylinders jacketed, 
and good ports, etc., ........ 

Compound engines as in general practice in the merchant service, 
with early cut-off in both cylinders, without jackets and expan- 
sion valves, ......... 

Triple- and quadruple-compound engines, with ordinary slide- 
valves, good ports, unjacketed, moderate piston speed, 

Fast-running engines of the type and design usually fitted in war- 
ships, and express with fast-running engines, .... 



Factor. 


0-94 




0-9 


to 0-92 


0-8 


to 0-85 


0-7 


to 0-8 


0-65 to 0-75 


0-6 


to 0-7 



Example. — To find the expected mean pressure in the cylinders of a 
marine engine using steam of 60 lbs. absolute pressure, the rate of expansion 
4, the clearance equal to one-tenth of the cylinder, and the pressure in the 
condenser 2 lbs., the valve-gearing specially adapted for an early cut-off, and 
the ports, passages, etc., of ample size ; compression commences at § of 
the stroke. The cylinders are jacketed. 

The effective rate of expansion is 

. 1+0-1 
r, = 4 x : — = 3-143. 



1+4x0-1 



P' = 



0-25 + 01 
01 



x 2 = 7 lbs. 



and the rate of expansion 3-5. Then, 

„ _ 1 + hyp. log. 3-5 . , ,, 
p o m = 7 x J * s =4-5 lbs. 



= 60 



3-5 

1 + hyp, log. 3-143 
3-143 



- 41 lbs. 



Expected mean pressure - 41 (1 + 0-1) - 60 x 01 - 2(1 + 0-5 - 0-1) 

- 4-5 (-25 + 0-1) = 35-3 lbs. 

If the effects of clearance and cushioning be neglected, the mean pressure 
= 60 X 0*5965 — 2, or 33-8 lbs. This is less than the result obtained by 
the more accurate calculation in this case, because the cushioning is small 
for so low a back pressure when compared with the clearance. 

The mean pressure in practice will be found now by multiplying 35*3 lbs. 
by 0-94, and is therefore 33*18 lbs. 

Example. — To find the expected mean pressure in the cylinders of a 
compound engine using steam of 100 lbs. absolute pressure, the cut-off in 
both high-pressure and lew-pressure cylinders being at half stroke ; the 
clearance in both cylinders is equal to one-tenth of their net capacity ; the 
pressure in the condenser is 2 lbs. ; the cylinders are jacketed, and the ports. 
etc., of ample size, no expansion valves. Compression commences at f the 
stroke. Cvlinder ratio 4. 



126 MANUAL OF MARINE ENGINEERING. 

Here the effective rate of expansion = 2 ^ - — - = 1-83 

r 1 + 2x0-1 

The theoretical pressure in the receiver = - — — x — = 27-3 lbs. 

l-o2 4 

The expected pressure in receiver =27-3 x 0-85 = 23*2 lbs. 

The steam is compressed in high-pressure cylinder to 

p. = i + tV x 23-2 = 81-2 lbs. 

1 + 1 

The rate of compression = * — ^^ = 3*5. 

its 

The mean pressure due to a rate of expansion 1-83, and an initial pressure 

of 100 lbs. 

, nA l + hvp. log. T83 n _ 
= 100 ", ., = 87 lbs. 

The mean pressure due to a rate of expansion 3*5, and an initial pressure 
of 81-2 lbs. 

= 81 . 2 l + hyplog.3-5 = 521bs 
3-5 
The theoretical mean pressure in high-pressure cylinder 

=- 87 (1 + 0-1) - 100 x 0-1 - 23-2 (1 - t-5 - 0-1) - 52 (0-25 + 0-1) 
= 58-22 lbs. 

The expected mean pressure in high-pressure cylinder 

= 58-22 x 0-85 = 49-5 lbs. 

The mean pressure due to a rate of expansion 1-83, and an initial pressure of 
23-2 lbs. 

= 23-2 L±_^yP' lo S- 1^ 3 = 20-2 lbs. 
1-83 

The mean pressure due to a rate of expansion 3-5, and an initial pressure of 
7 lbs. 

= 4-5 lbs. 

Then theoretical mean pressure in low-pressure cylinder 

= 20-2 (1 + 0-1) - 23-2 x 0-1 - 2 (1 - 0-5 - 0-1) - 4-5 (0-25 + 0-1) 
= 17-5 lbs. 

And the expected mean pressure in low-pressure cylinder 

= 17-5 x 0-85 = 14-87 lbs. 

Example. — To find the expected mean pressure in a compound engine as 
fitted formerly in merchant steamers ; the cylinders are unjacketed, the boiler 
pressure 80 lbs. (95 lbs. absolute) ; the cylinder ratio is 3-5, and the cut-off, 
effected by common slide-valves, is at half-stroke in the high-pressure 
cylinder, and 0-6 the stroke in low-pressure cylinder. The clearance in both 
cylinders is one-twelfth the cylinder capacity. Compression takes place in 
the high-pressure cylinder when the piston is 0*2 of its stroke from the end, 
and in the low-pressure cylinder at 0-3. 

Efficiency in this case taken at 0-75. 



CLEARANCE. 127 



The effective rate of expansion in high-pressure cylinder 



= 2 : V = 1-86. 
1 + T? 
The theoretical pressure in the receiver 

= IW * 3*TcF6 = 24 * 3 lbs 

The expected pressure in the receiver 

= 24-3 x 0-75 = 18-23. 

The fate of compression in the high-pressure cylinder 
_ 0-2 + 0083 
" 0-0*3 

and the steam is compressed to 18-23 x 3*4, or 62 lbs. The mean pressure 
due to a rate of expansion of 1-86, and an initial pressure of 95 lbs. 

= 95 * + hyP- frg- 1*6 = 80 lbs , 
1 "ob 

The mean pressure due to a rate of expansion 3-4, and an initial pressure of 

62 lbs. 

= 62 L±*!h**J+ _ 40 lbs. 
3*4 

Then theoretical mean pressure in high-pressure cylinder 

= 80(1+tV)-95x t V,- 1 8-23 (1 - 0-4 - T \)- 40(0-2-^) 
= 58 lbs. 

And the expected mean pressure in high-pressure cylinder 
= 58 x 0-75 - 43-5 lbs. 
The back pressure in the condenser is 2 lbs. 

The effective rate of expansion in low-pressure cylinder 

, 1 

1 * + T2 

- JL x ■ l \ = 1-58. 

0-6 J_ l_ 

+ 0-6 12 

The rate of compression in low-pressure cylinder 

0-3 + 0-083 

= 0-083 = 4-6 ' 

Steam is compressed in low-pressure cylinder to 4-6 x 2, or 9-2 lbs. 

The mean pressure due to a rate of expansion 1 -58, and an initial pressure of 
18-23 lbs. 

= 18 . 23 x 1 + hyP- Jog- 1*8 . 16 . 8 lbs. 
1 "Oo 

The mean pressure due to a rate of expansion 4-6, and an initial pressure 

9-2 lbs. 

= 9-2 x L±hyP^Li'_6 = 5 lbs. 
4-6 

Then theoretical mean pressure in low-pressure cylinder 

= 16-8 (1 + T V) - 18-23 x T V - 2 (1 - 0-6 - ^) - 5 (0-3 + T V) 
= 14-13 lbs. 

And the expected mean j/resfure in low-pressure cylinder 
= 14-13 x 0-75 = 10-6 lbs. 



128 MANUAL OF MARINE ENGINEERING. 

A Practical Method of Estimating the Expected Mean Pressure in a com- 
pound engine may be followed by first calculating the theoretical mean 
pressure due to the total rate of expansion, subtracting from it the back 
pressure in condenser, and dividing by 2 for a two-cylinder engine, by 3 for 
a triple, and by 4 for a quadruple engine. The result is the mean pressure 
in the low-pressure cylinder, when there is no loss from " drop." Multiply 
this by the factor in Table xviii., and again multiply the product by 0'8, and 
the result is the expected mean pressure when the ivork is equally divided between 
the two cylinders. 

Example. — To find the mean pressures expected in the cylinders of a 
compound engine, using steam of 90 lbs. absolute pressure, and expanding it 
six times ; the ports being of ample size, and the cylinders jacketed ; the 
cut-off in high-pressure cylinder effected by an expansion-valve, and the 
pressure in the condenser is 3. 

The mean pressure due to a rate of expansion of 6, and initial pressure of 

90 lbs., = 90 X -4653 = 41'8 lbs. 

The effective mean pressure . . . = 41*8 — 3 = 38 - 8 lbs. 

The effective mean pressure in L.P. cylinder = 19*4 lbs. 

The expected mean pressure in L.P. cylinder = 19*4 X 0'9 X 0'8 = 13'9 lbs. 

If the ratio of the cylinder is 3 5, then 

The expected mean pressure in H.P. cylinder = 13*9 X 35 = 48*7 lbs. 

Graphic Method. — It has been assumed in this chapter that steam expands 
in accordance with Boyle and Marriott's law — viz., pv = constant — on account 
of its convenient form and easy calculation. As a matter of fact, the results 
obtained in this way are sufficiently accurate for practical purposes, although, 
of course, they are not scientifically accurate, for steam does not expand in 
practice at a constant temperature as is assumed in Boyle's law, and, more- 
over, when expanding in practice, it is doing work, and therefore giving up 
some of its heat with a corresponding reduction in pressure. 

The simpler and more easily-worked methods of obtaining the division of 
power in compound engines are graphic — i.e., by means of geometrical 
diagrams based on the same physical facts as before. 

Draw a straight line AC (fig. 55) to represent the capacity of the low- 
pressure cylinder, including its clearance, which latter is represented by A B. 

Draw an ordinate, A K, representing the absolute pressure of the steam. 

AC 
Take a point, D, on A C, so that -r^- is the total rate of expansion required 

in the system. 

Then draw the expansion curve D Q' C". 

Take a point F' in K D, so that KF' represents the clearance of the 
high-pressure cylinder. 

Draw F' F parallel to A K. 

Now, K D' represents the capacity of the high-pressure cylinder, including 
its clearance at the point of cut-off in that cylinder. 

F D 

Then, take a point L in A C, so that ^t" = rate * cu t-° n m tue high- 

r Li 

pressure cylinder, and draw L L' parallel to A K. 



GRAPHIC METHOD. 



129 



A L will, therefore, represent the capacity of the high-pressure cylinder, 
including its clearance A F. 

Now, take point Non AC, so that AN represents the clearance in the 
medium-pressure cylinder. 

Take a point, V, in A C, and a point, Q, in A C, so that A Q represents 

the volume of the medium-pressure cylinder, including its clearance, A N, 

N V 

= the rate of cut-off in the medium-pressure cylinder. 



and 



NQ 



Draw V V and Q Q' parallel to A K. 

Take a point, S, in AC, so that =-— - the rate of cut-off in low pressure 
cylinder. 

Draw S S' parallel to A K, and S' R parallel to A C. 

Then P" D' L' L" F" is the theoretical high-pressure diagram. 

N' V Q' Q" N" is the medium-pressure diagram, and B' S'C'CB the low- 
pressure diagram. 



KF D' 




AFND3L 



o S 

Fig. 55. — Theoretical Diagram for Triple 
Expansion. 



BL O S C 

Fig. S.'xj.— Diagrams as in Practice, from 
Theoretical Expansion Diagram. 

In actual practice the indicator diagrams differ from the theoretical ones, 
for reasons already given.* They may, however, be inscribed within the 
theoretical diagram of the mean pressure, as shown in fig. 55a, which has been 
drawn in accordance with the method prescribed above, so that F" D'F" is the 
high-pressure diagram, VQN the medium, and SC'CB the low. In prac- 
tice, the area of the actual diagram of the high pressure is 77 to 80 per cent, 
of the theoretical. The actual medium-pressure diagram is 70 to 73 per cent, 
of the theoretical, and the low pressure 55 to 60 per cent. 

In fast-running engines the percentage is, of course, less than that 
obtained from the diagrams of slow-moving engines. 

If it is found by this means that the power is not sufficiently evenly 
divided, the cuts-off in the medium-pressure and low-pressure cylinders must 
be modified. For example, if the power in the low-pressure cylinder is too 
small, the cut-off point, S, must be moved nearer to A, so that the figure. 
B' S' C C E, is enlarged at the expense of the medium-pressure diagram. 
N'V'Q'Q-N". 

* For a refinement in method of construction, vide Appendix, 



130 



MANUAL OF MARINE ENGINEERING. 



Nominal Rate of 
Expansion. 



rH . CN 



r-o 

r- us 

e»eo 



rH US 






Coal per I.H.P. 



r-t t-t rH i-t r-( CM rH 



■ CM 

'cN 



NN IN CM 



CM 









02 

- 

= 

w 

-3 
•< 

> 

fa 
O 

09 

fa 

z. 

o 
!z 

fa 
o 

02 

< 

05 

fa 
o 

oa 

- 
P 
■/. 

a 
X 



I— I 

-1 



3 

z: 

a 
z; 
O 



- 



o 

a 

0) 



•a 

s 



o 

H 



ooo 
I- CO 



CO rH 

cot-^ 



a oo co ^i 
us co co o 
oa oo cn -4) 



■» oo_ 
lO co" 



f^ 00 

r- f-i 

an 



r* a co co 
rn -^ cm co 

COCO U5CO 



CO •* 

00 —• 

co h* 



a 

n 



IN rH CO 

h* os o 
os co cs 





O hi» 


mo 










COCO 






t~l- 


COOS 








fa 


~r co 
coco 


a us 
CN os 










CO o 

.-I CO 


• • 


• • 


rH US 
CN 


-r hh 
OSCN 


US 




• • 


h4 


CM rH 


CM H 










CN 






*-t 












oo <o 


00 US 




IC 




oa 


CO a 


COCN 


00 CN 






CO 




t- rH 


fa 


a o 


OOS 


.-ceo 


US 


US 


o .* 


hji OS 


US H* 


CO ■* 




OS CO 


CO 




-j. US 




CO CO 


C) US 


00 © 


o 


00 


!-< Hjl 


rH CN 


CN OS 


CNOS 


CN 


COIN 


us 




00 US 


-j 






























e>a rn 


*"' "* 


IN IN 


CN 


rH 


CN t-< 


CN 


IH 


r-t 


?H 










fa 


os a 


CO c 


1-100 


CO 


1-1 


i-l CO 


US CO 


O rH 


cot~ 


■* 0» 


t~rH 


r- 


<N 


US CO 


OS CM 


o o 


O CO 


rM 


o 


hJi CO 


«-l US 


CO O 


CO 00 


CN CO 


COCN 


CN 


OS 


C5CO 


a 


CO o 


h* OS 


cooo 


t~ 


o 


00 CN 


CO t~ 


CNOS 


CN OS 


CN rM 


CO CO 




US 


i>- US 






























us ■* 


CN ■-! 


CN rH 


rH 


CN 


rH r-t 


CO 


r-t 


tH 


CN 


.-1 


rH 







fa 

u 



S2 

t-00 

Vco" 



h* CO 
O CO 
t*-CN 



CO hji 
nfi CO 
US OS 



US OS 

CO us 

CO rH 



CO o 
00 CO 

coos 



00 CO 

ooo 



CN I~ 

oso 
at- 



CN CN IN rH 



P-t 1-1 F- rM CO 1-1 



t~CN h* C 

IN CO CO C 

t-r-t us.c 

CN i-T 



00 

o 

00 



a 

o 
o 

- 
a 

3 
z 

fa 



Referred 

to 
L.P. Cyl. 



COl^ 

oso 



r^ r^. 

COIN 



US t» h» IN 



CN 4* 



US CO 

■*cb 



fa 


OS TH 


H US 










cooo 




COCO 
00 f^ 


us 


CN 






■j 


J- CO 

1-* W-* 


CO rl 
CN CN 










CO CO 




H* rH 


CN OS 
IN 


rH 






fa 


OS OS 


CN CO 


OS CO 


1-t 




US 00 




HA 


us r- 
0000 


CO 

US CO 


!>• 


CO 


CO-H 


j 


COrH 


OS t~ 

r-t r* 




CO 

rH 


rH 


O us 
IN »H 


COCO 


CO o 


H* rH 


HO 
CM r-t 


t- 


^f 


us 9) 


fa 


COOS 


CNOS 


HJI 


CO 


O 


r-^ CO 


US 

us o 


r~ 


us r* 


IN rH 


us 


t> 


^H CO 


3 


Hjl r^. 
•<*" CO 


CN 1^ 

COCN 


■* CO 


HJI 

CN 


CO 
CO 


J>. us 

•*• CO 


CN CM 

CO rH 


Oco 

COIN 


-r^ US 

CO 


I. US 

** r-t 


us 

■*1 


oo 

CN 


coco 

CO CN 



fa 



oo 

us us 
a a 
rM *■** 



US us 

rH a 



an 
coco 



rH OcO 



8s 



rn« 



Cut-Off. 
HP. 



oo oo 



CO 

o 



Hjl 

OrH 
t^ US 



ao 
b 



Revolutions. 



00CO «N 

us hji r-t a 

rH rH CN) rH 



USCO -t 

U5 b 
-r us 
IN — 



$ 



CO 

a 



I- OS 

■* o 

IN M 



Vacuum. 
Inches. 



op 

IN 



Steam. 
Lbs. 



■rP O 

US O 

rH CN 



CM IN r-t —t 



US C3S CO rH 
** •** CO US 

rH rH CM CM 



t~o 

CM US 
CM 



r-t fXi 

t- a 

CM r-t 



CM rH rH rH. 



a 

a 

Zi 

r-^ 
J 
>• 



Stroke. 
Inches. 



fa 
hi 



fa 


-« 












-«« 








, 












HJ1 


00 




s 




S 










CO 




-J 


00 


co 


00 


00 


t» 


->- 


US 


us 


^ 


n 


^1 


us 


fa 




" -H« 


-«n 








-*« 






-*» 


Hn 




-Hn 






US 


r-t 


OS 


OS 


M 


Ul 


us 


CO 


o 


00 


o 


00 


"*■ 


3 


r~ 


us 


US 


us 


hji 


■** 


EC 


■<*« 


CO 


us 


CO 


CO 


S-J 


CO 



fa 
tq 



CM 



o . 
»-2 



S5 






= ■0 
E&Bh 



- - fa 



co co 

"a 



= ■0 
3 m 

EuOh 



"a o 

&Hr-l 



-Q 

CO . 

o ca 
faS5 



ajco 

= 13 
a ss 

-=.a- 



to* 

— T3 
3 « 

-- 



3 1) 

_ - 



O 

- 

o 



CD . 

c « 

faS5 



a 

S5 



co 

o. 
o 

= 

-a 
o 



s 

C3 
§ 
If 



s 

ce 



£ 



CD 

If 
O 
CD 

o 



a 

a 



■< 



co 

< fa 



= 
10 



CO 

fa 



C9 

rt 
hi 

o 

fa 



O 
H 



- >> 

ft -c 

£ ft o 

g a cu 

o a ft 

O 00 oo 



RESULTS OF TRIALS. 



131 



X 

71 



< 

X 

» 

a 
o 

a 

Z 
i-i 
o 

« 

a 
o 

o 

o 

H 

-3 

CO 

a 
P3 



-31 



Nominal Kate of 
Expansion. 


oo : r- mi c is a o i- o » ?s h 30 o io to 

oo ' ' * a — i m os a r» oo 71 a -i" to a CM m 71 io 71 o 

-h~H^H i-l-H7l-^r*FHr1.-l W 


d2 

a 

■x 

n 
?-. 

c 


o 

c3 

•9 

s 




OOOOO^^CCtU-aril^MOOMntCIlOIXOt-W 
£*-T<OC7X-fC7 3-H»-«-1'SDtOC7-f , 71-*C7ina71inXX00''1'a0 

Ht>c*L:xxtOH[*Hffl0^3Ni--£0niSH0OCin"l 
i» to -r -f 71 71 71 71 ii 71 co i-i hh i-i ■* i-h CO* i-j ih 


pj 


^i:c:i^OJ«!CO»ar-Nisc 71 in i-i .- x r^ x -i* x a m in 
a co tN a tkooi — i ■• w O i — niii:iooo<oasa«nt i 
T_«-»Sa naxt- o«o m in «siMC4HriHHq« « co io g; 

CM 94 iH iH iH i-i r-i 


■" — 


1 202 

682 
464 

255 
189 
104 

1,059 

472 


s 


ccicr. fritffiio^oHfOi-:i!Oi-aoiioOf»cM^ 
i-rnna»eoeiooiOrtNi5t»i-i3030ioartOssi»sc' 
-rco^-raxl-~ai»xacOi3-P71c7 71^'-H'-H'-<aco-T-rio^ 

I1MHH f-l IH 


[jj 


fioin^nHnX!o;ioori»i»aoHHOi'Knf3iMOO^ 

H0000^'tN«?|Hr4H(OnrHCr:^r3t^H«(DMNinOM 
7^3 a "^OS OO^^COXOOWLTTfll^JIHnHHOjniflMLTQ) 

©fef i-T 


1 

1 

o 

es 

• 

1 


0} 

Q 
C 

s 


a o 


13 

ooc;:iiki««h)oio»oi»3hm)cjh;iisio3 . 

5iHa^M»!o:irfN!Ooai]-<i'oo?ioi'-''i-r:HO • 


hJ 


co oo to vo us io ooim in 

-h 71 a oocpo^ 1 ,^:: w a 71 71 71 x -h go 13 to in ph i3 H 7> 


s >> 


-5 -C-f -i> ^h co — a 
■ • 3 n a ■ • • x ■ a • o • x • to • • • 

71 —1 71 tH 01 <-* I-I rl 


s 


XX 3 -H 00 7* 7J ^f CO lO 13 IS 3C0tCt~C0X[- -1 71 -4 H 

NCHoiiccin-iibiionncxHiiM^i^^lijji-a) 

C4CO-*C0COIlCOC070Ilin74COCO71CO-*C0COC0COT1.70 77CO7<M 


Pi 


to 3 h 13 h x o t* to r-t in in oo 77 r» in to 77 71 r- x 

71 71 73 t- X O 00 I- t- 71 t- CO rH r- I> 77 13 A- to >3 -t b C C -f X i- 

to X I- l~ to to X 13 13 to 13 to X X T X I- to 771 1-- t- X X X X to to 


Cutoff. 
11. P. 


to 

p° . . .OCO 77C — CT. -rxc:ini3tot~0 7ix •' 

?" . . . to to i3 to to i- eft to to in 71 1* in in to to o 7O 


Revolutions. 


to to in 77 in in 
ii<::i-----ioNi.»i-H 10 cs 71 71 x to m x m -c — -o 
1- ~. 1- 1- 1^ to Nl-XK 1- to C.OStXI--ffl M t~ 1- to -H -o CO . 

r*r* t-l —1 f-l ^H ' -H — 1 


Vacuum. 
Inches. 


-H.f •-:. . -H -„ ?>=!• P _i,H- 

-r i- in 13 • 1^ x • x 1- 1^ m m 1- to to 71 m 71 1^ 1^ x to t- X 

71717171 7171 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 71 


Steam, 
Lbs. 


o>inM77unioxfini>ocooc;rH-foocoooinoia 
■a- m os X -r to -r m 13 i- 3 x c m X' X x co 3 co x 71 c; x in 

I-HrH-HMClHplrtrtClHjlHHHHClClCICIrtllClHH 


as 

| 


•Z o 
'A — 


70-P-fOXr-Xi3C3t0071*1<tOCiCO-tt^-l'-fCOI>tOXCOunX 
to i3 13 to ■* i3 •* -tf CO CO CO T 71 CO CO CO 71 71 71 71 71 13 CO ■* CO ■* rH 




pi 

-a 






w 71 0: x c 71 c 13 ^< 13 m x to t- 1^ x 13 to t- 71 s> 1-1 a co x 
co a in x x x t- 1- 1- to -t -1- -r ia m ci -r co co 71 co 10 m -r 13 to co 


pj 


—II -:" -*l ^11 -*1 

l>. t~ — • -f X a O tC C7 a C 71 O C7 -t C7 I- 17 71 X 3 I- — 1 T i3 O X 

tO 13 -t 13 -f 13 1i -1- -*■ C7 -T C7 C7 C7 CO 71 71 71 71 i-l 71 C7 CO CO CO T rH 


pj 


44 --I -ii erHt -ii 

iHtoaccoi — -raxtca7iaa7i-txwco7i7ixa-f7iino) 

-1* C7 71 C7 C7 CO 71 CO 71 71 71 71 i-H -« 71 rH -H ~* -^ -H -h 71 ^, 71 71 71 rH 




/: . ~ f : f : 00 OO ED EC BE1 DO DO X X '/: x X 33 X X X - X X x' X X X x' X 

ri-ir x' X /. X X X H r- x' r- X X H x" X X X X X X X DO X DO 




DO 

< 

55 


1. •* C — - O ^ - •— T.- 1 ^ -*^ - - — 



132 



MANUAL OF MARINE ENGINEERING. 



co 

i-i 

o 

w 



o 

& 

o 
o 

w 
o 

H 

GO 

6 



o 

CO 

►J 

P3 
H 

o 

oo 
H 

& 

CO 

W 

P4 



H 







CO 


CNf-M-tt^o; f-; — ««— cc;l*ms-c 


i 




«3 


© 


No^isoot^ooiQtD'HnnoQOx't*^ ,^ 






-*^ 


en 


n « « l: a m 51 ^i x » t^ x h- 1; o -t k - - :i .« 






Q 












J5L 








>a x 


-i«fSCHMO«-t'Mt^M-;:i-'t'XI»5l» 




p* 


P-' 


^ ?i r! « n ; d r: - r. c ; s; oo s h x r. .; x cc ci _i. 




a 


si® 


- 51 -i i> k i a c ■+ « « ?: m ?i ?i - ^ l- c -^ j 




t-4 


^ 




HrtH l— ' 






cc 


a^OOa^ffl-^NOOtDOOSQODcaiOONOJQO 






«* 


cc 


G»tNN05a0r"»C0i--I^OC0e0«OC0r~©,J,0»-*i-i,i, 
r-^i-^tM «> t- o CD ^p ■* rj* eo ■* CO «M tN c5 g IO JO -.2 55 






ri 


"* 






0* 






m 






■> co t- r- iQ — tc- io cc 




« 


p5 


; jOfNr»t-(M^S?!0«««3t~«OKrti l O)«X 




A s 
9 a 


►4 


£ CO to 


HNHHffMOHHOWrti-iMM — H^tMilM 










ss a 










2P4 




' — r— 






X 85 


PM 


CO 

JO "* 


t~ r-< i-h co '-h os io co o io to us co eo io io © *h >o eo ■* 




C5 


ft 


h5 -# 


^^i ^^ '^ ^^ ^T *^ ^j^ *T^ ^T *T^ ^T ^T^ ^T ^4^ ^j ^T ^r^ *-*-• *T ^i^ ^j* 






M 




-f «!>• om :o ~ ci cc is io oono 




5 5 


• 


o op ci jp jo ci o Jf o •* '{5 oj -+ -* -+ — 'p jo cp t^ JO 




■^ o 


« w 


• 


iA- l?3 i^- cb io jo jo jo >h cb t> cb i> lc jo jo l-: io co i> o 




« H F 


4 








^ o 


i 




-t >~ '-C 




°; 




IS 


t^-r^t^CJCCCCOCiOOOJ— i -t -C X M » N O! 12 r- o 




D" 




o 


O 12) O JO CO IT5 JO t- i-"5 JO t— t^ X JO X OJ -f X r- UC JO ' 




tf* 






— _ 




B 






«:t*— (*l I'^ieet Jl — ^i — rt? -^?ir:'-j-~lci Hej 




a 




s t- 


r- x jo jo r- o r~ co x io t — !• -*■ cc jo io t— jo jo t> jo 




o 




>- ci 


ffl N N W W N (M N CI i>i CJ CI CI CI <M CI CI CI CI CI CI 




i > 










. 4 


i 


















c : 




m JO 


OOOOOlOONIMOtC O -+ O CC -t r- 25 JO JO io i2s 






: 


.0 X 


Xt^Xt^XXC^Xt^JOXXXt^Xt^OSXX— X 




ml 


i 


h^ 


- ' 




p 


i 








p," 




O 


OOOOOOOOQOO>0OOOO00.t-CC"-*O 




s 




© 


CI O 25 t- IO IO CC X X IO CC CI — i JO OJ JO CC CC JO CI CC 






o^ 






>5 
1 




— ' 








6 










o 
u 

t/1 


[/* CO 


O r~ — X X CC X CJ CI JO IO JO CC 25 CC t- X JO ~ : — o 




•sua< 


1—t **" 


JOlOCO-*-*JO-t<CC-*CC"*CCCCCCCC'Mf-<CC'CC - -OCI 






,— '— 






85 






JO t— -f JO -* -+ CI G5 -t CI 25 10 CC 25 IO 25 CI t f JO CI 






OJ 


CM O 


X X X t^ t^ JO t^ JO CO CO io 10 IO IO -*■ rt< CC iO CO IO CI 




tM 


-43 

3) 


. CO 05 

10 


^^^'^'o'^' 1 C5' r w F J5 ;- 2)'T2' F ^'w'T5'"25"C2;" 1 C5'"C"C"C2'* 1 C 




O 


g 


-tH <+- 


CCCGCCCCCOCCCCCCSSCOC 






2 


M O C 


7t ^t ~i ~Z 7*. 7i 7i ~ 7t 7t Tt 7t 7t ~ Tl ~ ^i 7Z Ti ~^ — 






Q 


rf ?) M S (M M O (M « LI M-'-tiN I.T O N l^ "t - t^ O "t - W 










"*-*-^lCC-^CCCCCCCCCCCICCCICICIC-l— ICCCCCI — 








>—■— 


, cc 




U. 




T3 






o 




a 






85 




2 
o 






© 










BH 




i— 


OOOOOOCCOOOOOOOOOOOOO 




*2 " J 




a 


^^T3^T3-^^-d-2)-3-c!-s-jr-3-c _ 3-a-c-2:-C'd 




fe 8 




O 






QG * 










3 c 


i 


ce 






C 8 




_o 






j6 


3 








W 




49 
> 


OOOOOOOOOOOOOOOOOOOOO 




W 




QflflPfiqaqqflQanaqqsqpqp 




85 
K 




3 






o 




1 







RESULTS OF TRIALS. 



133 



so 

o 

<! 
O 

I 

w 

a5 
X 

z 
o 

02 

< 

a- 
X 

I 

pj 



02 

S3 



02 

J 
D 
X 

a 



XI 



— 

0Q 
'- 



Nominal Kate 
Expansion. 



c 



o 



M O SJ O St»ONOW00 Ol CO 

it . . . x i~ oi . x rM- c o r- cp -■ c ci 

ob * * ■ ~ e5» «a ■ i' i: -t i) i ri i c s m 






o 
c 



e 
z 

a 



Referred 

to 
L.P.Cylr. 



Cut-off in 
H.P. Cylr. 



Revolutions. 



Vacuum. 



»OiQ»®'*Ot>'*t'M(NtO'*Ort"« 

L-»-t«oo»W'tos«i'aifli | *«ffi 

— i f- iO 00 Tf 00 rf< O f- — CO a". - -a IO — 01 IO o 
O -~ -»■ r^ iO oi oi — oi — i —i — i — < 



f i - : oi lo c r- co '" — t- W is t- — i aa i- t- r- 

srtc»-C'*C't-r-'* (< oo <m a os e» 

<*,« OOOHOJrolOOlO'Vt-lCrtN 

ofof— To^of — T 



o 

CD 



t^ n a o ■* cr. a o w i; i> a w oi -h • x a 
tj< eo -«* »,<» os t- l- c o ■* o •; -< fi ™ « h 
of oT i-T of i-T 



•fC^MLIMMOO-t-OM'**'- lM« 

— x ■«* -* x ■>* co o — i co —i eo oi t^ o ~, iQ co 
Ol^ O'VIOtDO! t- LS X «Q ^KDUHOgHH 



■XOCNOXCOr^ca 



13 o a o M3 oo o 



-riHTtorinsoaoaciOMOrtMi-i 
— eOTj<eOL'5 0eoeoeoeoeoeoeo-<*i"*oieoeO"* 



•-C O IO (MOO 

g <^i ©i © e» »Q o cp >h fyi t> c» ep »o M est •>* t> io 
S ^ <* e>i ir- o i-h eo o ■* eb o ■* A w i> A w.io 





,,; X X © O — iO!OHMOL''.05aih'OHL'5 00 

■^h-io'i^NihHWfflcfthobooiob 
| - J <NeoeOTi<-'#eoeoo-ieoeocooieoeooioicoeo 


. s a l- a o h o c ^ 


o X IO o t— O X —i L0> 



^»ki t^o fi » * cc m - t^Tt w : > i-c tj 

_: CC X C-- tOOOfflfflOOOKOOfflijItOCOf 



C- CO IO ; t" CO 1C >0 00 CO <N CO CO JO 

o © © 'ioioooooio 



oco 



c c c 



LOS 



LO 



X 

r~ 



oi a> ?o —i r— t— i io o-i •— — eo oi — ;a x '-a 

i>Mii^>;Mae^;- a «-t co 

— i ■— i i— i i— i eo eo — i -, — — 



eo 



COOOOOOOCOOJCICI 



= -*t^uto — — xt^^LOoiOLieot^xt^o-i 

M Ol Ol Ol Ol Ol Ol Ol 01 Ol <N 01 Ol Ol Ol Ol Ol Ol Ol 



j a o m k n o a o o o a o o o c o n s 
Steam, i ^-tCM't^TiioMrtxaxaoisot'O 

^ rt ^ H IM (M rt f) rt rt n ?1 -I rt »| rt ™ -. M 



Stroke. 



;K-*a-axxL-ai!'axx'*aX"*M 
- a 13 a 13 ?5 ■* ■* * « w « -i - ei « - el m 



a « x -t x a a x - k a « a a t^ a :i :i 

O0)00»t000l>®©'*IQ'<frtMlO«MM 



;M^'t- oi x c — a a « x a n •* k a 
— a lo >~ lo -t ■* -* Tt eo eo eo oj oi oi — 



Screws, Xumber of. 



= 



S"28 



;/ — "' "— -^tl 

:- a m - a m ii is a a a a x m oi oi o« c<i 
— -t eo eo eo oi eo oi o^ oi -^ ,__.m__h_h 



'— 



» a CD „ C& 

03 H 0Q EH c« ' 



o 



< M -; '^. A u c ^ > =-, d ^. x. S - < n a 



X X X 



J X X X X X X 



X X X X X 



x x i X X X X X X X S X X' X X X X 



> a o a' o" 6 6 b0 ' r ' -' M "^ a ^ a 6 > ^ 

' -3 - x - - -a .: r ' -a .S r "a .5 " ts d .H 

~ x 



134 



MANUAL OF MARINE ENGINEERING. 



CO 

w 

& 

o 
co 

< 

X 

w 

I 

w 
l-H" 
pm 

Ph 
H 

13 
« 

o 

I 

P5 
p 
o 

ft 
O 

CO 

rJ 

<1 

« 
H 

ph 
c 

to 

H 
pJ 

CO 

H 



X 
X 

PQ 
Eh 



Nominal Rate of 


















OS 








Expansion. 


















X 


OS 




! 










o 


-p 


i- 


01 


V.O 


to 


-t 


03 


-5 


lO 


X 






c- 






CO 


i~ 


-r 


ei 


on 


rH 




01 


-r 


CM 








I- 


X 


OS 


-r 


X 


oo 


O 


rH 


01 


CO 


rH 






c 




1ft 


r- 


o 


rH 


X 


t^ 


>o 


CO 


r 


00 


Ol 




c 


&■* 




^H 


rH 


rH 


" 
















>• 






OS 


1- 


to 


o 


lO 


I- 


to 


c 


01 


r- 


to 




o 


p. 




-r 


o 


00 


I- 


01 




-r 


00 


>o 


-T 


o 


o 


f-4 


i-4 




oo 
of 


to 

Cl" 


rH 

of 


of 


q 
of 


OJ 

rH 


OS 


to 


01 


o 


T 


W 


m 






















































!Z 


o 






00 


CO 


00 


-* 


-f 


X 


|s* 


01 


o 


oo 


01 


C 

pq 




p- 




cs 


rH 


-T 


OS 


-r 


rH 


OS 


or: 


i.O 


to 


rH 


i-i 




tO 

of 


lO 

of 


rH 

of 


rH 

of 


O 


00 


X 


to 


CI 


iO 


-f 


s 








OS 


on 


vO 


-- 


to 


_ 


I~ 


— 


01 


,. 


_,. 


K 


Ph 




OS 


to 


— . 


OS 


X 


01 


00 


—• 


X 


01 


X 


O 


►5 


M 
« 






CO 


OO 

oo" 


SO 

oo" 


CO Ol 

of, of 


rH 


p 

rH 


CO 


r-« 
rn 


to 










-r 


r~ 


00 


-t> 


o 


1- 


H 


r 


^ 


X 


to 






Ph 




to 


■<* 


X 


on 


t-- 


<M 


00 




oo 


c; 


01 






K 




l~ 


pH 


00 


■r^ 


CO 


» 


lO 


on 


00 


X 


to 




































-p 


CO 


M 


CO 


CM 


0! 


"- 1 














f'fc 


.: 


x 


rH 




00 


CO 




00 


OS 


to 










Php-? 




b 


-f 




OS 


o 




L- 


to 


-r 










i- 1 


113 


1- 




■* 


to 




IQ 


00 


rt 






pi 


OR 


OS 


0-1 


to 


X 


l~ 


X 


r* 


CO 


01 


01 


t^ 




en 




h2 


l~ 


OS 


o 


X 


p^ 


-r 


01 


uO 


X 


f. 


X 






►4 


hJ 


PH 


1^ 


1-1 


1-1 


ffl 


r^ 


CM 


rH 


rH 


r^ 


^ 


pl: 


cc 


OS 


■n 


t^ 


^ 


OS 


OS 


lO 


-f 


f 


t~ 


o 






,fi 


to 


X 


to 


OS 


1-* 


■* 


rH 


lO 


X 


t~ 


OS 


s 

o 

o 

K 


p- 

3 


A 


^ 


n 


rH 


rH 


rH 


Ol 


^ 


01 


^. 


rH 


rH 


rH 


Ph 


- as 

r= 


-p 


ioi 


101 


CO 
rH 


p 


•P 


01 

t'~ 


00 

o 


O 

on 




p 

00 


«*s 


rs 


hi 


H* 


00 


CO 


-r 


00 


00 


rf 


CO 


00 


Tj> 


oo 


Ph 


tn 


to 


50 
01 


Ol 


Ol 


p 
o 


OS 


rH 


■p 


CO 


CO 


00 








r4 


o 


l~ 


X 


c 


OS 


o 


rn 


uO 


X 


l- 


to 


ffl 












rH 




rH 


rH 










55 






























s 


Cut-off in 
H.P. Cylr. 




• 


■ 


• 


• 


• 


• 




OS 

to 

b 


b 


; 


: 






§5 

^H 


X 


oo 


liO 


o 


01 


to 


1~ 


lO 


lO 


rt 




Revolutions. 




oo 


-f 


-T 
rH 


01 


Ol 


01 




rH 


X 
00 


OS 

00 






o 


I- 


o 


VO 


o 


CO 


o 


o 


lO 


o 


rH 




Vacuum. 


c 


£~ 


§ 


uo 


to 


1~ 


ia 


-T 


I- 


to 


UO 


-p 






HH 


CI 


(M 


CM 


CM 


01 


01 


01 


CM 


Ol 


01 






00 


-p 


m 


to 


uO 


i- 


rH 


l> 


c 


to 


-p 






,Q 


l^ 


OS 


on 


OS 


-1" 


Ol 


l~ 


I- 


X 


o 


OS 






P-h 


01 


t-H 


^ 


<M 


CM 


CM 


01 


rH 


rH 


01 


1~l 




i 


Stroke. 


a: 




01 

-p 


01 


01 

Ht 


X 


o 

00 


-T 


g 


00 


X 


X 




Ph 




--. 




HH 








r-^l 


















I- 


00 


OS 


-P 


t>. 


01 


vO 


c~ 


-H 


X 


DC 

s 

Q 




M 


H^ 


X 


I' 


I* 


to 


to 


o 


-r 


-p 


Ol 


00 


CM 




Ph' 


e 


I-i 


1~ 


r*l 

CO 


OS 


-t 


t- 


oi 


uO 


l^ 


rf 


X 


te 




Hi 


HH 


X 


I- 


1^ 


o 


to 


\n 


-r 


-r 


01 


oo 


CM 


3 


s 




























P4 








-<5> 




-'^1 


-^i 


rtl 








-rP 




■^ 




rH 


OS 


VO 


o 


»o 


o 


X 


C 


00 


o 


1- 




H 


g 


M 


L- 


to 


to 


to 


lO 


lO 


CO 


-T 


01 


CO 


01 


Ph' 




*h;i 


IHM 


_., 






rb. 


-H-W 






^?l 


«H« 






s 


00 


00 


rH 


t- 


-f 


rH 


-T 


to 


-T 


OS 


X 






w 


M 


T 


-r 


-r 


00 


00 


CO 


Ol 


01 


rH 


rH 


rH 






C 






















Number of Screws. 




H 


O 


O 


o 


o 

to 





o 


c 





o 
-o 


o 


h 




















• 






»« 
-« 

fc 




j: 


tH 


S 


•o 


s 


T3 








> 


? 




O 


Es 


< 


CJ 


A 


rH 


H 


fr 


„ 


c 


X 






oi 


aj 


CO 


X 


X 


X 


03 


Ph 


tH 

S 


03 


x' 






ftj. 


^i 


s 


£ 


s 


l-H 

rr. 


s 


03 


03 


a 








£3 


H 


HH 
rH 


H 


K 


fc 


w 


OS 


03 


w 


M 



RESULTS OF TKIALS. 



135 



'J. 

w 
z 

o 
55 

pa 

o 

03 

-< 
Pi 

H 

1 

w 

Pi 
P 

P5 
Q 
O 
& 
C 

o 

05 

Pq 

o 

02 

H 
& 

03 

H 



I— I 

X 

pa 

PQ 
<^ 
H 



Nominal Bate of 
Expansion 



. 58 



*i p lA 85 r* 
te «>. w M us 



z 



w 

Z 

n 

Z 

o 



o 

s- 



o 

= 



si 
u 



o 



W 00 
5D 00 

•* of 



85 i- 1 50 
■Q =8 OC 

^ «o e 



* ti 

85 00 

-T SO 



CO 00 

M tl 

tl tl 

B-l" t f 



oooxi-^« — 2 — 

c-i «o o — >* * oc cr. it 

^*U30USrHf-lTHU3U3 



tl 
o 
tl 



tl 0» -r — -T tt >.t 
I- O 50 00 C 00 » 

-re ^< tH iH ri ,ia 



P- 

a 

Pi 



o 
o 



a 
z 

S 
z 






■So"! 



oc d oc 00 i- o o ti ti 

i.o i- :o CI iO i-* 58 CO -r 

!-1-i«©C0~r-<T-tr-«i8tf5 



00 



tl -r 

■> od 

lO 85 



tl 



e e 

T l- 



Hi 



85 
CO 



i-l 85 
-* CO 



j. 00 H O « 

n ~ 4* ce> o eg 



Pi 
i-4 



oo ti 

m O -H 

3 t, 05 



o 



oo _ 
85 © 



do 



- 



tt oc 

't 85 



SO 

i 



Pi 



■= — tl 

H -r CO 



H I» Li 



oo 
ti 



a ? o 

08 t— 13 

tl « to 



Ph 

X 



■A 



CO 

I- 



- 
x 



CO O rH >/5 

^1 O SO K 

so l~ -r I- 



Cut-off in H.P. 
Cylinder. 



Revolutions. 



-f t~ 01 

e is 9 

b o b o 



o 

S3 



50 t- 

SO L.O 

a b 



o p 

US t- CI 

W 58 I- 



US 

oo 



Vacuum. 



a © 



Steam. 



- 




00 
0-1 


I~ 


l~ 

tl 


US 
01 


wj 


01 

35 


o 
tl 

Ol 


o 

CO 
01 


c 
c 

Ol 


oo 



9 



o 



58 



Si 

C 

z 

3 

5« 



Stroke. 



=8 



Pi 

►J 



5 ■= 



l- 01 



I- -r tl 

01 tl -r 



cd io co id ia so 



0) 



Pi 



-«i 

CO 



-1*1 
"f SO 
tl 01 



tl I- tl 

CI — — 



O C 01 
tl 01 CI 



iber of Screws. 


B 
H 


"5i 
a 
33 


d 
•a 


6 


_c 


d 


d 

■3 


d 

T 


6 

•3 




• 


• 




• 


• 




• 


• 


• 


c£ 




d 




o 

5 




d 

Si 


>> 


H 


w 




cc 


CO 


oc 


03 


93 


to 


93 


93 


93 




00 


DO 


03 


CO 


co 


93 


93 


93 


93 



136 



MANUAL OF MARINE ENGINEERING. 



CHAPTER V. 



STEAM USED AFTER EXPANSION — TURBINES. 



The Turbine is essentially a Velocity Machine, inasmuch as the velocity alone 
of the steam is the active principle, and pressure has no part whatever in 
producing its movement. There are rotary machines of various kinds, in 
which pressure does produce motion, they are, however, not turbines. 

Turbines are of Two Elementary Kinds— the one works by means of the jet 
of steam issuing at high velocity from a generating nozzle acting impulsively on 
the blades or vanes of a rotor wheel secured on a shaft, as shown in fig. 51, 
thus causing it to turn round at a high rate of revolution. Such a machine 
was invented by Branca in 1630, and it is analogous in hydraulics to the 

undershot-mill wheel, and to the more 
modern and more refined Pelt on wheel ; 
since it is operated by the impulse of 
the steam particles on the vanes, it is 
known as an impulse turbine. The 
other kind is known as the reaction 
turbine, inasmuch as its rotor is caused 
to move in the direction opposite to 
that which a jet of steam issues tan- 
gentially from its circumference. Such 
a machine moves in consequence of the 
reaction of the flowing steam, and was 
the well-known steam engine of Hero 
of Alexandra. 130 B.C. ; it is analogous 
to Barker's mill in hydraulics. 

The Modern Reaction Turbine has, 
as a rule, no actual nozzles, but their 
equivalent is provided by making the blades of a special section and setting 
them at such an inclination, as shown in fig. 56, from, these channels the 
steam issues at so fine an angle of inclination that the component of the 
reaction resolved tangentially is large, and causes it to turn. The guide 
channels or the equivalent of nozzles, through which the steam is led into 
the rotor, may be formed in the same way as shown in this figure. 

The Modern Impulse Turbine also may have, and generally has, a series of 
expanding passages, the equivalent of nozzles formed by the blades in a similar 
way, or they are formed in a casting and fixed to the circumferential part 
of the stator, or fixed disc, attached to the turbine casing, as in fig. 57. which 
is an example of passages made in that way. 

Combination Turbines. — For many reasons the efficiency of the pure 
impulse and pure reaction turbine is too low for commercial success, so that 
all modern turbines are so constructed that both the elementary processes 
of impulse and reaction occur within each pair of stator and rotor sets of 
blades. Their construction has developed along two characteristic lines, 
in which the impulse or reaction process predominates respectively. In 




Reaction Turbine. 



COMBINATION TURBINES. 



137 



what is now called the impulse turbine expansion of steam and generation 
of velocity take place solely in the fixed nozzles ; the issuing jet acts directly 
on the leading half of the rotor blades, giving up part of its energy to them, 
and being then deflected through an angle of nearly 180° gives up the remainder 
by reaction. There is no change of pressure within the cell containing the 
moving blados, nor in the moving passages, except what is due to friction. 




Fig. 57. — Impulse Turbine Compounded for Pressure and Velocity. 



Consequently the nozzles may be isolated or in groups (v. fig. 59). The 
reaction turbine, as now made, is not unlike the above, but in it there is pro- 
vided a further drop in pressure, and the corresponding generation of velocity 
in the moving passages. In this case, since there is a drop in pressure in 
these moving passages, the fixed passages or nozzles must occupy the whole 
of the annulus, in order to avoid excessive leakage by short-circuiting. 



138 



MANUAL OF MARINE ENGINEERING. 



The Shape of Passages, and consequently that of blade section, depends 
on the drop of pressure required. If that drop is greater than 7^ to p., = 
0"58 p v the passage should converge and then diverge ; if p., is greater than 
- 58 Pi, convergent only. Thus in the example given (fig. 56) of simple 
turbines, large drop is provided for, and the passages are, therefore, con- 
vergo-divergent. Marine turbines being compounded by numerous stages, 
and having to run at low velocity, require convergent passages. The reaction 
turbine, having for convenience a similar ratio of expansion for each fixed 
and moving row of a given diameter, will have similar blading for both. 

Compound Turbines of both types may be multiplied in series, limiting 
the drop in pressure for each stage to a portion of the total " head." This 
is called compounding for pressure. This compounding of the impulse turbine 
may be supplemented by another kind, in which the velocity of the nozzle 
jets is expanded in more than one row of moving passages, the velocity of 



57a. 




Fig. 576. 

the latter being less than in the uncompounded stage. This is called com- 
pounding for velocity. Fig. 57 shows diagrammatically an impulse turbine 
compounded for pressure in three stages, each being compounded for velocity 
once. The velocity changes of the steam are shown in fig. 576, and it may be 
noted that the only function of the row of fixed blades between the moving 
rows is to deflect the stream. 

An impulse turbine compounded for pressure may have a series of rotor 
discs with blades set transversely and interposed between a series of stator 
discs having guide blades forming a set of nozzle-like channels. If these 
blades are all of the same height, gradual decrease in pressure through the 
series is provided for by starting with only a few passages in the first stator 
disc, the number in the next disc is larger, and a further increase made in the 
third, and so on, the increase being such at each step that the steam expands 
in accordance with the natural law. 



COMPOUND TURBINES. 139 

Expansion, however, may be provided for in quite another way ; the 
length of stator and rotor blades may be gradually increased in length, so 
that variation in area of the anulus is such as to permit of the steam expanding 
in the same way while it follows its course through the machine. In either 
case there is at each drop in pressure a corresponding generation of velocity 
at each stator, and a delivery of kinetic energy to each rotor disc, so that 
the expansion is continuous step by step from the initial pressure at entry 
to that at exhaust, and the total work done distributed through the whole 
of the stator discs. 

The object of this system of compounding is to reduce the mean velocity 
of flow to a minimum, so that the rate of rotation of the machine may be 
reduced down to practicable limits. So far as marine work is concerned, 
those limits must be such that a reasonable efficiency may be realised with 
the propellers revolved direct by the turbine. 

The impulse turbine is not necessarily compounded for pressure ; it may 
be compounded for velocity a sufficient number of times to use up that gene- 
rated in one drop from maximum to minimum pressures. The number of 
velocity stages, as they are called, depends on the velocity of the blades 
relative to the initial velocity of the jets. On account of the losses by friction 
and eddies this type is not commercially efficient ; if this were not so they 
would displace all other types for large sizes, inasmuch as they would be of 
less size, lower cost, and freer from the mechanical troubles so common in 
this instrument. 

The turbine, however, may be treated in just the same way as a compound 
or triple-compound engine by providing means whereby there are two or 
three separate stages in the operations of the machine, in each of which 
there is at commencement a partial drop of pressure with a corresponding 
generation of velocity, followed by the gradual drop in velocity and the 
imparting of kinetic energy by the usual steps above described ; at the 
end of the first stage, and before beginning of the second, there is another 
drop in pressure due to the passage through the second set of guide and 
expansive nozzles of larger capacity than the first ; the steam acquires 
thereby a fresh velocity, and gives up its energy step by step through the 
second stage ; and so on in a similar way through a third of similar con- 
struction, until the steam exhausted of practically all its potential energy 
enters the condenser. 

Fig. 57a shows diagrammatically the fall in steam pressure in a three- 
stage compound-impulse turbine, and fig. 57& the corresponding generation 
of velocity and its gradual extinction in the two steps in each stage, as shown 
in fig. 57. In practice there are usually more than two steps in each stage 
of a turbine, and on shipboard it happens usually that there is a high-pressure 
turbine driving its own propeller, and exhausting to one and generally two 
low-pressure turbines, each with its own line of shafting and propeller. There 
may be, however, three turbines in series, high, medium, and low pressures, 
as in a triple-reciprocating engine, each with its own line of shafting and 
screw. These arrangements are necessary for marine purposes that the 
revolutions may be as low as possible to admit of screws of such a reasonable 
diameter as to be serviceable and at the same time efficient. Fig. 58 shows 
a section of the Liungstrom turbine with its two interlaced rotors, which mo\e 
in opposite directions and irreversible Fig. 59 shows in section a Curtis L.P. 



140 



MANUAL OF MARINE ENGINEERING. 




Fig. 58. — Ljungstioni Turbine (Longitudinal Half-section). 



CURTIS TURBINE. 

,,Ll£ ... ._, 



141 




h: 



MANUAL OF MARINE ENGINEERING. 



and astern-going turbine for a warship. Figs. 59a and 59b give the section 
and elevation of the Zoelly turbine, as used in the German Service. 

The Design of Screw for a Turbine Ship must first of all be determined 
from the conditions imposed by the ship and her service, and the turbines 
designed to suit the revolutions af which such screws can 
be run consistent with good efficiency and safety. 




Fig. 59a.— Zoelly Marine Turbine. 7,000 H.P. at 650 r.p.ni. 







Fig. 59ft.— Marine Turbine of 7,000 H.P. (Zoelly.) 

The Efficiency of the Turbine * must, therefore, be determined at varying 
revolutions near to that rate at which the propellers may be run, and a 
curve plotted in the way usual to express it. The efficiency of the screw 
decided on by considering the above-named conditions should be plotted 
in a similar way and the curve formed, after which the combined efficiency 
can be calculated and expressed by its curve, and from it the designer can 
definitely and finally decide the exact speed of revolution at which the 
machines shall run. 

" It is only with direct-driven turbines that these considerations require to be made (r. Appendix A). 



SEA EXPKRIENCES WITH TURBINES. 143 

Power developed by a Turbine cannot be obtained direct from it. as that 
of a reciprocator is by means of the indicator, so some other means of finding 
it has to be adopted. When driving a dynamo to generate electrical current 
it is easy to determine the power exerted by a turbine by means of the 
electrical instruments used for measuring quantity and intensity of current. 
From data obtained in this way turbine efficiency has been deduced so that if 
the weight of steam used can be accurately determined, the power output 
can be calculated. To-day, however, marine engineers have a better means 
of gauging the capacity of the instrument driving the propeller in the torsion 
meter than even the indicator ever was. 

Prof. Rateau's Formula for Steam Consumption is as follows . — 

• • L - , ,, , 16-2 — 2-05 log P 
^team consumption in a turbine = 2* 1-3 -| , p . -• 

P is the initial pressure absolute. jj is the terminal pressure absolute. 

The Sea Experiences with Turbines in the cruiser " Chester," as related 
by Lieut. Yates, U.S. Navy, are interesting as well as instructive. This 
ship is fitted with Parsons' turbines of 19,000 S.H.P., driving four screws ; 
she attained a speed of 26*52 knots on trial. The lieutenant begins by 
emphasising the extreme importance of warming up the turbines thoroughly 
before attempting to run them, otherwise there is great danger of stripping 
the blades. This operation takes much longer than with a reciprocator, 
and should be from 3 to 3i hours when possible ; the turbine can be, of 
course, warmed quicker, but it is not advisable to do so. 

Water in the turbine causes no apprehension, and quickly disappears. 
" Whipping " of the rotors occurs in the H.P. turbine in bad weather, causing 
it to vibrate and groan slightly, and when the astern-going turbine is operating 
there are tremors. 

Manoeuvring was not difficult, for as many as 85 signals from the bridge 
were responded to in 50 minutes. It is found that the turbines do not 
adapt themselves to the new conditions imposed on them consequent on 
changes of speed made during manoeuvres at sea. 

A drop in boiler pressure from any cause generally is followed by wet 
steam, and the increase in blade friction due to it ; the loss of speed is thereby 
aggravated ; and a drop of 1 inch of vacuum caused a loss of speed of half 
a knot. Great care is necessary with the bearings, for if they get warm 
water cannot be applied as with the reciprocator ; there is nothing for it 
but to slow down, and sometimes to stop altogether, for a little heating 
causes sufficient expansion to make things much worse very quickly. The 
oil served, therefore, should be most carefully examined and kept in a high 
state of efficiency, and the oil used should be of the best and fittest for such 
service. Hot thrust bearings were not uncommon, due to the fine adjustments 
necessary to them. The wearing down of the bearings also is the cause of 
much trouble, even though it be as little as x^Vo °f an inch, for 
then the gland packings leak and make the engine-rooms unbearable if the 
ventilation is not really good. If the propeller blades are injured, and even 
slightly deformed, there is a marked effect on the performance ; but no 
cavitation was observable in this ship. Seeing, however, that the screws 
were quite small in diameter and well immersed, it is not very astonishing. 

The mechanical efficiency of these turbines was very satisfactory, for 



144 



MANUAL OF MARINE ENGINEERING. 




———PI 



Fig. 00. — Fottinger Torsion Metes 



••us 




Fig. 60a. — Fottinger Torsion Meter Diagrams (Reciprocator). 



PR. FOTTINGERS TORSION METER. 1^5 

when there was a small leak at the main stop valve of one, it kept turning 
at 100 revolutions per minute, while the vacuum held good ; and with steam 
shut off and no vacuum another shaft ran at half the revolutions of those 
then engaged in propelling the ship, due to " drag" on its propeller. 

Torsion Meters are now used on all trials of marine turbines, and by 
the engineers in charge of the machinery. As the name signifies, they are 
the measurers of the torsion or twist of a shaft — that is to say, they register 
the relative angular' movement of two transverse circular planes at a definite 
distance apart by a line in a plane through the axis, thus noting the points 
on their circumference and their distance apart when the shaft is subject to 
torsion. Suppose two circular discs to be keyed on a shaft 10 feet apart ; 
they are of considerable diameter, so that any small angular movement gives 
an appreciate circumferential one. If each has a radial line marked on its 
face so that these lines will be the traces of a plane through the shaft's axis 
when transmitting no pow r er, then when transmitting power they will be no 
longer in line one with the other, for the axial plane through the one line 
will be at an angle with that through the other. Moreover, experiment 
has shown that the angular movement will be in proportion to the amount 
of torque and, therefore, is a measure of it. Now, suppose one of the discs 
to have a light metal cylinder fitted to it, of sufficient length to come close 
to the other disc without actually touching it, and a line is marked on this 
cylinder longitudinally to indicate the position of the disc's radial line, any 
torsion of shaft will be seen at once by its displacement past the marking 
on the other disc. In a general way this displacement could not be observed 
when the shaft is revolving, and certainly could not possibly be noted ct 
the speed of revolutions of turbines. 

Amlser's Torsion Meter, however, is constructed on this principle, and 
its indications are easily and clearly seen and read by the ingenious method 
of causing an electric spark to give an instantaneous momentary illumi- 
nation of the index as it passes the eye, which produces the effect of an 
apparent stoppage of revolution, and admits of the reading of it quite easily. 

Dr. Fottinger s Torsion Meter is a mechanical apparatus, so arranged 
that it can register its movements on a sheet of paper, as does the common 
indicator. In this case (fig. 60) there is a mechanical connection with the two 
discs and sleeve by means of a system of compound levers, the outer end 
of the last one having a pencil or tracing pin, which rests on a sheet of paper 
laid on a fixed cylinder surrounding the shaft. If no power is being trans- 
mitted a plain line is made on the paper as the shaft revolves ; when power 
is being transmitted by a reciprocating engine a wavy line some distance 
from that base line is traced (v. fig. 60a), the waviness being due to the 
variation in twisting moment during each revolution of such an engine. With 
a t in bine under load there is no such variation in smooth water, consequently 
the meter traces another plain line parallel to the first, and its distance from 
it is the measure of the torque. The accuracy of this instrument depends 
very much on that of the mechanism, and seeing that what it rests on is in 
motion all the time the wear on it due to this will not improve it in that 
respect. Nevertheless, effecting, as it does, its own register, makes it a 
very useful instrument for observation, more especially is it the case when 
applied to reciprocators. There are, however, other instruments in general 
use which have very little or no mechanism, and such as there is 

10 



146 MANUAL OF MARINE ENGINEERING. 

permits of no doubt to arise as to the accuracy of the results derived 
from its use. 

The Hopkinson-Thring Torsion Meter has the great advantages over 
any apparatus produced hitherto that it requires only a very short length 
of shaft, and gives a direct reading of the torsion on the shaft. The zero 
point of the apparatus also can readily be ascertained by "barring" the 
shaft. The instrument offers also a simple means of determining the friction 
of the shafts themselves. 

The principle of the apparatus designed by Professor Hopkinson and 
Mr. Thring is a differential one, and consists in the observation of the twist 
between two adjacent points on the shaft by means of two beams of light 
projected on to a scale from a fixed and a movable mirror. The beam pro- 
jected on the scale by the fixed mirror is taken as the zero point, whilst 
the beam projected by the movable mirror indicates the amount of torque 
on the shaft. Both mirrors revolve with the shaft, but even at moderate 




Fig. 61. — Hopkinson and Turing's Torsion Meter Mounted Complete on a Shaft. 

speeds the reflections appear as continuous lines of light across the scale, and 
there is, therefore, no difficulty in taking readings. 

The torsion meter is shown in fig. 61, mounted complete on a shaft, whilst 
a diagrammatic arrangement of the complete apparatus is shown in end 
elevation and plan in fig. 61a. A collar, A, clamped to the shaft of which 
the torque has to be measured, is provided with a flange projecting at right 
angles to the shaft and an extension. 

A sleeve B (fig. 61a), provided with a similar flange and extension at one 
end, is clamped at its further end on to the shaft in such a manner that its 
flange is close to that on the collar A, whilst its extension overlaps that of 
the collar A. on which it is supported to keep it concentric. Both the collar 
and sleeve are quite rigid, and it is, therefore, obvious that when the shaft is 
twisted by the transmission of power, the flange on the sleeve B will move 
relatively to that on the collar A, the movement being equal to that between 
the two parts of the shaft on which these fittings are clamped. This move- 
ment is made visible by one or more systems of torque mirrors mounted 



THE HOPKINSON-THRING TORSION METER. 



147 



between the two flanges, which reflect a beam of light, projected from a 
lantern, on to a scale divided in a suitable manner on ground glass. 

Each system of torque mirrors consists of a mounting, pivoted top and 
bottom on one or other of the flanges, in which two mirrors are arranged 
back to back. This mounting is provided with an arm, the end of which 
is connected by a flat spring to an adjustable stop on the other flange. Any 
relative movement of the two flanges will turn the torque mirror, and thereby 



^-TORQUE MIRROR 





GLASS S&PXZr f^p 



Elevation. 



-zero mirrors 





Fig. (lire.— Hopkinson and Thring's Torsion Meter. 



cause the beam of light to move on the scale, the deflection produced being 
directly proportional to the torque applied to the shaft. 

With the arrangement described, a reflection will be received from each 
mirror at every half revolution of the shaft ; but where the torque varies 
during a revolution (as with reciprocating engines), a second system of mirrors 
may be arranged at right angles to the first system, so that four readings 
can be taken during one revolution ; or, if two scales are used, eight readings. 



148 MANUAL OF MARINE ENGINEERING. 

Fig. 61a shows how the beam of light reflected by the mirror when in its 
highest position passes through the upper part of the scale ; while the second 
reflection will occur when the mirror is in the position occupied by the zero 
mirror, the beam of light passing through the lower part of the scale. The 
position of the torque mirror in plan is such that the reflected beam strikes 
the scale to the right of the zero line, but when the shaft has made a further 
half revolution, the reflected beam from the other mirror will strike the scale to 
the left of the zero line. Obviously the deflection on both sides should be equal. 

The fixed mirror is attached to one of the flanges (in fig. 61a to the flange 
of the sleeve B). This must be adjusted so that the beam of light reflected 
from it is received at the same point on the scale as those from the movable 
mirrors when there is no torque on the shaft. To facilitate the erection and 
adjustment of the apparatus, the box containing the scale and carrying the 
lamp is fitted with trunnions, so that it can be inclined as required. 

If the position of the apparatus becomes altered relatively to the scale 
owing to the warming up of the shaft or from other causes, this is indicated 
immediately to the observer by an alteration in the position of the zero as 
reflected by the fixed mirror. Hence, the zero can be adjusted by moving 
the scale so that its zero coincides with the reflection from the fixed mirror. 
It will be obvious that it is not necessary to move the scale, as the mean of 
the two readings will be the same. It will readily be understood that a move- 
ment of the torque mirrors can only occur through a relative, movement of 
the two flanges, so that vibration of the shaft or of the ship will not influence 
the readings in any way. 

The constant of the instrument — viz.. the factor which, when multiplied 
cr divided into the product of the torsion-meter reading and the revolutions — 
gives the horse-power, may be calculated within 2 or 3 per cent., if the section 
of shaft within the instrument is uniform. But a direct calibration of the 
shaft with the instrument in position before the former is put into the ship 
should be made. This is effected easily by applying a known twisting couple. 
It is no inconsiderable advantage of this instrument that a direct calibration 
is established between the torsion-meter deflection and the torque on the shaft. 

In the Bevis-Gibson Torsion Meter, which is largely used in every-day 
work, light is also used to indicate the angular movement. In this case 
however, the discs are quite a part, and perforated with small narrow slits 
in such a way that when the shaft is running without load the light of an 
electric lamp behind the one disc can be seen through a sighting instrument, 
called a " finder," behind the other disc — that is, the lamp slit, its disc slit, 
the other disc slit, and the " finder " slit are all in line every time that part 
of the discs pass, when the shaft is running without load, so that on looking 
through the finder the flash of light is seen every time the slit passes the 
finder ; and since the revolution is rapid, it appears as a continuous illumi- 
nation. When power is being transmitted, the shaft is twisted ; the slits 
get out of line, and no light is seen through the " finder." The latter, how- 
ever, can be moved through an angle by means of a finely adjusted screw, 
etc., until the light is again picked up when looked at through the " finder " 
disc's slot at the time the lamp's disc's slot is passing the light slit. The 
angular displacement of the " finder " corresponds with the angular move- 
ment of the shaft, and the shaft horse-power can be determined thereby in 
the usual way. 



COLLIE'S TORSION METER. 



U9 



Collie's Torsion Meter (tig. 62) is also a mechanical contrivance of con- 
siderable ingenuity, by which the angle of twist is caused to be registered 
by a pointer and dial, not unlike the Bourdon pressure and vacuum gauge 
used on the L.P. valve box of a compound engine. In this arrangement 
two counter shafts are connected at the middle by the coarse-threaded end 
of one entering the threaded sleeve turned by and free to slide on the end 
of the other. Each shaft is driven independently by a Renald chain geared 
up so that it runs about three times to one of the main shaft. If the main 
shaft is not transmitting power, the sleeve simply revolves, and the pointer 





=3= 



=E3= 



=3= 




MAI* SHAFT 



Fig. 62. — Collie's Mechanical Torsion Meter. 



of the gauge is motionless in mean position. As soon as power is trans- 
mitted the shaft twists and one counter shaft moves in advance of the other, 
and forcing the sleeve to slide as a consequence of its screwed end moving 
on the male end of the other ; the pointer is thus caused to turn round so as 
to indicate the exact angular displacement of the length of shaft between 
the two driving wheels. Both these two meters, as, indeed, must all 
mechanical ones, depend on the accuracy of manufacture and the state of 
repair for the value of their indications ; it is evident that the magnification 
of the errors will be practically at the same rate as the magnification of the 



150 MANUAL OF MARINE ENGINEERING. 

indications. It was natural, therefore, for scientists and inventors to turn 
to the other means of indications and magnification, which had provided 
the means for measuring other forces with a delicacy that no mere mechanicaL 
contrivance can achieve. The mirror magnification of radius and conse- 
quently of the arc used in telegraphy no doubt suggested itself to others as it 
did to Prof. Hopkinson and Mr. Thring. 

The Denny- Johnson Meter differs from the others, inasmuch as linability 
of the points on the discs or otherwise is indicated by sound, instead of light, 
as transmitted to a telephone receiver. When the two points or projections 
in this case on the disc are in line, they are so close to two fixed projections, 
as to make virtually an electrical connection. When torsion takes place 
the connection is broken — that is, the meeting of the projections do not 
synchronise — and it is only by displacement of one of these fixed points 
equal to the relative circumferential movement of the point on the disc that 
restores synchronising contact with the consequent production of sound in 
the telephone. The amount of this movement is registered in the ordinary 
simple way, and from it the horse-power is estimated. 

Shaft Horse-power transmitted by a shaft can be calculated from the 
torque as follows : — 

T is the twisting moment or torque in inch-pounds. 

R, the revolutions made per minute by the shaft. 
The work performed) _ 2rxT _ 0-5236 TxR _ TxR 

per revolution j ~~W ° T U *°^ b ' *>•*•*— 33,000 ~~ 63,000 

The torque can be calculated from the angle of twist or torsion by means 
of the following formulre : — 

a is the arc at a radius r of the angle of torsion 9, - = fi ; I is the length 
and d the outer and d 1 the inner diameter of shaft. 

„ 10-2 xTxI,. r > 6 a 

P = M.X.cZ *" ( Rankme >- 360 = 2Vr ° C = ft? 

m , . ' x 2r. mt _ 10-2 xTx! 

That is p = -^- = 5W Then 5y - = - M x # > 

/• , fl __ 584 X T ><Jl for solid r \ ft - 584 T x l \f or hollow 

or(i.) U-- UrXd 4 | shafts, U ' M M,(^-(/ x 4 )/ shafts. 

That is T - /B84-xl ' 

M,. is the modulus of stiffness or rigidity of the material, which for steel' 
generally is 10 to 12 millions. With steel shafts of best make, experiments 
have shown the value of M,. to be 11,750,000 for solid, and 12,150,000 for 
hollow. In every-day practice. 11,250,000 is taken for solid steel shafts : — 

Ti t " Xl|4x 11,250,000 X d* 

Then T = m} = 19,264 — • 

Substituting this value of T in the formula for S.H.P. 

a tt -n m eta* x di R ° x d * x R n 1 tt- a 

S.H.P. = 19,264 -j- X 6 3^ = -3^^-T "olid shafts. 

S.H.P. = ^*_T_^^ f or hollow shafts. 
3-17 X I 



ORDINARY STEAM liNOINE INDICATOR 



151 



It may be observed that it is usual to experiment in the workshops with 
the shafts of every ship to ascertain beforehand what amount of torque is 
necessary to produce a degree of angular movement in a definite portion of 
its length ; from such experiments the modulus of rigidity is ascertained. 
It is then easy to construct a diagram from that particular shaft, by which 
the power may be read off for any angular movement indicated by the torsion 
meter, and the revolutions at the time of observation. Fig. 63 is such a 
diagram as to need no explanation. , • 

It was held at one time that the end pressure on a ship's shaft due to 
thrust seriously affected the register of torque ; it was supposed until lately 
to affect it to the extent even of 3 per cent. ; but more recent investigations 
by Dr. Hopkinson with more sensitive instruments seem to have removed 
the impression, and that practically end thrust may be disregarded now as 
a factor in the calculation of shaft horse-power. 




fuii line Curve Si'Ows Cranh effort Curse deduced from indicator Diagrsm. i-etters corresponding position ct KPCrenk 
Dotted, ., ,. .. „ obtained , Light Tors ion /deter. Numbers , „Si'oCa in Tonic' texrOtlt 

MEAN TwiSTInO MOMiii DlAORAU fOR a RecipROCATino EnOINS. 

h.t>. 



IP 




LP. 




0' 30' 60' 90' Ob' ISO' 180' 210' 240' 770' 30f 330" 3d' 

Fig. 63. — Crank Effort and Torsion Meter Diagram (J. Hamilton Gibson). 

The Ordinary Steam Engine Indicator is often spoken of as a defective 
instrument, and one not to be relied on to give accurate results. It is quite 
true that it is not exactly a perfect one, and that the power registered by 
means of it comes short of the actual amount developed in the cylinders, 
but, at the same time, while admitting this, it should not be forgotten that 
it renders another and a very good service to the engineer besides that of 
giving him the power, and one that is quite as important to him. It shows 
in a ready and rapid way whether the internal and unseen parts of the engine 
are in good working order, and efficient for their service. It is to him what 
the stethoscope is to a doctor of medicine. 



15*2 MANUAL OF MA1UNE ENGINEERING. 



CHAPTER VI. 

EFFICIENCY OF MARINE ENGINES. 

The Efficiency of an Engine should be measured by the cost of the useful 
work it does on service ; formerly the fuel burnt in the boiler from which 
the steam was supplied was taken as the measure of cost, and so long as the 
fuel was of standard quality and the efficiency of the boilers constant, it was 
a fair and ready means of doing so. But coal from the same pit may vary 
in calorific value even when freshly raised, and undoubtedly does so after 
exposure ; the method of ascertaining the weight permits of inaccuracy, 
although perhaps this is only slight ; the state of the atmosphere seriously 
affects consumption, for it is obvious that cold moisture laden air will require 
more fuel to raise it to the temperature of the furnace for combustion than 
is necessary for an equal amount of dry warm air ; the human element like- 
wise plays an important part in the production of steam, as may be evidenced 
when the single boiler of a ship is barely large enough for the engine require- 
ments ; in such a case a good and experienced stoker will keep the steam 
pressure and supply steady, whilst an indifferent one will use more fuel, 
and then fail to maintain full pressure. Further, atmospheric conditions 
seriously affect the quality of the draught, so that while with a good breeze 
the draught will be sharp and the combustion efficiently effected, on a hot 
sultry day with no wind it will be bad, and the fires require much forcing. 
Then, too, the condition of the boiler is not always the same ; the grate 
bars may be faulty, the tubes dirty, and the inside with more scale at one 
time than another ; but it is only fair to say that nowadays the scale should 
never be so very great as to cause serious differences. Taking all these 
things into account, it is obvious that the more correct method is to debit 
the engine with the weight of steam supplied to it rather than with the 
weight of coal. In old days, when there were no auxiliaries, but only the 
main engines, there was no call to differentiate as there is now the steam 
supplied from the boilers between that used by the main engine and that by 
auxiliaries, etc. To-day the amount needed for other purposes than that 
of driving the main engines is very appreciable, amounting to as much as 
14'5 per cent, of the total used in the " Lusitania " when running at full 
speed. There are in passenger ships centrifugal circulating and air and feed 
pumps constantly going, but they are virtually a part of the main engines ; 
steering engines, ash hoists, blowing fans, ventilating fans, refrigerators, 
electric lighting, sanitary pumps, etc., are also working during a considerable 
portion of the day, and bilge pumps, fire and wash-deck pumps, steam whistle, 
etc., take a good share also ; besides all these in cold weather steam heating 
will demand an extra expenditure of fuel, an important item in the North 
Atlantic. 



VERTICAL ENGINES. 153 

The main propelling machinery, therefore, must be debited only with 
the steam it uses and that used by such of the auxiliary machinery as is 
necessary to keep the main engines running. 

With such feed and other pumps as are now employed in the engine- 
rooms of important vessels, the water consumption can be closely measured, 
as they are capable of acting as meters of the water they pass by simply 
fitting them with counters to record the number of strokes they make ; the 
chief engineer can thus quite easily calculate the actual amount of water 
pumped from the main condenser and from the auxiliary condenser, and 
record thereby the steam consumption of the machinery as a whole, or of 
the main engines only. 

To show the cost of the useful work done by marine engines used to be 
somewhat difficult, and could be calculated only by using assumptions that 
were always somewhat uncertain. That is, until the torsion meter was used 
the work transmitted to the propeller was, except in some special exceptions, 
practically guessed at ; now, however, thanks to these instruments, we can 
measure what is called the shaft horse-power of both turbines and recipro- 
cators, and thereby compare the outputs by reducing them to a common 
denominator ; this being so : — 

Water consumed per S.H.P. (shaft horse-power) is the measure now of 
the efficiency of a marine engine, qud engine, without complications or 
explanations. But by this method the general efficiency is shown without 
any differentiating between the steam and the mechanical efficiency of 
the system. 

The Mechanical Efficiency, or the relative value of the engine as a piece 
of mechanism, is measured by comparing the shaft horse-power with the 
gross power generated, as shown by the indicator diagram ; 

Mechanical efficiency of a marine engine, therefore, is S.H.P. -h I.H.P. 

Mr. Denny found by using a torsion meter that the mechanical efficiency 
of some quadruple engines made by his firm was as high as 94 per cent. 

The late Mr. Mudd ascertained the power required to move certain triple- 
compound engines standing in the erecting shop without screw shafting 
was at working revolutions 45 I.H.P., or 5 - per cent, only of the gross power 
indicated (900) when working at those same revolutions in a loaded ship. 
This would show, then, the efficiency to be 95 per cent. ; but it must be noted 
that in this case no allowance is made for the extra friction on the valves, 
guides, etc., due to the greater pressure on them when at work at full power. 

The older horizontal jet-condensing engines had comparatively a low 
efficiency due to the general friction of the very heavy moving parts and 
the resistance of the two air, two feed, and two bilge pumps worked by them. 
There is good reason to believe that the mechanical efficiency of some of 
them was seldom over 75 per cent. Latterly, however, the well-made hori- 
zontal engines of Penns, Maudslays, etc., when running at the higher revolu- 
tions, and developing much more power than •formerly, and having surface 
condensers, had even 80 to 85 per cent, efficiency. 

Vertical Engines with surface condensers and slow running had an effi- 
ciency from 5 to 8 per cent, better than the horizontals of similar size, and 
working under similar conditions. To-day naval machinery and that of 
express steamers have an efficiency from 90 to 94 per cent, at full speed, 



154 MANUAL OF MARINE ENGINEERING. 

when with only the air pumps driven by it, and of the triple- and quadruple- 
reciprocating type. 

The Mechanical Efficiency of a Turbine has been determined approxi- 
mately by observing the electro-motive power required to revolve it at 
working speed by means of a motor, and taking it as the mechanical loss 
when working. Its efficiency is, of course, high, and from such observations 
made with similar machines in service on land, it may be taken that the 
marine turbine has a mechanical efficiency of about 95 per cent, in a general 
way, and that with the largest ones it may have a somewhat higher, probably 
96 to 97 '5 per cent. 

The Mechanical Efficiency of the Reciprocating Engine is not so high as 
this, although that of well-designed carefully made and balanced engines 
of high speed will not be far short of the 95 per cent, when running in good 
working order and free of all pumps, and well lubricated. Under these cir- 
cumstances its efficiency is probably 92 to 94 per cent, at least. It is, how- 
ever, somewhat misleading to express the mechanical losses as a fraction 
of the Total Indicated Horse-Power, inasmuch as that varies closely with 
the cube of the revolutions, while the losses vary more nearly as the revolu- 
tions ; so that, although an engine may show an efficiency of only 80 per 
cent, at 50 revolutions, it may be as much as 94 at full speed of 75 revolutions 

/50\ 3 
— that is, the loss is (,=v) X 20 — without any change in the adjustment of 

a single part. 

Mechanical losses really depend largely on the size of the engine. Now, 
in a general way the Nominal Horse-Power expresses fairly well the size 
of any engine, and, therefore, all other things being equal, it will be a suffi- 
ciently accurate assumption that frictional losses are proportional to N.H.P. 
Practice has demonstrated that about 70 per cent, of these losses will in the 
ordinary marine engine vary directly with the revolutions within reasonable 
limits, and further that the remaining 30 per cent, increase at a more rapid 
rate ; in fact, in modern engines total losses roughly vary as the cube root 
of the revolutions raised to the fourth power — that is, as R*. 

It is obvious that the mechanical efficiency of any engine will vary from 
time to time, and under some circumstance the variation may be very con- 
siderable. It is also well known that at very slow rate of revolution the 
apparent mechanical losses are uncertain, and always proportionately large. 
The only losses that can be considered here are those which inevitably occur 
with any engine when in quite a good state of repair and in really good working 
order. 

The Efficiency of Marine Engines, when well made, in this good state 
of repair, and in good working order, should be approximately in accordance 
with the following rule : — 

«..;. , N.H.P. xR. . »«. 

Junction horse-power = ■ — (// + # vR). 

„«. . LHP. - F.H.P 
Efficiency = - -yg-p 

Nominal horse-power = D X S -s- K, 
where D is the diameter of the L.P. piston, and S the stroke, both in inches. 



EFFICIENCY OF MARINE ENGINES. 155 

For the two-stage compound engine K = 15 "0. 
,, triple-stage ,, K = 126. 

., quadruple-stage „ K = 10"5. 

R the revolutions per minute. 

For diagonal paddle-wheel engines with air pumps only X == 1*5 ; y — 10. 

„ vertical screw engines, mercantile, with all pumps 

connected, . . . . . . . x = 1*0; y = 8. 

,, light quick-running screw engines with all pumps 

connected, . . . . . . x = 07 ; y = 7. 

., naval and express screw engines, centrifugal cir- 
culating, . . . . . . . x = 0'6 ; y = 7. 

., naval and express screw engines, with only air pumps x = 0'5 ; y = 6 - 5. 

,, special naval screw engines, no pumps . . . x = 0'3 ; y = 6'0. 

,, ,, forced lubrication, no pumps. . . x = O'l ; y = 5*0. 

Example 1. — A " tramp " steamer having an engine with cylinders 22 r 
37, and 62 inches diameter and 45 inches stroke, which at 85 revolutions- 
develops 2,300 I.H.P., what is the efficiency ? 

Here N.H.P. = 62 x 45 - 12 6, or 221. 

Frictional H.P. = — ^~- (8 + ^85) = 220. 

Efficiency = (2,300 - 220) -j- 2,300, or 0-004. 

Example 2. — A destroyer has two sets of engines, each having cylinders 19, 
28 - 5, and 43 inches diameter with 18 inches stroke at 370 revolutions ; the 
total T.H.P. is 4,200. 

Here N.H.P. each engine = 43 X 18 - 12'6 = 61. 

Friction H.P. = 6 -\^-° (6 + 0'3 ^370) = 184. 

Efficiency = (2.100 - 184) + 2,100 = 0"913. 

Example 3. — A paddle steamer having cylinders 56 and 110 inches diameter 
and a piston stroke of 72 inches develops 7,150 I.H.P. at 50 revolutions, 
what is the F.H.P. and efficiency ? 

Here N.H.P. = 110 X 72 - 15, or 530. 

F.H.P. = ^^° (19 + 1-5 V50) = 411. 

Efficiency = 7,1 ^^ U = 0*942. 

Example 4. — A naval ship has two engines, each with cylinders 43 and 
69 inches, and two L.P. each 77 inches, the stroke being 42 inches. 

Here the two L.P. cylinders are equivalent to one 109 inches diameter. 

At 140 revolutions each engine develops 12,000 I.H.P. 
„ 126 „ „ 8,000 „ 

,. 85 ,. „ 2,500 „ 



156 



MANUAL OF MARINE ENGINEERING. 



Now, 


N.H.P. 


(<0 


F.H.P. 




Efficiency 


(h) 


F.H.P. 




Efficiency 


(c) 


F.H.P. 




Efficiency 



109 X 42 h- 12-6 = 363 

363 x HO 



1.000 

12,000 - 514 
12,000 

363 x 126 



(7-0 + 0-6 x t/l40j = 514. 
= 0-957. 



1,000 

8,000 — 460 
8,000 

363 x 85 



(7 + 0-6 4/126) = 460. 
= 0-942. 



1,000 

2,500 - 298 
2,500 



(7 + 0-6 £/85) = 298. 
= 0-881. 



Example 5. — A triple-compound engine with cylinders 18J, 28, and 40 
inches diameter, the stroke 20 inches, and at 250 revolutions the I.H.P. 
is 1,150. There are no pumps, and lubrication is forced throughout. 



Here 



N.H.P. = 40 x 20 - 12-6 = 63-5. 
tion HP. = 63 1 ?J 5 ° (5 + 0-1 #250) = 89'5. 



Efficiency = 



1,000 
1,150-8 



1,150 



, or 0-922. 



Dr. Bauer states that from observations of and experiments with marine 
engines made in Germany the efficiency of small ones with an indicated 
horse-power not exceeding 50 is only 0'59, while those of 5,000 I.H.P. and 
upwards it was as high as 0"91 ; also that the efficiency varied roughly from 
059 to 0'91 in accordance with their power for other engines. He quotes 
some examples of engines whose actual efficiency had been determined, and 
they are as follows : — 

TABLE XXV. — Efficiency of Some German Engines. 



1 

German Engine. 


A 


BCD 


E 


V 


Indicated horse-power, 
• Efficiency, 


1,630 
0-885 


1,640 1,940 2,370 
0-910 0-911 0-920 

1 


2,890 ! 4,500 
0-911 0-935 



Mr. Denny found by torsion meter trials with a reciprocating engine 
that the efficiency at full power, 1,550 I.H.P., was 0'92 ; and in another 
case with a power of 1,950 it was as high as 0'935. 

The Results of Trials made with Triple-compound Engines designed for 
electric generating are shown in fig. 64, and are very instructive, although 



TRIALS WITH TRIPLE-COMPOUND ENGINES. 



157 



not made under quite the same conditions as those obtaining with a marine 
engine, for in these cases the load was an artificial one (brake), and somewhat 
arbitrarily varied. 

The following figures and diagrams show the results of some carefully 
made trials with two triple-compound engines, by Messrs. Belliss & Morcom. 
Birmingham. They are of their special enclosed type, having three cylinders 
and three cranks supplied with steam at 150 lbs. pressure, and. exhausting 
into a surface condenser whose pumps are operated by independent engines. 
The whole of the pins, bearings, guides, and working parts are lubricated 
with their special American mineral oil forced through them by a pump 
constantly at work connected to the engine. It will be seen that the friction 
per revolution, both with and without load, varies, and has two minimum 
and two maximum values ; that at full speed the frictional H.P. per revo- 
lution loaded is really only a little less than that when running free, whereas 
at "200 revolutions, or half speed, there is the greatest difference between 
them, that at full load being half that running free ; also that at dead slow 
speed, say 25 revolutions, the friction per revolution, both light and loaded, 
is nearly four times that of the loaded engine at 200 revolutions. 

11" — 17" — 24" 
Engine (No. 1401) ^y> 150 lbs. at Engine Stop-valve. 



Table of Powers, <kc, Under 


Load Con 


ditions. 






Revolutions. 


166 


300 


350 


400 


I.H.P. 

B.H.P., . . . . - . 
Difi'erence or friction, H.P., . 
I.H.P. per revolution, . 
B.H.P. ,, ,, 
F.H.P. ,, ,, ... 

B.H.P. , . 

i h p efficienc y> • • . • 


248-2 

240-4 
7-8 
1-496 
1-45 
•047 

06-87. 


384-15 
360-5 
2365 
1-280 
1-201 
•079 

03-9 7. 


408-1 
380-5 
27-6 
1-168 
1086 
•079 

93-25 7. 


429-4 
400 
29-4 
1073 
100 
•073 

93-15 7. 













Friction Powers Under No-Load Conditions. 



• 


Revolutions. 


56 


102 


113 


213 


313 


365 


Friction H. P., .... 
,, ,, per revolution, 

-t 


4-9 

•0875 


10-7 
•1049 


7-4 
•0655 


24-6 
•1153 


361 
•1153 


30-9 

■0847 



The above figures are embodied in the diagram (fig. 64). 
The no-load cards were taken non-condensing, but the load cards were 
taken condensing. 

This is probably the reason why the friction power shown at no-load 



158 



MANUAL OF MARINE ENGINEERING. 




EXPERIMENTS WITH TORPEDO BOAT. 



159 



is greater than that at full-load. The efficiency under load condensing is 
invariably slightly better than when non-condensing. 

It is also interesting to note how rapidly the rate of friction per revolution 
increased when the power was high and revolutions were quite small. The 
efficiency of the larger engine at full load and highest revolutions was - 929, 
and the friction losses were '047 I.H.P. per revolution at 170 ; 0"079 at 
300 revolutions ; 0"078 at 350 revolutions, dropping to 0"073 at 400. These, 
however, are engines having forced lubrication. 

Mr. Yarrow's Experiments with a Torpedo Boat to ascertain the efficiency 
of engines, propeller, and ship were made in a most careful and exhaustive 
way, and are most interesting and instructive. They show that the mechanical 
efficiency of the engines varied from - 766 at 9 knots to - 923 at 15 knots, 
and that the mechanical losses varied at a higher rate than given by the 
arithmetical progression of revolutions. The high efficiency of these quick- 
running engines at full power when carefully manufactured and adjusted 
is manifested by an inspection of the figures in the subjoined table. 

TABLE XXVI. — Yarrow's Experiments on Efficiency. 



Speed, • knots, 


9-0 10-0 


110 


120 


130 


14-0 


15 


Ind. horse-power, . 38-5 49-5 
Friction, etc., loss, 9-0 10-1 
Efficiency, . . 0-766 f 0-796 


67-1 
11-7 

0-827 


99-0 
13-5 

0-864 


143-0 
15-4 

0-892 


193-6 
17-4 

0-910 


255-2 
19-5 
0-923 



Mr. A. H. Tyacke, of Hull, made a series of trials with the engines of 
two ships manufactured by Earles Co. to ascertain frictional resistance at 
different speeds. In each case the engines were allowed to run without their 
piopeiler shafting connected ; indicator diagrams were carefully taken at 
various rates of revolution. At the same rates a set of diagrams were taken 
with the ship running with propellers connected. The first ship had cylinders 
12, 20, and 32 inches diameter, and a piston stroke of 21 inches. The second 
ship had somewhat larger engines, the cylinders being 12|, 22, and 36 inches 
diameter, and a piston stroke of 24 inches, and the highest revolutions only 
111, which is small for such small cylinders. Under these circumstances 
the efficiency is remarkably good, especially that of the latter. 

TABLE XXVII. — Tyacke's Efficiency Experiments. 



Revolutions per Minute. - 54 70 


S8 90 


106 


Indicated horse-power, . . 54-6 87-0 180-9 223-6 
Frictional horse-power, . 20-5 27-4 34-7 4(1-4 
Efficiency (I.H.P. - F.H.P.)- \ Q ,^ 1 . 6g9 . 808 j . gl9 
1 ri.l ., . . . . ) 

1 , 1 


306-0 
46-2 

0-849 



160 MANUAL OF MARINE ENGINEERING. 

TABLE XXVIIa.— Tyacke's Further Experiments. 



Revolutions per Minute, - 


40 


60 


93 111 


Indicated horse-power, 
Frictional horse-power, 
Efficiency (I.H.P. - F.H.P.) -f- I.H.P., . 


38-80 
1415 
0-635 


82-20 
22-39 

0-728 


322-00 
40-70 
0-873 


480-60 
47-05 
0-902 



Steam Efficiency is quite as necessary as mechanical efficiency to a success- 
ful marine engine, and has to be as carefully considered. First of all, the heat 
conditions under which it works are the raising of a pound of water from the 
temperature of the hot-well to a pound of steam with a sensible temperature, 
and a latent heat corresponding to the boiler pressure ; during the cycle of 
operations on the engine it is reduced again to water of the original tempera- 
ture. Its efficiency then will be measured by comparing the heat given up 
during the cycle when doing work with the total heat. If T is the tempera- 
ture of steam at P pressure, L its latent heat, and t the temperature and I the 
latent heat due to the pressure in the condenser — that is, the total heat of 
evaporation at boiler pressure minus the heat of the feed-water is the amount 
debited, and that amount less the total heat of evaporation at the pressure in 
the condenser is to the credit account — the efficiency is, therefore, 



Efficiency of steam 



(T + L - t) -{I) T + L - * - I 



T +L-t 



T + L-« 



The numerator of this fraction is the greatest amount of heat possible to 
be used with steam of the pressure, and multiplying it by 772 will give the 
maximum amount ci mechanical work possible in an engine working under 
the conditions imposed. Dividing the product by 33,000 will give the horse- 
power. 

For example, marine engine condensers may now be relied on to produce 
a vacuum of 28 inches, so that the back pressure is 1 lb. The total heat 
at this pressure is 102° sensible, together with 1,042° latent, or 1,144°. At 
100 lbs. pressure absolute the total heat is 327°+ 884°, or 1,211° ; at 150 lbs. 
it is 358°+ 862°, or 1,220°; at 200 lbs. it is 381°+ 845°, or 1,226°, while at 
250 lbs. it is 401 c + 831°, or 1,232°. 

The heat required to change a pound of water at 102° temperature to a 
pound of steam at these various pressures will be 1,109, 1,118, 1,124, and 
1,130 B.T.Us. ; in each case the rejection is 1,042°. 
Hence at 100 lbs. pressure the heat available for power is 1,109°— 1,042°, or 67°. 

The efficiency of the steam = 67 -J- 1,109, or 0*0604 — that is, 6 per cent. 
At 150 lbs. pressure the heat available is 1,118° — 1,042°, or 76°. 

The efficiency of the steam = 76-4- 1,118, or 0*068 — that is, 6*8 per cent. 
At 200 lbs. pressure the heat available is 1,124° - - 1,042°, or 82°. 

The efficiency of the steam = 82-4- 1,124, or 0*073— that is, 7*3 per cent. 
At 250 lbs. pressure the available heat is 1,130° — 1,042°, or 88°. 

The efficiency of the steam = 88 -*- 1,130, or 0*078 — that is, 7*8 per cent. 
In actual practice steam of 200 lbs. pressure enters the engine at a tempera- 
ture of 381° F., and leaves it at the condenser at about 120°, so that it gives up 
in the engine 261°, which multiplied by 772 and divided by 33,000 is 6*1 H.P. 



STEAM EFFICIENCY OF RECIPROCATORS. 



10 1 



The triple-expansion engine of the mercantile marine consumes 16 lbs. 
per horse-power-hour, or 0-267 per minute. In such an engine a pound of 
steam produces 3-75 I.H.P. 

Hence the steam efficiency of the engine is 3*75 -f- 6-1, or 0-615. 

An engine using steam at 250 lbs. pressure absolute receives it at a tem- 
perature of 401°, and rejects at 120°, giving up to the engine 281°, which is 
equivalent to 6-58 H.P. 

A quadruple engine under these conditions consumes 14-5 lbs. of steam 
per I. H.P. -hour, or 0-242 lb. per minute, consequently a pound of steam 
produces in it 4-13 I.H.P. 

Hence its steam efficiency is 4-13 -f- 6-58, or 0-628. 

A turbine using steam of 200 lbs. pressure and rejecting at 81° will absorb 
300°, equivalent to 7-07 H.P. 

Assuming it to consume only 12 lbs. of steam, the power generated in 
by a pound of steam is 4-99 H.P. 

The efficiency of this turbine is 4-99 -*- 7-07, or 0-706. 

The Steam Efficiency of the Best Turbines is as high as 72 per cent, when 
of large size, and under favourable circumstances ; that of turbines of good 
make and over 2,000 shaft horse-power can be taken at 64 to 66 per cent. 
With superheated steam those on shore of large size can be depended on to 
show an efficiency of 72 to 75 per cent. 

The Steam Efficiency of Reciprocators of large size, good design, and 
good construction, 60 to 63 per cent, is satisfactory. 

The following table gives the maximum amounts of work generated 
theoretically by a pound of steam during admission and expansion : — 

TABLE XXVIII. — Maximum Work done by 1 lb. of Steam Press. p v expanding 
Adiabatically to Press. p 2 , and exhausting to Condenser at 1 lb. Press. 



Initial pressure, 100 lbs. 125 lbs. 


150 lbs. 


175 lb3. 


1* 

200 lbs. 


250 lbs. 


Terminal p 2 . 


B.T.T7. 


HP. 


B.T.U. 


H.P. 


B.T.U. 


H.P. 


B.T y. 


n.p. 


B.T U. 


H.P. 


B.T.U. i H.P. 


1-0 lb. 


288-0 


6-80 


303-0 


714 


314-5 


7-42 


324-0 


7-64 


332-5 


7-84 


348-0 


8-21 


20 lbs. 


286-8 


6-76 301-8 


7-09 


313-3 


7-26 


322-4 


7-61 


331-3 


7-81 


346-8 


8-18 


30 .. 


277-2 


6-54 ! 292-6 


6-90 


309-2 


717 


313-5 


7-40 


322-2 


7-60 


337-2 


7-95 


4-0 .. 


266-2 


6-28 281-9 


6-65 


294-0 


6-93 


303-2 


7-15 


3120 


7-36 


328-2 


7-74 


5-0 ., 


258-0 


6-09 ■ 273-5 


6-45 


285-0 


6-72 


295-0 


6-96 


304-0 


717 


320-0 


7-55 


60 .. 


249-9 


5-90 266-9 


6-27 


278-0 


6-55 


287-9 


6-80 


297-4 


7-01 


313-9 


7-40 


70 .. 


243-9 


5-75 


259-3 


6-11 


271-9 


6-42 


281-8 


6-64 


290-8 


6-86 


307-8 7-26 


80 .. 


237-5 


5-61 


253-5 


5-96 


265-5 


6-26 


276-8 


6-53 


285-5 


6-73 


302-5 7-13 


9-0 „ 


2320 


5-47 


248-0 


5-85 


260-5 


6-14 


271-3 


6-40 


280-7 


6-62 


297-0 


7-00 


10 „ 


226-9 


5-35 


2421 


5-71 


254-0 


6-00 


265-5 


6-26 


275-9 


6-5) 


292-1 


6-89 


11 „ 


2221 


5-24 


237-1 


5-59 


249-8 


5-89 


260-7 


615 


270-5 


6-38 


287-1 


6-77 


12 ,. 


217-0 


512 


233 


5-49 


245-5 


5-80 


256-1 


6 04 


265-4 


6-26 


283-0 ' 6-67 


13 „ 


212-2 


5-00 


228-6 


5-39 


240-7 


5-69 


251-6 


5-93 


261 1 


6-16 


278-7 6-57 


14 ., 


207-4 


4-90 


224-1 


5-28 


236-4 


5-60 


247-2 


5-83 


257-3 


6-07 


274-4 6-47 


15 


2040 


4-81 


220-0 


519 


2320 


5-47 


243-5 


5-74 


253 


5-97 


270-0 <6-37 


10 „ 


200-6 


4-73 216-6 


511 


228-6 


5-39 


239-6 


5-64 


249-9 


5-89 


266-4 6-28 


17 „ 


197-4 


4-65 213-3 


5 03 


225 


5-30 


236 


5-56 


246-5 


5-81 


262-6 6-19 


18 „ 


194-3 


4-58 210-1 


4-95 


220 


5-23 


232-9 


5-49 


243-3 


5-74 


259-9 6-11 


19 „ 


191-3 


4-50 207-2 


4-88 


219-4 


517 


230 


5-42 


240-5 


5-67 


257-3 


6 06 


20 „ 


188-4 


4-44 204-4 


4-82 


217-3 


5- 12 


228 


5-38 


238-0 


5-61 


255-4 


6-02 


25 „ 


1760 


4-15 191-3 


4-51 


2(15-0 


4-83 


215-9 


5-09 


225-4 


5-31 


243 5-73 


30 „ 


164-8 


3-89 180-0 


4-24 


194-5 


4-59 


205-0 


4-83 


214-9 


5-07 


232-0 


5-47 


35 .. 


L55-2 


3-66 169-6 


4-00 


184-2 


4-34 


L95-2 


4-60 


205-2 


4-84 


.).)■>. o 


5-24 


40 „ 


146 -2 


3-45 L60-5 


:(-so 


175-1 


413 


I86 


4-40 


.196-0 


4-62 


21 30 


5-02 
























11 





162 MANUAL OF MARINE ENGINEERING. 

Steam Efficiency may also be ascertained by referring to Table XXVIII., 
where it will be seen that a pound of steam expanding adiabatically from 
200 lbs. absolute to 1 lb., when it enters the condenser, can theoretically give 
332-5 B.T.U., equivalent to 7-84 horse-power. 

A Turbine practically works on these conditions with a steam consumption 
of about 11-5 lbs., or 1 lb. of steam is good for 5-24 H.P. In this case, then — 
The efficiency = 5-24 + 7-84, or 66-8 per cent. 

A triple-compound engine using steam at 200 lbs. pressure absolute ex- 
panding to 8 lbs. (nominal expansion rate 25), and exhausting at 1 lb. to the 
condenser, requires 15 lbs. of steam per H.P.-hour, so that 1 lb. is good for 
4 H.P., and theoretically under these conditions develop 285-5 B.T.U., equi- 
valent to 6-73 H.P. (v. Table XXVIII.). In this case— 

(a) Efficiency = 4 -4- 6-73, or 0-594, or 59-4 per cent. 
But 1 lb. of such steam if expanded to its full extent is shown to be good foi 
7-84 H.P., then— 

(b) Efficiency = 4-4- 7-84, or 0-51, or 51-0 per cent. only. 
A Quadruple Engine working with steam at 250 lbs. absolute, expanding 
to 8 lbs., or nominally 31 times, exhausts to the condenser at 1 lb., consumes 
14 lbs. of steam per H.P.-hour, so that 1 lb. is good for 4-286 H.P. But 1 lb. of 
steam is theoretically good for 302 -5 B.T.U. or 7 -13 H.P. under these conditions- 
Then efficiency = 4-286 -f- 7-13 = 0-601, or 60-1 per cent. 
Taking the full value of 1 lb. of such steam as 8-21 {v. Table XXVIII.)— 
(b) Efficiency = 4-286 -=- 8-21 = 0-52— that is, 52-0 per cent. only. 

High Pressure : its Advantages and Disadvantages. — Before the intro- 
duction of the compound engine for marine purposes, the boiler pressure had 
been as high as 60 lbs. in quite large steamers, in H.M. service, with the 
non-condensing engines of 200 N.H.P. ; some of these were fitted in certain 
battle-ships during the 1855 Russian war ; but after the compound engine 
secured the confidence of all classes of steamship owners, that pressure was 
very much exceeded with beneficial results. 

Prior to the general use of the triple-compound engine 90 lbs. was a very 
common boiler pressure, and many ships' boilers were made for a working 
pressure of 100 lbs., and a few for as high as 110 lbs. The triple-compound 
engine itself was for a long time worked with steam of 150 lbs. ; then 165 lbs. 
became a fashionable pressure, to be soon superseded by 175 lbs. ; and 
now even 200 lbs. is general, although the economic gain with triples by 
going from 150 to 200 lbs. is questioned by some engineers who have made 
careful observations of all the conditions involved in the change.* 

The objection to still higher pressures is rather of a practical nature, 
but can be safely overcome, since steam superheated to 600° F. may now be 
use*l. Steam at a pressure of 250 lbs. absolute has a temperature of 401° F., 
or nearly that of the melting-point of tin. It will, therefore, affect the 
condition of some of the metals with which it comes in contact, rendering the 
surfaces brittle and in a bad condition to withstand severe rubbing. More- 
over, common unguents are vaporised, and the walls of the cylinders become 
too hot to condense their vapour when exposed to very high temperatures ; 
but the heavy mineral oils now used, however, have a boiling point of 

•The N.E. Coast Inst. E. and S. Standard Specification of 1917 gives 180 lbs. for cargo steamers 
"it!) triples. 



FRICTION OF THE PISTON. 163 

700° F. The difference in expansion of different metals is so considerable, 
that the utmost care must be exercised in the design and manufacture of the 
cylinder, etc., to prevent racking, which causes leakages and breakages. The 
pressure necessitates considerable thickness in all cast-iron work ; and 
neglecting all considerations of weight or cost, this alone constitutes a source 
of objection and danger, inasmuch as the sudden exposure of thick masses 
of this metal to high temperatures on one side only is sure to distort, and 
very likely to fracture it. The liability to leakage is, of course, greater 
with the higher pressure, independent of temperature, and the danger to 
the attendants, from even small explosions, is very much increased. Again, 
in order to expand steam at this pressure, so as to obtain from it the maximum 
efficiency, it will be found necessary to use a series of cylinders ; and although, 
as will be shown later on, the engine with four stages is not without virtue, 
a larger number will not commend itself to engineers generally for engines 
of the smallest and largest sizes. In practice, it has been found that an 
engine can be worked with steam of a pressure less than that usual, but 
superheated to the temperature corresponding to the higher pressure, and 
yet be more economic and efficient than a similar one working with that 
higher pressure of steam. 

Efficiency of the Engine as a Machine. — The marine engine suffers loss, 
in common with all machines, from certain physical causes beyond the 
absolute control of the most skilful designer, and engineers can only aim at 
mitigating the evil, without entirely overcoming it. The chief cause of loss 
of energy is, of course, friction (1) of the piston, (2) of the stuffing-boxes, 
(3) of the guides and slides, (4) of the shaft journals, (5) of the valves and 
valve-motion. Another source of loss is that from the resistance of the 
pumps ; and, finally, the inertia of the moving parts, which have a reciprocal 
action, as in the piston and rods, may be the cause of further waste of energy. 
Unless the momentum is balanced, or the energy imparted at one part of 
the stroke and stored in the heavy moving masses, is given out wholly by 
the end of the stroke, a serious loss ensues, and the mechanism has to sustain 
the strain of forces which might be otherwise usefully employed. 

1 . The Friction of the Piston in the vertical engine in smooth water depends 
on the pressure of the packing on the sides of the cylinder ; so that, if the 
piston were solid and simply a good fit, there would theoretically be no 
friction, and in practice none beyond that due to the viscidity of the unguent 
and to the pressure on the sides of the cylinder from the rolling and general 
motion of the ship. This is what is required in a piston, and that one is 
most nearly perfect which is capable of moving steam-tight in the cylinder 
with least pressure of the packing, and so approximates to the condition of 
a solid one. Resistance due to this cause has been reduced to a minimum 
in modern engines by care in manufacture and skill in design ; the cylinder 
is now truly and smoothly bored from end to end, the metal, which should 
be hard and close-grained, soon becomes polished and glazed, and in the 
best possible condition for smooth working ; the packings of the piston are 
metallic, and the methods of pressing out the packing ring such as to ensure 
a uniform and evenly spread pressure of small magnitude. The loss from 
packing the pistons too tightly may become very great, and too much care 
cannot be exercised in attending to this most important part of the engine. 

In horizontal and diagonal engines, the weight of the piston pressing 
on the side of the cylinder sets up friction, and as the cylinder wears in 
consequence more on the bottom than the top, it gets out of shape, thus 



164 MANUAL OF MARINE ENGINEERING. 

necessitating more pressure on the packing-ring to maintain steam-tight 
contact, and thereby increasing the friction. The arrangements necessary 
for carrying the weight of the piston prevent all the better forms of piston 
rings and springs from being adopted in horizontal engines ; so that the 
friction of pistons alone renders the engine less efficient than the vertical 
form. The most important improvement effected of late years, tending to 
the better working of diagonal engines, has been making the piston solid 
and of steel (v. fig. 90); in this way the weight has been considerably 
reduced, and the strength increased very materially ; the prevention by this 
plan is better than the cures attempted by guide-rods, etc. 

2. The Friction of Stufling-Boxes. — Since the use of the higher pressures 
of steam this has become a very important factor in the consideration of 
engine efficiency. It may be extremely variable, as it depends so much on 
the care and judgment of the attendants, that, however carefully these parts 
may have been designed and constructed, their good working is entirely 
beyond the control of the designer. The manufacture of good and reliable 
metallic packing at reasonable rates has removed to a large extent this 
difficulty, so that now the glands of the high-pressure and medium-pressure 
cylinders are almost invariably fitted with some satisfactory form of metallic 
packing ; and even those of the low-pressure cylinder are so treated in all 
important steamers, especially those of large size, although some engineers 
still prefer good vegetable packing for them. That the resistance may be 
very considerable is proved by the fact that the old trunk engines could 
be slowed down and nearly stopped by tightening the trunk glands, and in 
the ordinary piston-rod engine the speed may be appreciably retarded by 
the same process. The glands should never be so tight that the rods are 
rubbed absolutely dry in passing through them. For the efficient working 
of the engine it is better that a faint leakage of vapour should pass out with 
the rod, as that is generally an indication that the packing is just tight 
enough, besides which the moist vapour lubricates the packing and keeps 
it soft when of vegetable composition. 

3. The Friction of the Guides and Slides is now, perhaps, the least im- 
portant in the everyday experience of the losses to which an engine is liable ; 
as those parts may be said to design themselves, and are now of ample surface. 
and they are in such a position as" to command the attention of the engineer ; 
they are also, as a rule, easily lubricated. In most classes of engines, the 
piston-rod guide is the chief one for consideration, and since, from the form 
of the rod-end, it is nearly impossible to give too small a surface to the shoe, 
it is seldom found, even in ill-designed engines, to give much trouble ; but 
in certain forms of engine this is not always the case, so that care has to 
be taken both in designing and attending to the guides. The maximum 
pressure on the piston-rod guide of a marine steam engine is usually from 
two to three-tenths of the load on the piston, and supposing the ratio of 
maximum to mean pressure to be l - 50, and the coefficient of friction, under 
good circumstances, probably not less than 0*05, then 

_ . L , ., AA1 . ftni . load on piston 

Resistance of guide = 01)1 to 0-015 X z-~- > 

1 'OO 

which means that from two-thirds to one per cent, of the energy of the piston 
is of necessity consumed in overcoming the friction of this one guide. 

Improvements in the manufacture of, and choice of suitable metals for 
guides, have very sensibly diminished the loss from this cause ; besides 



LOSS FROM FRICTION. 1G5 

which the pressure on the guides has been in modern engines reduced by- 
making the connecting-rod longer in proportion to the stroke. When cast-iron 
has become, by rubbing, glazed on the surface, there is no material better 
for guides. However, as some engineers will not wait for, or do not trust to, 
this state of metallic surface, but prefer the rubbing surface to be of a softer 
nature than the rubbed, it is not unusual to find white metal fitted to the 
" shoes." That some of the older engines were inefficient from loss at the 
guides, is proved by the rapid wearing of the old bronze shoes ; the work 
necessary to convert so many cubic inches of metal into powder being the 
measure of avoidable loss at that point. 

4. The Loss from Friction at the Shaft Journals, etc., may be also very 
considerable, as the load on the piston is transmitted from the crank-pin 
to them, in addition to that caused by the weight of the shaft itself and the 
connections. The same may be said of the crank-pins, which have pressing 
on them the whole of that load, in addition to the weight of the rods. This 
friction may be very severe, especially in fast-running engines. It was 
formerly held that friction was independent of velocity, so far as movement 
through a fixed distance is concerned ; that is, if a body be moved through 
10 feet, the friction is the same if the movement takes place in 1 second or in 
10 seconds ; but if time be taken into account, the friction of moving the body 
ten times over the 10 feet in 10 seconds is ten times that of moving it once 
in 10 seconds. Since M. Morin made his experiments further investigations 
have been most carefully made by Mr. Tower and others, which show that 
friction does vary with the velocity, probably as the square root of the 
velocity ; hence for modern piston speeds a larger allowance of surface is 
required than formerly. In a marine screw engine making seventy revolutions 
per minute, the friction is at least seventy times that of one revolution ; 
and, consequently, if a paddle engine, having the same size of cylinders, 
•and working with the same pressure of steam, makes only thirty-five revolu- 
tions per minute, its friction of journals will be half that of the screw engine 
per minute. As the first screw engines working without gearing were gene- 
rally designed by men whose experience had been gained with the slower 
working paddle engines, it is not astonishing to find that the bearings were 
not always sufficient for the work on them, and perhaps the speed of the 
rubbing surfaces prevented the lubrication from being so efficient as had 
been the case previously, and so aggravated the evil. Again, the old paddle 
engine and geared screw engine had cylinders of longer stroke compared 
with their diameters than had the direct-working screw engine, and as the 
diameter of the shaft depends on the area of piston and length of stroke 
combined, while the pressure on the bearings is affected only by area of 
the piston, the diameter of the shaft might remain the same, although the 
size of the piston had been very much increased. Now, since most of the 
old rules for length of journals took cognisance of the diameter of shaft 
only, although the pressure on the journals might have been doubled, there 
was only the same surface to take it. 

For example, a paddle engine of 5 feet stroke might have the same 
diameter of shaft as a screw engine of 2 feet 6 inches stroke, each having 
the same cylinder capacity ; but the engine with the short stroke would 
have a piston area twice that of the long stroke, and consequently with the 
same steam pressure there would be double the load on the journals, and 
tins with, generally, double the number of revolutions of the shaft. 



166 MANUAL OF MARINE ENGINEERING. 

Friction and Surface. — It was formerly an axiom that friction was inde- 
pendent of surface; but there must always have been limitations to such 
a statement. It meant that so long as the moving body was supported 
on a surface sufficiently large to prevent the unguent from being squeezed 
out, the resistance due to friction was the same fraction of its weight, what- 
ever the area. Practical engineers, however, never had unbounded belief 
in this axiom. It is now clearly understood that when two surfaces are 
separated by a stratum of oil, or other unguent, they do not touch one 
another, and, therefore, each is sliding on the unguent. It is also generally 
supposed that the particles of unguent assume a globular form in the process, 
so that they form a kind of ball-bearing surface. Anyway, so long as the 
unguent is maintained in position, the resistance is small and is constant. 
It is, however, well known that when the same unguent is used with different 
metals having apparently equally smooth surfaces, there is a difference in 
the coefficient of friction, as the fraction is called ; and, further, that certain 
metals will not work at all well on others — e.g.. soft steel on bronze. It is 
evident, then, that either the metals affect the unguent, or they do not really 
present equally good surfaces. It is known that oil acts chemically on 
copper, and, therefore, on copper compounds, while it has no action on tin 
and antimony. This may possibly account for the different behaviour 
of white metal and bronze. 

Modern experiments have shown that surface has its influence, even 
within the safe limits ; that the coefficient is higher with very light loads 
per square inch than heavier ones, so that when an engine is running quite 
light, the percentage of loss from friction is higher than when running full 
load. In the case of a marine engine running slow, the pressure per square 
inch on the guides, etc., will be much less in consequence of reduced load; so 
that the coefficient of friction will be, from that cause, somewhat higher. On 
the other hand, as the engine is moving slower, and inasmuch as friction varies 
as JV, the coefficient should be less, and so one may balance the other. 

Tower found that at 90° F.. 

. . v'V 

Coefficient of friction = 20c -p— 

Where V is the rubbing velocity in feet per minute, P the nominal pressure 
in lbs. per square inch, c a coefficient depending on the lubricant, which, 
for sperm oil, is "0014, rape oil "0015, mineral oil -0018, and olive oil '0019, 
the oil supply being liberal and constant. 

For example, the coefficient for a guide on which the pressure is. say, 
100 lbs. per square inch, the piston speed 900, and on which mineral oil is 
freely used, 

F = 20 x -0018 ^j^ = -0108. 

If the oil is only sparingly supplied, as with a syphon lubricator, the coeffi- 
cient may be four times this, or '0432, which is very nearly what Morin 
stated to be the best result with well lubricated surfaces. 

Temperature also affects' the friction ; the coefficient is reduced when 
the temperature rises so as to render the unguent fluid and not sticky. On 
the other hand, further rises of temperature tend to make it too fluid and 



LOSS FROM THK PUMPS. 167 



P 



thin. With some mineral oils, raising the temperature from 60° to 120 
caused the friction to be quadrupled. 

The aim of the engineer must be, therefore, to prevent metallic surfaces 
from coming into actual contact, for then the friction would be very severe, 
and soon cause the surfaces to abrade, and even, as is seen sometimes in the 
case of cast iron, to strike fire. The lubricant should be introduced at the 
points where the pressure is least, in order that those where greatest may be 
well lubricated. Friction at the journals, too, must always be a source of 
anxiety, though not nearly so much so now as formerly. Crank-shafts are 
more truly turned, although even now there is room for improvement ; the 
foundations of the engines are more stiffly made, so that there is no springing 
at the bearings ; the bearing surface is pro rata larger, and white metal is 
universally fitted to both crank-pins and main bearings. 

5. The Friction of Valve Motions and of the valves is very considerable 
at all times, and may be severe when the valves are running dry. Even 
when the pressure on flat valves is partly relieved by frames, etc., on their 
back, the load on the rods was sometimes so great as to bend them. With the 
large increase of piston speed of necessity have come larger valves. This, 
with increased boiler pressure, forced most makers of marine engines to 
revive the piston valve for the high-pressure cylinders, and latterly to fit 
them to the medium-pressure and even to the L.P. cylinders. On the 
other hand, a few eminent builders stuck to flat valves for all their 
cylinders, and that, too, with a success that was somewhat remarkable, 
but it was with lower pressures and slower rates of revolution than now 
obtain. 

Attempts have been made from time to time to use Corliss and other 
quick-shutting " drop " valves instead of slide valves. So far the results 
have been disappointing, and the increased efficiency due to quick cut-off 
and small clearance which such valves and gear give to land engines is now 
not yet attainable to the same extent with marine compound engines. But 
the marine engine may have its efficiency very much improved by some 
other form of valve which, while giving plenty of egress for the steam to 
the condenser, very much reduces the clearance now obtaining in low-pressure 
cylinders. 

6. Loss from the Pumps. — In all engines, whether condensing or non- 
condensing, there is a loss of efficiency from the feed-pump when it is worked 
by the engine. It is true that the work done by it in forcing the water into 
the boiler is stored energy, and, therefore, not lost, but its low efficiency 
is a source of loss. It is now, however, the common practice for the feed- 
pump to be worked by an independent engine. This is convenient, and 
has advantages which will be discussed elsewhere, but it must not be over- 
looked that the cost in steam is greater than by the old method, as the 
independent engine consumes at least double the amount of steam per I.H.P. 
that the main engines do. 

To some extent the air-pump may be said to store energy, for it takes 
water from under a pressure of, say, 2 lbs. per square inch, and places it at 
the disposal of the feed-pumps under a pressure of 15 lbs. The chief work 
of the air pump is to withdraw air and other gases from the condenser at the 
pressure in it, and deliver them at atmospheric pressure. This, except as 
a means to an end, is all lost energy, while the friction, etc., of this pump 
adds to t»he loss, and so reduces the efficiency of the engine. 



168 MANUAL OF MARINE ENGINEERING. 

Air-pumps, especially in warships, are sometimes worked by an inde- 
pendent engine (v. figs. 124 and 127) ; the same objections apply to this pump 
being dealt with in this way as to the feed-pumps, and there is not the corre- 
sponding advantage in practice, although there might be if the pump were 
regulated to run exactly to the requirements of the condenser. 

From carefully made experiments with land engines, there is good grounds 
for estimating the power absorbed in driving the air pumps of a marine 
as varying from 5 per cent, of the gross I.H.P. in small engines to about 
1 per cent, in large ones. If the air pump does not exceed one-twentieth the 
L.P. cylinder, it will be only from 2'0 to 07 per cent. 

In an engine of 1,500 I.H.P. the air pump will absorb 3 per cent, of the 
power developed, while in one of 3,000 I.H.P. it will be T35 per cent., and 
in one of 5,000 I.H.P. 0'82 per cent. 

The circulating-pumps render nothing for the large amount of energy 
put into them. and. viewed in this light only, would seem to be a great im- 
pediment to the engine. As a means to a most valuable end it is otherwise 
appraised ; as a necessary adjunct to the surface-condenser it has conduced 
very much to the economy of the modern marine engine. Besides all the 
energy lost in the driving of this pump the greatest waste of the steam engine 
is taken by it, for its water conveys away all the latent heat required to 
make the steam. It has been suggested that by turning the stream of water 
towards the stern some propelling effect would be obtained, and some of this 
loss be prevented. So far no engineer has gone out of his way to effect 
this small economy. 

The power absorbed in circulating the water through the condenser in 
steam engine installations on land is heavy ; much more so than that on 
board ship, where the condenser itself is commonly below the water-level, 
and there is no " head " but that due to resistance in tubes, etc., to be 
pumped against. The variation found in these land installations is, how- 
ever, instructive, for in cases of 1,500 I.H.P. the circulating pumps absorb as 
much as 10 per cent., at 3,000 I.H.P. stations it is 4'5 per cent., and at those 
of 5,000 I.H.P.. 3"2 per cent. On shipboard, with the discharge valve at or 
below the water line, 1'7 to 2 - 5 per cent, of the main engine power is sufficient 
in the tropics, and about a half in temperate zones. In merchant ships, 
with the discharge well above the water line, the loss is 2*2 to 3'0 per 
cent, in the tropics. 

7. Inertia of Moving Parts. — There should be no losses from this cause 
if proper provisions are made, for all the energy stored in them in getting 
them to their maximum motion should be usefully employed and absorbed 
before coming to rest. In the older engines, as in many modern ones, con- 
siderable loss arose from the fact that much of the energy stored in the 
pistons and other heavy moving parts was not usefully employed, but was 
wasted in vibrating the engine itself and the ship. This, however, is fully 
dealt with in Chap. xxix. There will, of course, still be small losses due to 
friction on bearings and guides from the inertia of the moving parts after 
the engine has been balanced, for the balancing cannot bring all the forces 
into one plane. 

It will be seen by the foregoing that the energy lost in a marine engine 
is not a definitely fixed quantity, and is dependent neither on design nor 
on construction alone, but rather on the degree of care exercised by those 
having the charge of it. That much may be saved by good and careful 



INERTIA OF MOVING PARTS. l(j ( J 

designing and workmanship is evident, but, within certain limits, a good 
engine may prove less efficient than a really inferior one from mere lack of 
proper attention from those in charge, and especially if they are not supplied 
with good and suitable lubricants. 

The Losses due partly to Mechanical Defects and partly to Physical Causes 
are those which cannot be classed as belonging to the engine as a machine, 
nor to it as a heat engine simply. The most important of these is conse- 
quent on the employment in its construction of materials having a high 
power for conductivity of heat. The steam pipes were generally of copper, 
and when of steel are comparatively thin, so that there is much loss of heat 
from their surfaces. Careful covering with material having a low conductivity 
does much to prevent this waste, but, with the high pressures now used and 
consequent high temperatures, unless the lagging is very thick and the pipe 
flanges well covered there is still great loss. The loss, too, is not limited 
to the mere heat which escapes, for if saturated steam — that is, steam con- 
taining as much water as it can — is robbed of any of its heat on its road to 
the cylinders it deposits some of its water ; this water obstructs the free 
passage of steam through valves and passages, and, till forced through the 
escape valves or drain cocks, obstructs the pistons themselves. The 
efficiency of the engine as a heat engine is also materially affected by the 
presence of water in the cylinders. The losses are very manifest when 
the speed of the engine is, as often happens, materially checked by water 
coming with the steam. The gain from superheating the steam for the 
marine engine arose, and will again be found to arise, largely from the 
steam being quite dry rather than from the higher temperature at the 
cylinder. 

Liquefaction takes place on expansion in the cylinders in spite of steam 
jackets and superheating, and a small amount of water to lubricate the 
internal parts is a good thing, especially as the use of oils is very restricted 
or wholly done away with nowadays. Notwithstanding, the cylinders should 
be as carefully " lagged " as the steam pipes, and the covers and flanges 
of the high-pressure cylinder should be as carefully covered as any other 
part. 

It is very doubtful if steam, during expansion in the fast-running marine 
engine, can take any appreciable heat from the jackets ; that such jackets add 
somewhat to the efficiency of the engine must, however, be admitted when 
they are properly drained ; it is probable, therefore, that the gain is one of 
mechanical efficiency, from the fact that the cylinders are freer of water with 
steam jackets in use. 

If this argument is based on fact, a jacket to the low-pressure cylinder 
might be of some use, and even outweigh the loss arising from the reheating 
of steam about to enter the condenser at exhaust. 

Superheating the steam so that it might continue wholly in the vapour 
state till it reaches the condenser has been the hope of some sanguine in- 
ventors. Superheating the steam with heat that otherwise is lost is obviously 
the most desirable thing as a means of economy. Superheating, however, 
must have its limits, for in economising steam there may be great loss of 
energy, as. for example, by excessive friction of the high-pressure piston, 
valves, and rods from want of a lubricant when the steam is so dry as to 
absorb it all in the vaporous state. As a matter of fact the steam may leave 



170 



MANUAL OF MARINE ENGINEERING. 



the marine boiler with a large amount of superheat, and even on reaching 
the cylinder still have a fair margin, but the piston will not have moved 
many inches before it is saturated steam. If water deposit in the high- 
pressure cylinder is to be avoided, the superheater will require to have 
special heat supply or its own fire, and unless very special means are 
adopted for lubricating the piston and valve will work badly and cut the 
surfaces. 

Superheating went out of fashion on the advent of steam of 60 lbs. pressure, 
partly due to the fact that its temperature, 307° F., was nearly as high as 
superheating had then gone, and to the restrictions put by the Board of 
Trade on superheaters and their somewhat rapid decay. 

Metallic packing was unknown at that time, and the gain in economy by 
the compound engine over the old surface condensing 30-lb. pressure ones 
satisfied fully the enterprise of the good folks in those days of good freights 
and cheap fuel. 

By superheating the steam for electric light installations on shore great 
reductions in consumption of fuel have been effected (fig. 64). No doubt 
similar results may also be obtained by employing satisfactory superheating 
apparatus on board ship, as shown below, but great care must be taken in 
the design and construction of the piston and valves of the first cylinder 
into which such steam enters, or the loss may exceed the gain in steam 
efficiency. 

Experiments with Superheated Steam in Modern Times* have been made, 
and the results of some very interesting ones were given by Mr. Felix Godard 
in a paper read at the Inst. Naval Architects. These trials were made with 
the screw steamers " Garonne " and " Kance," each 300 feet long, 40 feet 
beam, and 25 '5 feet deep. Mean draft of water 21 feet, gross register 2,700 
tons. 

The " Garonne " has no superheater, but her total heating surface is 
equal to that of the " Kance." together with the surface in that ship's super- 
heater. That is, the total surface exposed to heat is the same in each ship. 
Their engines have cylinders 23 inches, 36 inches, and 59 inches diameter 
and 42 inches stroke. 





'■ Garonne." 


"Ranee." 


Boiler pressure, ..... 


178 lbs. 


177 lbs. 


Steam temperature, .... 


378 3 F. 


518' F. 


Heating surface of boilers, . . . 


3,767 sq. ft. 


2,982 sq. ft. 


Superheating surface of boilers, . . 


none 


785 sq. ft. 


Revolutions per minute, . . . 


72-3 


75-4 


Indicated horse-power, .... 


1,104 


1,304 


Coal consumed per l.H.P. per hour, 


1-12 lbs. 


0-90 11). 


Coal per mile on service, 


1 .VI lbs. 


12(5 lbs. 


Increase in power, .... 




18-1 per cent. 


Decrease consumption per I.H.P., . 


. . 


20-1 „ 


Saving of fuel on voyage, 




18-2 „ 



Two other ships were fitted in the same way — that is, one, the 

* Vide Appendix E for more recent experiences. 



EXPERIMENTS WITH SUPERHEATED STEAM. 



171 





* 


V 




Steam Consumptions of Various Engines 
Using Superheated Steam. 

A = Vertical compound fast running, 295 I.H.P., 




17 






t 


vacuum 20 inches, boiler pressure 175 lbs. 


V' 










B = Three-crank compound slow running, 5500 




Q 










I.H. P. , vacuum 28 inches, boiler pressure 


-C 










150 lbs. 


k 












C = Triple-expansion, 670 l.H. P., vacuum 26 














inches, boiler pressure 150 lbs. 


O. 












D = Triple-expansion high speed, 1925 I.H. P., 


X 




b 








vacuum 26 inches, boiler pressure 183 lbs. 


mi 


|| 


\ 










E = Triple-expansion, 3000 I.H. P., high speed, 


^ 


\ 












J 












vacuum 26 inches, boiler pressure 180 lbs. 














F = Triple-expansion slow running, 3000 I.H. P., 


j 






V* 










boiler pressure 199 lbs. 


1 


15 
















2 




sc 










^ 




7 




















• 




















c 








^ \ 














<J 






















k. 






















i» 


14 






















a 






















> 






















1- 


















s 




^w _ 








13 












^ N 


* 


t. 
























^^ 


s 


V. 








1 




















. s 


V 








i 


















\n 


» > 


t 




























v> 


X 




^3 
























^^^x- 


V s 


N S 




N 


b 


























. V 






% 










12 
























% 


j^ 








fj 






















^ 


^^ 










10 


































c 


JC <0 *° 80 ,0 ° "0 MC IM ,80 3oo J3C J40 JtO J»0 ' #oo 



Sci foerheof de c rees fob 1 
Fig. 64 . 



172 MANUAL OF MARINE ENGINEERING. 

" Gtiaclaloupe," had the same quantity of heating surface, 13.509 square feet, 
as the heating and superheating surface together of the other, the " Peron." 
The temperatures were 378° as before, and only 460° with the superheaters ; 
the indicated horse-power was practically 6,585 and 6,750 with 88 revolutions, 
the speed of the " Peron" being on service 16 - 95, against the 166 of the 
"Guadaloupe." These ships were 430 feet long, and of 6.800 tons gross 
register. 

Another source of loss is the resistance due to the friction of the steam 
in passing through the pipes, passages, and valves ; and although here again 
there is not a total loss, still it is not compensated for in the way that the 
engineer desires. As friction causes heat, so the friction of the steam along 
the surface of the pipes and passages generates heat ; but since this heat is 
not allowed to escape, it is taken up by the steam, and tends to superheat 
it. The loss due to this cause in the steam-pipes is probably very small, 
especially when the pipes are of such a size that the velocity of steam through 
them is not excessive, and Mr. D. K. Clark found that it is inappreciable 
when the velocity is not more than 130 feet per second with very dry steam, 
and 100 feet per second with ordinary dry steam. The greatest loss is when 
the steam has to pass through narrow orifices where the perimeter of such 
orifices is large compared with the area, as is the case at the steam ports of 
a cylinder ; this is called " wire-drawing " the steam, and there is always 
a loss of pressure from this cause, even when the area through which the 
steam passes is equal to that of the section of the pipe through which it 
has previously passed. When the area is reduced, and the perimeter is large, 
the loss is, of course, still greater, and hence the loss at the ports of a cylinder, 
where the cut-off is early by means of a common slide-valve, is very con- 
siderable, and may amount to as much as 10 per cent, of the pressure, unless 
the ports are very large, or the travel of the valve exceptionally long. To 
obviate such an evil, the double and treble-ported valves are used, and 
other plans adopted, whereby increased area of opening to steam may be 
obtained. If care is taken so as to avoid all unnecessary obstructions to the 
passage of the steam in the pipes and stop-valves, and there be sufficient 
opening of the port at the beginning of the stroke, the loss of initial pressure 
should not exceed 2| per cent., and in some marine engines there is no appreci- 
able fall of pressure from the boiler to the cylinders. There is also a loss 
of energy when the steam enters the cylinders due to sudden change in 
velocity, which will be from 150 feet to 15 feet per second in large engines, 
and even greater in small engines, where the piston velocity is very much 
less than 10 feet per second. This cannot be avoided in any way, as it is 
practically impossible to increase the piston velocity to even one-half that 
of the steam ; and it would be excessively inconvenient to increase the 
area of ports, etc., so that the velocity of steam should more nearly approach 
that of the piston. But as the loss from this cause is very slight indeed, 
no extra cost expended in attempting to avoid it would meet with an adequate 
return. 

A considerable amount of heat is lost by the radiation from those parts 
which are alternately exposed to the hot steam and to the atmosphere ; and 
this was especially great in trunk engines, where the surface of the trunks is 
very large, and, being hollow, of course have the inner surface giving off heat 
as well constantly. Fortunately, the surfaces of the piston-rods, trunks, etc., 



ANOTHER SOURCE OF LOSS. 173 

soon become highly polished, and so do not radiate the heat so quickly as 
they would were they rough. This loss, too, cannot be avoided, or even 
reduced to any appreciable extent. 

Finally, there is the loss due to the heat conducted from the cylinders, 
pipes, etc., to the other parts of the engine with which they are connected, 
and which pass it away by radiation at their surfaces. 



174 MANUAL OF MARINE ENGINEERING. 



CHAPTER VII. 

ENGINES, SIMPLE AND COMPOUND. 

Elementary Steam-Engine. — The steam-engine, in its most elementary form, 
has only one cylinder, into which the steam is admitted at each end 
alternately, so as to move the piston backwards and forwards, and having 
performed its work is then allowed to escape into the atmosphere. Although 
from certain causes this was not the form of steam-engine as first invented, 
it is nevertheless the most simple one, and by taking it as the origin, the 
genesis of the marine steam-engine can be better explained. As the engines 
for marine purposes are still largely those having cylinders and pistons, it 
will be unnecessary in this chapter to deal with any other forms. 

Genesis of the Compound Engine. — The exhaust steam issuing from an 
engine having a late cut-off and an initial pressure of, say, two atmospheres 
(or 15 lbs. above atmospheric pressure), would attract the attention of an 
observant engineer from the force with which it emerges from the exhaust- 
pipe, and would naturally lead him to inquire how so great a waste of energy 
might be avoided. It would be clear to him that there was sufficient 
remaining in it to do useful duty after it had accomplished its work in the 
cylinder. Being acquainted with the steam-engine of Watt, he would suggest 
that, instead of allowing it to escape into the atmosphere, it might be con- 
ducted to the cylinder of a condensing engine, which could work with a 
steam pressure of one atmosphere, and while operating the piston of this 
second engine, would cause no more back pressure in the first cylinder than 
before. Such an arrangement would be a combination of a high-pressure 
and a condensing engine, and hence it was called a compound engine. 

The engineer to-day would suggest that the steam be taken to a low- 
pressure turbine, as is often done, and the turbine exhaust to a condenser, 
whereby a better use would be made of it than the Watt engine did. 

The idea of a compound engine, however, was due to the genius of Hornblower, a 
Cornish engineer, who in 1781 took out a patent, in which he claimed: — "I use two 
steam vessels, in which the steam is to act, and which in other steam-engines are called 
cylinders. I employ the steam, after it has acted in the first vessel, to operate a second 
time in the other, by permitting it to expand itself, which I do by connecting the 
vessels together, and forming proper channels and apertures whereby the steam shall, 
occasionally, go in and out of the said vessels, etc." 

Arthur VVoolf, in his patent taken out in 1804, states that " if the engine be con- 
structed originally with the intention of adopting my said improvement, it ought to 
have two steam vessels of different dimensions, according to the temperature or the 
expansive force determined to be communicated to the steam made use of in working 
the engine ; for the smaller steam vessel or cylinder must be the measure of the larger. 
. . . The small cylinder should have a communication, both at its top and bottom, 
with the boiler which supplies the steam. . . . The top of the small cylinder 
should have a communication with the bottom of the larger cylinder, and the bottom 
of the smaller one with the top of the larger, with proper means to open and shut 
those alternately by cocks or valves, etc., . . . and both top and bottom of the 
large cylinder should, while the engine is at work, communicate alternately with the 
condensing vessel"." He proposed to use steam at a pressure of 40 lbs. 



EFFECTS OF INCREASE OF PRESSURE. 175 

Expansive Engine. — If the observer, however, happened to be better 
acquainted with the expansive force of steam than with the use of a turbine 
and condenser, he would suggest that the steam should be cut off at such 
an earlier part of the stroke as would ensure its pressure, at emission, being 
only slightly above that of the atmosphere, and all available energy abstracted. 
Such an engine would naturally be called expansive, in contradistinction to 
the elementary engine, working without expansion. Any further attempt at 
increased expansion would prove less fruitful, as the steam, expanded below 
the pressure of the atmosphere, will fail to escape into it when opened to 
exhaust ; besides which, the " back " pressure on the other side of the piston 
would, during the latter part of the stroke, be greater than the forward. 
Then it is that by connecting the exhaust pipe to the condenser, in which 
the pressure would be 10 or 12 lbs. below that of the atmosphere a higher 
rate of expansion, could be obtained. 

Effects of Increase of Pressure. — If an engine is to work economically, so 
far as steam is concerned, it has been stated that the terminal pressure, before 
admission to the condenser, should be as low as possible consistent with good 
working. 

It is now easy to maintain a vacuum of 28 inches in a modern condenser, 
but 25 inches was usual with the jet condensers. As some engineers may 
still prefer to work their engines with only 24 to 25 inches of vacuum (for 
the sake of obtaining warm feed-water), let it be assumed, for the sake of 
argument, that 24 inches is the vacuum in the condenser. When the full 
benefit of expansion is required, the terminal pressure should not exceed 
7 lbs., which will be 3 lbs. above the back pressure. With a turbine ex- 
pansion may go on till the difference is less than 1 lb. 

In order to appreciate fully what is encountered in making advances in 
boiler pressures, it will be well to compare two engines working under the 
conditions set out above. Suppose these two engines to have each one 
cylinder of the same diameter and stroke, the boiler pressure of the first 
to be 2 atmospheres, or 30 lbs. absolute, that of the second 3 atmospheres, 
or 45 lbs. absolute, the terminal pressure in both cases to be 7 lbs., and the 
back pressure 4 lbs. The cut-off in the first will be ~ v of the stroke, or a 
rate of expansion of 4*285, and the mean pressure with an initial of 30 lbs. 
is 174 lbs. ; deducting 4 lbs. of back pressure, the effective mean pressure 
will be 134 lbs. In the second engine, the cut-off will be T 7 T of the stroke, 
and the rate of expansion 6'43, and the mean pressure with an initial of 
45 lbs. is 20 lbs. ; deducting, as before, 4 lbs. for back pressure, the effective 
mean pressure is 16 lbs. The effective initial pressures will be 30 — 4, or 
26 lbs., and 45 — 4, or 41 lbs., respectively. 

Since the cylinders are of the same capacity, and the terminal pressures 
are the same, each engine consumes the same weight of steam ; but the 
total heat of evaporation of steam from the temperature corresponding to 
4 lbs., and at that corresponding to 45 lbs., is 1,130° F., while at that corre- 
sponding to 30 lbs. it is 1,121° F., there will be, therefore, an expenditure 
of fuel to obtain the steam at 45 lbs. slightly in excess of that at 30 lbs. As 
this, however, amounts to less than 1 per cent., it may be neglected, and 
the cost of the steam assumed to be the same in both cases. It will be seen 
then that, with this advance of boiler pressure, there is an advance in mean 
pressure, and the gain in power amounts to nearly 20 per cent. ; but the 



176 



MANUAL OF MARINE ENGINEERING. 



initial load on the piston of this more economic engine is 57 per cent, higher 
than that on the more wasteful one, and consequently the rods, framing, 
etc., must be 57 per cent, stronger ; moreover, the shaft will be increased 
in size, and the cylinder and passages must be stronger. Altogether, then. 
the engine will become heavier and more costly, as the boiler pressure is 
increased, while the expenditure of fuel is less per horse-power. 

To render the comparison strictly fair, it would be necessary to take two 
engines of equal power, so that if the stroke of the pistons is the same, their 
areas will be inversely proportioned to the mean pressures, and, conse- 



1JV4 
16 



of 41, or 34'5 lbs. 



quently, the initial loads will now be 26 lbs. and 
which gives an excess of 31*3 per cent. 

Progress Made by Early Marine Engineers. — The early marine engineers, 
however, did not advance on these lines as a rule, for nearly the same rate 
of expansion was observed at full speed with steam of three atmospheres 
as had obtained with steam of atmospheric pressure ; consequently very 
little benefit was derived in economy, compared with what might have 
been the case had they done differently. The chief result accruing from 
the increased boiler pressure was in practice the larger indicated horse-power 
developed by engines of the same size as formerly, partly due to the aug- 
mentation of mean pressures, and partly to the increased piston speed resulting 
from them. 

Engines of as much as 200 N.H.P., working with steam of 60 lbs. pressure, 
were supplied to the Navy by Messrs. Penn and Messrs. Maudslay as early 
as 1853 ; but a period of more than fiiteen years elapsed before the Admiralty 
again employed such pressure in any larger ship than a gun-boat. 

By the following table a comparison can be made of four typical simple 
engines working with steam at different pressures, and their performance 
under the varying conditions : — 



TABLE XXIX. 



Load on safet}' valves, 


lbs. 


30 


45 


60 


60 


Diameter of cylinder, 


. ins. 


50 


50-8 


511 


38 


Initial absolute pressure, 


lbs. 


45 


60 


75 


75 


Cut-off, 


. stroke 


6 

Tff 


tV 


A 


« 

TO 


Back pressure, 


lbs. 


4 


4 


4 


4 


Mean effective pressure, . 


• >> 


3677 


35 66 


3515 


63-95 


Terminal ,, 


• >i 


27-00 


18 00 


15 00 


45-00 


Maximum load on piston, 


• »! 


80.4S3 


113,456 


145,550 


80,514 


Mean ,, ,, 


• »J 


72, ISO 


72,180 


72,180 


72,180 


Ratio of max. to mean, . 


• 11 


1115 


1572 


2016 


1-115 


Weight of steam used, . . 


• !> 


893 


609 


515 


S68 



Examples (1), (2), and (3) show conclusively that by increasing the boiler 
pressure and rate of expansion, so as to obtain nearly the same mean pressure, 
the weight of steam used is considerably diminished and the terminal pressure 
considerably reduced, but that there is' an increase in the maximum load 
on the piston proportionate to the absolute pressure in the boiler, so that 
the ratio of maximum to mean pressure of example (3) exceeds that of example 
(1) by more than 80 per cent. 



RECEIVER COMPOUND ENGINE. 177 

Although such an increase in weight of machinery as would be neces- 
sitated by so great an increase in load, may not be of much importance in 
some ships, in others it would be prohibitive ; for when the power required 
for certain speeds of ship becomes large compared with the displacement, 
it requires the utmost care in design to keep down the weight, so as to admit 
of the engines being carried by the ship on the required draught of water. 
For this reason, in actual practice, it was found advisable to use steam of 
under 50 lbs. pressure (above the atmosphere) in very fast river steamers. 
or even in high-powered steamers for Channel service of moderate size, on 
account of the limited speed of piston obtainable with the paddle-wheel, 
until, by means of forced draught, constructing the ship and engines as 
much as possible of steel, and a special design of light compound engines, 
pressures of 100 to 125 lbs. could be employed in such ships ; now in some 
few cases triple-expansion engines using steam of 175 lbs. pressure are fitted 
in paddle steamers of bigh speed. 

Example (4) is given that the effect of two widely different boiler pressures 
may be compared, when the rate of expansion is the same. The mean pressure 
is, of course, much higher, and, but for the back pressure being constant, 
would bear the same proportion to that at the boiler pressure of 30 lbs. as 
the initial absolute pressures — viz., 5 to 3. The weight of steam used is 
very little less than that of 30 lbs. pressure, and, owing to the reduction in 
the size of the piston, the maximum load in this case is practically the same 
as in example (1). The pressure at exhaust is exceedingly high, being 45 lbs. 
absolute, or 30 lbs. above that of the atmosphere, so that it is capable of 
doing considerably more work, if admitted into another cylinder of larger 
size, than that of the first ; and even if admitted into one of the same 
size (provided it finally exhausts into a condenser), more work will be 
obtained from the steam than if it is allowed simply to escape into the 
atmosphere. 

Receiver Compound Engine. — Now, suppose that an engine working 
under the conditions set out in example (4) (so far as pressure and cut-oft 
are concerned) exhausts into a steam-tight space, so that there is back pres- 
sure in front of the piston equal to the pressure in this receiver of the exhaust 
steam ; and suppose, further, that the steam is taken away by anothei 
cylinder from the receiver at the same rate as it is supplied by the cylinder, 
there will then be a constant mean pressure maintained there. For the sake 
of fixing the application to example (4), suppose, again, the pressure in its 
reoeiver to be 30 lbs. absolute, then the mean pressure in the cylinder will be 
67 # 95 — 30 = 37'95 lbs. only. Now, suppose a second and larger cylinder 
to be supplied with steam from the receiver at such a rate that there is no 
change of pressure in it (this being accomplished by so arranging the cut-off 
in the second cylinder, that the weight of steam taken by it equals the weight 
of steam exhausted from the first one) ; the cut-off may be determined from 
the formula p v = const. ; so that if V be the volume of the second cylinder 
and v that of the first, 45 lbs. the terminal pressure in the small and 30 lbs. 
the initial in the large cylinder. Then, cut-off in the second cylinder = 

— -^ = o X ^, and if the ratio of V to v is 3, the cut-off in the second cylinder 
oU v ^ v 

is h stroke, and the rate of expansion in it 2. 

The mean pressure, with an initial pressure of 30 lbs. and a rate of 



178 MANUAL OF MARINE ENGINEERING. 

expansion 2, is 25*38 lbs. ; and allowing for a back pressure of 4 lbs., the 
mean effective pressure in the second cylinder is 21*38 lbs. 

Since the area of the second piston is three times that of the first, the 
work done in the second cylinder is equivalent to what might be done by 
one of the same area as the first, with a mean effective pressure three times 
as great, or 64*14 lbs. per square inch. It will be seen from this that the 
total work done by the combined cylinders is the same as would be done by 
the original cylinder with a pressure of 37*95 + 64*14, or 102*09 lbs. per 
square inch ; hence we find that there is a gain of nearly 60 per cent, by 
the introduction of the second cylinder. So far, the compound engine would 
be undoubtedly more economical than the expansive engine, as exemplified 
in examples (1), (2), and (4), but less so in this particular instance than 
example (3). 

Expansive and Compound Engines Compared. — To examine the relative 

economy of a compound and of a simple expansive, engine, it is necessary 

that they should both work with the same boiler pressure and the same 

rate of expansion. Now examples (3) and (4) satisfied the first condition, 

and if the second is also satisfied, then they may be compared. The rate of 

expansion in example (4) is 1*666, and since the volume of steam in the 

second cylinder at the end of its stroke will be three times that in the first 

at the same period, the total expansion effected by both cylinders will be 

3 x 1*666, or five times. The cut-off in example (3) was two-tenths the 

stroke, and therefore its rate of expansion is five ; so that these two examples 

may be compared as to the efficiency of the steam. The effective pressure 

of the compound system may be referred to the large cylinder, in the same 

way in which it was referred to the small one, and will be that Actually on 

the large cylinder, together with that on the small one divided by the ratio 

of their capacities ; hence, effective mean pressure referred to the large 

/ 37*95\ 
cylinder is (21*38 -1 5 — ), or 34*03 lbs. per square inch. It will be seen 

that this is 1*12 lbs. less than that obtained in the simple expansive engine, 
and therefore a loss has occurred somewhere in the compound system. 

Suppose, now, that the cut-off in the large cylinder is so altered that 
the pressure in the receiver is 45 lbs., so that it receives steam at the same 
pressure as that which exhausts from the small one ; in this case there will 
be no " drop " in the pressure from commencement of exhaust to the end in 
the small cylinder. 

The mean effective pressure in the small cylinder is now 67*95 — 45, or 
22*95 lbs. per square inch. 

4-0 *w 
The cut-off in the large cylinder = j^^n, or £ the stroke, which gives 

3 as the rate of expansion. 

With an initial pressure of 45 lbs. and rate of expansion 3, the mean 
pressure is 31*5 lbs. ; allowing 4 lbs. for back pressure, the effective mean 
pressure is 27*5 lbs. 

Referred to the large cylinder the effective mean pressure of the system 

22*95 
is now 27*5 -\ 5 — , or 35*15 lbs. on the square inch, or exactly the same as 

that obtained in the simple expansive engine. 

Effect of " Drop " in the Receiver.— It is seen from the above, then, when 



DIRECT EXPANSION COMPOUND ENGINE. 179 

no " drop " occurs there is no loss of efficiency ; but that when the pressure 
in the receiver is less than the terminal pressure in the small cylinder, there 
is somehow a loss of effective mean pressure. This arises from the steam 
being allowed to expand from the small cylinder into the receiver without 
doing work. But it is known that, when this takes place, the steam becomes 
somewhat superheated ; for, inasmuch as the loss of pressure has occurred 
without conversion into external work or loss of heat in any other way, it 
must appear in some other form. Although this loss is not wholly recovered, 
it must be to some extent reduced by the benefit which the steam derives 
from the superheating in expanding in the large cylinder. 

Division of the Work. — It will be also seen that, as the ratio of the 
cylinders' capacity is 3, and the effective mean pressures 22*95 lbs. and 
27 '5 lbs., the work done in the small cylinder to that in the large is as 22*95 
to 27*5 x 3, or nearly 1 to 3*6 ; while in the former case it was as 37*95 to 
21*38 x 3, or nearly 1 to 1*7. 

Therefore, with an earlier cut-off in the large cylinder, more work is 
developed in it than is the case when with a later cut-off ; moreover, with 
this ratio of cylinders, in order to get the highest efficiency of the steam, 

the ratio of the work done is as 1 to 3*6 ; and the initial pressure on the 

75 4.5 

large piston is 41, and on the small ^— — , or 10, as against 71 on the 

expansion engine of equal size ; and even if the compound engine were 
arranged with one cylinder above the other, the combined initial pressure 
would be 41 -f- 10, or 51, as against 71 of the simple expansive. 

Direct Expansion Compound Engine. — The compound engine may, how- 
ever, work without any intermediate receiver, if the pistons are arranged 
to move simultaneously, either in the same or opposite directions. To 
consider this case — suppose the cylinders to be side by side, and the pistons 
to move in opposite directions, as originally proposed by Woolf, so that 
when the small piston has receded one-tenth of the stroke, the large one 
has advanced by exactly the same amount, and the space between them 
is 0*9 v + 0*1 V ; the volume of steam at commencement of exhaust is 
v, and the pressure at that period, as before, 45 lbs. ; the volume at any 

point of the stroke (va), or the space between the pistons 

n ,, 10 — n n 

= io v + -Io- t; = io (V - v)+?; ' 



and since V = 3 v, space between the piston at n-tenths of the stroke 

The pressure at this point = 45 -*- f^+lj. The pressures at every 

tenth of the stroke from to 10 will be 45, 37*5, 32*14, 28*12, 25*0, 22*5, 
20*45, 18*75, 17*3, 16*07, 15*0 ; the mean of which is 24*78 lbs. per square 
inch. Deduct this from 67*95, and the effective mean pressure in the small 
cylinder is 43*17 ; deduct 4 lbs. from 24*78 lbs., and the effective mean 
pressure in the large cylinder is 20*78 lbs. 

The effective mean pressure of this system, referred to the large cylinder, 



180 MANUAL OF MARINE ENGINEERING. 

is now (20"78 -| o~)> or 35*17 lbs. per square inch, which is the same as 

that obtained by direct expansion in one cylinder — example (3). 

It will be observed, however, that the ratio of the power exerted by the 
small cylinder to that exerted by the large one. is as 43" 17 to 3 X 20*78, or 
1 to 144, being nearer an equal distribution of the work than in either of 
the cases of the intermediate receiver engine. This latter result is caused 
principally by the decreased back pressure in the small cylinder. 

In actual practice there ar3 certain causes which materially modify the 
results shown by both of these forms of compound engine, as illustrated in 
the foregoing. 

In the receiver compound engine, the pressure in the receiver is not 
constant, because of its limited size ; the difference in the periods of exhaust 
and admission, and the cushioning after cut-off in the L.P. cylinder by the 
small piston cause considerable oscillation. 

The direct expansion engine is only nominally without a receiver, as 
the space between the small piston and the large one is often necessarily 
considerable (v. fig. 75), from the size of the communication pipes and the 
valve-box of the large cylinder. The valve of the L.P. cylinder cuts off 
some time before the small one ceases to exhaust, causing cushioning in the 
latter and in the spaces ; the small cylinder also commences to exhaust 
before the large one can take steam. 

Direct expansion compound cylinders have, however, been arranged so 
that one cylinder communicates directly with the other, without any inter- 
vening space, by placing the cylinders side by side, and causing the pistons 
to operate on cranks set opposite one another (v. fig. 187). But such engines 
have, for other reasons, proved generally unsuitable for propelling. 

Requisites in the Marine Engine. — A marine engine must be (1) when 
required readily started, stopped, and reversed ; (2) it should have a turning 
moment or torsion as nearly uniform as possible ; (3) it must be able to work 
continuously for long periods without stoppage from any cause ; and (1) 
and lastly, it must be economical. 

The first condition is a sine tjud non, and is generally fulfilled by having 
two or more cylinders operating on cranks at suitable angles. 

The second condition is absolutely necessary when weight is a serious 
consideration, and is very fairly satisfied by the two or more cranks at proper 
angles, and the cylinders so designed as to divide the work nearly equally 
between them. 

The third condition will depend very much on the variation in stress, so 
that the engine with small initial pressure in each cylinder compared with 
mean pressure, is more likely to fulfil it than one with large initial pressures. 

The fourth condition, which is of the utmost importance to the ship- 
owner, is very well complied with by the triple- and quadruple-compound 
engines in all sizes, and by the turbine and combined turbine and reciprocator 
in large sizes. 

With two cylinders and cranks at right angles, there must inevitably 
be some amount of "' drop," if the work is to be evenly divided when the 
power is as great as usually developed by a marine engine at service speed. 
The crank of the H.P. cylinder should lead — that is, should be in advance 
of the other crank by 90°, or such other angle as it is deemed best to set 



THEORETICAL EFFICIENCY OF VARIOUS MARINE ENGINES. 181 

:he cranks at. When this is the case, the small cylinder begins to exhaust 
just alter the crank of the L.P. cylinder has got well over the centre, and 
tends to maintain a constant pressure on the large piston through the earlier 
portion of its stroke, and at cut-off the pressure in the receiver is not much 
below its average pressure. If, on the other hand, the crank of the large 
cylinder leads, exhaust takes place only a little before cut-off in the large 
cylinder, and causes a hump in the indicator-diagram, showing an increase 
in the amount of " drop," and that with no diminishing in the mean back 
pressure in the small cylinder. Engines having the low-pressure engine 
crank as the leading one were also generally unhandy. 

The first triple-compound engines were, as a rule, designed so that the 
high-pressure crank " led," or was in advance of the medium-pressure crank 
by 120°, and the medium-pressure crank 120° in advance of the low-pressure 
irank ; but in modern practice it is usual to fit the low-pressure crank in 
advance of the medium-pressure crank, etc., as in this way there is less vari- 
ation in temperature in each cylinder, although the load on the pistons is less 
during the first half of the stroke and greater during the second half than is 
the case when the high-pressure crank leads, and the engine is not appreciably 
less *' handy " (v. fig. 164). 

The first cylinder of a compound system is called the " high " pressure, 
and the last the " low," from their association with the condensing and 
non-condensing engine. For convenience in speaking of them, they are 
designated by the initials H.P. and L.P. Hitherto, in the chapter, all com- 
parisons of the compound with the simple expansive engine have been made 
on the supposition that the expansive engine has only one cylinder of the 
same capacity as the low-pressure one of the compound system. To render 
the comparison perfectly fair, it will be necessary to take such cases as may 
be found in actual practice. In triple-expansion engines the middle cylinder 
is called the " medium-pressure," and sometimes the " intermediate," and 
designated by the initials M.P. ; and in quadruple engines the third cylinder 
is called the 2nd M.P. 

Comparative Theoretical Efficiency of Various Marine Engines.— (1) A 
single-ciflinder expansive engine : rate of expansion, 5 ; initial pressure, 80 lbs. ; 
absolute back pressure, 4 ; area of piston, A. 

Mean pressure . . = 41*76 lbs. 

Effective mean pressure . = (41*76 — 4) = 37*76 lbs. 

Effective initial load on piston = (80 — 4) A = 76 A lbs. 

Efficiency of the system . =1*00. 

(2) A single-crank tandem compound engine : rate of expansion, 5 ; initial 
pressure, 80 lbs. absolute ; area, L.P. piston, A ; ratio of cylinders, R, 
generally in practice, 4. 

Effective mean pressure referred to L.P. piston . — (41*76 — 4) 

= 37*76 lbs. 

R 

Terminal pressure in H.P., and initial pressure in L.P. = -= X 80 = 64 lbs. 

o 

A 

Effective initial load on H.P. piston . . . = (80 — 64) - = 4 A. 

L.P. „ ... =(64 -4) A = 60 A. 

Efficiency of the system = 1*00. 



182 MANUAL OF MARINE ENGINEERING. 

TotaJ load on crank is, therefore. 64 A, against 76 A with the single- 
cylinder expansive engine. 

As in actual practice there is invariably a drop, which will amount to as 
much as 10 lbs. in an engine of this kind, the initial pressure in the low- 
pressure cylinders being decreased by that amount, and that of the high- 
pressure increased. So that, actually, the total loads will be as 56*5 to 76, 
or a saving in the compound engine of over 25 per cent, of the load put on 
the rods, framing, etc. ; and also enabling a large reduction to be made in 
the diameter of the shafting. The engine will work much more steadily, 
owing to the ratio of maximum to mean pressure being so largely reduced '. 
and the handiness very much increased from the cut-off in the high-pressure 
cylinder being so late as y^ the stroke. The friction of the two cylinders, 
etc., is, however, considerably greater than that of the one, but this is set off 
by the reduction in friction on the guides and journals ; and the friction on 
the valve of the single cylinder exposed to high-pressure steam will be 
more than the combined friction of the two valves, the small one of 
which only is so exposed, while the expansion valve (which is necessary 
to the single cylinder for so early a cut-off) will also increase its loss from 
friction. 

The compound engine compares more favourably with the simple expan- 
sive when both have two cylinders and two cranks. Then each engine has 
the same number of working parts of necessity, and the simple expansive, 
besides having the usual slide valves, each of which is exposed to the boiler 
pressure, has. an expansion valve to each cylinder, in order to effect so early 
a cut-off. The following examples will show the results of the two systems 
under the same circumstances : — 

(3) A simple expansive engine having two cylinders, each of whose 

A 

pistons has an area of -^ inches ; rate of expansion, 5 ; initial pressure, 

80 lbs. 

Mean pressure . . . = 41*76 lbs. 

Effective mean pressure . . = 41'76 — 4 = 37'76 lbs. 

Effective initial load on piston . = (80 — 4) -^ = 38 A lbs. 

•a 

Effective mean load on both pistons = 37*76 x A lbs. 
Efficiency of the system . . = TOO. 

(4) A compound engine having two cylinders, the ratio of whose piston 
area is 3, and the area of low-pressure piston, A ; rate of expansion, 5 ; 
initial pressure, 80 lbs. ; pressure in receiver, 21 lbs. 

The cut-off in high-pressure cylinder to effect this rate of expansion = i 
or 0*6 stroke. 

The cut-off in low-pressure cylinder to maintain 21 lbs. pressure in the 

receiver = -^ 5- = 0*76 the stroke. 

21 x 3 



THEORETICAL EFFICIENCY OF VARIOUS MARINE ENGINES. 



183 



Effective mean pressure in H.P. cylinder 

T P 

)J »J .Li. A . , r 

Effective initial load on H.P. piston . 

T P 

>> »> AJ.A . ,, 

Effective mean load on both pistons . 
Efficiency of the system . 



72-48 - 21 = 51-48 lbs. 
= 20-32 - 4 = 16-32 lbs. 

= (80 -21)g= 17-3 x A lbs. 

= (21 - 4) A = 17-0 X A lbs. 

= 51-48 X I" + 16-32 x A 

= 33-48 x A lbs. 

= 0-887. 



It will be seen that the initial load on each of the pistons of the com- 
pound engine is very nearly equal in this case, and is less than half that on 
each piston of the expansive engine. The work done is very nearly equally 
divided between the two cylinders, but falls short of that done by the expan- 
sive engine by more than 11 per cent. 

The compound engine, therefore, is nominally not so economic in steam, 
but is subject to much lighter loads, and to less variation in load and tem- 
perature. 

To test the merits of the compound system carefully devised experi- 
ments were made by the British Admiralty, and by the Government of the 
United States, which show that, although the difference between the 
coal consumed per I. H.P. in the two systems was not very great, the com- 
pound engine, on the whole, is more economical. The best known of these 
experiments was the trial between the gunboats " Swinger " and " Goshawk," 
the latter having compound engines, with cylinders 28 inches and 48 inches 
diameter and 18-inch stroke, the former expansive engines, having two 
cylinders, 34 inches diameter and 22-inch stroke, and those of the sister ships 
given below. In these and in many other cases it was demonstrated that 
while on trial trips the coal and water consumptions were always less with the 
compound engines, it was really the results of prolonged trials on service that 
proved conclusively the real and practical superiority of the compound 
system, which was not alone in consumption of fuel per I.H.P., but in 
mechanical efficiency, whereby the consumption per voyage was marked, 
and the reduction in wear and tear even more so. These arguments were, 
of course, convincing to the shipowner. 



Diameter of cylinders and stroke. 


"Sheldrake." 


"Moorhen.' 


"Mallard.'' 










Exp., 2 cyls., 


Exp., 2 cyls., 


Comp. 




34" x 21". 


34" x 21". 


31" - 48" x IS". 

1 


Boiler pressure, lbs., 


62-35 


53-0 


63 5 


41 


58-4 


59 


Vacuum, . - ins., 


20 


24 


23-3 


24-4 


23-75 


25'1 


Revolutions per minute, .... 


115-5 


84-7 


121-3 


92-9 


124 8 


98-8 


Speed of piston, - - feet per minute, 


404 


295 


424 


324 


374 


296 


Indicated horse-power, .... 


367 


137 


387 


180 


398 


213 


Speed of ship, - - . - knots, 


9 251 


7 232 


9-634 


7-899 


9-894 


8413 


Water consumed per I. H.P. per hour, 


21-5 


25-4 


201 


24-6 


1712 


1766 f 



184 MANUAL OF MARINE ENGINEERING. 

Further Comparison of Efficiency of Engines. — The compound engine with 
three cranks and having two low-pressure cylinders and one high, possesses 
advantages beyond those of the two-cylinder two-crank arrangement. It 
was, no doubt, first chosen principally to avoid excessive diameter of low- 
pressure cylinders for very large power, and next because of the uniformity 
of the twisting moment on the shaft with the three cranks at angles of 120° 
apart. But, further, in this engine the work can be fairly equally divided 
between the cylinders without those disadvantages previously shown to 
exist in the two-cylinder engine working under this condition. To divide 
the work equally, only one-third will be allotted to the high-pressure cylinder, 
and one-third to each of the two low-pressure cylinders, and this can be 
done by maintaining a considerably higher pressure in the receiver than 
obtains in the two-cylinder arrangement. The " drop," therefore, is con- 
siderably less ; and since each low-pressure piston has only half the area 
of th^t of the two-cylinder engine, the initial loads under these conditions 
are not abnormally large. The increase in receiver pressure reduces the 
initial load on the high-pressure piston, and hence its diameter may be 
increased, so as'to get increased expansion in it and decrease in "drop" 
without increasing the initial load beyond that on the high-pressure piston of 
a two-cylinder engine of equal power. 

(5) An engine having two low-pressure and one high-pressure cylinders 
working on three cranks. — To fully appreciate these advantages, suppose 
such a three-cylinder engine to be working under the same circumstances 
as that of example (4), page 182, so that the area of each low-pressure piston 

A A 

is ^r, and of the high-pressure piston », the rate of expansion 5, and the 

initial pressure 80 lbs. absolute. 

As in the previous example the cut-off is 0'6 the stroke. Suppose the 
pressure in the receiver to be 32 lbs. absolute, 

DA V O'fi 

Then the cut-off in L.P. cylinders = -^ ~- = 056 the stroke. 

The effective mean pressure in the H.P. ) _ /70-48 — 321 -- 40"48 lbs 

cylinder, . . . . . J 

The effective mean pressure in each L.P. ) _ /oq-iq __ a\ — 24-19 lb- 

cvlinder . . . . . J 

\ 
Effective initial load on the H.P. piston = (80 — 32) ~ = 16 X A lbs. 

each L.P. „ = (32 - 4) ~ = 14 X A lbs. 



The effective mean load on all three f = 40-48 X g + 2 x 24'19 X g 
pistons .... 



A , „ _ .„ A 

W48 X 
( = 37-68 x A lbs. 



Efficiency of the system . . = 0'998. 

It is seen by this that the initial load on the high-pressure piston is 74 per 
cent., and that on each low-pressure piston 17A_ per cent., less than the corre- 
sponding loads of the two-cylinder engine of the same size ; that the gain of 
efficiency of the steam is 124 per cent., if no allowance is made for possible 
superheating of the steam on expanding into the receiver ; the " drop " in 



TRIPLE-EXPANSION COMPOUND ENGINE. 185 

this case is (54 — 32) or 22 lbs., as against 33 lbs. in the two-cylinder engine. 
It is seen, then, that theoretically this engine is nearly equal in steam effi- 
ciency, both to the simple expansive, and to the compound direct-expansion 
engine. On the other hand, the friction of three engines must be set against this, 
besides the extra cost of manufacture and the space occupied by machinery. 

Experiments in the Mercantile Marine with engines having cylinders with 
ratios better suited to the steam pressure and other conditions demonstrated 
much more emphatically the superiority of the compound engine in economy 
of fuel and, perhaps, more decisively the other advantage of that system, 
for it was soon found that the wear and tear and oil consumption of expansive 
engines having a working pressure of 50 to 60 lbs. The two-cylinder and 
three-cylinder receiver type of compound engine soon became popular, and 
remained so until steel boilers could be made for such high pressures as 
to require an extension of the compound system. 

Triple-Expansion Compound Engine.— The compound engine having one 
high-pressure, and one low-pressure, with a medium-pressure cylinder, is the 
one most commonly found in use, if steam of over 120 lbs. pressure is to 
be used economically. To ensure economy, the steam must be expanded 
down to about 10 lbs. absolute, the initial loads on the pistons moderate, 
and the " drops " not excessive. The low-pressure cylinder may be rather 
smaller in size than that of the ordinary iit'o-cylinder compound arrangement, 
due to the increased efficiency of the steam from the high rate of expansion, 
and the greater referred mean pressure. 

(6) For example : — To determine the particulars of a triple-compound 
engine on this system to be equal to that set out in example (4), page 182, 
the initial pressure being 127 lbs., and the rate of expansion 10. 

The mean pressure in a single cylinder, with a cut-off at T ^ the stroke, is 
41*91 lbs. ; deducting from this 4 lbs. for back pressure, the mean effective 
pressure is 37'91 lbs., or nearly the same as that of example (5), page 184. 
Suppose the cut-off in the high-pressure cylinder is 0"6 the stroke, then the 
ratio of the high-pressure to low-pressure cylinder must be 6 to effect a rate 
of an expansion of 10. The ratio of the medium-pressure cylinder to high- 
pressure cylinder may be taken as f, and, consequently, the ratio of low- 
pressure to mean-pressure cylinder is \J\ The pressure between the 
high-pressure and mean-pressure cylinders is to be 50 lbs., the cut-off in the 
mean-pressure cylinder will, therefore, be 0*61 the stroke. The pressure 
between the low-pressure and mean-pressure cylinders is to be 21 lbs. ; the 
cut-off in low-pressure cylinder must, therefore, be 0*6 the stroke. 

The effective mean pressure in H.P. cylinder = (115 — 50) = 65 lbs. 

MP. „ = (45-4-21) = 24-4 lbs. 

L.P. „ = (19 — 4 = 15-0 lbs. 

\ 
The effective initial load on H.P. piston . = (127 — 50) '-. = 12'8S A lbs. 

M.P. „ . = (50-21) A ^ = 12-1 A lbs. 

L.P. „ . = (21 -4) A = "17-0 A lbs. 

The effective mean load on all three J = (65 x g ) + <£4"4 x £ A) -f 15 A 

pistons ) . ... 

r I = 36 X A lbs. 

Efficiency of the system . = ^rx = 0"949. 



lfJB MANUAL OF MARINE ENGINEERING. 

It will be seen that in this case, owing to " drop," there is a loss of nearly 
2 lbs., but the work is fairly divided between the three cylinders, and the 
initial loads are by no means' high ; the drop from the high-pressure cylinder 
is only 26 lbs., and that from the mean-pressure cylinder 9| lbs. This engine 
then effects an expansion of 10, and is, therefore, a very economic one ; it 
will have a very regular motion, and share in all the benefits of a three-crank 
engine ; and the stresses on the working parts will be very moderate. 

(7) To see how far this is true, it is only necessary to compare these results 
with those from a three-crank engine having one high- and two low- pressure 

cylinders, each L.P. having a piston area of -• 

A 
Suppose the high-pressure cylinder to have a piston area of ^, the cut-off 

in the high-pressure cylinder to effect an expansion of 10, must be 0'4 of the 
stroke ; the receiver pressure will be 42 lbs., and the cut-off in low-pressure 
cylinder 03 of the stroke. 

The effective mean pressure \ ._ ^ _ ^\ = 55 lbs 
in the H.P. cylinder .J 

The effective mean pressure ) _ ^y.y _ ^\ _ 23-7 lb« 
in each L.P. cylinder . / 

The effective initial load on { = (12 - _ ^ A = ^.^ x A ^ 
the H.P. piston . . J 4 

The effective initial load on ) = „ 2 _ 4) A 19 . Q x A lbg 
each L.P. piston . . J v ' 2 

The effective mean load on \ = fa A\ /^ x A\ = ^.^ ^ Albg 
all three pistons . -J V 4/ \ 2/ 

The initial load on the high-pressure piston is here 65 per cent, larger than in 
the preceding case, and that on each low-pressure piston 56 per cent, larger 
than on the mean-pressure piston, and llf per cent, above that on the low- 
pressure piston ; but the efficiency of the steam is somewhat higher in the 
latter case, the drop from the high-pressure cylinder being only 8'8 lbs. The 
ratio of maximum pressure to mean in the high-pressure cylinder is 1*54, and 
in the low-pressure cylinder P6, which are about the same as those of the old 
expansive engine working with a boiler pressure of 45 lbs., and cutting-off at 
- 3 of the stroke. It may not be always advisable to set the cranks at angles 
of 120° in a three-crank compound engine ; their precise position should 
depend on the power developed in each cylinder, and the relative twisting 
efforts at any period. 

The success of the triple-expansion engine is now so well assured, and 
all doubts as to its efficiency and good working are so effectually dispelled 
as to require no further discussion. It does not differ in any essential feature 
from the ordinary compound engine, and its success was in no small measure 
assured by the fact that most makers of the new type departed as little as 
possible from their previous practice in its general construction. A few 
years' experience demonstrated that the triple-expansion engine is more 
economical than the ordinary compound engine ; that the wear and tear 



INCREASED PRESSURE OF STEAM. 187 

is no more, but rather less, when three cranks are employed, than with the 
two of the ordinary compound ; and that boilers of the common marine 
design could be made to work satisfactorily at a pressure of 150 lbs. per 
square inch, and even higher, while, with ordinary care, their durability and 
good continued working are not less than those of similar boilers pressed 
to 60 lbs. per square inch under similar circumstances. Speaking generally, 
the consumption of fuel is 20 per cent, less with a triple-expansion engine 
than with an ordinary compound engine working under similar circumstances. 
That is, a triple-expansion engine, supplied with steam at 150 lbs. pressure, 
uses 20 per cent, less weight of water per I.H.P. than an ordinary compound 
engine supplied with steam at, say, 90 lbs. pressure, both engines being 
equally well-designed, manufactured, and attended to. Also, that a triple- 
expansion engine is more economical than an ordinary compound engine when 
both are supplied with steam at the same pressure, for all pressures of 95 lbs. 
and upwards, and especially so in the case of large engines. Hence, it may 
be taken that the superior economy of the triple-expansion engine is due 
to two causes, viz. : — (1) To the superior pressure of steam used and the 
higher rate of expansion thereby possible ; and (2) the system whereby large 
initial loads and large variations of temperature in the cylinders and large 
" drop " in the receivers are avoided. 

Increased Pressure of Steam is obtained by a very slight increase of con- 
sumption of fuel, and the efficiency of steam rapidly increases with increased 
pressure ; hence, steam of high pressure is more economical than that at 
inferior pressures. For example : — 

• (i.) The total heat of evaporation of 1 lb. from 100° and at 274° F. (corre- 
sponding to 45 lbs. pressure absolute) is 1,097 thermal units. 

(ii.) From 100° and at 320° F. (corresponding to 90 lbs. absolute) is 1,110 
thermal units. 

(iii.) From 100° and at 353° F. (corresponding to 140 lbs. absolute) is 
1,120 thermal units. 

(iv.) From 100° and at 377° F. (corresponding to 190 lbs. absolute) is 
1,127 thermal units. 

Suppose in each case the steam to be expanded to a terminal pressure of 
10 lbs. absolute, the rates of expansion will then be 45, 9, 14, and 19 respec- 
tively ; and the mean pressures corresponding to these initial pressures and 
rates of expansion will be 25 lbs., 32 lbs., 36 lbs., and 39 lbs. respectively. If 
the volume of a pound of steam varied exactly in the inverse ratio of the 
pressure, these figures would represent the relative values of the efficiency of 
the steam at the various pressures. But, taken exactly, the relative values 
are 25, 333, 38'5, and 42*6, thus showing that a pound of steam at 90 lbs. 
pressure is capable of doing 33 per cent, more work than a pound at 45 lbs. ; 
a pound of steam at 140 lbs. pressure, 16 per cent, more than a pound at 
90 lbs. ; and a pound at 190 lbs. pressure, 10"6 per cent, more than a pound at 
140 lbs. pressure, or 28 per cent, more than at 90 lbs. pressure. In other 
words, an engine using steam at 140 lbs. pressure should, apart from any 
practical considerations, consume 16 per cent, less fuel than one using steam 
at 90 lbs. ; and, again, that an engine using steam at 190 lbs. should consume 
28 per cent, less fuel than one using steam at 90 lbs., and 10'6 per cent, less 
fuel than one using steam at 140 lbs. 

Looking to see how far practice agrees with these results, it is found that 



188 MANUAL OF MARINE ENGINEERING; 

the ordinary compound engine, using steam at 140 lbs., is only a little more 
economical than one using steam at 90 lbs., while the triple-expansion engine, 
with steam at 140 lbs. pressure, gives an economy rather greater than theory 
shows should be due to increased pressure. It follows, then, that there is some 
other cause operating to produce the economic results shown in every-day 
practice with this system, for there is now no question that the saving 
in fuel effected by a triple-expansion engine, using steam and expanding 
11 or 12 times, is about 20 per cent, of that used by an ordinary com- 
pound engine of the same power, using steam at 90 lbs. and expanding 8 
to 9 times. 

It used to be maintained by the opponents of triple-expansion engines 
that, if the ordinary compound engine is designed with a long stroke, it is 
as economical. In order to see how far this is true by practice, it is sufficient 
to examine the diagrams of the engines of the s.s. " Northern," whose 
cylinders were 26 inches and 56 inches diameter and 60-inch stroke, used 
steam at 130 lbs. absolute, and indicated 1,235 H.P., which show a con- 
sumption of 154 lbs. of water per I.H.P. per hour ; and those of s.s. " Draco," 
whose engines had cylinders 21 inches. 32 inches, and 56 inches diameter 
and 36-inch stroke, using steam at 125 lbs. absolute, and indicated 618 H.P., 
and showed a consumption of 14*1 lbs. per I.H.P. per hour. Or, again, 
by comparing the performance of the " Draco " with that of the " Kovno," 
whose cylinders were 25 inches and 50 inches diameter and 45-inch stroke, 
using steam at 105 lbs. pressure absolute, and indicating 809 H.P., with 
a consumption of 16"6 lbs. of water per I.H.P. In these cases, the consump- 
tion of the " Kovno " was 17*73 per cent, in excess of that of the " Draco," 
and 7"58 per cent, in excess of that of the " Northern " ; and the " Northern " 
consumed 9*4 per cent, more than the " Draco." 

It is, however, needless now to multiply cases, as it is a matter of common 
observation that the saving in fuel is from 20 to 25 per cent., and it may 
safely be taken that the triple-expansion engine, using steam at 165 lbs. 
absolute pressure, uses 20 per cent, less fuel than an equally good ordinary 
compound engine of equal power and working under similar circumstances, 
using steam at 100 lbs. absolute pressure. 

As the three-crank triple-expansion engine is now accepted as the most 
suitable for marine practice generally, it is instructive to compare it. so far 
as practical considerations are concerned, with the ordinary compound. 
For that purpose, suppose, now, two engines are to be taken — viz., a triple- 
expansion and an ordinary compound — to develop equal powers with the 
same stroke of piston and same diameter of low-pressure cylinder. The 
initial pressure in the one case is to be 100 lbs. absolute, and in the other 
165 lbs. absolute. Let the number 14 represent the area of the low-pressure 
piston in each case ; the mean pressure referred to the low-pressure piston 
is to be 24, and the efficiency of the systems equal, so far as the steam is 
concerned. The area of the high-pressure piston may then, without fear of 
controversy, be taken as 4 for the ordinary compound, and 2 for the triple, 
and the area of the medium-pressure piston of the triple as 5. If the referred 
mean pressure is equally divided in each case between the cylinders, the 

14 

mean pressure in H.P. of the compound engine will be -j- X 12, or 42 lbs. ; 

in the triple-expansion engine the mean pressure in the H.P. cylinder will 



TRIPLE-EXPANSION COMPOUND ENGINE. 189 

14 14 

be tt X 8, or 56 lbs. ; and in the M.P. cylinder -r- X 8, or 22'4 lbs., and 

2 J 5 

the following shows the relative work done, viz. : — 



Ordinary Compound Engine. 

High-pressure cylinder, 4 x 42 or 168 
Low „ 14 x 12 or 168 



Triple-Expansion Engine. 
High-pressure cylinder, 2 x 56 or 112 
Medium „ 5 x 22 4 or 112 

Low ,, 14 x 8 or 112 



That is, the average load on the rods, columns, guides, etc., is 50 per cent, 
more with the ordinary compound engine than with triple-expansion. 

To obtain a mean pressure of 24 lbs., with an initial pressure of 165 lbs. 
absolute, and a pressure in the condenser of 2 lbs., the rate of expansion is 
14, with an efficiency of 06 ; and with an initial pressure of 100 lbs. the rate 
of expansion is 7. 

On examining diagrams taken under these circumstances from actual 
engines, the following is to be observed : — 

Initial pressure in the high-pressure cylinder of the compound engine, 
98 lbs. ; back pressure, 23 lbs. ; effective initial pressure, 75 lbs. per square 
inch ; or load on the piston, 75 X 4, or 300. In the low pressure the initial 
pressure is 2*2 lbs. ; back pressure, 4 ; giving an effective initial pressure of 
18 lbs. per square inch ; or load on the piston, 18 X 14, or 252. 

The initial pressure on the high-pressure cylinder of the triple-expansion 
engine is 160 lbs. ; back pressure, 63 lbs. ; effective initial pressure, 97 lbs. ; 
or the load on the piston, 97 X 2, or 194. In the medium pressure the 
initial pressure is 70 lbs. ; and the back pressure, 21 lbs. ; effective initial 
pressure, 49 lbs. ; or the load on the piston, 49 X 5, or 245. In the low 
pressure the initial pressure is 18 lbs. ; and the back pressure 4 lbs. ; giving 
an effective initial pressure of 14 lbs. per square inch ; or a load on the piston 
of 14 X 14, or 196. Thus showing the loads in the case of the triple-expansion 
engine to be much less, notwithstanding the higher pressure of steam 
employed. 

This, too,' is shown in actual practice by comparing the initial loads on 
the engine of the s.s. " Northern," whose cylinders are 26 inches and 56 inches 
diameter and 60-inch stroke, indicating 1,242 H.P., and supplied with steam 
at 130 lbs. pressure absolute, with those of the triple -expansion engine of 
the s.s. " Ariel," whose cylinders are 23 inches, 35 inches, and 60 inches 
diameter and 57-inch stroke, indicating 1,527 H.P., and supplied with steam 
at 165 lbs. pressure absolute. 

The " Northern's " high-pressure piston sustains an initial load of 
530 X 100, or 53,000 lbs. ; the low-pressure, 2,463 x 24, or 59,112 lbs. The 
"Ariel's" high-pressure piston sustains 415 X 100, or 41,500 lbs.; the 
medium-pressure, 962 X 60, or 57,720 lbs. ; and the low-pressure, 2,827 X 18, 
or 50,886 lbs. — notwithstanding that the engines are larger and develop 
nearly 25 per cent, more power with higher boiler pressure. 

The more even distribution of pressure also very materially affects the 
resistance of the slide-valves, and so tends in every way to reduce the losses 
due to mechanical causes. 

Experience has shown that the wear and tear of the triple-expansion 
engine with three cranks is very considerably less than that of ordinary 
two-crank compound engine of the same power and stroke, and no doubt 



190 MAVUAL QF MARINE ENGINEERING. 

this is due to those causes already shown to exist with this class of engine, 
as compared with the expansive engine. 

The three-crank triple-expansion engine has, however, shown another 
valuable quality, and one which may easily be surmised from the foregoing 
reasoning — viz., that a much higher indicated power may be developed with 
a low-pressure cylinder, equal in size to that of a common compound engine, 
without any increase in the initial loads on the pistons. This may be shown 
by comparing the performance of the engines of the " Eldorado," whose 
cylinders were 26 inches, 40 inches, and 68 inches diameter and 39-inch 
stroke, supplied with steam at 165 lbs. absolute pressure, with that of the 
engines of the " Juno," having cylinders 35 inches and 69 inches diameter 
and 39-inch stroke, supplied with steam at 100 lbs. absolute pressure. When 
running at 72 revolutions, the " Eldorado's " engines develop 1,572 I.H.P., 
and the " Juno's " 1,249 I.H.P., or 26 per cent, more power from the triple- 
expansion than from the ordinary compound, although the low-pressure 
cylinder was 3 per cent, smaller. The initial loads on the pistons of the 
" Eldorado " are 54,060 lbs., 66,568 lbs., and 61,727 lbs., as against 69.264 
lbs. and 71,041 lbs. on those of the "Juno." 

Similar results can be shown with four-crank engines of various sizes ; 
and to extend the question, it may be taken as approximately correct that a 
referred mean pressure may be used in a triple-expansion engine 50 per cent, 
higher than with an ordinary compound engine without any serious difference 
in the stresses on the working parts and frame-work. It is for this reason 
possible to manufacture a triple-expansion engine at the same price per 
I.H.P. as an ordinary compound engine. The pcopelling efficiency of the 
three and four-crank engine is especially noticeable when running at low 
speeds, and it is no doubt on this account they were best for naval purposes 
where so much cruising is done at comparatively slow speeds. They are 
also capable of being worked at much fewer revolutions without stopping 
on the centres than a two-crank engine, which is highly advantageous in 
navigating intricate channels and docks and during a fog, as steerage power 
is kept without much " way " on the ship. They are also, when well con- 
structed and properly adjusted, almost noiseless, and cause little or no 
vibration, which is an advantage in every ship, but especially in yachts and 
passenger steamers. 

The Compound Systems of Cylinders are now admitted on all hands without 
any further controversy to be superior to the simple-expansive one, not- 
withstanding that in the past at each of its stages of development this claim 
has been contested by those who seemed unable to grasp the fact that 
the determining factors in each particular controversy were practical rather 
than theoretical ones, and much of the same nature as the one which caused 
Watt to remove the condensation of the steam from the cylinder to the 
condenser. Moreover, it was by practical tests rather than argument in each 
of these stages that engineers were weaned from their love of the old to their 
faith in the new system. It is unnecessary now to dwell on the subject, 
or to recapitulate further the early experiments made to prove that the 
compound engine was superior to the simple expansive, as measured by the 
water consumption per I.H.P. ; or those later ones, whereby the compound 
was shown to be inferior to the triple-expansion engine ; nor even to those 
later still, when the advocates of a further subdivision claimed for their 



COMPOUND SYSTEMS OF CYLINDERS. 191 

quadruple-expansion engine a distinct improvement on the good results 
achieved by the triple. 

So far no one has ventured on a quintuple stage system, although engines 
with five cylinders and cranks have been made, besides which the increase 
in boiler pressure now possible with cylindrical boilers has warranted such 
an advance. Although the controversy has ceased, and a compound system 
of cylinders of some kind is always found in a modern marine reciprocating 
steam engine, it is well that the reasons for such an adoption should be clearly 
stated and well understood, for after all it is somewhat in the nature of 
promulgating a paradox to proclaim that if steam is expanded through all 
the ports, passages and pipes of a compound system of four stages, the 
economic result is better than if expanded in a single cylinder in the simplest 
way possible. 

The fact is that, theoretically, the compound system is wrong so far as 
efficiency of steam expansion is concerned ; nevertheless, in practice, the 
same conditions which caused the compound engine to triumph over the 
expansive have established the superiority of the triple over the compound, 
and given an advantage to the quadruple over the triple. The system with 
its extensions have permitted the safe employment of higher steam pressures 
with their higher efficiencies without making increases in the initial loads 
on the rods, crank-pins, bearings, guides, etc., and, more important still, 
without the large variations in temperature in the cylinders so objectionable 
when steam expansion is to be effected without loss. 

When the boiler pressure was 30 lbs. (45 lbs. absolute) the steam at entry 
to the cylinder would have a temperature of about 270° F. ; during exhaust 
the temperature in the cylinder would be only about 150° F., or a difference 
during the cycle of 120°. The expansive engine using steam at 75 lbs. 
absolute would have a range throughout the system of 155°, while in each 
cylinder of the compound engine it will be only about 70°, allowing for the 
usual drop of pressure between them. In the triple-expansion system using 
steam of 195 lbs. absolute pressure, although the variation throughout is 
about 230°, in each cylinder it is not much more than 70°. 

In the case of a compound engine, the L.P. cylinder is about the same 
size as the single cylinder of an expansive engine ; or, supposing the latter 
to be a two-crank engine, each of its cylinders will be half the capacity of the 
low pressure of the compound, and so each piston is half the area of that of 
the L.P. piston. The initial pressure per square inch on the pistons of the two 
engines will be the same if the boiler pressures are equal (say 60 lbs.) ; but 
the area of the high-pressure piston of the compound would be only about 
0*36 of the low-pressure, and consequently the ratio of forward load on it 
would be under these conditions as 36 to 50 of that on each of the expansive 
engine cylinders, but the back pressure of these latter would be only about 
3 lbs., so that their effective initial load would be (60 + 15 — 3), or 72 lbs. 
per square inch, while the high-pressure piston of the compound would have 
a back pressure of about 22 lbs. absolute ; consequently the effective maxi- 
mum load would be in this case (60 -f- 15 —.22), or 53 lbs. per square inch 
only. The total loads will be then measured relatively as — 

Compound, H.P. piston, . . 36 x 53 = 1908 
Expansive, each piston, . . . 50 x 72 = 3600 



192 MANUAL OF MARINE ENGINEERING. 

The L.P. cylinder of the compound engine will have a forward initial 
pressure on it of about 18 lbs. absolute, and a back pressure of 3 lbs., the 
effective initial pressure being thus 15 lbs. Then the following holds good 
for comparison as being the maximum loads : — 

On each of the cylinders of the expansive engine the 

load may be taken as 50 X 72 = 3600 

That on the HP. cylinder of the compound engine, 36 X 53 = 1908 

That on the L.P. cylinder of the compound engine, 100 X 15 = 1500 

But not only is the range of temperature greater with the cylinders of 
the simple engine, but all the steam that enters must pass over a surface, 
which, for 45 per cent, of the cycle time, is exposed to the cooling action of 
the condenser ; moreover, those ports, passages, valves, cylinder ends, and 
pistons have a surface three times the area of the corresponding parts of the 
H.P. cylinder of the compound system, over which all its steam passes, and, 
moreover, its exposure is only to that of the receiver, whose temperature 
will be about 230°, as against 140° of the condenser. The difference will be 
greater still with a modern condensing apparatus, providing a vacuum as 
high as 29 inches, with a temperature during exhaust probably as low as 
110°. Under such conditions the condensation on entry to the cylinder, 
and during expansion in it, would be great ; and, besides, beyond the loss 
of heat, the efficiency of the engine would be lowered by the impeding action 
of the deposited water. Any re-evaporation that is possible would take 
place too late in the stroke to be of any practical use in the cylinder of the 
simple engine ; but in the case of the compound, however, it would be of 
considerable use in the second cylinder, although no value may have accrued 
from it in the H.P. cylinder where the transformation took place. The loss 
by condensation at the valves, whether they be of the flat or piston type, 
is more considerable in practice than might have been anticipated. The 
inside or exhaust side of the L.P. valve of a compound, as of both of those 
of the simple engine, is exposed always to the cooling effect of the condenser, 
while the hot steam is on the other side; under these circumstances, and 
inasmuch as the metal of these valves is comparatively thin, the condensa- 
tion is really quite considerable ; but it is even more so in the simple 
engine, as the steam is then of boiler pressure or nearly so, and consequently 
of much higher temperature, so that transmission is the more rapid due to 
the greater difference of temperature. 

In fact, the difference of temperature is the all-ruling factor in this 
as in some other engineering problems ; for, beside the above, on the practical 
side there is always the additional risk of fracture of cylinder and valves, 
and, in the case of a compound system, the straining action on it of the 
cylinders, if they are cast or rigidly bolted together. The solution of the 
latter difficulty, however, is easy, being simply to avoid rigid connections, 
and provide that each cylinder shall be quite free to expand or contract as 
circumstances may cause it. 

Fig. 65 is interesting and instructive, showing, as.it does, the economic 
process of expanding steam of high pressure through the four cylinders 
of a quadruple-expansion engine made by Messrs. Richardson & Westgarth. 
The stages can be clearly traced on the combined diagram, and the tem- 
perature and pressure changes marked. 



COMPOUND SYSTEMS OF CYLINDERS. 



193 




o 

'So 
a 
K 

c 
_o 

"55 

S 
c8 
n 

M 
O 



'■si 

to 



— 
S 



s 






pq 



«3 >> 

ft 

h 
O 

c9 

c 



SO 



■LjOu/ jod sqf g/ */wj" 3jnsssJ,j 



13 



194 



MANUAL OF MARINE ENGINEERING. 



Fig. 66 is also instructive, as by it can be seen the various stages of 
progress in the use of steam on shipboard, and the effect of adding a L.P. 



uadrofit 




At nqsghertc _Pr* insure 



—~~1 



Fig. 06. — The various stages in the use of Steam expansively in Marine Engines. 

turbine to a compound cylinder engine studied, and how the attenuated and 
apparently feeble steam at exhaust is capable of producing in a suitable 
generator quite a large amount of power. 



HORSE -POWER. 195 



CHAPTER VIII. 

TTORSE-POWER — NOMINAL, INDICATED, AND SHAFT OR BRAKE. 

When the steam engine began to replace other motors, it was soon found 
necessary to introduce some unit by which its power could be expressed 
without using such high numbers as foot-pounds ran to, as to place it beyond 
the grasp of ordinary minds. As the engine was frequently taking the place 
of horses to operate mining and other machinery, it was only natural that 
the work performed by a horse should be taken as the basis for this unit 
of measurement. The number of units of work performed by the most 
powerful dray horses in a minute was found to be 33,000, so Watt chose this 
as the unit of power for his engines, and called it " horse-power," and this 
has continued to be the standard ever since, both for land and marine engines. 
Watt found that the mean pressure usually obtained in the cylinders of 
his engines was 7 lbs. per square inch. He had also fixed the proper 
piston speed at 128 X v/stroke per minute, and his engines were arranged 
to work at this speed, so that he estimated the power which would be de- 
veloped when at work to be 



Area of piston x 7 X 128 ^stroke -f- 33,000. 

The power so calculated was called " Nominal," because the engine was 
denominated as of that power, and in practice that power was actually 
obtained. But when the boiler could be constructed so as to supply steam 
above the atmospheric pressure, and the engine ran with more strokes per 
minute than before, the power actually developed exceeded the nominal 
power, and from the name of the instrument by which the pressure of steam 
in the cylinder was obtained it came to be called the " Indicated " Power. 

The discrepancy between Nominal and Indicated Power became in time 
so great, that for all scientific purposes the former ceased to be of value. 
It remains, nevertheless, in general use among manufacturers and users 
of engines, because it better conveys the commercial value and size than 
does the developed power ; for since the area of the piston is usually the 
only variable in the expression, it follows that the size of the cylinder, and 
therefore the size of all the other parts, must vary directly with the 
Nominal Horse-Power. But since Indicated Horse-Power depends on three 
functions — viz., area of piston, speed of piston, and the pressure of steam — ■ 
the value may be changed by altering the value of one or more of these, which 
alteration may be material without affecting the commercial value. For 
example, an engine may be caused to run at a much higher number of revolu- 
tions, even so as to double its Indicated Power, with hardly any additional 
cost whatever in construction to the engine itself. 

The Admiralty modified Watt's rule to suit the practice of the early 



196 MANUAL OF MARINE ENGINEERING. 

days of steam-navigation, by substituting the actual«|>iston f.peed for the 
arbitrary one, and so the old rule was for 

Admiralty Nominal ) _ Area of piston X speed of piston X 7 
Horse-Power J — 33,000 

The Admiralty, however, dropped the use of the expression altogether, but, 
before doing so finally, used it in the modified sense of being one-sixth of the 
Indicated Power. 

In the Mercantile Marine the rule for Nominal Horse-Power is by no 
means uniform. Before the introduction of the compound engine and 
increased boiler pressures, it was usual to allow 30 circular inches per N.H.P. 
— i.e., the rule was- — 

_ T TT _. (Diameter of cylinder in inches) 2 , , .. , 

N.H.P. =- -^ X number oi cylinders. 

For two-stage compound engines — D being the diameter of the low-pressure 
cylinder, and d that of the high-pressure — 

But neither of these rules took into account the length of the stroke 
or the boiler pressure, although it was generally understood and' roughly 
standardised. The general adoption of the triple-compound engine, however, 
caused the question of stroke to be removed from the field of competition, 
and there is now more uniformity in practice. Besides which, in some large 
centres of marine engineering, all the makers have a standard set of sizes 
of cylinders for N.H.P. to which they adhere. In order to meet difficulties 
some engine-makers have adopted a rule for Nominal Horse-Power, based 
on the capacity of the cylinder, and in so doing have nearly met the require- 
ment on which the continuance of the expression depends — viz.. that it is a 
measure of the commercial value of the engine. The power per revolution 
depends on the capacity of the cylinder so long as the mean pressure is the 
same ; but since small engines are usually worked at a higher number of 
revolutions than larger ones, the power developed by the former will bear a 
larger ratio to the Nominal Power than will be the case in the latter. 

A simple and fair rule for N.H.P.* is deduced as follows :— D is the diameter 

of low-pressure cylinder, d that of the H.P., d x that of the M.P., and d 2 that 

of the second intermediate of a quadruple engine ; S is the stroke, all in 

inches, and D X 0-65 is the standard length of stroke ; in a triple compound 

/D\ 2 /D\ 2 /D\ 2 

i^j = 2-7 ; (-7-1 =6 or thereabouts; in a quadruple compound ( -,-J =8; 

* X * 

J — 4; ( -7- J =2 or thereabouts. 

N.H.P. (Triple) - — + ^ 



N.H.P. (Quadruple) = 



30 

d 2 + d* + d. 2 + D 2 

30^ 



• The JJonril of Trade Kulc now in force is X.1I.1*. = (3 H 4 D- \/S'» x VP + 700. 

H is the total heatins surface. D the diameter of L.P. cylinder, S the stroke in feet. an<! P the load 
on safety valve in lbs. per square inch. 



ESTIMATED HOESE-POWER. 197 

Any change of stroke should give a proportionate change of N.H P., 
hence 

N.H.P. = ^ + <*i 2 + D 2 - S 



30 D x 0-65 

Substituting the values of d, d-^ and d. 2 in terms of D, then 

N.H.P. = — ^K — f° r a compound engine. 

D x S 
N.H.P. = for a triple-compound engine. 

D x S 
N.H.P. = " for a quadruple-compound engine. 

Lloyd's N.H.P. — No Nominal Power, however, can be any guide to the 
•capabilities of the engine, unless the power of the boilers is also in some way 
expressed or understood ; and as it is not easy to imagine how the former 
can be introduced into any expression which shall effect the latter, or vice 
versa, the suggestions of Lloyd's Committee remained unfulfilled, but the 
Register now contains a statement of the leading particulars of the boilers, 
and for purposes of levying the fees for the Survey and Registration of 
Machinery, Lloyd's Register employ the following rule, 

Lloyd's nominal horse-power = p x z ( 1 — ), 

where p is the boiler pressure; D is the diameter and S the stroke of L.P. 
piston, both in inches; and H is the heating surface in square feet. 

The value of x is 15 for the ordinary boiler with natural draught, but 
with forced or induced draught x is 12. 

The value of z is 0'34 when the boiler pressure is under 160 lbs. ; when it 
is 160 lbs. per square inch or above z is 0*393. 

That there is need of uniform practice in naming the power of engines, 
is apparent to everyone having to do with steamships, and the Board of 
Trade Department, which registers the power, has so far limited its efforts 

in this direction to the old formula, N.H.P. = * J. . As this takes 

60 

no cognisance of stroke, it was never satisfactory ; the Department, however, 
is still satisfied with it for its purpose, but it is surely time to find some other 
rule which shall determine the rating of engineers and such other matters, 
as well as be a fair indication of the power the ship possesses to propel her. 

Estimated Horse-Power.* — As it is desirable that a power be named for 
an engine which shall enable the lay mind to judge of its capabilities, pro- 
bably the better plan would be to revert to the principle of Watt, who, as 
has been shown, attempted to specify the power which the engine was actually 
expected to develop ; such a rule, therefore, should give approximately 
the Indicated Horse-Power. It would, of course, be far better to register 
the I.H.P., but as it is not always possible to obtain this, the next best method 
js to estimate it, and call it the Estimated Horse-Power, or E.H.P. 

» N.E. Coast I. E. and S. recommend the following : — 

E.H.P. = " .,' — . D in inches, S in feet. 



198 MANUAL OF MARINE ENGINEERING. 

The following rule will give approximately the horse-power developed 
at full speed by a two-stage, triple-, or quadruple-expansion engine made 
in accordance with modern practice : — 

«, xr t> D ' 2 x Jp x R x S . 

^ HP = 7^00 

Where D is the diameter of the low-pressure cylinder, p the absolute pressure,. 
R the number of revolutions per minute, S the stroke of piston in feet. 

For Example. — To estimate the Indicated Horse-Power of an engine 
having cylinders 30 ins., 48 ins., and 80 ins. diameter and 48-ins. stroke, 
revolutions 75, and boiler pressure 165 lbs. 

^ xr t> 80 ' 2 X s/180 X 75 x 4 

EHR = 7^00 

= 3297. 

Many other rules have been propounded for N.H.P., some of which are 
ingenious, but impracticable, while others fail to give results of any value 
whatever, so that neither class needs notice here ; but it may be mentioned 
that when non-condensinn engines were more used in steamships than they are 
at present it was found necessary to have a special rule for them, which was 

D 2 x ^S 



N.H.P. = 



20 



D being the diameter of the cylinder in inches, and S the stroke in feet. 

Indicated Horse-Power may be defined as the measure of work done in 
the cylinder of a steam-engine, as shown from the indicator-diagrams, and 
only falls short of the actual work by such small losses as are caused by 
the friction of the pin or pencil against the paper, the friction of its working 
parts, and that in the pipes or passages connecting the indicator to the 
cylinder. The latter discrepancy is by far the most important, and is some- 
times serious in very long stroke engines, where the indicator pipe is several 
feet long. The others, in the hands of a skilful operator, are not so serious, 
certainly not in modern marine engines to the extent stated by Mr. Him, 
who says he found the Indicated Horse-Power, owing to losses in the diagram 
from the friction of the indicator, to correspond with the useful xvork done 
by the engine. All the same, it should not be forgotten that with such an 
instrument as the indicator, the nearer it is to the steam in the cylinder the 
better. There should be no pipes, if possible, and, if any, they should be 
fairly large. 

The Indicator Diagram. — The diagram itself shows only the pressure 
of steam acting on the piston at any and every part of its stroke ; but from 
it may be calculated the mean effective pressure acting during that stroke, 
and it is assumed that the particular diagram measured is only a sample 
of what might have been taken at every stroke, so that the mean pressure 
thus calculated is the force acting on the piston during the whole period 
of its motion in which the power is taken — usually one minute. Hence, 
Indicated Horse-Power = area of piston in inches X mean pressure in lbs. 
per square inch X number feet travelled through by the piston per minute ■*■ 
33,000. 

This, of course, applies only to double-acting engines, as in single-acting 
tMgines the pressure is acting only half the time on the piston, and hence. 



MEAN PRESSURE. 



199 



instead of taking the number of feet travelled through by the piston per 
minute as the multiplier, — the length of stroke in feet X number of strokes 
per minute should be substituted. 

Mean Pressure. — The mean pressure is usually obtained by dividing the 
indicator-diagram by a number of equidistant ordinates perpendicular to 
the atmospheric line, and so placed that the distance of the first and last 
from the extreme limits of the diagram is half the distance between two 
consecutive ones ; the sum of their lengths, intercepted by the diagram, 
divided by their number, gives the mean length, and this, referred to the 
scale on which the diagram was drawn, will give the mean pressure. To 
illustrate this : — Fig. 67 is an indicator-diagram whose length, A X, is, say, 
5 inches, and taken with a spring requiring a pressure of 30 lbs. per square 
inch to compress it 1 inch ; so that if M L is 2 inches, it represents a pressure 
of 60 lbs. ; and if B L is 2| inches, it signifies that, at the point L, the pressure 
on the piston was 2| X 30 lbs., or 75 lbs. per square inch above the line 
A X, which, in this case, shall be the line of no pressure, and hence is 75 lbs. 
absolute, or 60 lbs. above the atmospheric pressure. Now, for convenience 
of division, let there be 10 ordinates enclosing 9 spaces — since there is to 




ABODE 

Fig. 67. — Indicator Diagram. 

be a half space at each end, there will be in all equal to 10 spaces — so that 
the distance between the ordinates is 5 inches -f- 10, or half an inch. Measure 
off A B = } inch ; B C, C D, D E, etc., each = | inch, and at B, C, D, E, etc., 
draw perpendicular lines, cutting the diagram at M L, N, etc., Y Z. Then 

(M L + O N + etc +YZ)-rl0 = x inches, and a; X 30 is the 

mean pressure of the diagram. 

This diagram is from one side of the piston only, and, when one only is 
obtainable, it is sometimes assumed to represent both, and the mean pressure 
thus obtained used to calculate the power ; but it seldom happens, although 
it is much to be desired, that the mean pressure is precisely the same on 
both sides of the piston, consequently, any result obtained in this way is not 
satisfactory. If the effective area of the piston is the same on both its sides 
— that is, if there is the same area on which the steam acts to propel the 
piston forward, on the one side as on the other — the mean pressure found 
from the diagram taken from the one side, may be added to that found from 
the diagram taken from the other, and divided by 2 to give the true mean 
pressure per revolution. 



200 MANUAL OF MARINE ENGINEERING. 

Professor Rankine showed that the diagram should be divided from 
A to N bv ordinates equidistant apart, and the mean obtained by the 
following rule : — Let n be the number of such divisions (usually 10), 
b , 6j 6 2 6 3 . . . b n the length of the ordinates intercepted by the diagram : 
then 



Mean length = (-"-±-5? + b x +b 2 + etc., . . . + b n -i} -s- 



n. 



The chief objection to this is that, in actual practice b n would be always 
without value, and b either without value, or so difficult to measure as to 
cause differences of opinion as to its value. 

Another, and a very ready way of obtaining the mean pressure, is by 
measuring the area of the diagram by a planimeter, and dividing it by the 
length A N, the result being the mean breadth as before, and this multi- 
plied by the scale of lbs. of the spring will give the mean pressure. This 
is, of course, the quickest plan, and the most accurate, as being mechanical ; 
.and where many diagrams have to be calculated with despatch, it is very 
advisable to have a good planimeter. Special instruments are now made 
for this purpose. 

In whatever way the mean pressure be measured, it forms the basis of 
calculation of actual energy, or, as it has been called, Indicated Horse- 
Power, and is therefore of the utmost importance, since most modern 
formulae bearing on marine machinery and marine propulsion are based on 
I.H.P. Hence, any error in taking the diagrams must lead to errors in 
design from calculations by formulae based on false premises ; this should 
always be borne in mind by the operator, on whose skill and care a good 
and true diagram depends as much as on a good instrument. It would be 
a very valuable quality in an indicator to be able to give the useful ivork 
of the engine, as was stated by Mr. Hirn to be the case generally ; but it is a 
quality which no such instrument can possess, inasmuch as, with the same 
cylinder performance, there may be a great variety of actual performances 
of the engine, depending on the efficiency of the various parts, and the indi- 
cator only gives this cylinder performance. Could the effective power be 
easily obtained as it is with electric generating engines, a great benefit would 
be conferred on masine engineers in making calculations, etc., and in deter- 
mining the best make and design of engine. The precise power absorbed 
in overcoming the resistance of the working parts of a marine engine could 
not be measured ; for if the engine be allowed to run without load, it is 
not running in the same state as when running with its load ; and any diagrams 
taken then cannot be taken as the loss from friction, etc., on the guides, 
journals, pistons, and valves when running with the increased pressure on 
them due to the increased pressure on the piston with the load. Since the 
efficiency of an engine very much depends on the resistance on these parts, 
any calculation or formula which excludes, or does not give due allowance 
to this, is misleading. Hence, to assume that the power absorbed by an 
engine in overcoming its resistance is measured by the power indicated 
when running without load is not correct — as it is possible for an inefficient 
engine to show a high efficiency when tested in this manner. It is. however, 
true that a large portion of the resistance is the same, or practically the 
same, when the engine is running at the same number of revolutions with 



SHAFT HORSE-POWER. 201 

and without load. The frictional resistance of glands, pistons, pumps, 
tunnel-shaft journals, high-pressure piston valve, and sundry small gear, 
is the same per revolution at any speed, whatever be the load, or power 
developed. The total resistance or non-useful work of engines, therefore, 
probably varies nearly directly as the revolutions. 

Shaft Horse-Power is a term now often used and well known to the marine 
engineer, and is likely to be more so than the I.H.P., which hitherto has been 
in such general use to express the performance of an engine. 

Inasmuch as the indicator was of no service to the maker of turbines, 
nor could any instrument which merely shows the pressure of steam be a 
means of determining the power developed by a velocity machine, some 
other method had to be devised for that purpose. When a turbine was 
driving a dynamo it was easy to calculate the mechanical power of the driver 
by the measure of the electrical output from a dynamo whose efficiency was 
known. The brake horse-power of a turbine could be found by causing it 
to operate on a water brake, and so for some considerable time that was the 
only way in which turbine power was determined. It is true when there 
were two sister ships, whose actual resistance was known, an approximation 
to the power developed by the turbine was made by means of that of the 
reciprocating engine in the other ship, as given by the indicator. This, 
however, was not a satisfactory state of things, especially as I.H.P. is looked 
on with suspicion, and that not without justification, for even with a slow- 
running engine the human element is a factor involved in the accuracy of 
the I.H.P., while with a very fast-running engine the diagram made by the 
best of indicators handled by a man is influenced by the skill of the operator. 
There is, however, an indicator, the product of Prof. Hopkinson's genius, 
that does give a diagram which is free from suspicion, by which the actual 
cycle of pressure in a cylinder may be viewed as it really is, and from it the 
power may be determined. But it is as inapplicable to the turbine as the 
Richards or other indicator for the reasons given above. 

It was necessity, therefore, which stimulated invention and caused to 
be brought forth the torsion meter, whereby the power transmitted by a shaft 
can be determined by observing the angle of twist of a definite length of that 
shaft. There are various kinds of such meters, all of which show a consider- 
able amount of inventive genius, mechanical knowledge, and skill. In some 
of them measurement is mechanical, made by a self-registering instrument, 
in others the eye is employed, and in others the ear, to fix the amount of 
distortion at any given time when the shaft is transmitting energy. It is 
certainly most desirable to eliminate the human element if possible, but in 
doing so at present other factors of an equally undesirable kind are intro- 
duced (v. Chap. v.). 

Shaft Horse-Power calculated as on p. 150 : — d is the diameter of a shaft, 
I the length under observation, both in inches, T is the torque in inch-pounds, 
6 is the angle of twist in the length I, and Mr is the modulus of rigidity, which 
is about 11,750,000 in solid shafts and 12,150,000 in hollow ones. 

584xTxL .. , , fa 

6 in degrees = — ^ n — tor solid shafts 

6 Mr X d i 

= nr-pjj j-f- for shafts with a bore of diameter d,. 

Mr (d 4 — d x 4 ) 



202 MANUAL OF MARINE ENGINEERING. 

The Shaft Horse-Power at revolutions R = r= X tt^ = TxRt 63,000. 

The formula in daily use is therefore (v. p. 150) 

S.H.P. = ° d * X 7 R for solid and d (rf 4 - df)-^—. for hollow shafts. 
Q X I y X t 

Q is usually taken at 3-27 ; this assumes the modulus to be 11,250,000, 
which is below what is generally found for steel shafts of very good quality. 

Example 1. — What is the total H.P. of a ship having three shafts 
6 inches in dameter revolving at 600 times per minute, and twisting 0*4° in 
40 inches ? 

S.H.P. of each shaft = M *%"*:«» = 2,378. 

The total power of the ship is, therefore, 3 x 2,378, or 7,134 H.P. 

Example 2. — A tunnel shaft is revolving 300 times per minute, and is 

10 inches diameter ; the angle of twist in 40 inches is 0*33°, what power is it 

transmitting ? 

xj o tt Tj °' 33 X 10 >°°0 X 300 _ ... 
Here S.H.P. = ^-^ = 7,645. 

The angle of twist is at all times very small, seldom exceeding 1'3° in 
120 inches of length at full power, so that the pair of discs of a torsion meter 
are necessarily of large diameter to give accurate and fine readings, especially 
as in many ships it is possible to use only a very short length of shaft. When 
possible, however, it is certainly desirable to cover as long a portion. of a 
tunnel shaft as possible. 

The shafting of a ship, however, has another load on it besides that of 
transmission of the power generated by the engine, for the thrust block is 
usually close to the engine room, and consequently the whole of it from the 
screw to the block has to resist the thrust. This may amount to as much as 
to equal 20 per cent, of the torque force, but its influence on the twist is really 
very slight, probably only about 1*5 per cent, at most ; generally it is so 
small as to be negligible, as stated by Prof. Hopkinson, from a considerable 
experience gained in testing his torsion meter. Each shaft should be and is 
tested by levers and weights to ascertain its actual resistance to torque, and 
to definitely determine the real and exact amount necessary to twist the 
shaft through a definite angle. From the observations so obtained, it is 
easy to make a diagram, to which reference may be made on trial when 
the torsion meter is showing the value of d to obtain the corresponding S.H.P. 
per revolution, so that by merely multiplying by the number of revolutions 
per minute the full S.H.P. is determined. 

It is, however, not sufficient for a marine engineer to know only the power 
developed by the engine or turbine ; he must be acquainted equally well 
with the power taken and delivered at every critical point of the ship and 
machinery, so that he may make up a balance sheet which shall show on the 
one side the maximum gross power the engine has given out, and on the 
other side the disposal of the same, so that nothing is missing and unaccounted 
for. In fact, the economies of engineering are of the very highest importance, 



THRUST HORSE-POWER. 203 

and cannot be too carefully studied by all concerned from the designer to the 
engineer in charge. Each and every part of the machinery of a ship must 
be in such a state as to be working at the highest rate of efficiency, and how 
to ascertain that rate exactly must be the constant care of those in charge 
of it and responsible for its good and economic working. 

Thrust Horse-Power is an expression that will be more often used in the 
near future than in the past, inasmuch as means may be soon provided where- 
by the actual thrust on a line of screw shafting will be as easily shown as the 
pressure in the condenser now is. Inventors have for some time turned 
their attention to devise some simple and easy method of doing this, and 
one of them, Mr. Heck, fully disclosed to the members of the Institution 
of Naval Architects (Transactions, 1909) two or three such methods that are 
within the bounds of practical politics, while not quite satisfying engineers 
that they are the ones to be finally adopted. They are all ingenious and 
capable of giving results fairly free from inaccuracy. So far, these and some 
others are based on the principle of showing, by means of a hydraulic ram or 
its equivalent, the pressure per square inch in the chamber, or the total 
load on the equivalent ram, and how to make allowance for the friction of 
packings, etc. A more recent idea is to make the flange couplings hollow 
and elastic, fit them together, water-tight, and fill the space between them 
with water whose pressure is indicated by a gauge in the usual way. The 
thrust block might also be used to give its own indications of thrust (as also 
suggested by Mr. Heck), if it were mounted on roller bearings or suspended 
on a stirrup and its thrust taken by a pair of hydraulic rams with chambers 
connected, and the water in them acted on a gauge and spring or loaded 
resistance of some kind, as the hydraulic brakes on a gun carriage. 

The Gross Power of a Steam Engine is, of course, that generated in the 
cylinders by the steam pressure on the pistons acting through the space 
traversed by them. This is known to the marine engineer as the Indicated 
Horse-Power. 

Shaft Horse-Power is that transmitted through the tunnel shafting from 
the engine to the propeller, and ie, therefore, the net product of the engine 
or the gross power less that absorbed in moving the engine and its appur- 
tenances, ealled the Friction Horse-Power. 

The mechanical efficiency of the engine is, therefore, S.H.P. -4- I.H.P. 

Brake Horse-Power is also that transmitted through the shaft to a resistance 
capable of absorbing it just as the propeller does that from the marine engine. 
In this case the brake not only takes the power, but indicates exactly the 
torque or twisting moment from which the B.H.P. may be calculated. Brake 
horse-power should and does coincide very closely with S.H.P. 

The Thrust Horse-Power is that exerted by the screw in pushing the ship 
forward, and is measured by multiplying the actual thrust in pounds by the 
number of feet moved through by the ship in a minute and dividing by 33,000. 

If S is the speed in knots per hour, and T the thrust in pounds, 

™ . v, T x S X 6,080 . 

Thrust horse-power = 00 ^^ ^— = 1 X o h- 6Zb. 

r 33,000 X oU 

This measures the capacity of the screw as a propeller, consequently 

The efficiency of the screw = T.H.P. h- S.H.P. 



204 MANUAL OF MARINE ENGINEERING. 

Indicated Thrust is an expression introduced by Dr. William Froude ai 

a measure of the thrust of a propeller. 

I.H.P. X 33,000 

Indicated thrust = - ., , , — , ,. 

pitch of screw X revolutions 

Tow Rope Horse-Power is that necessary to propel the ship at the required 
speed if freed from what are called the augmented resistances, the chief of 
which is due to the action of the screw itself on the sWrn of the ship to retard 
her motion. If Tr is the tension on an imaginary tow rope from the ship 
to the towing agency, and S the speed of the ship in knots per hour, then 

Tr X S 6,080 Tr X S X 101-3 
Tow rope horse-power = ^qq X -^ - = - -33^ 

This is the net-horse-power of propulsion ; hence — 

Tr X S 
Propulsive efficiency = — ^ 5- I.H.P. 

Net Horse-Power can be calculated, as already shown, by estimating the 
skin resistance and the residuary resistance, adding them together, multiplying 
the sum by the feet passed through in a minute, and dividing the product 
by 33,000 lbs. 

If Sr be the skin resistance and Rs be the residuary, both in pounds, and 
S the speed of the ship in knots per hour, then 

Net H.P. = < Sr + R „l * 101 3 . or Sr + R? 



33,000 ' 326 

Piston speed and revolutions enter largely into the calculation of horse- 
power, and, therefore, important factors at every stage of a design. The 
modern engineer does not permit himself to be hedged in and bound by the 
arbitrary rules of former generations, nor indeed would they have been 
to the extent they did had they had the benefits of the knowledge, experience, 
and materials now enjoyed. Fertina lente was the policy of every progressive 
as of every prudent engineer in early days, and may with advantage form 
a portion of that of modern ones, who now benefit by the failures as well as 
the successes of the pioneers of the profession. The original users of the screw 
propeller cannot be accused of fearing high revolution of that instrument, 
and had those who followed them so soon attempted to make engines which 
could revolve at the speeds of these screws when directly coupled to them, 
such wretchedly low efficiencies as exhibited in the machinery and propeller 
of the "Greyhound " under Dr. William Froude's analysis would not have been 
possible. 

To-day both piston speeds and revolutions of engine are much higher 
than prevailed with the old compound engine. The better distribution of 
torque with triples and quadruples, and the balancing of the inertia forces 
have enabled this to be done successfully, so that to compete with the turbine 
still higher revolutions may be attempted in the future with engines specially 
designed for it. 

Experience has shown that heavy marine pistons may run safely at a 
mean velocity of 900 feet per minute, and, in some instances, the pistons 



PISTON SPEED AND REVOLUTIONS. 005 

of some large vertical engines in first-class cruisers have reached a velocity of 
even 1,000 feet ; while higher speeds still have been attained in torpedo-boat 
destroyers, whose pistons move, when run at express speed, at a velocity over 
1,200 feet per minute. Although there is no difficulty in causing a piston to 
move at even higher speeds than these, it is doubtful if there would be any 
advantage in doing so, besides which the risk of causing serious damage to> 
the cylinders, and precipitating a break-down without any warning, is very 
great. There is no doubt that a well-fitted piston, moving in a smooth and 
true cylinder at a speed of 1,000 feet per minute, will work well so long as 
the rubbing surfaces receive some lubrication from the moisture of the steam 
or the oil injected, and there is not the slightest fear of danger under these 
circumstances ; but if, with a little priming, scum is carried into the cylinder* 
and causes abrasion of the rubbing surfaces, an immense amount of mischief 
may be done in a few seconds. Moreover, when the cylinders wear a little 
out of shape from one cause or other, so that the packing-rings will have 
lateral motion, the danger increases with the velocity of the piston. In the 
Navy engines supplied with steam from water-tube boilers have no internal 
oil lubrication beyond what passes in on the rod surfaces ; so that in destroyers,, 
whose pistons are moving at a speed of 20 feet per second, there is only the 
moisture from the steam to lubricate them. 

Although the revolutions of a screw engine may be, within certain limits,, 
as few or as many as the designer chooses, experience or prejudice has fixed 
very closely in practice the limits beyond which it is not considered expedient 
to go. In the days of the geared engine, the screw revolved three or four 
times to one of the engine, and no objection was raised to the small screw 
and the high number of revolutions ; later such a thing was deemed very 
objectionable — on the ground of excessive speed of piston and excessive 
friction in journals. The slow-moving engine was quoted as a proof of the 
economy of slow piston speed and small friction without being a real founda- 
tion for the argument. 

The fine lines of the older steamships admitted of the small screw, which 
was the accompaniment of the engine, by necessity geared. Bluff ships, as 
now built for mercantile purposes, require a much larger screw for the same 
power of engine and dimensions of hull than formerly obtained ; and it is not 
to the slowness of the pistons that they owe their economy, but rather to the- 
small number of strokes per minute made by them in turning the large screw. 

An engine requires a certain power to be expended in moving it through 
one revolution to overcome internal resistances ; if the number of revolutions; 
is 80 per minute, this power will be double that at 40, and, roughly, will 
vary directly with the revolutions. But the resistance of the propeller, caused 
by friction of the water on the surface of the blades, will increase roughly 
as the square of the revolutions, so that the power expended to overcome 
this resistance at 80 revolutions is eight times that required at 40 revolutions. 
If now the screw can be so altered with respect to pitch that, at 40 revolu- 
tions, the same speed of ship is obtained as at 80 revolutions, the indicated 
horse-power will be found to be considerably less ; and although the coal 
consumed per I.H.P. will not be less, and may possibly be more than before, 
the consumption per day will be considerably less. Now, although this 
economy is co-existent wfth decreased piston speed, it is not due to it. 

The object of a high rate of piston velocity is to decrease the piston area 



20t) MANUAL OF MARINE ENGINEERING. 

and that generally for the sake of reducing the size of the engine. But an 
increased velocity may be obtained either by increasing the stroke of piston, 
or by increasing the number of revolutions ; if the former method is adopted 
there will be no decrease in the size of engine ; but, on the contrary, an 
increase in space occupied and in the weight. If a high piston speed is 
obtained by a high number of revolutions, a smaller cylinder will suffice for 
a certain indicated horse-power than if the same piston speed were obtained 
by length of stroke alone. In other words, engines which are required to 
develop a certain power in a minute will vary in size of cylinder inversely 
as the number of revolutions per minute, all other things remaining constant; 
and if the cylinders are of the same diameter, the stroke will vary inversely 
as the number of revolutions. 

The piston speed of many engines is gOA'erned entirely by circumstances 
beyond the immediate control or will of the designer. An example of this 
is the case of the paddle-wheel engine with vertical oscillating cylinders. 
If the position of the shaft is determined by the structural arrangements of 
the hull, as is often the case, then the diameter of the wheel is fixed, and the 
speed of ship fixes the number of revolutions to be made by the wheel ; the 
length of stroke of piston is limited by the distance from the centre of the 
shaft to the floors or keelson of the ship. Further, if the engineer is free to 
decide the position of the shaft, any attempt to increase the piston speed by 
placing the shafting high is frustrated by the fact that, the higher the shaft 
the larger will be the wheel, and consequently the fewer the revolutions. If 
the engine is inclined, then the designer may fix the diameter of the wheel 
to suit the revolutions which he deems most advisable, or he may fix the 
position of shaft to suit the ship's structure, and still be free to choose the 
stroke of piston. 

Again, the horizontal engine had to be designed so as to accommodate 
itself to the space allotted to it in the ship, which means that only a limited 
length of stroke was permissible. The revolutions, however, in this case 
could be varied considerably ; but there is, after all, a limit to the number ; 
beyond this limit any increase will result in very little gain in speed, and a 
very certain loss of efficiency. If the screw is of comparatively small diameter, 
owing to the shallow draught oi the ship, a higher number of revolutions than 
usual is absolutely necessary to project a sufficient mass of water back to 
propel the ship forward with the necessary velocity ; and it is the medium 
number, or that number at which the engine can be run without loss of 
efficiency so as to obtain the maximum speed of ship that is so difficult to 
decide, and which can only be determined with any degree of certainty by 
experiment. 

One great feature which places the vertical engine so much above all 
the other forms of screw engine, as an economic and good working machine, 
is its superior length of stroke. Power for power, the vertical engine always 
has exceeded the horizontal in this respect ; and although in the practice of 
the past there was no very great difference in this matter between the two 
types, the tendency is now to make the stroke as long as is possible or con- 
venient in the engines of all ships. 

The advantages of the long stroke are due to the corresponding decrease 

in piston area. Two engines of the same power, and working at the same 

timber of revolutions, must have the same volume of cylinder ; or, to speak 



REVOLUTIONS. 207 

more correctly, the pistons must sweep out the same volume if their efficiency 
is the same. The crank-shafts will be of the same diameter, and the crank- 
pins, also, practically of the same dimensions. Now the one with the long 
stroke will have smaller pistons than the other, consequently the total pressure 
on the pistons will be smaller — and, in fact, is inversely proportional to the 
stroke ; consequently, the pressure on the guides, crank-pins, and journals will 
vary in the same way, and the friction on them correspond also. The lateral 
pressure of the piston packing rings will vary with the diameter, so that any 
reduction in diameter will produce a corresponding reduction in the friction. 

But, perhaps, so far as .economy in working is concerned, there is no more 
important consideration than the reduction in clearance space effected in 
the low-pressure cylinder by the reduction in piston area. The steam ports 
will be nearly the same in section, whether the engine be long or short stroke ; 
but the space between the piston and cylinder-ends is very considerably 
reduced, and will vary inversely as the length of stroke, because the axial 
distance of piston from the cylinder-ends is constant. 

The Rate of Revolution oi Marine Engines * at full speed varies roughly 
inversely as the square root of the nominal horse-power, which for this 
purpose may be taken as N.H.P. = D x S v K. 

D is the diameter of the low-pressure cylinder (or the equivalent, if there 
are two) in inches. 

S is the stroke of piston also in inches. 

K is a coefficient of 15 for two-stage compound, 12-6 for triple-compound, 
and 10-5 for quadruple-compound engines. Then — 

Rate of revolution per minute . . . = Q -j- -y/N.H.P. 

For the ordinary cargo boat, Q . . =1,200. 

For express steamships, Q . . . = 1,800. 

For naval and very fast express ships, Q . = 2,250. 

Example (a). — The proper rate of revolution for an express steamship 
having cylinders 30 inches, 45 inches, and 70 inches diameter X 42 inches 
stroke. 

Here N.H.P. = 70 X 42 -*- 12-6, or 233. 

Revolutions per minute = 1,800 -e- V233, or 118. 

Example (b). — Rate of revolution for a warship having engines 33 inches, 
52 inches, 64 inches, and 64 inches diameter X 48 inches stroke. 

Here D = y/2 X 64 2 , or 90-5. 

N.H.P. = 90-5 X 48 - 12-6, or 345. 

Revolutions per minute = 2,250 s- V345, or 121. 

Revolutions. — Although there is a very considerable range for choice of 
number of revolutions of the engines of most merchant steamers, there are 
certain well-defined limits beyond which very few practical engineers go. 

Very few screw engines are now worked below 75 revolutions per minute 
when in good condition ; and it is at this speed that most of the engines of 
the large mail steamers are kept running on the voyage so long as the weather 
permits. The engines of warships, for two very good reasons, work at much 
higher speeds. Their machinery must be light, and go into a small space, 
so that it is necessary to make an engine of certain dimensions to suit these 

* The N.E. Coast Inst. E. and S. recommend the following as the rule for rate of revolution of caruo- 
ship engines when on voyage : — -, oa 

N = 32 (S + 4) -=- S or ~ + 32. 



208 



MANUAL OF MARINE ENGINEERING. 



conditions, and cause it to develop the requisite horse-power by running 
at a higher number of revolutions. The speed of a warship is much higher 
in proportion to its size than is that of the merchant ship, while the draught 
of water is no more, and often less. For these reasons the screw of the war- 
ship is small for the power to be developed, so that even if large engines 
were admissible to drive the screw, they would be of small advantage, as 
they would have to move at a high rate. It will be seen, then, that small 
fast-running engines are a necessity, and especially is this so with modern 
warships, whether armoured or unarmoured. The latter must be as fine as 
possible, and every ton of weight saved to obtain the very high speeds which 
their service demands ; the former demands every sacrifice to save weight in 
machinery, for the sake of adding it to the armour and armament. 

Since a warship has so seldom to steam at full speed, and when she does, 
it is only for a short period, the short-stroke fast-running engine is not so 
very objectionable, and rigid economy is quite a secondary consideration in 
war questions. 

The following table gives the number of revolutions at which naval 
engines are run on their trial trips : — 

TABLE XXIXa. — Rates of Revolutions of Screw Propellers. 



Type of Engine. 


Description of Ship. 


Power of each Engine to 


a Screw. 


Revs, per Minute. 


Reciprocators, 


Battleships and lst-class 










cruisers, 


19,500 to 15,000 IE 


130 to 14u 


>• 


2nd-class cruisers, 


4,500 to 6,500 


>» 


140 to 146 


11 


3rd-class cruisers, . . 


3,500 to 5,000 


ii 


225 to 245 


91 


Scouts, 


7,000 to 8,500 


it 


210 to 202 


91 


Destroyers, . ... 


2,000 to 3,500 


ii 


360 to 400 


»» 


Torpedo boats. 


. . 






11 


Large Atlantic expresses, 


15,000 to 20,000 


ii 


76 to 80 


11 


Large mail steamers, , 


5,000 to 15,000 


ii 


90 to 80 


19 


Large cross-channel 










steamers, . . . 


3,000 to 4,000 


ii 


175 to 150 


>1 


Small cross-channel 










steamers, . 


2,000 to 3,000 


»> 


200 to 175 


» 


Large twin-screw cargo 










steamers, 


2,200 to 4,500 


19 


80 to 90 


M 


Large single-screw cargo 










steamers, . 


2,500 to 4,500 


l> 


80 to 70 


!) 


Medium single-screw cargo 










steamers, . 


1,000 to 2,500 


>> 


90 to 80 


9) 


Yachts and small craft, . 


250 to 500 


Jt 


150 to 250 


Turbines, 


Battleships, . . . 


5,500 to 7,500 S.H.P. 


320 to 300 


»> 


Large cruisers, 


10,000 to 20,000 


91 


200 to 175 


91 


Second-class cruisers, 


6,000 to 12,000 


11 


500 


11 


Scouts, 


4,500 to 7,500 


11 


750 to 600 


H 


Destroyers, . 


2,500 to 5,000 


11 


940 to 750 


.. 


Atlantic expresses, largest, 


15,000 to 17,500 


»» 


190 to 165 


91 


Express steamers, large, 


4,000 to 6,000 


»» 


300 to 190 


»> 


Cross-channel, large, 


3,000 to 5,000 


11 


600 to 450 


*» 


Yachts and small craft, . 


1 ,000 to 2,000 


11 


1,000 to 750 


,, geared, 


Expresses, . 


2,500 to 3,000 


*1 


300 



LENGTH OF STROKE. 



209 



The stroke of horizontal engines varied from 18 inches of the gunboat to 
54 inches of the large armour-clad, and the vertical engines of the Navy vary 
from 18 inches in the "destroyers" to 51 inches in the first-class cruisers 
and battleships. Latterly, with the four-cylinder engines, the stroke has 
been 42 to 48 inches in the large ships. 

Length of Stroke. — For very many years there existed a standard scale 
for the stroke of the vertical engine of the mercantile marine, and although 
there was no written law which guided engineers in the choice of this important 
dimension, it was so well known that only the diameter of the cylinders was 
mentioned in speaking of the size of engine, and in most of the rules for 
nominal horse-power used by manufacturing engineers in their dealings with 
shipowners, no direct allowance was made for length of stroke. With the 
walls of the cylinders only jacketed the best relation of diameter to stroke 
was as 1 to 1*5. So the ratio of stroke to diameter of H.P. cylinder is usually 
1-7 and the ratio of stroke to diameter of L.P. cylinder 0-9 to 0-6. 

The following Table gives the stroke corresponding to the different 
powers ; the one column giving the standard, and the other the stroke a3 
existing in ordinary e very-day practice in the mercantile matin e : — 



TABLE XXIXb. 



N H P 


Standard 


Stroke, as in 


N.H.P. 


Standard 


Stroke, as in 


1 


Stroke. 


common practice. 


Stroke. 


common practice. 


! 

20 


12 ins. 


15 ins. to 18 ins. 


140 


33 ins. 


33 ins. to 42 ins. 


30 


15 ins. 


18 ins. to 21 ins. 


160 


33 ins. 


33 ins. to 42 ins. 


40 


18 ins. 


20 ins. to 24 ins. 


180 


36 ins. 


30 ins. to 45 ins. 


50 


21 ins. 


24 ins. to 30 ins. 


200 


36 ins. 


36 ins. to 48 ins. 


60 


24 ins. 


24 ins. to 27 ins. 


250 


39 ins. 


39 ins. to 54 in 6 !. 


80 


27 ins. 


27 ins. to 30 ins. 


300 


42 ins. 


42 ins. to 54 ins. 


1U0 


30 ins. 


30 ins. to 36 ins. 


400 


45 ins. 


45 ins. to 60 ins. 


120 


30 ins. 


30 ins. to 39 ins. 


500 


48 ins. 


48 ins. to 66 ins. 



TABLE XXIXc. — Eevolutions and Piston Speeds according to Rules 
recommended by n.e. coast inst. e. and s. for cargo steamers 
on Service. 



Stroke 


Revolutions. 


Piston Speed. 


Stroke. 


Revolutions. 


Piston Speed. 


Feet. 


Per Minute. 


Feet 


Feet. 


Per Minute 


Feet. 


1-50 


117 


352 


3-25 


72 


464 


1-75 


103 


368 


3-50 


69 


480 


2-00 


96 


384 


3-75 


66 


496 


2-25 


89 


400 


4-00 


64 


512 


2-50 


83 


416 


4-25 


62 


528 


2-75 


78 


432 


4-50 


60 


544 


3-00 


75 


448 


4-75 


59 


560 



14 



210 MANUAL OF MARINE ENGINEERING. 



CHAPTER IX. 

GENERAL DESIGN AND THE INFLUENCES WHICH AFFECT IT. 

The General Design and Arrangements of Marine Engines are to-day charac- 
terised by simplicity perhaps more than anything else, notwithstanding that 
multiplicity of parts and connections gives the modern engine-room an air of 
complexity that was wanting in the older ships. Forty years ago there 
were, as a rule, only two cylinders to the main engines : both air and cir- 
culating pumps, as well as the feed and bilge pumps, were worked from the 
pistons, direct in the Navy, and in that way or by means of levers interposed 
in the mercantile marine. There was, as a rule, only one auxiliary feed 
pump, and in large ships one fire engine in addition, which did duty for a 
bilge pump as well as for deck service when required ; merchant cargo ships 
with ballast tanks had also the second auxiliary pump for emptying them, 
and arranged to act for general purposes. Steam starting gears- were being 
used generally with very large engines, but never with small ones. Steam 
steering gears were equally rare, and there was, of course, no electric light 
engine, nor any refrigerating plant. Distillers had been introduced for 
giving fresh water as auxiliary feed to the boilers by Hall, of condenser fame, 
thirty years before this period, but the idea had not taken on. There were, 
however, in general use the well-known Normandy distillers for producing 
drinking water ; every Naval ship had a set, and they were to be found 
in the better-class passenger steamer. The mercantile engineer had, for the 
exercise of his talents and the occupation of spare time, the care of the steam 
winches, just as the naval engineer had in some ships the charge of the turret 
and turntable engines and ammunition hoists, etc., outside his own domain. 
Very few ships had more than one main engine or a steam launch or pinnace, 
and no ship had air compressors. All these things have been added one after 
another, until the care and anxiety of the chief engineer is no longer centred 
and concentrated on the main engines, but now it is rather the very numerous 
parasites of them and of the crowd of outside machinery — machinery so 
necessary to the well-being of the modern warship and express steamer — 
that gives him the most concern. But, in spite of all this, there is about 
the main engines, especially if they be turbines, a simplicity that is com- 
mendable under such circumstances. The day for fancy design and odd 
arrangements of machinery has gone ; in their place there is now only a 
stereotyped system that marks the final stage of natural selection and the 
survival of the fittest. The engine of one modern engine builders differs 
very little in essence from that of another ; the turbine, it is true, has its 
variation in principle as well as in arrangement, but it, too, is gravitating 
to the stereotyped design, although makers may still each have his own, 
differing in form of details and fittings. 



DESIGN AND ARRANGEMENTS OF MARINE ENGINES. 211 

The accepted and approved type of reciprocating engine is, of course, 
the vertical inverted direct-acting one, whether it be worked by steam or by 
the internal combustion of oil or gas. The cylinders are placed in line each 
over its own crank, and, as a rule, they are separated one from another ; each 
has its columns braced and secured so as not to be dependent on its neighbour 
for stability or steadiness. The connections for steam transmission are by 
pipes so fitted as to expand and contract freely under thermal or pressure 
forces without acting or reacting on the cylinders and so cause them bad 
alignment. At the same time the general structure of columns is braced 
together, so as to give it, as a whole, additional stability against external 
disturbance, as from the shock or strain of rolling, pitching, and collisions. 
Small mercantile engines, however, are still made with the cylinders cast 
or bolted together in pairs, but the tendency is in the direction of separating 
all of them, especially when the stroke of piston and consequently their 
length is greater than that usual with the high-speed land engines of equal 
power. Each cylinder now operates on a single complete piece of crank 
shaft running in two bearings coupled to the next cranks ; this is so in all 
but very small engines (under 100 N.H.P.) or in larger engines, where weight 
and space have to be cut down to the minimum, when the shaft is then in one 
piece. The valves are now seldom placed on the cylinder outer sides, but 
generally on the fore or aft side, so that the valve spindle is immediately 
over the shaft, and consequently capable of being driven by the double 
eccentrics and link motion direct without the interposition of weigh shafts 
and levers. Steam reversing gears are fitted to all but the very smallest 
engines, and are generally of the push-and-pull type introduced and still 
supplied by Brown Bros., of Edinburgh, for the mercantile marine, while 
the " all round " worm and wheel gear is most frequently used in the Navy. 
The former is decidedly the handier, and can now be obtained at quite as low 
a cost as the all round gear, but the latter can be used to turn the main engines. 
Similar push-and-pull gears, of various designs, are employed to operate 
the very large stop valves of the reciprocating and turbine engines, as they 
<lo also the heavy change valves of the latter, whereby the turbine is reversed 
by changing the incoming steam from the head-going to the stern-going 
part of it, or when the exhaust steam from the L.P. cylinder is diverted to 
a L.P. turbine. The circulating pump is now, in all but the small mercantile 
engines, of the centrifugal type driven by an independent engine ; in all 
sorts of fast-running engines the air pump is also, as a rule, absent, and it, 
too, has its separate engine. With turbines where high vacuum is a necessity 
for high economy of steam, the double-stage air pump of Mr. Weir is often 
employed ; there is still a prospect of the centrifugal pump being employed 
in series on this service, as it has been used for boiler feeding and delivering 
•water generally against a high " head " on land. 

The Admiralty long ago initiated, and the mercantile marine followed 
in express steamers, the practice of removing the whole of the feed and bilge 
pumps from connection with the main engines. The merchant ship, how- 
ever, was really first in the employment of the independent feed pump with 
automatic regulating gear, whereby its speed was only such as to give just 
the necessary supply of feed demanded by the stokehold watch-keeper. 
Now, in all important steamers, and certainly in all turbine-driven ships, 
the feed bilge fire and general service pumps are absolutely separate and 



212 MANUAL OF MARINE ENGINEERING. 

independent of one another ; moreover, they are generally in duplicate to 
avoid the risk of stoppage of the main engines in case of the failure of an 
auxiliary. 

Formerly the Condenser was a huge cast-iron vessel, forming an integral 
and important part of the engine structure or framing, to which it generally 
gave a massive foundation for the guide columns, and contributed to the 
stability of the framing (v. fig. 71), as well as a firm support for the bearings 
of the pump beam's weigh shaft. With the advent of the three-crank engine 
came the tendency to separate the condenser from the engine frames, 'and 
as this kind of engine became lengthened out, and; furthermore, when there 
came a four-crank engine to stay, it was no longer expedient to have it with 
such long tubes as would be entailed with a condenser body the whole length 
of the engine. Moreover, the addition of three and four sets of columns 
or column feet on it made it a costly and somewhat risky casting, as well 
as a heavy portion of the weight of the machinery. To-day the condenser is 
generally a distinct and separate vessel placed close to the L.P. cylinders, 
and of a form suitable to its own particular service, and in no way subser- 
vient to any other such service as was formerly exacted from it when it formed 
part of the engine superstructure (v. fig. 68). The cylindrical form was the 
common one, on account of its natural strength and cheapness, but now the 
demand of condenser experts have caused the cylindrical to be rejected for 
one with the heart shape cross-section, notwithstanding that the general 
requirements of such experts can be carried out sufficiently well in the cylin- 
drical shell by fitting it with the form of special baffle directors (v. fig. 116), 
or in the rectangular contraflo design of Mr. Morison (v. fig. 117). 

The framework of the engine is now much as was usual formerly in it? 
general design, but greater care is exercised to give both general and local 
stiffness to every part, which, no doubt, is an improvement which tends tG 
the better working of all engines in every way. 

The Lubricating Arrangements are much more extensive and complete 
nowadays than formerly obtained, and the necessity for a regular steady 
and positive supply of oil to every bearing guide and pin is now recognised 
and provided by means of small pumps worked by the engines, or from over- 
head tanks, which deliver a steady stream at a pressure high enough to force 
the lubricant into every part requiring it. 

The Design of an Engine is influenced by External Causes, which are not 
always seen or even acknowledged, but nevertheless are often in active opera- 
tion all the time. There is even in the engineering and shipowning world 
also that which is generally elsewhere called fashion, exercising its powers, 
and imposing on the designer conditions with which it is often futile to struggle. 
Even the better general and technical knowledge possessed by experts to-day 
does not always preserve them from following a lead which is more or less of 
a blind nature, and from starting out in directions quite contrary to the 
convictions based on their experience, and for no other reason apparently 
than because someone else has done so with all the appearance of success. 
The trial and error system still prevails largely, in spite of the technical and 
scientific training now so freely obtainable everywhere at quite low cost. 
Model experiments are always interesting, often instructive, and, when 
properly understood, may be good and safe guides for directing the pro- 
ceedings of the work-day engineer, but such models, even when telling truth, 



DESIGN OF ENGINE INFLUENCED BY EXTERNAL CAUSF-S. 



213 



do not always tell the whole truth ; what they suppress often comes out as 
a ghastly truth when the full-sized working engine is produced ; what was 
a trifling matter and hardly tangible in the miniature becomes in the preat a 




a 

a 
o 

'to 

a 
a 

x 
© 



H 

© 
a 
*>» 

u 
C 

o 



o 

■J 
© 

-4-3 

«B 
m 
c4 

© 

CQ 

a 
© 

fi 
O 

"x 

3 
<33 

'3 



© 



00 

o 



terrible incubus, producing the most damning consequences ; what was a 
pigmy in the small engine and quite easy of control becomes a very Franken- 
stein" in the large one, irreducible, and carrying all before it, and spoiling 



214 MANUAL OF' MARINE ENGINEERING. 

everything. It behoves everyone, therefore, to use models and model 
experiments with great discretion. 

The Supply of Materials exercises a powerful influence on the designer 

of engines generally, but especially on him who has to cut down to a minimum 

the weight of and space occupied by machinery in the way the marine engine 

builder is compelled to do. When Brunei designed the " Great Britain " 

there was no steam hammer in existence and no forge capable of making 

so large a shaft as that required for the ship ; consequently he had to be 

content with one of cast iron when at that stage ; fortunately for him, 

Nasmyth invented just then the steam hammer, and made one in time to 

produce a wrought-iron shaft, with which the first voyage of that ship was 

made. Cast iron was largely used for all parts of marine engines, and designs 

were made accordingly. When copper was boomed up to £80 per ton* and 

even higher by an enterprising but somewhat short-sighted syndicate, the 

equally enterprising and versatile marine engineer adopted steel and iron for 

piping, and had other things made of cast steel which formerly had been 

exclusively of bronze. Steel castings, doubtful and often unsatisfactory 

as they may be, have served their turn in changing the design of many 

parts of an engine, but what has effected the greatest departure from old 

practices to that which prevails to-day is the cheap and unlimited supply 

of excellent mild wrought steel. By its means the pressure of steam possible 

in cylindrical or tank boilers has been raised from 100 to 240 lbs. per square 

inch, and their possible diameter increased from about 14 feet to 18, while 

their cost has been reduced very considerably. The internal plates of such 

boilers used formerly to .be made of " Lowmoor quality," at a cost of £27 

per ton and upwards ; similar plates in the superior metal (mild steel) and 

of very large sizes can be purchased now for a third of that sum ; shell plates 

of iron, 1| inches thick, could be bought, but they were narrow and not very 

long ; moreover, if of the full area the mill could turn out, they were very 

costly. Now, plates up to 50 feet long and 12-5 feet wide can be rolled, up to 

a thickness of 1*8 inches ; shell plates are often If inches thick and 30 feet 

long, so that one, or at most two, plates when single-ended are sufficient ; 

circular-end plates are made up to 13 feet in diameter. Forgings of every 

description and size can be made of this material, and, if needed, steel of a 

higher tensile, say 40 tons ultimate, with quite a good amount of stretch, 

can be obtained also in large size and at moderate cost ; for special purposes 

at costs by no means prohibitive vanadium, nickel, or other high-class steels 

are made, and supplied of a quality and fitness beyond reproach. 

Rolled bars of excellent steel can be obtained of any diameter up to 
15 inches, and square bars to 6 inches ; rectangular section bars can also be 
had in various sizes up to 15 inches by 2 inches, and 11 inches by 2.V inches,, 
so that steel caps for bearings, rod ends and other similar purposes can be made 
in wrought, steel at a cost very little in excess of that of castings in iron. With 
the self-hardening tool steel and high-speed lathes and machine tools now in 
use, these rolled bars can be converted into bright ones at a trifling extra cost. 
The framework of an engine in wrought columns and tie bars is, therefore, 
now a much less expensive luxury than it was formerly, even for large engines. 
Then, too, the other sectional steel now obtainable in such variety has per- 
mitted designers to fashion engine beds and their framing in ways not possible 
without it, whereby great saving in weight is effected, and inasmuch as little 

* Now £150, due to war conditions. 



THE INFLUENCE OF TONNAGE LAWS. 215 

* 

or no pattern-making is required for such designs, it is an economic method 
when the engine required is a special one with few or no repetitions of it 
expected. 

The method of obtaining sectional material (introduced by Mr. Dick) 
by the extrusion of the zinc bronzes through dies has permitted of multiplying 
varieties without necessitating the expense involved in cutting rolls. This 
and the other ways in which these high-class tough, strong bronzes mav 
be used have had no small influence also on the design of the smaller special 
engines. 

Such things permit of refinements in design not possible in the days 
when engineers were limited to the choice of wrought iron of 20 tons ultimate 
tensile strength, or, by paying a high price, 25 tons at most. Crank and 
straight shafts of almost any size can be forged and machined at prices that 
were impossible for even small ones a few years ago. Even cast iron has been 
improved by selection and mixing, that without paying fancy prices for any 
of the good brands, the tensile and bending tests of castings are equal to the 
very best given by Fairbairn & Whitworth. Aluminium has not yet seri- 
ously influenced the marine engine designer ; that it will in the near future 
is certain ; its lightness alone will attract him, and probably alloys will be 
found which, while not adding seriously to their weight, will improve their 
strength and resistance to corrosion. 

The Influence of Tonnage Laws on the hulls of ships is well known and 
obvious, but it is not limited to them. It pervades the whole of a ship more 
or less, and perhaps more so the machinery space than elsewhere. Latterly 
that influence has been more potent and insidious in its effect on marine 
engine design than believed to be possible years ago. Steamships have had 
from the earliest days of their construction some consideration given for 
the disadvantages under which they were worked compared with the sailing 
ship. The space occupied by the machinery has always been deducted 
from the gross tonnage, inasmuch as it could take no cargo : later on, in order 
to encourage shipowners in making the spaces, not only habitable, but health- 
ful for those in charge of, and those labouring at the machinery, special 
allowances were made for the light and air spaces to engine and boiler rooms, 
as well as for those rooms themselves. If the actual total of the spaces 
which are allowed off the gross tonnage of a ship amount to 13 per cent, of it, 
the actual deduction permitted, so as to arrive at the net or register tonnage 
is no less than 32 per cent. In the days of low freights, costly fuel, etc., 
it is highly necessary to keep the register tonnage down, as on it the ship is 
taxed. Now, whereas in Naval ships the space allotted to machinery is 
always small for it, as indeed it was formerly only too often the case in the 
mercantile marine, until it was found that the enlarging of it enabled a con- 
siderable saving in working expenses to be effected at little cost and with 
inconsiderable drawbacks. Formerly with the two-cylinder engine often 
of quite small power, but requiring large supplies of fuel, it was quite impos- 
sible to arrange for a reduction of tonnage measurement so large as 13 per 
cent, without a much too serious sacrifice ; nor, indeed, did then the exigencies 
of the times press for such drastic measures to obtain it, consequently it was 
only tug boats and express steamers having very large engines in proportion 
to their size that enjoyed these liberal concessions, as indeed they alone 
were intended so to do ; consequently to the majority of steamers there was 



216 MANUAL OF MARINE ENGINEERING. 

* 

no gain by making the engine and boiler rooms larger than required by the 
bare necessities of the case, except that, until the amendment of the Merchant 
Shipping Acts, light and air spaces were not measured into gross tonnage, 
although they were included in the deductions from it ; consequently every 
ton of space devoted to that purpose caused a virtual reduction of two tons 
in the register tonnage. 

Since the space occupied by the machinery could not be laden with cargo, 
every foot of it was a foot less space in the cargo holds, it was made as small 
as possible, and the engineer designer had to exercise his wit to devise the 
most compact engine to occupy the least possible space. Hence the popularity 
with shipowners of the single-crank engine, either with one cylinder or two 
compound ones tandem ; of the diagonal engine with a cylinder in each wing 
operating on a single crank (fig. 69), or with a third cylinder vertical, as in 
fig. 70. The compound engine with one cylinder vertical and the other 
horizontal, both operating on the same crank, found favour, as did also the 
engine with one pair above them and the other pair of cylinders behind 
operating on the same pair of cranks with levers, as revived by Mr. M'Alpine, 
latterly with the idea to provide a naturally balanced engine. Besides these 
typical instances, there were other ways in which the engine was treated by 
the ingenious, as, for example, in the way of special valve gears, etc., to 
reduce still more thereoy the space occupied by it (fig. 71). 

A further effect of the tonnage law, however, was more serious, inasmuch 
as it retarded the adoption of the better forms of triple- and quadruple-com- 
pound engines, and encouraged the manufacture of those that were worse ; 
the fitting of valve gearing of almost fantastic design, which displayed an 
exaggerated inventiveness, while it gave endless trouble and anxiety to those 
in charge of it, was entirely due to the desire to save space and reduce the 
size of the engine-rooms. The early three-crank triple engines were often 
treated in this way ; by shipowners they were objected to, although only 
too anxious to benefit by the three-stage system, on the ground that they 
required too much room ; by the engine builder they were subject to such 
cutting and squeezing that the crank bearings and pins were reduced 
past the just minimum, and the valve gears were obstructions, and pre- 
vented the proper attention necessary to such working parts ; moreover, 
they required such an expenditure of oil as to detract from the economy of 
fuel (r. fig. 71). 

The two-crank triple, which in its way worked and did good service, was 
not so good an engine mechanically as the three, nor was the two-crank 
quadruple so good as the present four-crank one, although it was better in 
some ways than the two-crank triple. It was, however, due probably to the 
tonnage question that the quadruple engine came so soon as it did into the 
field of practical engineering, inasmuch as the claim that won it most attention 
and patronage at the first was the small space the two-crank engine occupied 
compared with that required for the three-crank compound. To-day all this 
is changed ; every ship must have the 32 per cent, reduction ; and, further, 
in some cases, since the law says that where the spaces which may be deducted 
amount to over 13 per cent, of the gross, the rate of reduction shall be one and 
three-quarter times the actual amount. In some very fully-powered ships the 
actual spaces may amount to even 54 per cent, of the gross ; then the net or 
register tonnage is really only 5 # 5 per cent, of the gross. There are, however, 



THE INFLUENCE OF TONNAGE LAWS. 



217 




218 



MANUAL OF MARINE ENGINEERING. 




THE INFLUENCE OF TONNAGE LAWS. 



219 



limitations set now so that the allowance may not amount to a public scandal, 
as it did quite recently to the Dock Companies, Harbour Boards, etc. Now it 
would seem as if an engine-room could not be too large, but, since the Board 




o 

O 

« 
> 

"5 

> 



a 

— 
02 



a 

"So 

c 

a 
_o 
'35 

c 

eS 
PU 

X 

? 

£ 

s* 



-4 

a 

u 
a 

6 
35 

s* 

J5 
E* 






220 



MANUAL OF MARINE ENGINEERING. 



of Trade will not permit it to be wholly measured off tonnage unless the size 
of the engines themselves will warrant the space allotted, there is a direct 
premium existing for making an engine as long as possible, and also to extend 
it athwartship liberally likewise. Consequently the four-crank engine (fig. 




72), triple as well as quadruple, is the favourite in all express steamers, 
because ostensibly it is so much easier to balance. 

The " Joys," " Marshall," and other special gear for driving valves placed 
in front or rear of the cylinders are consequently no longer necessary and so 



THE INFLUENCE OF THE BOARD OF TRADE. 221 

gone, and in their place the old and well-tried pair of eccentrics with their 
link motion is revived in their best forms, so that they are now with ample 
bearing surfaces, sufficient breadth of strap, etc. ; they are, moreover, quite 
accessible. The condenser, for perhaps the same reason, sometimes is placed 
in the wings and the pumps of various kinds scattered about, each with its 
allotted and ample space. No longer are tbey and all other auxiliary engines 
and machines squeezed into odd corners and niches, which were narrow 
and inaccessible often, and always cramped and confined. 

All this may not be altogether the result of the tonnage laws alone, but 
it is pretty certain that if there were not such good allowances off gross tonnage 
there would not be such liberal spaces, and if there were not roomy spaces, 
then the engines would have to be cut to fit them, such as they might be. 

The Influence of the Board of Trade on marine engine design and practice 
in other directions has been, and is still, very great and, on the whole, bene- 
ficial. The better construction and design of the cylindrical boiler was 
largely due to the action taken by Mr. Thomas Traill and his staff officials of 
the Board about 1873 in formulating Rules and Regulations, which, while 
being somewhat arbitrary and often unnecessarily rigid in application, were, 
nevertheless, one of the chief means whereby the marine boiler has been so 
free from accident, slight and serious, for so long. The investigations carried 
out by the late Mr. Peter Sampson when assistant to Mr. Traill brought to 
light an enormous amount of useful information for the guidance of engineers 
generally, as well as for their own, in drawing up the Rules. That they then 
viewed all steel with a suspicion that did not seem warranted was the subject 
of much regret by those who felt sure that good steel could and would be 
produced at a cost which must, in course of time, drive wrought iron out 
of the market, if it was given the same freedom as accorded to wrought iron 
of every make and sort. Nevertheless, it must be admitted now that in thus 
keeping all steel under strict surveillance, much that was bad was prevented 
from coming into general use, and all makers of the material were thereby 
compelled to exercise the greater care in manufacture and treatment, which 
now permits of the greater freedom in its use. At the same time, there has 
been, and still is, thoiigh in a lesser degree, the regret that more encourage- 
ment is not given to those manufacturers who honestly strive to provide 
good and safe material for engineers' general use, having virtues superior 
to those of the common sort. In a general way, perhaps, low tensile steel 
is safer than the higher kinds, but it does not, therefore, follow that no high 
tensile steel can command the same confidence reposed in it that obtains with 
the " best mild steel." In this respect their progress has been slow in the 
mercantile marine. The high tensile steels are employed very sparingly 
compared with what might have been the case had the restrictions on their 
use been exercised in accordance with their real merits, instead of in com- 
pliance with the policy of the " Board." It is, of course, freely admitted 
that a Government Department has to be very discreet in its actions when 
dealing with one manufacturer and another, especially seeing that the desire 
is generally of all concerned for the practice by all officials to be uniform and 
absolutely impartial ; that the treatment of all shall be on the same lines 
without differentiation, baseless or otherwise. In the use of an old and 
tried material like cast iron, however, there are not the same conflicting 
causes for singular treatment by the Board of Trade and its officials ; yet, 



222 MANUAL OF MARINE ENGINEERING. 

while at one time nearly the whole of the engine was made of cast iron, and 
some of the most important parts are still formed of it, where it could be 
squeezed out, it has been by the policy of both Admiralty and Board of Trade 
of late years to do so. Yet this material has some distinct virtues which render 
it not only a convenient but a safe one for the composition of certain parts. 
Under tensile stress it stretches more than steel does up to the elastic limit ; 
to resist compression it has no equal, so that some structures which are liable 
when under load to be subject to a considerable amount of compression 
are really better made of it than of steel. Notwithstanding this, however, 
the tendency is to avoid its use, largely due to the policy and rules of these 
Government Departments on the ground of its comparative brittleness and 
liability to crack. The continued testing by expensive methods of the steel 
material, both in the cast, forged, and rolled state, as now used in engine 
construction, has tended to retard the use of it for a considerable period 
after steel makers had found the means of producing it cheaper than wrought 
iron. Since very little, if any, reduction in scantlings was claimed by engineers 
when substituting steel for wrought iron in many parts, it was of little or no 
consequence what its ultimate tensile strength was ; it was, however, impor- 
tant, and is still necessary, to be assured that it is tough and ductile. This 
could be quite certainly and satisfactorily ascertained by submitting samples 
to a simple cold-bending test, and further to punching and upsetting tests, 
such as applied to rivets. Had such simple and inexpensive means been 
adopted, this superior material might have taken the place of common wrought 
iron, and even of cast iron, in many places long ago. 

The restrictions of Government Departments, as a rule, tend to stag- 
nation in engineering practice, while fulfilling the function for which they 
exist to protect the public from the dangers that might arise if greater freedom 
were given to engine builders. But neither at the Admiralty nor at the 
Board of Trade is there now raised that dead wall to rational and desirable 
progress ; indeed, the Admiralty during the past two decades have become 
almost too progressive for the economic working of those who serve them. 

The Balancing of Engines and Avoidance of Vibration is a distinguishing 
feature in latter-day marine engineering practice, and the advent of the 
turbine on shipboard has rather accentuated the necessity for the exercise 
of the art of balancing than abrogated it ; for, although that instrument 
itself is free from the inertia defects of the reciprocating engine, it requires 
nevertheless the most careful of balancing itself ; but to compete with the 
turbine-driven steamer, that one having reciprocating engines must have 
them so that they run with an absence of vibration as far as is possible. 
Thanks to that veteran engineer and practical scientist, Dr. Otto Schlick. who 
gave us so freely the benefit of his years of patient labour and careful research, 
we understand now, as we did not before, the reason for and the causes of 
all vibration in a steamship, and, better still, he had devised means for 
correctly ascertaining them and curing their effects, for which we were more 
indebted to him. Formerly it was not uncommon to attribute all the vibra- 
tion of a screw ship to the action of the propeller ; that some of it might be 
due to the momentum of the moving parts of the engines was, of course, 
apparent to all engineers having even an elementary knowledge of dynamics ; 
attempts of a feeble and tentative kind were made to check them by fitting 
balance weights to the shafts opposite the cranks of the horizontal engine, 



BLADES OF A SCREW. 223 

or by casting with the turning wheels balance weights of segmental form. 
No doubt the inertia effects of the pistons and rods were to some extent 
neutralised by these balance weights, but it would appear that what the 
makers of these engines really aimed at was merely to balance the weight 
of the crank-pin, arms, and connecting-rod ends rather than the horizontal 
forces due to the movement of the pistons, etc. Balance weights were very 
seldom fitted to vertical engines, notwithstanding that from these statical 
considerations they were really the more needed in them than in the hori- 
zontal. 

Nowadays balance weights are freely used in all classes of engine and 
for all sizes, but the application of them is effected with more discretion 
and judgment than formerly prevailed, so that much better effects are 
obtained with considerably less material and cost. The four-crank engine, 
balanced on the system introduced by Dr. Schlick and perfected in this 
country by Messrs. Tweedy and Yarrow, is without added weights, and now 
in general use ; an almost perfect balance with an absence of vibration is 
obtained by this system with the special arrangement of the angles of cranks 
to suit the momenta of the moving parts. 

Whether the engine be a three-, four-, or five-crank one, it must be satis- 
factorily balanced in every warship and express passenger steamer, and even 
in the cargo steamers, which may and often do convey passengers, it is 
desirable to avoid unnecessary vibration. 

That Vibration arises in part from the Action of the Screws is only too true 
now as it was formerly, for the same causes exist in full force to-day not- 
withstanding our better knowledge of the subject, thanks to Dr. Schlick, 
for, after the engines have been most carefully and perfectly balanced, there 
often is manifest a residual vibration or trembling, which, while it may be 
quite local and not general, is nevertheless disagreeable. Moreover, the same 
defects are observable in as pronounced a manner in turbine-driven steamers 
notwithstanding the uniformity of torque and absence of any unbalanced 
inertia forces. 

There are various ways in which a screw may set up vibrations, all of 
which, even the smallest, is capable of creating the evil if its application 
is intermittent and regular and its periodicity coincides with that of the 
natural vibration of the ship as a single structure, or of any part of it which 
is free to vibrate. Under such circumstances of synchronism quite small 
and insignificant forces are capable of producing enormous results, as is well 
known to all engineers, and the most familiar illustration is perhaps that 
of a disciplined force crossing a bridge, who, if walking in step, are liable to 
set up dangerous oscillations in the most substantial of structures. 

If the Blades of a Screw are of Different Pitch, especially at or near the 
tips, the pressure on opposite blades is not the same, the propeller will be 
running out of balance, and as each blade should take its due proportion of 
thrust to run in perfect balance, such differences in pitch or blade surface 
Avill easily cause vibration to be set up as the true centre of pressure will not 
coincide with the shaft axis, but have an orbit of its own, the centre of 
which is on that axis. The desire for the utter absence of vibration and the 
economy of high efficiency have induced some engineers to have the blades 
of propellers chipped and ground, so that, not only are the working faces 
smooth and true to pitch throughout, but each blade is of equal pitch with 



224 MANUAL OF MARINE ENGINEERING. 

the others. In spite of the time and cost of carrying out this refinement, 
there is ample justification when the revolution is high and the ship a fast 
express steamer, or for good gunnery. For a cargo steamer it is, of course, 
quite desirable that the blades shall be of equal pitch, and, as far as possible, 
by care in moulding, to have a uniformity, but there is not the same warrant 
for chipping and grinding in their case. The same remarks apply to screws 
that are damaged by bending or breaking of the tips. It is, however, aston- 
ishing how badly damaged a screw may be and yet do its work fairly well. 

The Screw is always working in Water disturbed more or less by the passage 
of the ship herself, as also by the currents set in motion by the suck of the 
screw. The former, known as wake currents, cause a difference of pressure 
on the upper and lower blades of the screw ; the result is, therefore, pretty 
much the same as in the case of a screw with a blade out of pitch, except 
that in this case the real centre of pressure has no orbit but remains constant 
outside the shaft axis. The upper or surface current in wake of a ship 
has more forward motion than lower ones, and moreover is often quite a 
distinct layer of water flowing over practically still water, each blade in 
turn comes into it suddenly, and receives a sudden accession of load and a 
tendency to retardation of motion ; if the period of blade stroke on this 
stream coincides or synchronises with the periodicity of the ship, it will soon 
set up quite sensible vibrations, and, if there is no break in the timing, they 
may become quite violent. If the synchronism is not perfect they will 
damp out and disappear, only to reappear later with equal violence. This 
kind of disturbance causes horizontal vibrations, which were very pronounced 
in ships having only two blades to a single screw ; it was, and is, still noticeable 
with single-screw ships with four-bladed screws, especially when the tips are 
broad. 

The Proximity of Screws to Portions of the Hull is also a common cause 
of vibrations, especially of the small or residual ones, that are often as trouble- 
some as the larger ones. This is due to more than one agency. If the blade 
tip passes closely to a fixed obstruction, such as the stern frame, or the hull 
of the ship in the case of multiple screws, its resistance or thrust pressure 
is for the moment changed, and as each blade in turn does so, there is pro- 
duced a series of impacts, which in themselves may be slight and almost- 
imperceptible, yet may in their collective action, when synchronising with 
some massive portion of the ship, produce in it serious activities. There is 
always a film of water of quite sensible thickness dragged on by friction 
with the skin of the ship at a velocity not much inferior to the ship itself. 
If, therefore, the blade tip is so close as to come into that envelope of inert 
water from water that was flowing past the screw at the rate of 2,000 to 
3,000 feet per minute, it would be almost as bad in its effect as if it struck 
a solid, and the blow each time a blade tip passed through would be, and 
probably is, very considerable, and the result is manifest on Dr. Schlick's 
interesting pallograph diagrams. 

Then such obstructions as the stern frame and brackets, spectacle pieces, 
etc., act as brakes to the spiral currents from the screw tips, so that as each 
blade passes near it there is the change in direction of flow, which, although 
very slight in its effect, may in the cumulative form be very real. 

The Power of Sucking in Air, so beautifully and clearly demonstrated 
by Professor Flamm, and experience by each observer of the screw in 



THE INTRODUCTION OF STEEL CASTINGS. 225 

everyday practice, may, and probably does, lead to a considerable amount of 
vibration. It is well known that when air is drawn into the screw race there 
is a diminution of thrust, and there may be, and often is, a momentary 
cessation of it ; if this becomes periodic, and the effects cumulative, a most 
unpleasant form of hull disturbance will be experienced. This air suction 
occurs the more readily with screws whose blade tips are near the surface, 
and even when there is a fair amount of immersion in still water the suction 
or feed to the propeller when running fast produces a hollow or depression 
of the surface, so as to render air "spouts" easy of formation. The same thing 
happens when there is quite a gentle swell on, so that the heave and pitch 
of the ship causes the blade tips to approach too close to the surface. 

Cavitation may also be a cause of Vibration when the screws are quite 
thoroughly immersed, especially when the screw is driven by a steam engine 
with a torque of great variation, or with an internal combustion engine 
with the same or worse defect, so that the propeller has sudden and severe 
acceleration in angular velocity, whereby the pressure per square inch is 
increased beyond the limit of good working, so that cavities are formed 
by the reduction in pressure behind the blades, permitting of air separating 
from the water and forming masses of bubbles, which, when liberated, may 
have the same effect on the screw as the air drawn down from above by 
" suction." 

AH these Things may cause Vibration, and always are sources of loss of 
efficiency. The former can be generally reduced to a minimum, if not alto- 
gether damped out, by causing the engines to run at a rate of revolution 
that precludes synchronism with any of the important masses of the hull. 
The efficiency of propeller and hull can be little effected by such means, but 
doubtless the hull that does not vibrate is likely to permit of more gross 
power being applied to propulsion than one that does. 

The Necessity for Perfect Balance of Engine and a quiet-running Screw 
is imposed on the designer, and he must study the questions involved as 
carefully as formerly he had to do those involving the safety and economic 
running of the machinery on shipboard. 

The Auxiliary Machinery must also be free from Vibration and unevenness 
from the power to produce it ; for to-day, with turbines and the beautifully 
balanced reciprocators, it would be grotesque to find the ship vibrating from 
a donkey pump's unbalanced forces ; and now that air pumps, as well as 
all other pumps, are separable from the main engines, it is incumbent on 
their makers to supply them free from a vice, from which the main engines 
have been eradicated. The makers of electric-power generating engines have 
been compelled long ago to balance them so that they create no nuisance 
to the neighbourhood in which they are situated by themselves vibrating 
or causing any tremble in the neighbouring houses ; therefore, as a rule, the 
makers of these engines who have had such shore experience can be trusted 
to supply them on shipboard quite free from vice. 

The Introduction of Steel Castings was hailed with delight by engine designers 
thirty-five years ago, as it raised the hope that the weight of machinery 
generally and the cost of many of its parts would be very materially reduced, 
and that in other parts- the risks run consequent on the necessary employment 
of cast iron, and even with cast bronze in their manufacture, would cease. 
To-dav the engine designer is still living largelv in the same hope, inasmuch 

15 



226 MANUAL OF MARINE ENGINEERING. 

as the sound, perfect casting at moderate cost in a metal strong and tough 
and capable of working at quite a high temperature remains yet to be pro- 
duced in quantity and promptly. British steel founders can, and do, make 
huge castings, which for ship work are invaluable ; they also supply engineers 
with some very fine and useful ones, but they had to look to another country 
for that soundness and uniformity in quality so much to be desired by those 
who are responsible for and take the risks of such things. It is a matter oi 
great regret that there lacks sufficient certainty in the home product to 
encourage its more extended use for such parts of an engine as must be 
sound or machined all over. This is a disappointment to those who have 
looked year after year for the improvement so long hoped for. 

The Use of Aluminium in engine construction has been somewhat delayed 
from somewhat similar causes, in spite of the fact that the material was 
sold at such a low price as to permit of its free use in engine construction. 
Now, however, the very beautiful castings supplied for motor car machinery 
must induce the marine engine designers to use them more freely in their 
products. The metal now used is an alloy, and is only a trifle heavier than 
pure aluminium ; the castings are sharp and sound in outline and clean in 
surface ; the tensile strength is sufficient, being as good as ordinary bronze. 
It will not, however, withstand the action of sea water as does bronze, so that 
it cannot be used for the parts exposed to it, but there are very many other 
parts which are quite free from contact with sea-water, and others where a 
coat of paint will give sufficient protection. 

Duralumin, a 90 per cent, alloy of aluminium introduced by Vickers, 
Ltd., has such good qualities as soon to command the attention of the 
marine engine designer, inasmuch as it is quite strong (28 tons tensile with 
15 per cent, extension), and will resist shock ; it is very light (sp. gr., 2-8) 
and malleable, and can be treated as the zinc bronzes are, by drawing out into 
bars of various sections, also rolled into plates and sheets, and sold at quite 
moderate prices, so as to be extensively used in the construction of air-craft, 
motor cars, etc. ; it cannot, however, be used for castings as yet. 

In the Future the Reciproeator will probably undergo more changes, 
necessary to its success in competition with the. steam turbine ; and also that 
it may hold its own with the oil engine, it may be necessary to follow some 
of the features peculiar to that engine. For example, the oil engine at present 
has certain limitations, which hedge it in completely. The steam engine 
maker may impose on himself, if he chooses, some of them with advantage. 
For instance, the huge cylinders of the present triple and quadruple engine 
may give place to a multiplicity of smaller ones ; so small may they be indeed 
that the low-pressure ones may have relatively such large ports and valves 
as will enable them to benefit usefully by the high vacua now attainable in 
condensers on board ship at a trifling extra cost. 

Superheating of steam to an extent now practically impossible also may be 
followed with complete success in small cylinders, including a marked reduc- 
tion in the consumption of fuel. There may be even modifications in the 
methods of reversing the engines, whereby there will be effected savings in 
first cost, as well as economy in working costs The flexibility of the reci- 
procating steam engine is a strong and marked feature in the eyes of the 
marine engineer, especially of him who has to work and be responsible for 
the machinery of a ship : it should count likewise with owners, underwriters, 



FUTURE OF THE REC1PROCATOR. 227 

•and " all those who go clown to the sea in ships " as a fair set off for such 
•savings in fuel as claimed, even though they be substantial ; especially 
•shoidd this consideration weigh heavily in the counsels of citizens of a country 
having an abundant supply of excellent coal but little oil, and that little 
limited to one or two remote districts. To have to import food into a country 
is regrettable, but to import fuel where it can be avoided is folly. If our 
ships are to be largely fitted with internal combustion engines, or have boilers 
capable of burning oil fuel only, its state in case of a war with a country 
having a powerful fleet will be a perilous one, and lead to a worse disastei 
than a temporary shortness of food. 



228 MANUAL OF MARINE ENGINEERING 



CHAPTER X. 

THE CYLINDER AND ITS FITTINGS. 

The Cylinder is the most Important Part of the reciprocating engine, for on; 
it every other part is more or less dependent. The capacity of the engine 
for developing power — that is, for converting the energy of the steam into 
mechanical work — depends on its size and efficiency. The efficiency of the 
cylinder for this purpose will depend largely on the design of the various 
ports, passages, and valves, while its mechanical efficiency will be affected 
by design and the quality of the material and workmanship of the cylinder 
and its parts, inasmuch as the losses in this member of the machine are chiefly 
due to friction aggravated by thermal conditions not present in other portions 
of mechanism. It is, therefore, very highly necessary that the utmost atten- 
tion is bestowed in the first place on all the problems involved in the deter- 
mination of size, the general design of it, and all its working parts, and the 
greatest care in the manufacture, both in the workmanship and the selection 
of materials. Finally, in the working of the engine, judgment as well as care 
is necessary in fitting and adjusting them, and constant attention during 
service, that the highest efficiency may be obtained. 

The Size of the Cylinders can be calculated on first principles in quite a 
simple way, but to determine the best size, or that most suitable for each par- 
ticular case, requires something more ; it requires special consideration of 
all the circumstances involved in the case. In a general way, it may be 
taken for granted that the greatest economy, both in prime cost and in working 
expenses, is to be attained, with the smallest capacity of cylinder in which 
the maximum horse-power guaranteed by the builder can be developed. 
There are, of course, exceptions to this as to other equally useful practical 
rules, such exceptions must, of course, receive exceptional treatment. 

The marine engine differs in many important respects from the land 
engine of similar design ; in one respect the conditions of service are very 
dissimilar, for whereas the land engine usually runs at a fixed rate of revolu- 
tion when employed for continuous work, as in driving mills, dynamos, etc.. 
the marine engine varies in revolution as the speed of the ship changes. 
Further, the load on a land engine may, and often does, vary from nothing 
to the maximum power with only a very slight, and that a temporary, vari- 
ation in rate of revolution ; whereas in the marine engine the gross horse- 
power developed varies roughly as the cube of the rate of revolution — that 
is, if a marine engine is slowed down by throttling or " linking up " to half 
the rate of revolution at full power, the horse-power will be only one-eighth 
of it. Even a reduction of 10 per cent, in speed of ship and revolution 
involves a reduction in power of no less than 27 per cent. Now, as the marine 
engine, even in express steamers, is seldom worked at full power, that 



ECONOMY IN PRACTICE. 229 

developed on service is generally well below the full capacity of the engine ; 
the periods of full power development are few and of short duration, and 
of such a nature is the service that demands them, that the cost of producing 
the power is insignificant compared with that of the general working, so that 
if extravagant by comparison with the ordinary expenditure, it is, never- 
theless, quite warranted. In other words, so long as the increase in power 
is obtainable on demand, the cost of it is of little consideration ; conse- 
quently the efficiency under those conditions may be quite low, if by these 
temporary sacrifices the gain in efficiency under the normal conditions of 
service is thereby attained and substantial. High efficiency during the long 
periods of service is, therefore, the first consideration of the designer, as well 
as that of the purchaser of the engine, and just as electrical engineers on 
shore demand engines which can on an emergency, and for a short time, 
develop something like 20 per cent, more than their normal maximum output, 
and call it " overload," so the marine engineer should be content with such 
overloads on the same terms for trial trips and spurts at sea, instead of 
Tequiring a larger engine, which will really run the greatest part of its life at 
only 75 per cent, of the full power at which it can run, to do that full power at 
maximum efficiency. 

To the Engines of Warships the same argument may be applied in a general 
way, but to them some other considerations are applicable which may modify 
the decision of the designer. Here, in order that the utmost power may be 
obtained under the more restricted conditions of weight and space, there 
may be need of the highest efficiency, in order to get the greatest possible 
output of power from the boilers. But, judging by modern practice in such 
ships, the small cylinders, with the necessary low rate of expansion for them 
at full power, prevails to a greater extent than in even the mercantile marine, 
because a warship seldom runs at even so small a reduction of speed as the 
10 per cent. ; with her a reduction of speed of 40 per cent, is common, and 
that means only about 22 per cent, of the full power is required. 

The Larger the Cylinders are the more costly they must be to manu- 
facture and maintain ; moreover, the clearance losses and those from con- 
densation will be greater in the larger than the smaller cylinders. Further, 
the pipes and connections, the rods, valve gears, columns, shafts, and framing 
vary with the size of the cylinder, many of these parts being influenced by 
the maximum rather than the mean pressure or load on them. It is, of course, 
obvious that for a given power any decrease in cylinder capacity must be 
accompanied by a decrease in rate of expansion, all other things being the 
same. But all other things are not the same ; the steam pressure from the 
boilers and the pressure in the condenser are the same, but the steam efficiency 
in the smaller set of cylinders may be such that the actual mean pressure 
at the same normal rate of expansion under working conditions is considerably 
greater than in the larger ones, due to less wire-drawing, radiation, etc., 
cylinder condensation, and decrease in back pressure due to larger ports, 
etc. 

Economy is found, therefore, in Practice with rates of expansion lower 
than theory indicates, and consequently of late years great increases of boiler 
pressures have not been accompanied by the corresponding increases in rate 
of expansion such as was formerly anticipated. Hence, for the full powers 
.required for trial speed the rate of expansion is expressed by p -5- 15, where 



230 MANUAL OF MARINE ENGINEERING. 

f is the absolute pressure of the steam supplied to the engine. Following. 
this rule — 

195 
Rate of expansion of triple-stage engines steam 180 lbs. = — = 13. 

„ quadruple „ 210 lbs. = ~l = 15. 

ID 

225 lbs. = ~ = 16. 
lo 

On service such ships will work with rates of expansion of 16 in the triples 
and 18 to 20 in quadruples. 

In Naval Service and Express Cross-Channel Steamers the rates of ex- 
pansion are somewhat less, so that their measure is made by dividing the 
absolute initial pressure by 18. For the triple-expansion engines of such 
ships with steam at 200 lbs. gauge, or 215 lbs. absolute — 

215 
Rate of expansion = -r^- = 12. 

lo 

The Service Speed of Short Distance Express Steamers is about 5 per cent- 
under the maximum or trial trip speed, and since the mean pressure referred 
to the L.P. cylinder will vary approximately with the square of the speed,, 
the mean pressure on service will be (0 - 95) 2 , or - 9 of that on trial. Itt 
designing the engine, therefore, it should be borne in mind that it is desirable- 
to have maximum efficiency with a mean pressure 10 per cent, below that 
of the maximum possible from the cylinders, and as a corollary the cylinders 
must be large enough to develop 11 per cent, more power than wben on 
service — that means the "overload" is 11 per cent, in pressure, but it will 
be 16*7 per cent, of the I.H.P. 

Instead of Speed Margin with such Ships, it would be better to have a 
power one, for it is seen that with only 5 per cent, of the former there is 
16'7 of the latter, and, after all, it is more important in a general way to 
have a reserve of power, which will ensure maintaining the service 
speed. 

With Cargo Steamers (here is a Disturbing Element not experienced with 
the warship or passenger steamer — viz., the difference in draft of water 
with different cargoes, and the possible difference in speed of revolution — 
in fact, it is equivalent in effect on the engines of a change in size of propeller. 
The engines can now move at a much higher rate of revolution with a smaller 
mean pressure of steam, and unless checked by throttling or "linking up." 
proceed at a rate far in excess of that anticipated by the designer, and pro- 
vided for by him. Under such conditions the propeller is very liable to 
sudden and severe racing, in spite of governors and hand gove*nin£. Such 
engines must, therefore, for safety sake, have in themselves that which goes 
toward restraint under such conditions. 

The Indicated Horse-power of an engine is the product of the mean pres- 
sure, the piston area in square inches and feet passed through by the piston 
in a minute divided by 33.000. It may be said, therefore, that the power 
depends on the area of the piston on the speed of piston and the pressure on 



DIAMETER OF CYLINDERS OF A MARINE ENGINE. 231 

it. With a given engine the two latter may be varied arbitrarily, inasmuch 
as the pressure may be increased or decreased by the throttle valve at will, 
or may be altered permanently by changing the cut-off of the distributing 
valves on the cylinders ; the speed of piston, all other things remaining the 
same, being dependent on it will vary with the mean pressure. But with 
the same mean pressure the piston speed may be permanently reduced by 
increasing the pitch of the screw. When designing an engine the mean 
pressure aud speed of piston intended must be decided on in order to cal- 
culate the area of piston. As a matter of fact, it is really the rate of revolu- 
tion that must be decided on as a basis of calculation for a reciprocating 
engine, just as it is for a turbine. With a vertical engine under ordinary 
conditions the length of stroke may be generally of any reasonable amount, 
so that the real limit, after all, is the practical one of how fast may a piston 
move with safety and consistent with continuous good working. 

It must not be overlooked that the maximum velocity of movement of a 

piston is ^, or 1*571 times the mean, so that if the mean speed is 1,000 feet 
A 

per minute, the velocity of the piston at about the middle of the stroke is 

at the rate of 1,571 feet. 

The Diameter of Cylinders of a Marine Engine is ascertained by first deter- 
mining that of the low pressure from the referred mean pressure and speed 
of piston decided on, and making the high pressure of such a size as to permit 
of that mean pressure with the boiler pressure provided. The generality 
of engines now in use are of the compound type, triple or quadruple. The 
rate of expansion will be chosen with regard to the conditions under which 
the engine has mostly to work. In a general way it is true, as already stated, 
that the smaller the cylinders are the better, both in prime cost and economic 
working. It has been shown that the consumption of steam in Naval reci- 
procating engines at full speed per I.H.P. per hour is somewhat higher than 
at a somewhat reduced one, while, on the other hand, at low speeds, notwith- 
standing the higher rate of expansion obtaining, the consumption is often 
even greater than at full speed. This means that with the comparatively 
small cylinders the economy at an extravagant rate of expansion is better 
than that in the same cylinder at a high and presumably economic rate. 

In the mercantile marine the lowest rate of steam consumption per horse- 
power is also often at a higher rate of expansion than that at the maximum 
power, but not much more so. With the cargo steamer, where economy 
of fuel is of prime importance, and the engines are run nearly always at full 
speed, the consumption of steam per I.H.P. is a minimum, or nearly so at 
that speed. Mr. Parsons found that the steam consumption of the triple- 
expansion engines of the s.s. "Vespasian" at 70 revolutions was 16*9 lbs. 
per horse-power hour, at 65 revolutions it was 17 '6 lbs., and from that the 
rate gradually rose as the revolutions fell, till at 50 revolutions it was as much 
as 19 - 8. At 65 revolutions the rise was small, as it was only 17 '6 lbs., and, 
assuming the same rate of slip, the following comparative results are true. 
The total consumption of water at 70 revolutions was 17,250 lbs. per hour, 
while at 65 revolutions it was 14,200 lbs. For the same distance, therefore, 
if at 70 revolutions the consumption is 17,250, while when done at 65 revolu- 
tions it will be only 15,140 — that is, as 1*14 to 1*0 — or a saving of 14 per cent, 
in spite of an increase per I.H.P. of 4*14 per cent. 



232 



MANUAL OF MARINE ENGINEERING. 



Results op Trials of Various Ships, showing Consumption of Steam 
and Fuel in Lbs. per H.P.-Hour. 



H.M.S. Hn , 
Max. I.H P.. , 18,500 I.H.P. 
Engines Recipros. 


H.M.S. Cn., 
23.500 I.H.P. 
Recip. Trpls, 


H.M.S. Tz., 

9,800 I.H.P. 
Reeip. Trpls. 


U.S A. Dlw., 
28,500 I.H.P. 
Recip. Trpls. 


H.M.S. As., H M.S. Hlr 
24,000 I.H.P. 10,400 I.H.P. 
Recip. Trpls. Recip. Trpls. 


I.J.N ilk.. 

27.000 S. If. P. 
Curtis Tribs. 




Steam. 


Coal. 


Steam. 


Coal. 


Steam. 


Coal. 


Steam. 


Coal. 


Steam. 


Coal. Steam. 

1 


Coal. 


Steam 


Oil. 


Full power, ' 17-2 
50 to 75 %. 15-2 
20 to 10%, 15-1 

1 


1-80 
1-76 
1-94 


16-0 
14-4 
15-3 


1-99 
2-04 
2-00 


13-91 
15-45 
16-25 


2-65 

2-28 
2-06 


• 
21-0 
18-7 
22-0 


... 


• 
19-9 
15-37 
16-95 


2-03 I .. 

1-85 

1-88 


1-407 

1-49 

1-64 


13-77 

14-76 
20-57 


1-10 

1-27 
1-80 



* This consumption was for all purposes. 



Max. S.H.P., 

Engines, 


R.M.S. La., 
64.600 S.H.P. 
Direct Turbs. 


U.S.A. Nvd. 
23,3Q0 S.H.P. 
Geared Turbs. 


U.S.A. Fda. 
40,500 S.H.P. 
Direct Turbs. 


U.S.A. Pkn., 
11,700 S.H.P. 
Curtis Turbs. 


U S.A. Esn., 
17,800 S.H.P. 
Geared Turbs. 


\ 

R.M.S. Ok., i U.S.A. Clfa. 

6,900 S.H.P. 29,000 S.H.P. 

Mixed R. & T. 1 Electric Turb. 




Steam. 


Coal. 


Strain. 


Oil. 


Steam. 


Coal. 


Steam. Oil. Steam. 


Oil. 


Steam. 


Coal. 


Steam. 


Oil. 


Full power, 
50 to 75 °/ , 
10 to 20%, 


12-77 
13-90 
21-23 


1-43 
1-56 
2-52 


15-40 
16-16 
22-95 


1-32 
1-30 
1-72 


13-40 
23-77 


1-59 

2-23 


14-49 
15-50 
21-49 


1-42 
1-40 
1-67 


15-76 
17-40 
28-75 


1-23 
1-22 
2-03 


14-12 
14-10 
15-20 


• • 


11-90 
11-10 
14-60 





It is to be observed that, whereas a marine engine slows down as the 
rate of expansion increases, a similar engine on shore driving a mill or dynamo 
will run at constant speed. This is important, because economy of steam 
consumption is very much influenced by the rate of flow through the cylinders, 
just as it is in the turbine, the low consumption of which is in no small measure 
due to the rapidity with which the steam passes from entrance to the con- 
denser. In a reciprocating engine, when at 150 revolutions, and of the triple- 
expansion type, the steam takes 0-8 second to pass through the system, 
but at 75 revolutions the time is 1-6 seconds. 

Mean pressure is modified in quite a satisfactory way by linking up ; 
for example, the s.s. " Al." was intended to work at sea with about 1.000 
H.P. ; on trial trip at full speed 1,232 I.H.P. was developed. The following 
shows how this reduction was effected by simply notching up to the extent 
of two turns of the hand wheel, and what were the consequences : — 

TABLE XXX.— Trials of a Typical Cargo Steamer. 





Steam. 


Vacuum. 


Revs. 


Mean Pressures. 


Indicated Horse-Power. 


Total. 


High. 


Medium. 


Low. 


High. 


Medium. 


Low. 


Trial Runs, 

full, 
Sea spocd, 

linked, . 


145 
145 


25 
25 5 


64-2 
581 


50-4 
452 


24 51 

22-9 


7 91 

7 20 


3857 
3134 


434 5 
367 2 


4121 
339 9 


1 : 232 3 
1,020 5 



The coal was weighed during these trials, and the consumption was found 
to be at the rate of 1*51 lbs. per horse-power at full, and 1*55 lbs. at 
the reduced speed. Now, had smaller cylinders been provided, and the 



DIAMETER OF THE L.P. CYLINDER. 233 

revolutions raised from 58 to 64, the rate of consumption would not have 
exceeded the 1'51 lbs., but probably have been less. 

The Back Pressure in the L.P. Cylinder should be not more than 15 lbs. 
greater than that in the condenser — that is, with a vacuum of 28 inches it 
should not exceed 2"5 lbs. absolute. In a well-designed low-pressure cylinder 
at moderate piston speeds the vacuum in it at exhaust should be 95 per cent, 
of that in condenser, when that is not more than 28 inches ; 93 per cent, is 
common experience with good engines when working with a vacuum of 
27*0 inches at full speed. For purposes of calculation, therefore, an allow- 
ance of 90 per cent, will be quite on the safe side. With modern air pumps 
and a good condenser 28 inches can be maintained even in the tropics ; 90 per 
cent, of this is 25*2 — that is, the back pressure is about 2i lbs. If the ship 
is likely to see much service in the tropics, 27 inches will be safe, and 90 per 
■cent, of it is 24*3 inches, so that the back pressure then will be 2 "85 lbs. With 
a turbine the back pressure in it is very nearly that in the condenser top, as 
there is practically no exhaust pipe, and certainly no valve obstructions. 
For general traders the back pressure should be assumed at 3 lbs., and for 
those other ships casually visiting the tropics 2A lbs. is sufficient allowance 
for calculations. 

Example (1). — To ascertain the probable mean pressure referred to the 
L.P. cylinder of a passenger steamer whose service will be through the tropics, 
the boiler pressure to be 185 lbs. 

Here the initial absolute pressure is 200 lbs. 

The rate of expansion 200 -5- 15 = 13-33. 

Referring to Table xvii. for this rate of expansion, the ratio — = 0*269. 
Back pressure is 3. Pi 

Then theoretical mean pressure = (200 X 0-269 — 3) = 50*8 lbs. 

The actual mean pressure will be found by multiplying this amount by 
the factor given for such engines on (Table xviii., p. 125), say, 0*7. Thus : — 

Probable mean pressure = 50 - 8 x - 7, or 35'56 lbs. 

The Diameter of the L.P. Cylinder may be found* then as follows :— When 
p m is the referred mean pressure. Let d be the diameter in inches, R the 
number of revolutions, and S the stroke in feet. 

T , , ■ . d 2 I.H.P. x 33,000 

Then area of piston = «- = ^ x r x 2S • 

flufe-Diameter L.P. cylinder = / I-H.P. X 21,000 = 7 I.H.P. X ,42,q gT 

V pm X S X R y p m X piston speed 

If the stroke is not decided, S X R is half the mean piston speed. 

The volume of steam at cut-off for a rate of expansion E must be = capacity 
•of L.P. cylinder -•- E. 

The equivalent cut-off in the H.P. cylinder for this will be expressed 
as a fraction of the stroke = ratio of cylinders -=- E. 

If the cut-ofi is to be early, as it should be, to avoid excessive " drop/ 

• The X.E. Coast Inst. E. aiul S. Rule for this in cargo ships is : — 

;j HP v -'l-a 
D — \/ * * ; the I.H.P. is that developed while on voyage, 

b + 4 



234 MANUAL OF MARINE ENGINEERING. 

it will be seen that the ratio of cylinders must be smaller than if a later cut- 
off were admissible or desirable. 

Then, if x be the cut off as a fraction of the stroke, and clearance is 
neglected 

Then the ratio of L.P. to H.P. cylinder = E X x, 
or x = cylinder ratio -e- E. 

Taking the cut-off at 0*55 of the stroke, the ratio of L.P. to H.P. will be 

0'55 X rate of expansion, and if that is taken as absolute pressure -=-13 for 

economic engines. 

^ , -^ . , TT . TTT , ,-. - 55 X absolute working pressure 
Rule. — Katio of L.P. to H.P. cylinders = — — „ 

absolute working pressure 
° r = 23^6 ' 

Example. — To find the sizes of cylinders for a ship whose engines are to 
develop 3,600 I. H.P. on the conditions named before, with the piston speed 
at a mean of 700 feet per minute. 



Here the diameter of cylinder . . = \ - '_ _ „ z^-—- = 78 inches. 

,K -56 x 700 



/3.600 > 42,000 

"V"35-£ 



0'55 X 200 
Ratio of L.P. to H.P. cylinder . = — = 73. 



/78 2 
Diameter of H.P. cylinder . .. = */— = 



15 

28-8. 



The Size of Intermediate Medium-Pressure Cylinder to avoid drop should 
be larger than it would be if determined by such practical considerations 
as low initial loads to avoid shock, variation in temperature, etc. The true 
mean between the L.P. and H.P. cylinder would be as follows : — Where d 
is the diameter of the H.P., and D that of the L.P. cylinder. 



/ jo TY2 

Diameter of M.P. cvlinder 



=V" 



This, however, is too large, and not in accordance with practice ; the 
following rule may be followed, therefore, with triples : — 

Rule. — Diameter of medium-pressure cylinder = a/ — , 

or, Simply „ „ = {d + D) x 0*45. 

For the engine in the above Example, 



Diameter medium cylinder =a/ — =48*1. 



/28 



For such diameters of cylinder in an ordinary passenger steamer the stroke 
would be 48 inches. Then the engines would have cylinders 28|, 48. and 
78 inches diameter, and 4 feet stroke, the revolutions are then 700 -=- 8, or 
87*5 per minute. 



ARRANGEMENTS OF CYLINDERS. 235 

The Arrangement of the Cylinders of a marine engine is now a much more 
important, as also a more interesting, problem than was the case formerly/ 
when there were only two of them. To-day, with the multiplicity of cylinders 
and cranks that are commonly found, especially in oil internat combustion 
engines, the arrangement is governed by circumstances quite beyond the 
ken of the older engine builder, who was never troubled with problems 
involving the nice balancing of the moving parts so as to avoid vibration, 
or to study the economy of steam piping involved in determining the sequence 
of cylinders so as to get the best flow of steam from boiler to condenser with 
the least expenditure on balance weights. 

Now the designer has to face such questions, as well as all that is involved 
in the division of cylinders to bring them within reasonable limits of size 
when the power to be developed is large. Large cylinders can be made as 
well now as formerly, but modern engineers prefer to have two of moderate 
size in the place of the one of 130 or even 140 inches diameter fitted by their 
predecessors, even though the highest pressure in them is only half that 
in these older and larger ones. Moreover, the solid steel pistons of to-day 
are lighter and certainly safer than the old hollow cast-iron ones, and in the 
vertical engines they are so very much easier to examine, overhaul, or remove 
than those of the horizontal engine of old days. Nevertheless, the tendency 
is to split the L.P. member of a compound system into two or more portions, 
for thereby a triple-stage engine may have four cranks, and be balanced on 
the Schlick system, and each L.P. cylinder can have a larger ratio of port 
and steam passage section, whereby the back pressure in it is less than that 
obtaining in the single large cylinder, and thereby benefit by a high vacuum 
in the condenser. 

The Size of the Steam Ports in a Cylinder do not vary in practice with 
the square of the diameter, but at a somewhat less rate, inasmuch as the 
transverse measurement may be in proportion to the diameter, while the 
longitudinal or axial breadth does not increase so rapidly, therefore the 
larger is the cylinder the smaller is the ratio of maximum port to piston 
area ; consequently with the same speed of piston the flow of steam is of 
necessity higher through ports and passages. If, therefore, the number 
of cylinders be multiplied, so as to keep their sizes comparatively smaller, 
the higher will be the steam efficiency. The mechanical efficiency, however, 
will probably be considerably less, on account of the multiplicity of working 
parts, glands, etc. On the whole, however, within reasonable limits and 
with modern workmanship and material, the general efficiency of the engine 
with the larger number of cylinders is higher than any older one with the 
less. 

There are other influences at work to-day, which incline the marine engine 
builder to more subdivision of cylinders. The oil engine, with its very 
high initial pressures and temperatures, requires not only a very strong and 
simple cylinder, but one of quite a limited diameter. Since the single- 
acting two-cycle engine is likely to be the type for competition with the 
turbine and reciprocator on shipboard, it is more than ever sure to have 
cylinders of limited diameter. The Diesel oil engine (fig. 49), using heavy 
oil of sorts, and without special ignitors, is looked to as the one specially 
suitable among internal combustion engines for marine purposes. This 
engine, with its initial pressure of 550 to 650 lbs. per square inch, and liable 




No. 1. 




H.P. 



No. 3. 




No. 2. 



LP. 



M.P. 



H.P. 







No. 4. 



H.P. 




No. 5. 



Fig. 73. —Various Arrangements of the Cylinders of Triple- Expansion Engines. 



ARRANGEMENTS OF CYLINDERS. 237 

to a temperature exceedingly high, requires to be quite moderate in size and 
strong in construction. 

Even the steam engine of to-day is subject to quite high pressures, and 
when the steam is superheated it may have a temperature of 600° F., with 
a pressure of 230 lbs. per square inch, which requires similar care and 
limitations as the oil engine, although in a lesser degree, however, to be 
observed in the design of the H.P. cylinder. Under these circumstances 
lubrication of the internal working parts is at all times uncertain, and never 
to be depended on as being positive, hence the smaller the cylinder, the less 
liability to serious derangement. 

The Arrangements of Cylinders possible for a Triple- and Quadruple-Stage 
Engine may be studied by referring to their diagrams on figs. 73 to 74. The 
early triple-expansion engine of Dr. Kirk had always three cranks, with the 
natural sequence of cylinders (v. fig. 76). The very earliest triple-stage 
engine, however, had only two cranks, with the H.P. and M.P. cylinders in 
tandem over one and the low pressure over the other, as in fig. 73. No. 1. 

Inasmuch as the older compound engines were made with the L.P. rods 
and valve gear much heavier than those of the high pressure, it was found 
better, when tripling such engines, to fit the third or new H.P. cylinder over 
the low pressure (fig. 73, No. 2) ; the engine so treated was not only safer 
but better balanced and easier to start than when the new H.P. was placed 
over the original H.P. cylinder. 

When the larger two-stage compound engines were tripled two new H.P1 
cylinders were fitted one over each of the old ones, as shown in No. 3, then 
the load was evenly divided on the cranks and the combination very quick 
in handling. Some of the old and large two-stagers were tripled by the 
addition of a complete new H.P. engine before the old ones, and coupled to 
them in the best way circumstances permitted. 

The old three-crank two-stage compound engines were tripled by adding 
a new H.P. cylinder tandem to each low pressure, and some new engines of 
large size have been made on this plan. 

No. 1 design of fig. 73a is one rather for treating an existing expansive 
engine having two equal-sized cylinders than to copy for a new engine. If,. 
however, large power were wanted in small floor space, but with unlimited 
height, it is quite a suitable arrangement. The same may be said for No. 2. 
on the same page when larger engines still are wanted on a limited floor 
space. In this case, however, there is a simplicity in design and limitation 
in dimensions that are attractive. In fig. 73a, Ex. 2, there are three cylinders- 
in line of equal diameter ; above them tandem-wise are three other cylinders- 
in line and of equal diameter. Of these the first is the high pressure of the 
system, the next two are the medium ones, and the lower three are the L.P. 
cylinders. The ratio of M.P. to H.P. is thus 2, and as that of the L.P. to the 
H.P. is 6. that of L.P. to M.P. is 3. As a concrete example, the L.P. cylinders, 
may be taken as each 60 inches diameter, the H.P. and two M.P. cylinders 
are therefore = 4x3x 60 2 = 42'4 inches diameter. 

Diagram 3 (fig. 73a) shows the four-crank three-stage compound 
arrangement of cylinders, with the two low pressure of equal weight of moving 
parts at the ends, as is the common practice. Diagram 4 shows a variation, 
with the two L.P. cylinders next one another, with a valve box common 
to the two and their cranks at right angles, so that the flow of steam isi 



238 



MANUAL OF MARINE ENGINEERING. 




M.P. 



M.P. 



H.P. 




No. 1. 



No. 2. 



LiR 



VIP 



HP 



LP 



-^^ 



=#r^*; 




No. 3. 
Four cranks, L. P. rods reduced. 



MP 



L,P. 



L<P 




Y 

s 



^^w 



H\R 




1 



Four cranks, all rods same size. 
Fig. 73a —Various Arrangements of the Cylinders of Triple- Expansion Engines. 



ARRANGEMENTS OF CYLINDERS 



M.P. 



H.P. 





No. 1. 



No 2. 



lSt\M.P 2ndiM.fi 




No. 3. 
Four Sets of Valve-Gear. 



LP 



2ndW.P 



lst\M.P 



H\R 



T^gS^ l | / J i ^^\i^, ^ 7D ,^\ 



3 if 



m 



Q&m 



zSx 



/Vo. 4 
Two Sets of Valve-Gear. 

Fig. 74.— Various Arrangements of the Cylinders of Quadruple-Expansion 
Two-, Three-, and Four- Crank Engines. 



240 



MANUAL OF MARINE ENGINEERING. 



practically continuous through quite short exhaust pipes, and the flat L.P. 
valves are fitted with a balancing frame between them. 

The Arrangement of Cylinders of the Quadruple Engine were originally 
as shown on diagram 1, fig. 74 and fig. 75, in order that a compact engine 
taking little floor space should have an advantage over the three-crank triple. 
Moreover, the four-stage compound system was capable of a good arrange- 
ment of valves when in tandem, as may be seen in fig. 75. Another tandem 
arrangement is shown in diagram 2 suitable for large engines and three- 
cranks. As in the case of the triple-stage six-cylinder engine, there may be 




Fig. 75. — Cylinders of a Two-Crank Quadruple- Expansion Engine. 

with the quadruple-stage only two sizes of cylinder, for here the first of the 
lower three may be the second M.P. cylinder, while the second two ol the 
upper three are the first M.P. cylinders. Then, if the ratio of L.P. to H.P. 
is 8, then the upper ones will be just half the diameter of the lower ; the 
ratio of second M.P. to first M.P. will be 2, and of L.P. to second M.P. cylinder 
also 2. Thus, if the lower cylinders are each 60 inches diameter, the upper 
ones will be 30 inches each ; quite a simple arrangement. 



DIAMETER OF L.P. CYLINDER. 241 

Designs 3 and 4 show the cylinders as arranged for four cranks, the latter 
being the natural sequence of cylinder, while with the former the smaller 
cylinders are outside and the three last in sequence, so that the rockin^ 
action is reduced by the arm of the couple being a minimum. 

Figs. 69 and 70 are examples of multi-cylinder engines operating on 
one crank, the axis of each engine being virtually in the same transverse 
plane as the others. Considerable fore and aft space is saved by such 
a design, and although very much out of balance statically, such engines 
will work satisfactorily, and are quite suitable and useful for tug boats and 
river craft working in smooth water, where space and prime cost are of great 
importance. They are largely employed on the American rivers and harbours, 
where the tonnage question does not affect the designer. 

Fig. 283 is another example of compactness, whereby four cylinders are 
caused to operate on two cranks without being in tandem. In this case, 
however, the engines are balanced both statically and against inertia forces. 

The Ratio of Cylinders in Practice depends somewhat on the service of 
the ship. The cargo boat, with triple three-crank engines and a working 
pressure of 185 lbs. the ratio of L.P. to H.P. cylinder is about 6-5 * while 
with a higher boiler pressure up to 200 lbs. it is 7-0. With express steamers 
and such pressures the ratios will be from 5-5 to 6-0 ; in Naval ships the ratio 
is no more than 6-3 when the working pressure is over 200 lbs., and 7-0 when 
250 lbs. Quadruple engines are limited to service in the mercantile marine 
and with them the cylinder ratios are from 8 to 9, the latter being the rule 
with boiler pressures up to 230 lbs. per square inch. In small high-speed 
craft, as torpedo boats, destroyers, and other similar ships, the ratio with 
180 to 200 was only 4='5 to 5-0. • 

To determine the Ratio of the L.P. to the H.P. cylinder in any compound 
system, the following simple rule holds good : — When p is the absolute pres- 
sure of the steam at H.P. valve-box, ratio of cylinders = -p ~ K, where in 
the mercantile marine K is 27 for cargo steamers, and 34 for express steamers. 
In the naval service K is 35 for cruisers and battleships, and 42 for scouts 
and destroyers. The economy of these vessels at low speeds is quite good, and 
even at high are not so very extravagant considering the low steam efficiency. 

Having determined the Diameter of the L.P. Cylinder necessary for the 
power required from the engine, and deduced from the conditions and cir- 
cumstances of the particular case the diameter of the H.P. and M.P. cylinders, 
the designer can proceed with the dimensions and arrangements of all that 
pertains to them. It is, however, of prime importance that the pipe through 
which the steam is supplied to the engine from the generators, and that 
through which it passes away from it to the condenser are of adequate size, 
and as short as circumstances will permit, for otherwise there may be a drop of 
pressure more than is desirable from the boiler to the valve chest, and, what 
is worse still, a more serious drop from the L.P. cylinder to the condenser. If 
the ratio of L.P. to H.P. cylinder is 7-0, the loss of a pound of pressure at the 
L.P. cylinder will require an increase of 7 lbs. mean pressure at the H.P. 
cylinder to make compensation. Further, if the referred mean pressure can 
be raised from 33 to 34 lbs. by a good condenser and adequate exhaust pipe, 
the gain in mean j)ressure is 3 per cent., and the gain in power somewhat more 
if the engine is quickened by it as it would be. In all reciprocating engines the 

* X.E. Coast Inst. E. and S. have adapted 7-5 in their standard specification for 180 lbs. 

16 



'242 



MANUAL OF MARINE ENGINEERING. 



flow of steam to and from them is intermittent more or less ; the higher the 
rate of revolution the smaller is the variation due to it. With slow-running 




60 

a 

'35 

e 

c3 

& 
X 

"3< 



a 

eS 



05 
Si 

H 

eg 



-8 

s 



o 



I— 



_ 



engines the H.P. valve-box may be with advantage of such a size a* to act 
as a receiver, and so prevent the wire-drawing action at entry ot steam being 



MAIN STEAM PIPE. 243 

excessive ; but there can never be the steady flow that is possible with a turbine, 
whereby there is no check to the stream with its consequent pulsation that 
must detract from the efficiency of the reciprocator. Then there must be 
likewise adequate means for conveying the steam from cylinder to cylinder 
with least loss in the process. There must be, of course, some drop in pres- 
sure at every stage, as without it there could be no flow of steam, but the drop 
should be small, and only a means to an end, and not a cause of loss beyond 
the actual needs. If the flow is practically continuous, the sizes of pipes 
and passages may be quite moderate compared with what they must be 
in a slow-running engine with intermittent demand. This must be remem- 
bered, that at the top of the stroke the acceleration of piston velocity is more 
rapid than at or near the bottom, and that with an early cut-off the velocity 
is less than the mean, so that calculations based on mean piston velocitv 
mil give ample areas of cut-off, etc. 

Since with slide valves at both steam entry and exhaust the orifice is an 
«xpanding and contracting one, while the section of the channels or passage? 
is constant, and considering that clearance space is detrimental to economy, 
the latter need not be of the same area as the ports ; or, what perhaps is the 
safer dictum, the ports should always have a larger area than the passage 
sections (v. fig. 77). Also the passages should be made as short as possible. 
and as free from turns and corners where eddie3 may set up as the general 
design permits ; especially should this be followed with the L.P. cylinder 
where the loss from clearance is irrecoverable. 

Drop Valves have been used successfully with engines of the marine type 
on shore ; these, however, are not required to reverse, and do not as a 
rule work a 24-hour day or a 7-day week ; moreover, a stoppage from any 
cause is not likely to be fatal at any time. If such valves could be trusted 
•on board ship, and arranged for reversal, much of the steam and thermal 
losses of the marine engine would be avoided, and besides, the pretty 
indicator diagrams produced be without the drawbacks, for which, from the 
marine engineer's point of view, they are inadequate compensation. Judging 
by what is occurring in the motor-car world, the drop valve is not the desider- 
atum of the oil engine builder, for one by one they are reverting to slide valves. 

Main Steam Pipe. — The main steam pipe, which supplies a cylinder with 
steam, should be of such a size that the mean velocity of flow through it 
does not exceed 8,000 feet per minute. When this is not exceeded, the loss 
of pressure between the boiler and the valve-chest is very slight indeed. 
If, however, the valve-chest is large, the cut-off in the cylinder is before 
half -stroke and the rate of revolution moderate, the area of transverse section 
of this pipe may be smaller than given by this rule, for as stated the 
piston speed is below the mean velocity at the early part of the stroke, and 
the space in the steam-chest acts as a reservoir for steam, so as to keep up 
a steady supply during admission. If the space is not less than one-half 
the volume swept through by the piston at cut-off, the velocity of steam 
in the pipe may be assumed to be 9,000 for engines of 150 N.H.P. to 250 N.H.P., 
and 10,000 for those above that power ; for smaller engines, owing to the 
comparatively larger resistances of small pipes, it is not advisable to take 
a higher speed than 8,000. On the other hand, if the cut-off is later than 
half stroke, and the valve-box small, the assumed velocity should be at 
least 10 per cent, less than that given above. 



244 



MANUAL OF MARINE ENGINEERING. 





Fig. 77.— H. P. Cylinder (Naval). 



LOW-PRESSURE CYLINDER. 



245 





Fig. 77n.— L.P. Cylinder (Naval) with Triple Steam Parts. 



246 MANUAL OF MARINE ENGINEERING, 

Taking 8,100 feet as the mean velocity, S the mean speed of piston in 
feet per minute, and D the diameter of the cylinder, then, 

«'.' - r ■ ' ■ /D' X S D ,_ 

Diameter of mam steam pipe = W ft ,, , =-^. v». 

Example. — To find the diameter of the main steam pipe to a cylindei 
4o inches diameter and 5 feet stroke, the revolutions at full speed to be 60 "per 
minute. 

Here S = 2 X 5 X 60 = 600, and D = 45 inches. Therefore, 

4.5 

Diameter of main steam pipe = — s/600 = 12*25 ins. 

When the main steam pipe is abnormally long, as is the case with large 
ships with long boiler rooms, the divisor should be 85 to 87. 

Area through Stop and Throttle Valves. — Although the loss of pressure 
at the valve-box is often attributed to want of sectional area in the main 
steam pipe, it is more frequently due to contracted area past these valves. 
The friction through a number of small openings is considerably more than 
through one of an area equal to the collective areas of those openings, especially 
if the sum of the perimeters of the latter largely exceed that of the single 
opening, and the " loss of head " will be large if due allowance is not made. 
For this reason there should be always an excess of area around valves and 
other obstructions to the free passage of steam, and the passages leading to 
and from them should be as easy as possible, so as to avoid violent changes 
both of direction and velocity of flow. 

Steam Ports and Passages. — Since, in most engines, the steam has to exhaust 
through the same ports and passages by which it was admitted, their size 
must be governed by the proper flow of emission rather than of admission. 
The area of section of steam ports should be such that the mean velocity 
of flow at exhaust should not exceed 6,000 feet per minute. The ports should 
be somewhat larger than the section of the passages, especially when certain 
kinds of valves are used, which will be dealt with later on ; as a rule, they 
have nearly the same area as the sectional area of the passages. To avoid 
excessive clearance, however, the capacity of the passages should be as small 
as possible consistent with free flow of steam, and as this depends greatly 
on their sectional area, so that the reduction in capacity can only be attained 
by making them as short as possible. Not only will the evils arising from 
clearance be avoided, but the loss through resistance be very materially 
lessened by shortening the distance between the valve face and cylinder,, 
and their cooling effect on the incoming steam temperature. 

Area of steam ports and of section through passages 

Area of piston X speed of piston 
67000 - 

_ (Diameter of cylinder)' 2 x speed of piston 
' 7,686 

Opening of Port to Steam. — It is advisable so to design the valve, etc.,. 
that the opening for admission of steam to the cylinder is sutticient to avoid 



EXHAUST PASSAGES AND PIPES. 



247 



any serious loss by " wire-drawing ; " but in actual practice, unless special 
gearing is designed so as to give a quick motion to the valve at the instant 
of cut-off, there is very considerable loss of pressure shown on the indicator- 
diagram ; and, what is worse still, from deficient opening, the loss is generally 
not limited to the period near to cut-off, but during the whole time of admission. 
The ordinary valve gears do not give that quick motion, either at opening or 
at cut-off, which is such a desideratum. Separate expansion valves and special 
valve-gearings admit of such a motion, and consequently the opening to 
steam with them may be smaller than when cut-off is effected by the ordinary 
slide-valve and link-motion ; they are, however, the cause of more loss than 
gain, and are now never used. 

Hence, when only common valves and gear are to be used, the area for 
opening to steam when at its greatest should be such that the mean velocity 
of flow does not exceed 10,000 feet per minute. In actual practice the amount 
of opening is often much less than that given by the above rules, but it always 
results in loss of pressure in the cylinder throughout, and excessive " wire- 
drawing " previous to cut-off. and, therefore, should not be less but 
greater. 

Exhaust Passages and Pipes. — The area of section of exhaust passages 
should be such that the mean velocity of steam does not exceed 6,000 feet 
per minute, and if the distance from the cylinder to the condenser is com- 
paratively great, a much larger area is advisable. There should not be a 
greater difference than 1J lbs. between the pressure in the cylinder and that 
in the condenser when exhausting, even with a high vacuum. 







TABLE 


XXXI. 






Piston Speed. 


Diameter 


Diameter* 


Area 


Opening 


Area * 


Feet per 


Main Steam 


Exhaust 


Main Steam 


to Steam 


Exhaust 


Minute. 

• 


-r D. 


-r-D. 


+ A. 


-r A. 


-i- A. 


•200 


0158 


182 


025 


020 


0333 


250 


0-177 


204 


0313 


0-025 


0417 


300 


0194 


0-223 


0-0375 


0-030 


0500 


350 


0-209 


241 


0437 


0-035 


0583 


400 


0-224 


0-258 


0500 


Q-040 


0667 


450 


237 


0274 


00562 


045 


0750 


500 


0-250 


288 


0625 


050 


0833 


550 


262 


0-302 


0-0687 


055 


00917 


600 


0-274 


0316 


00750 


0-060 


1000 


650 


0285 


329 


00812 


065 


01083 


700 


296 


0-341 


0875 


0-070 


01167 


750 


306 


353 


0937 


0-075 


01250 


800 


0-316 


0-365 


o-iooo 


0-080 


01333 


850 


326 


376 


0-1062 


085 


01417 


900 


0-335 


387 


01125 


0-090 


0-1500 


950 


0-344 


397 


01187 


095 


1583 


1000 


353 


0-400 


01250 


o-ioo 


01667 



The exhaust passages from the high-pressure cyliuder of a compound 
engine to the next cylinder should be such that the flow of steam does not 
exceed 4,500 feet per minute, in order that the difference between the pressure 



* N.E. Coast Tnst, E 
3.900 feet, and from L.P. 



and S. recommend for exhaust from H.P. cylinder 3.600 feet, from M.P. cylinder 
cylinder 4.500 feet, and difference iu pressure 1} lbs. at working sea speeds. 



'248 



MANUAL OF MARINE ENGINEERING. 



during admission in the medium-pressure cylinder and exhaust in the high- 
pressure cylinder may not be excessive, and also from the medium-pressure 
to the low-pressure cylinder the nominal rate of exhaust should not exceed 
4,500 feet. The pressure in the receivers is not sensibly constant, as it is in 
the condenser, being subject to sudden fluctuation when the high-pressure 
valve opens to exhaust and the medium-pressure valve opens to lead, and 
so on. 

Table xxxi. gives the relation between the various passages, etc., 
and the piston in accordance with the foregoing rules, and is based on the 
assumption of a mean velocity of flow of 8,000 feet per minute for the main 
steam pipe, 10,000 for opening to steam, and 6,000 for exhaust ; A is the 
area of piston, D its diameter, and ratio of D to length not more than 60. 

In naval and other ships, the engines of which are only occasionally, and 
for short periods, run at full speed, so that high efficiency at those times is 
not of first consideration, the exhaust-pipe to the condenser may be of such 
a size when short that the flow is 7,300 feet in large engines and 6,500 feet 
in smaller ones ; the exhaust-pipes from cylinder to cylinder may be corre- 
spondingly reduced, as may also the ports, etc. 

TABLE XXXII. — Weight of Steam in Lbs. at 100 Lbs. Pressure 
Delivered per Minute through Pipes with a Drop of 

1 Lb. only. 



t 








By E. 


C. Sickles. 










Diameter of Bore. 






Length of Pipe. 




i 


















26 Ft. 


50 Ft. 


76 Ft. 


100 Ft. 


125 Ft. 


150 Ft. 


its n. 


1 inch, 


4-40 


310 


2-54 


2-20 


1-96 


1-79 


1-60 


1J .. 






9-TG 


6-90 


5 63 


4-88 


4-36 


3*93 


3-6I) 


n » 






13 04 


9-29 


7 69 


6-52 


5-83 


5-33 


4-95 


2 „ 






30 54 


21-60 


17 60 


15-20 


13-60 


12 50 


11 -5s 


2i „ 






50-80 


35-90 


29-30 


25-40 


22-70 


20-80 


19-2,i 


3 „ 






... 


65-17 


... 


46 00 


• > • 


.* . 


35-00 


Si „ 






• •• 


98-30 


... 


69-50 


• •• 




., 


52-70 


4 „ 






... 


1381 


*•• 


97 60 


• •• 




, , 


73-80 


44 .. 






... 


187 9 


• •• 


132-9 


• •• 




# , 


102 


5 „ 






... 


255-6 


• • • 


180-7 


• •• 




,, 


136-8 


6 „ 






... 


419-4 


• •• 


296 5 


• •• 




.. 


229 1 


7 „ 






■ • • 


618-8 


• •• 


437 


• •• 




.. 


330-0 


8 „ 






... 


890-0 


• •• 


629 


• •• 




, , 


472-0 


9 „ 






• •• 


1206 


• •• 


5530 


• •• 




, , 


644 


10 „ 






... 


1592 


• • • 


1126 


... 




, . 


851-0 


11 „ 






... 


2046 


... 


1447 


• •• 




. , 


1093 


12 „ 


... 


2575 


... 


1887 


... 






1428 



Cylinder Liner. — In order that a suitable material may be supplied to 
resist the rubbing action of the piston without wearing away, and one that 
shall be capable of taking and retaining a polished surface, so as to minimise 
the friction of the piston, and which, when worn, may be easily and cheaply 
renewed, an inner bush or false barrel is fitted, usually called the cylinder 



CYLINDER LINER. "249 

liner. This liner should be made of a hard, close-grained metal having con- 
siderable strength, but not so hard as to resist the action of a cutting tool or 
file, and capable of taking and keeping a polish when rubbed by the piston- 
rings lubricated with soft water ; it should also be such that the expansion 
caused by heat is practically the same as the cast iron of which the cylinder 
itself is made. It is usual to make these liners of cast iron, strengthened, 
closed, and hardened by mixing with it certain kinds of pig iron, or by the 
addition of a small quantity of steel {vide Chap. xxx.). The Admiralty 
prefer the liners to be made of steel, hammered or rolled to a proper size 
for machining ; and some engineers use cast steel. Although the steel gives 
good results, it can be equalled by the specially-made cast iron, so far as 
good wearing is concerned, but, of course, steel exceeds cast iron in tensile 
strength ; this latter quality was necessary to a higher degree for the hori- 
zontal engine than for the vertical engine ; the Admiralty were justified in 
going to the expense of the steel, however, as it enable them to fit much 
lighter liners than would be admissible if made of cast iron. In the merchant 
service, with the vertical engine the cast-iron liner does exceedingly well, 
and it is not likely to be superseded by steel, even if this material can be 
manufactured much cheaper than at present. Liners are usually made with 
an inside flange at the bottom end (tig. 78), which tits into a recess in the 
cylinder end, and is secured there by screw-bolts. The upper end is turned 
for a few inches, so as to fit tightly into the cylinder shell at that part. The 
joint at the cylinder bottom is made with red-lead paint, while leakage 
between the liner and the cylinder shell is prevented at the other end by 
stuffing a few rounds of asbestos rope or Tuck's packing into a recess formed 
for that purpose, and preventing it from coming out by securing a flat wrought- 
iron ring to the liner so as to cover the packing. Sometimes in lieu of a 
stuffing-box, the outer edge of the liner and the edge of the turned part of 
the cylinder shell are champhered so as to form a groove ; into this groove 
a turn of Tuck's packing or asbestos rope is pressed with a ring as before. 
Some engineers, preferring to rely on metallic contact, turn a slight recess 
instead of champhering the edge of the liner, and caulk into it a ring of soft 
copper. The Admiralty method of making a steam-tight joint at the outer 
end of the liner is by means of a flat copper ring covering the joint and secured 
to the liner, as also to the cylinder by means of iron rings and screw bolts. 
The copper ring is deeply grooved between these iron rings so as to permit 
of a slight movement of the liner endways with respect to the cylinder {vide 
fig. 78a). The liners are sometimes secured without a flange at the bottom, 
by screwing studs through the cylinder shell and liner, and making the ends 
steam-tight as before. 

The space between the liner and shell should not be less than 1 inch, and 
may be filled with steam so as to prevent condensation in the cylinder, and 
heat the steam during expansion. If the cylinder has to be jacketed, this 
is really a better plan of doing it than by casting the cylinder and inner 
cylinder together, as was very generally done formerly. Independently of 
the advantage derived from the harder metal of which the liner may be 
made, compared with that which is suitable for so intricate a casting as a 
cylinder, there is another very great advantage to the manufacturer. Since 
it is a necessity that the inside walls of the cylinders be sound and free from 
sponginess, as well as blowholes, a casting may be condemned for a defect 



250 



MANUAL OF MARINE ENGINEERING. 



which in no way detracts from its strength or usefulness, excepting that it does 
not admit of the piston working on it steam-tight. If a liner is to be fitted, 
a little sponginess, or even a blowhole in the cylinder casting is of no con- 
sequence, and therefore the extra cost of fitting a liner does not, as a rule, 
exceed the reasonable premium which would be allowed for assuring good 
and sound castings ; and this is especially so in the case of large cylinders. 

False Faces. — For the same reason that liners are fitted to the cylinders, 
the cylinder face, on which the valve slides, needs a false face. This is usually 
made of hard, close-grained cast iron, of the same quality as the liner, and 
secured to the cylinder (fig. 78) by brass screws having cheese heads sunk 





Fig. 78. — Section through Cylinder. Fig. 78a.— Admiralty Method of Fitting Liners. 

in a recess, so as to be considerably below the surface. Care should be taken 
to lock these screws, so that they cannot slack back ; the simplest way of 
doing this is to cut a slight nick in the side of the recess, and caulk or drift 
the metal of the screwhead into it, after the screw is tightened in place. 

False faces were sometimes made of hard gun-metal or phosphor-bronze 
in the engines of warships. The superior strength of these metals over 
cast iron admits of the face being much thinner, but besides being much 
more expensive, there is great risk of damage to the cylinder itself, owing to 
the greater expansion of the bronzes by heat : they never permit of such a 



DOUBLE-PORTED VALVES. 251 

good working surface as does cast iron ; and even if danger from this is slight, 
some difficulty has been experienced in keeping the joint between the two 
metals steam-tight. 

By connecting the recesses for the screwheads with grooves cut in the 
face, the rubbing surfaces are well lubricated, and a considerable amount of 
relief given to the valve itself by the reduction of the effective pressure on it, 
caused by the steam flowing through these grooves, etc. 

The corners of the ports, both in the false face and cylinder face, should 
be well rounded, as the casting is very apt to crack at them if they are sharp. 

The Width of the Steam Ports, in the direction parallel to the cylindei 
bottom, is usually 0*6 to 0*8 of the diameter, but engines of longer stroke 
than usual require a larger proportion than this to obtain the necessary port 
area without having excessive length (measured in direction parallel to the 
axis). It is obvious that, at the cylinder bore, the width of port cannot 
exceed the diameter, and must really be somewhat less ; in actual practice 
it seldom exceeds 0'8 of the diameter ; but at the cylinder face it may, and 
sometimes does, exceed the diameter, the length being such that the area of 
section in the passage is uniform throughout. 

Piston Valves. — If the very broad cylinder face is bent into the form of 
a cylinder, there will be the same area of orifice, while the space occupied 
in direction of the width of the port is less than one-third of that required 
for the flat face. The valve for such a face must be cylindrical, or composed 
of two circular discs or pistons having the same depth of edge as there 
would be of bearing surface at each end of the ordinary slide valve. 
Such a valve (fig. 138) is called a piston valve, and besides possessing the 
advantage of occupying little space, has the more valuable one of being free 
from lateral pressure, requiring no balancing or relief, and moving with the 
least resistance of any slide valve. For these reasons, the piston valve is 
an exceedingly good form when high pressures of steam are used and for very 
large engines. It is a very general thing for engineers to fit a piston valve to 
the high-pressure cylinder of compound engines of all sizes and types ; many 
makers also fit the medium-pressure cylinder with piston valves, and a few 
fit them to all three cylinders, especially of large engines, and of engines 
running at high revolutions, as in torpedo boat destroyers. The earliest 
compound engines of Woolf, made in the early years of the nineteenth century, 
had piston valves. 

" Drop " Valves, with some form of Corliss valve gear, may take the place 
of piston valves in larger engines, at least if superheated steam comes into 
more general use. They are extensively used on the large slow-speed marine 
type of engine on shore, and in the slow revolution paddle steamer in 
America (v. fig. 30, p. 73). 

Double-Ported Valves. — Although there is of necessity only one opening 
of the steam passage into the cylinder, there may be two or more openings 
through the cylinder face into the steam passage. The combined area of these 
openings need not materially exceed that of the section of passage, but it 
is better to make it so ; as each will be open to steam by the same amount 
as the single port, if the valve has tbe same travel, lap, etc., the total opening 
is in this case double that of the single port for a double-ported face, and 
treble for a treble-ported face. When the face is treble-ported, the valve is 
generally arranged so as to admit steam through all three ports, but to exbaust 



252 MANUAL OF MARINE ENGINEERING. 

through two only, as there is seldom any difficulty in getting full opening to 
exhaust. 

Steam Jackets. — It is not necessary here to enter into the question of the 
economy of steam-jacketing. If the economy is doubted, it is, at least, 
.certain that it admits of the cylinders being gently warmed before starting, 
without moving the working parts. It was customary at one time to form a 
jacket around the cylinder by casting it with two thicknesses of metal ; but 
this was often inconvenient, and always risky to the moulder. Now, as a 
loose liner is fitted, the jacket is automatically provided. When the engines 
are of short stroke, the ends present almost as large a surface to the steam 
as do the walls, and should be steam-heated if the jacketing is to be effective. 
For strength of structure, too, all large cylinders should have hollow bottoms 
and covers, as stiffening them with ordinary webs is not always sufficient, 
.and is in some cases of iron castings even a source of danger. If the pressure 
is on the same side as the webs, they are safe and do add to the strength 
of structure ; if on the opposite side, then they are in tension, and their outer 
edge liable to extreme tension ; so that if there be a nick or other such defect 
from which to start a crack, or if subject to a sudden application of the stress, 
the outer edge is apt to crack, which will develop and spread into the main 
body of metal, finally causing serious damage. This arises from the fact 
of cast iron possessing so low a power of resisting stress in tension, compared 
with its power against compression. Care should be taken to thoroughly 
drain the steam jackets, and to this end no webs should so be placed as to 
stop the flow of water to the drain-cocks. Steam, when supplied to the 
jackets of the low-pressure cylinder, should not much exceed the pressure 
that is in the receiver ; for this purpose, a reducing valve is fitted between 
the boiler supply pipes and the jacket. 

Boring Holes. — The diameter of the boring hole depends generally on the 
size of the boring bar employed, and should not be less than one-fifth the 
diameter of the cylinder. When there is a single piston-rod, the stuffing-box 
is formed in the cover of the boring hole. When there are two or more 
piston-rods, the boring hole should be large enough to admit a man ; and, 
therefore, not less than 14 inches diameter, and, when possible, 16 inches 
•diameter. The doors are sometimes made with the flange fitting into a 
recess inside the cylinder, so that the piston-rod, having a butt end, may be 
drawn from the cylinder with the piston ; when this is required, the boring 
hole must be of sufficient diameter to admit of the piston-rod end drawing 
through it. 

Auxiliary Valves. — To render engines handy, so that they may be started 
from any position of the cranks, it is necessary to arrange the gear so that 
the main valves may be operated by hand, or else to fit smaller valves, which 
may be readily worked by the engineer ; these are called auxiliary valves, 
and should have a port area equal to - 002 the area of the piston. It is 
usual to fit such valves to the low-pressure cylinder of compound engines. 
Pass cocks, by which steam can be admitted to the valve boxes of inter 
mediate cylinders, should be fitted, and by their means the use of the L.P. 
auxiliary valve is often avoided. 

These auxiliary valves are usually only flat plates, without even an exhaust 
cavity ; but for very large engines they should be piston valves, or some 
form of balanced valve. 



COLUMN FACINGS AND FEET. . " 253 

Escape or Relief Valves. — These are simply spring-loaded safety valves, 
to allow of the escape of water caused by priming or condensation when the 
piston drives it to one end of the cylinder. They are fitted to each end of 
the high-pressure cylinders of all fast-running marine engines. It is suffi- 
cient in vertical engines if there is only an escape valve at the bottom of the 
medium-pressure and low-pressure cylinders. The diameter of these valves 
should be one-fifteenth the diameter of the L.P. cylinder of a compound 
engine. In the Navy it is usual for all large cylinders to have a fair of escape 
valves at each end. 

Drain-Cocks. — These should be placed wherever any water is likely to 
accumulate in the cylinder and casings, and their size 0*023 the diameter of 
cylinder + h mcn - They should be connected to a pipe leading into the 
condenser bottom ; for if led to the bilge the engine-room is filled with steam 
when open, and the receiver and low-pressure cylinder will seldom drain — in 
fact, during the greater part of the stroke, instead of letting water out, they 
let air into the low-pressure cylinder and spoil the vacuum. Care should be 
taken when the drain-pipe is led to the condenser that the water, etc., does 
not impinge on the tubes, or even on the condenser sides, so as to do serious 
damage, and a non-return valve fitted to prevent water getting back. 

Receiver Space. — The space between the valve of the high-pressure 
cylinder and that of the medium-pressure cylinder, and that between the 
valves of the medium-pressure and the low-pressure cylinders, should be 
from 0'6 to l'l times the capacity of the exhausting cylinder, when the 
cranks are set at an angle of 120°. When the cranks are opposite or nearly 
so, this space may be very much reduced. The pressure in the medium- 
pressure receiver should never exceed 0*7 the boiler pressure, and is gener- 
ally much lower than this. It is usual to fit a safety valve to the low-pressure 
receiver, loaded by weight or spring to a pressure of 20 to 30 lbs. per square 
inch ; otherwise, owing to the large flat sides between the two cylinders, and 
in the valve-box when a flat valve is employed, great risk of explosion would 
be run. This safety valve is usually of the same size and design as the 
cylinder escape valves. The receivers of three-crank compound engines need 
not be so large as the above, as the cranks are usually at angles of 120° ; 
in the case of triple-compound engines with the medium-pressure leading 
the high-pressure, a smaller receiver will do. 

Column Facings and Feet. — It was very usual at one time to form merely 
facings for the jointing of the cylinder to the frames and columns ; but as 
this necessitated the use of studs, or else driven bolts with the heads inside 
the cylinders, it is now abandoned, distinct projections or feet being cast to 
the cylinder bottom, having flanges corresponding to those on the columns 
or frames, so that they may be connected by driven bolts, which are always 
accessible. The only objections to this method are, that it is more expensive 
to mould, and a certain amount of risk is run of getting the casting sound 
and strong where the feet meet the main casting. The former should be 
disregarded in considering so important a part as the cylinder, and the latter 
is always avoided by a good moulder. 

Great care should be exercised in designing these column feet, for through 
them the load due to the steam on the cover (which is greater than that 
on the piston) is transmitted ; and as the load is a recurrent one and always 
applied suddenly, verv ample section of metal should be provided to sustain 



254 MANUAL OF MARINE ENGINEERING. 

it. The area of section through these feet should be such that the stress 
does not exceed 600 lbs. per square inch — that is. the area in square inches 
is not less than that given by dividing the maximum load on the cover in 
pounds by 600. The webs from the flanges of the feet should be well spread 
over the cylinder bottom and towards the sides, so as to distribute the strain. 

Holding-down Bolts. — The bolts connecting the cylinder to the columns 
or frames should be such that the stress on them does not exceed 4,000 lbs. 
per square inch, taking the section at the bottom of the thread, and when 
there is a large number of comparatively small size it shoidd not exceed 
3,000 lbs. per square inch (v. Chap, xxviii.). 

Horizontal Cylinders. — In addition to the facings or feet for connecting 
to frames, additional feet were necessary for the cylinders of horizontal 
engines to rest on and be secured to the engine bed. These feet, too, had 
webs so arranged as to distribute the strain caused by the reaction from the 
weight of the cylinder, pistons, etc. The front part of the cylinder was 
rigidly bolted down, while the back end, especially of long cylinders, was 
held down only, and free to move horizontally when expanded by the heat. 
But since cast iron will expand only one-tenth of an inch in 8 feet, by an 
increase of 180° Fahr. of temperature, there was seldom need to make any 
special provision, beyond boring the holes for the back bolts rather larger, or 
making them slightly oval in the cylinder feet. 

Oscillating Cylinders. — The chief peculiarity of these cylinders is the 
method of supporting by trunnions, which also serve as steam and exhaust- 
pipes (vide fig. 27). Half the load on the piston is taken on each trunnion; 
and .since they are of such ample diameter, it is sufficient to assume that the 
metal is subject only to shearing stresses, and therefore the area of section 
should be such that the stress does not exceed 950 lbs. per square inch. The 
diameter of the trunnions is governed by the size of the exhaust-pipe, since 
the steam must exhaust through one of them, and it is usual and convenient 
to make them both of the same size. In the case of a compound oscillating 
engine, the trunnions of both cylinders should be of the same size, which 
will depend on the size of exhaust of the low-pressure cylinder. The trunnions 
of the high-pressure cylinder, being so much larger than is necessary to accom- 
modate the steam-pipe, allow of a space between the outer or working part 
and the inner part or stuffing-box, which, if left open to the air, is well venti- 
lated, and so reduces the heat on the bearing due to steam of high temperature. 

The length of the trunnion journal or bearing should be such that the 
pressure per square inch on the area, made by the multiple of its diameter 
and length, does not exceed 350 lbs. ; generally it is from one-third to one- 
half of the diameter. 

The trunnions have interposed between them and the cylinder body a 
belt, which conveys the steam to and from the valve-boxes. This belt should 
be very strong and well ribbed to the body of the cylinder, immediately 
above and below the trunnions, and when the cylinder is fitted with a liner, 
it is better to form the outer shell in the shape of a beer barrel, so that the 
belt projects inside and not outside, as it would be were there no liner ; the 
strain from the trunnions is then at once taken by the cylinder sides without 
the intervention of webs and ribs. 

The cylinder valve faces of oscillating engines should be set so that the edge 
next the steam entrance should be the nearest point to the cylinder — that is, 



CYLINDER COVER STUDS AND BOLTS. 4 255 

the plane of the cylinder face touches a cylinder whose axis coincides with 
the axis of the cylinder-bore at this edge. When this is so, the lead of the 
steamway into the valve-box is short and easy, and the opening into the 
exhaust-belt on the side opposite is large, without causing the valve-spindle 
centre to be unnecessarily far out from the cylinder. 

Cylinder Covers. — Like the cylinder end or bottom, the cover has to be 
strong enough to take the full steam pressure, but as a rule it has no load 
to distribute to any other part. The same remarks as to webs, etc., equally 
apply to the covers, and all above 24 inches diameter for high-pressure 
cylinders, and 40 inches diameter for low-pressure cylinders, should be made 
hollow with two thicknesses of metal. Those of vertical engines are better 
made in that way for all sizes, inasmuch as it is necessary to fill in the spaces 
between the webs, when they are so made, to prevent the lodgment of water, 
etc., and it is usual to add a false cover, either polished or cast with a pattern 
to give a good appearance. This can always be accomplished by casting the 
covers hollow. The cylinder covers of naval engines and for large express 
steamers are made of cast steel, and formed with ribs having bull-nosed 
flanges to strengthen them. In small naval engines, such as those of destroyers, 
the cover?, especially of the medium-pressure and low-pressure cylinders, are 
often made of manganese or other strong bronze, which has an elastic limit 
higher than that of ordinary steel, and permits of them being cast much 
thinner and lighter. Covers, when of steel or bronze, are sometimes cor- 
rugated as well as coned to get the necessary stiffness without the complica- 
tions of ribs. The depth of the cylinder cover at the middle should be about 
one-quarter of the diameter of the piston for pressures of 80 lbs. and upwards ; 
that of the low-pressure cylinder cover of a compound engine should be 
at least - 15 its diameter. Since, however, the size of the piston-rod is the 
best measure of the pressure on the cover, it is better so to design the cover 
that its depth at the middle is not less than T45 times the diameter of the 
piston-rod. The depth of the cover at the edge depends on the steam port ; 
a recess being formed for the steam-way, and the inside of the cover otherwise 
being parallel to the piston. 

It is the custom with some engineers to end the cylinder a little beyond 
the extreme travel of the piston, the steam port-opening being then in the 
same plane with the cylinder flange ; the cover has a large recess in it, 
and its flange so extended as to enclose the port-opening (fig. 79). The 
advantage of this method is the decreased length and weight of cylinder, and 
being able to secure the cover in way of the steam port direct to the 
main casting, instead of to the comparatively weak bridge of metal across the 
port. On the other hand, however, the cover occupies considerably more 
room, and not being of circular form at the flange, cannot be turned so easily. 
For larger engines this plan is a very good one, and may be adopted with 
advantage, but for small ones and those of moderate size it is not always 
so convenient as the older one. 

Cylinder Cover Studs and Bolts should be made of the best steel, and of 
such a size that the stress on them does not exceed 5,000 lbs. per square inch 
of section at the bottom of the thread, as they are subject to sudden and 
intermittent loads and severe and sudden shocks when priming occurs, also 
to considerable wear and tear from the frequent removals of the covers for 
examination of the pistons. In large engines it is usual to fit the cylinder 



256 MANUAL OF MARINE ENGINEERING. 

covers with manholes for purposes of examination ; as for such engines larger 
studs may be fitted, a higher stress is permissible if desired, so that the section 
may be such that with the maximum pressure there is a stress of 5,500 lbs. 
per square inch. From one or two causes the resistance to pressure on the 
cover may not be evenly distributed over the whole of the studs, so that a 
good nominal margin of safety should be allowed in such a very important 
part ; this is especially so when a large number of small studs are fitted ; 
in this case, when of less diameter than l£ inch, the allowance should not 
exceed 4,500 lbs. per square inch ; under f inch no more than 3,500 lbs. 

Cylinder Flanges. — The width of the cylinder flange need not exceed 
three times the diameter of the bolts or studs, but if the former are fitted 
this allowance is not sufficient to give space for the heads. Studs are 
now nearly always fitted to marine cylinders in great measure for this 
reason. 

Clearance of Piston. — If both cylinder end, piston, and cover were accu- 
rately turned, as is very desirable and required by the Admiralty, and the 
brasses did not wear, a very small amount of space would suffice for clearance 
between the piston and cylinder end ; but as it is usual to leave these parts as 
they come from the foundry, and the bearings, however well made, may wear 
in course of time, it is necessary to make due allowance for this. Small 
engines up to 50 N.H.P. require an allowance of f inch at each end for rough- 
ness of castings, and T g- inch for each working joint — that is, for any part 
between the piston and the shaft journals where wear can take place ; engines 
from 50 N.H.P. to 100 N.H.P., ~v mcn au d ;;% mcn for each working part ; 
engines from 100 N.H.P. to 200 N.H.P., \ inch and ^ inch for each working 
part ; engines over 200 N.H.P., f inch and \ inch for each working part. 
Fast running engines should have a larger allowance, or be turned as above 
recommended. This allowance is usually called the stroke clearance. 

For example, take the case of a vertical direct-acting engine of 120 
N.H.P. ; the parts which wear so as to bring the piston nearer to the bottom 
are three — viz., the shaft journals, crank -pin brasses, and piston-rod 
gudgeon brasses, so that the clearance at top wiJl be | inch, and the clearance 
at bottom § inch + 3 X ^, or § inch in all. Here the total clearance in the 
cylinder is 1£ inch. 

Valve-Box Covers. — The covers are usually formed of a flat plate stiffened 
by ribs or webs. So long as the stiffening webs are on the same side as the 
pressure this form is satisfactory, especially when the covers are not very 
large ; but when they are large and it is inconvenient to web them on that 
side they should be made hollow, with the back rounded like a hog-back 
girder. These covers in Naval ships and express steamers are now very 
generally made of cast steel. 

The studs and bolts with which these doors are secured should be arranged 
in accordance with the rules laid down for those of cylinder covers. 

Small Doors and Covers. — As it is essential to examine from time to time 
the internal working parts, and as this examination is more for the sake of 
seeing that everything is in good order, rather than in the expectation of 
having to execute repairs, and generally there is little time at the disposal of 
the engineer for these purposes, small doors should be fitted to the large 
heavy doors, secured with only a few studs, so as to be quickly taken off and 
refitted. These may for this reason, with advantage, be of cast steel or of 



LAGGING AND CLOTHING OF CYLINDERS. 



257 



steel plate pressed to shape. There should be, when possible, a doorway in 
the bottom and cover, and to the valve casing of the low-pressure cylinder, 
large enough to admit a man. To the valve-boxes of all cylinders there 
should be peep-holes, through which to ascertain the leads and cut-off of the 
valves, and to press the flat valves to the cylinder faces should they have 
become blown off. 

Lagging and Clothing of Cylinders. — All hot surfaces, from which loss of 
heat may arise by radiation, should be covered with a non-conducting sub- 
stance. Felt was formerly employed and well suited for this purpose, being 
enclosed in polished teak or mahogany lagging, secured with brass bands 
wherever in view in the engine-room, and by pine lagging or canvas when not 
in view. Sheet-steel is, however, now used as being more enduring than 
wood, and when carefully fitted and highly "barfed " looks as well as polished 
wood, and it lasts much longer. 

Cement of kinds, silicate, cotton, and asbestos are specified by engineers for 
the cylinder covering, but silicate is very objectionable, on the ground that 
the dust coming from it through the lagging, owing to the vibration, etc., is 




Fig. 79. — Shortened Cylinder with Port in Cover. 



very apt to get into the bearings and guides, and cause serious trouble. 
Asbestos fibre and magnesia may, however, be used with advantage, 
especially on the high-pressure cylinders ; a cement made of this fibre, etc., 
has been found very successful ; a cement, consisting largely of carbonate of 
magnesia, has also proved an excellent covering for these and other hot 
surfaces. It is very necessary to cover up all hot surfaces when using steam 
of such high temperatures as are now common, and considerable gain is 
obtained by putting muffles on the cylinder covers and all metallic-exposed 
surfaces. The value of the different non-conducting materials are given in 
Chap. xxvi. 

The following rules are for the scantlings of the cylinder and its 

connectioiis : — 

D is the diameter of the cylinder in inches. 

p the load on the safety-valves in lbs. per square inch. 

p x the absolute pressure of steam at H.P. valve-box. 

f a constant multiplier = thickness of barrel + '25 inch. 

17 



258 



MANUAL OF MARINE ENGINEERING. 



Thickness of metal of cylinder barrel or liner, not to be less than 
p x D -J- 3000 when of cast iron.* 



Thickness of cylinder barrel = 
„ liner 



D 



6000 
= 0-8 x j 



(p + 50) + 0-2. 



For purposes of calculation p may be taken as follows : — 
H.P. cylinder p =» boiler pressure, or p x -15 



M.P. 

M.P. 

M.P. 2 

LP. 

LP. 



>> 

>> 

»> 



triple p = 0-6 
quadruple p = 07 
„ p = 0-45 
„ P = 0-22 
triple p =025 



boiler pressure 



it 



Thickness of liner when of steel 

metal of steam ports 

,, valve-box sides 

,, covers 

cylinder bottom 



5> 



covers 



cylinder flange 

„ cover flange 



»» 



i« 



valve-box flange = 1 "0 
door flange 
face over ports 



false face 



= 065 x /. 

= 06 x /. 

= 0-65 x /. 

= 0-7 x /. 

= 1*1 x /, if single thickness. 

= 0*65 x /, if double ,, 

= l'O x /, if single 

= 06 x /, if double 

*/• 
x/. 

x/. 

x/. 

x/. 

x /, when there is a false face. 

x /, when cast iron. 



= 14 
= 13 



>> 
•i 



0-9 
= 1-2 
= 10 
= 08 



For torpedo boats and gunboats, destroyers, and other such ships where 
extreme lightness is a necessity, and full power is only developed at intervals 
and for a short time, the scantling may be reduced by 25 per cent. 

Pitch of Studs or bolts in cylinder-cover or valve-box door in inches 



should not exceed 



V 



t x 130 



, t being the thickness of the cover or door 

flange in sixteenths of an inch, p, the pressure per square inch in pounds 
on it. 

Flat Surfaces. — All flat surfaces of cast iron should be stiffened by webs, 



< 2 x 50 
V 



These 



or stays of some form, whose pitch should not exceed \J 

webs should be of the same thickness as the flat surface, and their depth at 
least 2-5 times the thickness. 

The L.P. Cylinder Body or Barrel should be stiffened by external flanges or 
webs, at about 12 times the thickness of metal apart ; these webs should 
be 1-5 X / thick, and stand at least 0*75 X / beyond the surface of the 
cylinder. Some engineers, however, prefer to do without these stiffening 
webs, and make the cylinder somewhat thicker instead. The low-pressure 
cylinder, however, differs from the others in size, and in being exposed to 
external pressure in excess of the internal ; therefore it should have webs, 
especially when of large diameter. 

•When made of exceedingly good material, at least twice melted, the thickness may 
be 8 of that given by the above rules. 



STUFFING-BOXES AND GLANDS. 



259 



Stuffing '-Boxes and Glands. — For obvious reasons, it is useless giving 
any definite rules for the sizes of these, as they will differ from different 
circumstances, and many of the parts do not vary when others are varied. 




Fig. 80. - Combination Metallic Packing. 



T 



? 







"7% 
YS////, 



-*\- 



C3 I 




i 



i 



Fig. SOre 



Fig. 806. 



The following Table of sizes gives such as are found in good practice, and 
will be of more use than any abstract rules : — 



260 MANUAL OF MARINE ENGINEERING. 

STUFFING-BOXES, etc. 
TABLE XXXIII.— Stuffing-Boxes for Elastic Packing (Fig. 80a). 



A 


B 


c 


50 lbs. 


D 2 

100 lbs. 


I>, 

150 lbs. 


E 


F 






to 




60 


eft 


ti 






n 


o 


5 

o 


M 

no 


a 


s . 

.— ia 
.at- 


.5 


"8 

5 ei ^ 


c -6 * 




^ 




"o + 










06 d -i_ 


t- o 


u 


•i rt 


*.< 


3"* 


~« 


»-•< 


o 9 ?^ 


55S** 


5 - 


6 
4a 


o a 




°x 


°x 




l<n-« X 


2 *; 


ft) 

a 


2 


So 


5® 


.eo 

4) " 


5* 

a 


2*5 


2 « o 


5 S 


5 


i£ 


a 


q 


o 


o 


o 


03 


a 


Ins. 


Ins. 


Ins. 


Ins. 


Ins. 


Ins. 


Ins. 


Ins. 


Ins. 


i 


§ 


H 


1§ 


If 


If 


1 


3 


2 of f i 


i 


§ 


11 


14 


If 


2 


1 


s 
T3 


2 ., | 


1 


A 


n 


If 


2 


2i 


1 


i 


9 i 


H 


A 


2 


U 


24 


2| 


I 


i 


2 „ 4 ' 


l| 


T 


2| 


11 


2i 


24 


I 


I 


2 „ 4 ' 


18 


A 


2i 


2 


2f 


2| 


1 


i 


2., § 


li 


4 


24 


24 


24 


21 


1 


i 


2 ,, • . 


if 


1 

7 


2| 


2i 


n 


3i 


H 


i 


2 „ | 


2 


• 


34 


24 


34 


34 


U 


I 


2 „ | 


2i 


A 


31 


2| 


34 


3J 


H 


5 


2 „ I 


2i 


§ 


3| 


3 


3| 


4i 


14 


5 


| 2 ., | 


2f 


§ 


4 


3i 


4 


4| 


if 


A 


2 ., 1 
1 3 „ 2 


3 


H 


*S 


3i 


4£ 


5 


IS 




12., 1 
|3., I 


3i 


li 


4| 


31 


4£ 


5| 


IS 


1 


II:: 1 ! i 


»i 


i 


5 


4 


5 


5| 


2 


§ 


12 „ H 
3 ■„ 1 


3| 


I 


5i 


4* 


5£ 


6 


24 


1 


1 2 „ U 
|3 „ 1 


4 


13 

Iff 


5| 


4* 


54 


6! 


2J 


1 


12..H 
(3 ., 1 


H 


i 


6i 


4| 


6 


7 


21 


7 


J3 „ li 


5 


1 • 
Tff 


6| 


5 


64 


71 


2§ 


7 
1» 


/2 „ 11 

13., H 


54 


1 


74 


5| 


7 


84 


21 


A 


4„ 14 


6 


H 


8i 


6i 


71 


9i 


3 


7 
Iff 


4 „ )4 


6J 


IS 


8f 


6| 


8i 


9| 


H 


4 


4 „ 14 


7 


n 


9i 


7 


8| 


104 


34 


4 


4 „ li 


74 


n 


10 


74 


91 


in 


3f 


4 


4 ., li 


8 


if 


10| 


8 


10 


12 


4 


9 

T8 


4 ,. U : 


84 


13 


Hi 


84 


104 


12g 


4i 


9 


4 „ 11 


9 


i} 


12 


81 


n 


13| 


41 


s 

18 


4 ., l! 


94 


14 


12* 


91 


n§ 


144 


4| 


1 


4 „ 1| 


10 


n 


13 


92 


124 


14f 


45 


i 


4 „ 11 



Iii the case of the cylinder, it is usual to make the stufiing-boxes, for 
uniformity's sake, for the low-pressure and medium -pressure piston-rods of 
the same depth as that of the high-pressure rod. The stuffing-boxes of the 
valve-spindles, too, are usually exceptionally deep, on account of their 



STUFFING-BOXES. 261 

liability to leak, and the trouble of packing them. The packing of the 
stuffing-boxes when in the cylinder covers of vertical engines is very liable 
to give trouble with steam of high pressure from the want of moisture ; the 
lubricant affects only the top layers of packing, and keeps them soft, 
while the bottom ones get hard and charred. Metallic packings (v. fig. 80) 
are the best for use with steam of high-pressure, and although some patent 
vegetable, packings work very well, these latter are gradually being superseded 
by the former. The metallic packings are generally arranged in a series of hoops 
of triangular section, the pressure on the rod being caused either by a second 
set of hoops outside the first, causing a wedging action on the gland being 
pressed home, or else by an arrangement of springs or spring clips. The 
newer and better forms are now giving great satisfaction, and have taken the 
place of vegetable and asbestos in the high-pressure cylinder, and the medium- 
pressure cylinder also. Many engineers fit metallic packing in the low-pres- 
sure cylinder glands, others strongly object to do so, much preferring good 
vegetable packing on account of the excessive moisture in this cylinder. 

Metallic packing is, however, that now always used in marine engines 
of 50 N.H.P. and upwards for piston-rods, and generally for all rods exposed 
to steam above 3 inches diameter. 



262 MANUAL OF MARINE ENGINEERING. 



CHAPTER XL 

THE PISTON — PISTON-ROD — CONNECTING ROD. 

The Piston is essentially only a disc, strong and stiff enough structurally to 
withstand the pressure of the steam on it, and fitting steam-tight in the 
cylinder. The piston in this simple form is seen in the Richard's Indicator, 
and is often so fitted to small engines. 

In the early days of steam-engine construction, when there existed no 
machine capable of truly boring out a cylinder, the bore was not perfectly 
cylindrical, nor the sides very smooth, and, consequently, unless some form 
of elastic packing was interposed between the piston and the cylinder sides, 
it could not work steam-tight. It was customary to form the piston with 
a recess on the rim, into which rope or junk was coiled, just as it was usual 
to do with all pumps. This packing could not be examined or renewed 
without drawing the piston from the cylinder, a tedious operation at all 
times ; to remedy this the recess was made without a flange at the top or 
side of the piston next the cylinder cover, but a false flange or loose ring 
was bolted to the piston so as to retain the junk packing in place, and admit 
of its being removed or added to without removing the piston from the 
cylinder. This ring was called the " junk-ring," and retains that name, 
although junk is no longer used to pack pistons. After a few weeks' work 
the cylinder was rubbed smooth and fairly true, when the piston would work 
steam-tight with very little friction, and with moist steam of low-pressure and 
temperature the packing lasted a considerable time. 

A solid piston — that is, one without packing — is really the best for good 
working, so long as it remains steam-tight ; but as there is always some 
slight amount of wear, especially when the cylinder is fresh from the boring 
mill, and leakage past the piston is most serious, more particularly when the 
engine is standing still, it is necessary to have some means of adjustment, 
whereby the piston is maintained a steam-tight fit in the cylinder. 

In lieu of the vegetable packing, which is not admissible with steam of 
high pressure, engineers now fit metallic rings, called " packing-rings," in 
various forms, which are pressed outwards against the side of the cylinder 
by their own elasticity or by springs. These rings are maintained in position 
steam-tight by the junk-ring as of old. 

When these metallic rings are once in place so as to fit closely to the 
cylinder sides, there is no need of further lateral pressure until by wear the 
piston becomes slack, and steam permitted to pass it. However, nearly all 
existing pistons are automatic in this respect, and the consequence is that 
the packing-rings press so tightly on the cylinder sides, that the loss by 
friction seriously impairs fche efficiency of the engine ; and it is only when 
the ring or cylinder is considerably rubbed away, that the piston works with 



PISTON SPRINGS. 263 

ease. From these causes many really good pistons have been condemned 
after having been made to cause serious damage. 

Perhaps the first remove from the primitive piston is to be found in the 
form usually fitted in locomotives, and generally known as Ramsbottom's. 

Ramsbottom's Rings (fig. 81). — The late Mr. Ramsbottom, of the L. & 
N.W.R. Co., was the first to pack pistons by one or more narrow metal rings, 
turned somewhat larger in external diameter than that of the cylinder bore, 
and which, after being cut across so as to be capable of being compressed 
to suit the bore of the cylinder, are fitted into recesses turned in the piston 
edge. The rings fit easily into these recesses, and as they are so placed 
that no two of the joints are in a line, the piston is practically steam-tight, 
and works very well in locomotives and other quick-working engines of 
small size : but for large engines, and engines undergoing the same vicissi- 
tudes as those on shipboard, there is an objection to this form of piston. It 
will be seen that the rings cannot be examined or removed without drawing 
the piston, and that there is no means of preventing steam from passing 
where the spring is cut across, besides which the rubbing surface is very small, 
and the spring is always exertion; its maximum effort. The first of these 
objections is overcome (fig. 81a/ by fitting a junk-ring, having cast with it 
a spigot or ring, which goes down into the recess around the piston for the 
packing-ring, and made steam-tight ; into grooves turned in the outer surface 
of this spigot the Ramsbottom rings are fitted. 

For small engines these rings are made of steel ; for such engines as may 
be standing unused for many days, engineers prefer to fit hard bronze rings. 
For larger engines, where the section may be three-quarters of an inch square 
and upwards, the rings are better of tough and hard cast iron, which works 
very well indeed. 

Common Piston-Rings (figs. 82, 85a, 86, and 87) consist only of a single 
hoop made of very tough, close-grained, cast iron, made on the same principle 
as the Ramsbottom rings, but fitted steam-tight between the piston flange 
and the junk-ring, and free to move laterally. This packing ring is usually 
turned to a diameter about 1 per cent, in excess of that of the cylinder, and 
either cut across diagonally, or formed so that one end has a tongue fitting 
into a recess in the other (fig. 87), a brass cover-piece being fitted behind the 
gap, so as to prevent steam leaking into the space behind the rings. The 
ring is then fitted to the piston flange steam-tight by scraping both surfaces ; 
the ring is raised by interposing very thin pieces of paper between it and the 
flange, and the junk-ring is then fitted steam-tight to the piston and packing- 
ring by scraping, etc. Some makers of pistons pro/ess to turn the piston 
and rings so accurately as to require no scraping, and with the perfection of 
the modern lathe and the high-class tool steel now used this is possible with 
care ; other engineers prefer to grind the rings tight after coming from the 
lathe. In whatever w T ay the object is attained is of small moment compared 
with the necessity of having the ring perfectly steam-tight between the flange 
and junk-ring. 

Piston Springs. — When the piston is of comparatively small diameter, the 
elasticity of the packing-ring itself is sufficient to keep it steam-tight against 
the cylinder sides for a very considerable time after it is fitted ; and even 
larger rings may be made of sufficient strength to do this, but they would 
then be open to the same objection as raised against the Ramsbottom rings. 



264 



MANUAL OF MARINE ENGINEERING. 



The old method of pressing the ring out by means of dished springs or coach- 
springs, as shown in fig. 85a, is now seldom used in new engines ; the objec- 
tions to it are the uneven and unknown pressure exerted, and the reaction 
of the piston itself, from the fact of the springs pressing on it. It was a very 
difficult thing to set every spring so that the pressure on the ring was uniform ; 



Fig. 81. 



Fig. 82. 




Fis. 85a. 




Figs. 81 to 87. — Various Forms of Piston Packings. 

and the range of action of this form of spring is very limited, so that, although 
the ring might be very tight when first fitted, after a few days running it 
might be passing steam. The surface taking the pressure too was small, 
and the springs were apt to bed themselves into the ring, and in doing so 



VARIOUS FORMS OF PISTON PACKINGS. 265 

wear through their curved ends. These defects were partially remedied by 
adding to each one or more subsidiary springs on the principle of coach- 
springs, but that only tended to aggravate the other evil spoken of — viz., 
the reaction of the piston itself. 

When a piston is moving through its course, and guided therein by tbe 
rod at one end and the tail rod (or back guides in case of a horizontal engine) 
at the other, it should be quite free laterally from the packing-ring, which 
may follow its course freely. When the bore of the cylinder is quite true, 
and its axis coincides with the line of motion of the piston centre, it is of no 
consequence if the springs do bear on the piston ; but if the cylinder wears 
somewhat out of truth in either direction, it is important that the spring-ring 
■shall follow the sides of the cylinder freely ; it cannot do this if the springs 
react from the piston body. 

Cameron's Patent. — Fig. 86 shows a piston-ring pressed out with a corru- 
gated ribbon of steel ; the lateral pressure here is obtained by the resistance 
of the spring to being bent into a circle, and by the pressure exerted by the 
corrugations when the ends of the spring are pressed apart. This spring 
exerts an almost uniform lateral pressure on the packing-ring without touching 
the body of the piston, and by making the packing-ring comparatively thin, 
it will adapt itself to the shape of the cylinder when worn. The pressure 
on the ring can also be easily and nicely adjusted by packing pieces between 
the ends of the spring. One great advantage of this spring is that it can be 
fitted to any piston without condemning any of the parts beyond the springs. 

Mather and Piatt's Patent. — It was found that metallic packing-rings not 
•only wore sideways, but also on the edges, so as to become slack between the 
flange and junk-ring ; a very slight amount of play with heavy rings 
causes a very large degree of slackness from the continual concussion on 
change of motion at every stroke. To obviate this the ring was formed with 
inside flanges, as shown in fig. 83, and split into two, a spiral hoop, having 
three or four turns, being coiled inside the rings, whose action is to press the 
packing-rings outward against the cylinder sides, and up and down against 
the flange and junk-rings. This form has been generally very successful, and 
pistons so fitted have worked very well indeed ; but there is the objection 
that no adjustment of the spring is possible, and it is always exerting its 
maximum effort. The chief part of the elasticity of the spring, however, is 
exerted in pressing the rings against the flange and junk-ring, and the friction 
so caused helps to prevent undue pressure, on the cylinder side, so that in 
practice it is not found that there is excessive side pressure when first fitted, 
nor lack of it when the cylinder is worn. These springs are made of steel, or 
very strong cast iron, cut out of a ring of either metal. They are also some- 
times cast to the form required. 

Buckley's Patent consists of two rings, of section as shown in fig. 84 ; a 
spiral coil of steel wire is bent into a circle, and inserted between the two 
packing-rings. Pressure is exerted in the same way as in Cameron's spring, 
and tends to press the packing-rings both outwards and against the flange 
and junk-ring. This form of piston is still often used at the present time, 
and when properly adjusted works very well. The spring, however, is a very 
stiff one, and requires but little end displacement to exert a very considerable 
pressure on the sides of the cylinder. 

Qualter and Hall's Patent. — This piston has two packing-rings of tri- 



266 



MANUAL OF MARINE ENGINEERING. 



angular section, with a third ring inside and between them, as shown in fig. 
85. so that on this inner ring being pressed outwards, it exerts a wedge action 
on the two packing-rings so as to force them against the flange and junk-ring. 
Coach-springs are employed to press the ring outwards, which are each held 
in a brass frame having a tapered piece on the back, which fits into a recess 
in the piston and against a tapered cotter ; this cotter can be pressed down 
by a set screw in the junk-ring, and any required pressure is imparted to the 
ring through the spring in this way. This piston has, therefore, the advan- 
tage of being capable of adjustment without removing the junk-ring, and the 
adjustment can be made to a nicety with very little trouble, and also, in 
case of the engines not being required for some weeks, the pressure on the 
springs may be relieved until required again. 




Fig. 88. — Restrained Packing-rings (Admiralty Plan). 

Rowan's Patent and MacLaine's Patent each consists of two strong rings 
of square section, or of U section, pressed outwards by springs at the cross 
cut, and held against the flange and junk-ring by wave springs fitted in a 
groove between them. This arrangement has been found to work well. 

Restrained Packings. — In the early days of piston-making designers were 
chiefly concerned in providing means for keeping the packing-rings on tbe 
cylinder walls ; as pressures of steam increased they still experienced the 
same anxiety lest there should be loss by leakage. To-day we have less 
anxiety on that score, but a more serious, how to avoid the loss due to the 
rings being pressed unduly hard on the cylinder by the steam now used 



FORGED- AND CAST-STEEL PISTONS. 



267 



getting behind them. We cannot prevent the steam from getting behind 
the rings, so it is necessary to restrain them ; this was first done by securing 
the ends or locking them together by the same device that set them. With 
the very high pressures now employed even that is not sufficient, so it is the 
practice in the Navy to fit solid restraining-rings, as shown in fig. 88, to keep 
the cut rings in place when fitted, as in right-hand illustration of same. 

Body of the Piston. — Pistons of small size are usually made of a single 
thickness of metal without, of course, any stiffening webs or ribs (v. fig. 89). 




Fig. 89.— Forged -steel Pistons. 

Pistons of very considerable size have been made of cast steel in this way ; 
they are in the form of a cone, and shaped to suit the cylinder end (v. fig. 90) ; 
by that means they have the requisite degree of stiffness; originally they 
were made in this form to save weight, being for fast-running horizontal 
engines, but now they are used extensively in all fast-running engines. Cast- 
steel pistons are frequently used in the mercantile marine in engines of all 
sizes ; chiefly, however, when in those running at high speeds, and always in 
those of large size. They may be used with advantage in most engines, and 
are not much more costlv than hollow cast-iron ones, especially when of large 




Fig. 90. — Cast-steel Pistons. 

size. The pistons for very light, fast-running machinery are usually made of 
forged steel, and are very thin at the flange and near it. Solid cast-iron 
coned pistons can be used with advantage. 

Pistons of ordinary marine engines above 20 inches diameter for H.P., 
and 40 inches diam. for L.P. cylinders, are usually made of cast-iron cellular 
— that is, with two thicknesses of metal stiffened or connected by ribs and 
webs, and either by the thickness of metal or by the depth of body made 
strong enough structurally to safely withstand, not only the load due to 



268 



MANUAL OF MARINE ENGINEERING. 



steam pressure exerted on it and transmitted to the rod, but also the shocks 
to which it is liable when priming occurs. 

The piston body also must be so designed that it may be safely cast, for 
in the early days of large pistons it was not at all an uncommon thing for a 
piston to crack in cooling, or do so mysteriously afterwards. For this reason 
any rules must of necessity be empirical which set out the thickness of metal 
of the different parts of the body ; but care must always be exercised by the 
designer that no one part is too small for the stresses to which it is subject. 
For example, there must be sufficient metal in the immediate neighbourhood 
of the piston-rod boss to resist the tendency to force out the centre by shearing 
the metal. Again, the piston may be taken as consisting of a number of sectors, 
and by considering one of such small sectors loaded with the pressure on its area 
at the centre of gravity of its figure, the bending moment at any section may be 
found, and the thickness of metal tried whether it be sufficient for the purpose. 

For the section of an ordinary piston having a single rod, the following 
table gives the multipliers for obtaining the thickness of metal and sizes of 
the different parts. 

Details of Construction of the Ordinary Piston. — Let D be the diameter of 
the piston in inches, p the maximum effective pressure per square inch on it, 
x a constant multiplier, found as follows : — 

D ._ 

X = 50 ( v V + !) ' 

For high-pressure cylinders p may be taken at half the absolute boiler 
pressure ; for medium-pressure cylinders a quarter that pressure ; and for 
low-pressure cylinders half the pressure, divided by ratio of low-pressure to 
high-pressure cylinders. For quadruple engines the value of p in the first 
medium-pressure is 03 the press., and in tb* second is 016 the absolute pressure. 

^Hollow Cast-Iron Pistons. 
The thickness of front of piston near the boss 

>> 5 » >» 5 j rim 

,, back ,, 

,, boss around the rod 

,, flange inside packing-ring 

„ at edge 

,, packing-ring 

,, junk-ring at edge . 

„ ,, inside packing-ring 

,, ,, at bolt holes . 

„ metal around piston edge 

The breadth of packing-ring . 
,, depth of piston at centre 
,, lap of junk-ring on the piston . 
„ space between piston body and packing-rin 
„ diameter of junk-ring bolts 
„ pitch „ „ 

„ number of webs in the piston . 

., thickness ,, .... 



. =0-2 


X X. 


. = 0-17 


X x. 


. = 0-18 


X X. 


. = 0-3 


X x. 


. = 0-23 


X x. 


. = 0-25 


X x. 


. = 0-15 


X x. 


. = 0-23 


X x. 


. = 0-21 


X x. 


. = 0-35 


X x. 


. = 0-25 


X x. 


. = 0-63 


X x. 


. = 1-4 


X x. 


. = 0-45 


X x. 


g=0-3 


X x. 


. = o-i 


X x + 0-25 in 


. = 10 diametets. 


I) + 20 
12 * 


. = 0-18 


X x. 



solid packings. 269 

Solid Pistons. 

Cast Steel. Cast Iron 

Thickness near boss . . . . = - 26 x x = 052 X x. 
„ rim . . . . = 0-15 x x = 022 x x. 

Forged-Steel Pistons. 

Thickness near boss . . . . . . . = 02 x x. 

„ rim = 0*1 x x. 

When made of exceptionally good metal, at least twice melted, the thick- 
nesses of cast-iron pistons may be as much as 20 per cent, less than given by 
the rules ; but, on the other hand, if made of other than really good metal, 
they should be thicker. The piston should be made of good metal always,, 
and for fast-running engines it is better made of steel. The packing-ring 
was sometimes made thicker in the part opposite the cut than given above, 
in order to have sufficient elasticity of itself to press steam-tight against the 
cylinder ; but it is better to let the springs perform their function wholly, 
and leave the ring to act only as the packing. 

Junk-ring Bolts. — When screw-bolts are used to hold the junk-ring in 
place, they are either screwed into a brass nut let into a recess in the side 
of the piston (v. fig. 85a), or else screwed into a brass plug, which has been 
screwed tightly into the piston. The former plan is most general, and ha& 
the advantage that if the bolt thread is torn away the nut can be easily 
replaced, and owing to the length of body the bolts cannot slack themselves 
back. Some engineers, however, prefer to screw the bolts directly into the 
cast iron, making the tapped hole as deep as possible ; and although it may 
be supposed the bolts would set fast by rust, practice has shown that such 
does not take place, nor do the cast-iron threads wear quickly away. 

Studs are often used instead of screw-bolts (v. fig. 88), but although. to> 
some extent, possessing advantages over the latter, they are not so con- 
venient, and have all to be withdrawn when any refitting of the packing 
and junk-rings is necessary. 

Safety-rings and Lock Bolts. — The vibration of the junk-ring has a tend- 
ency to slack back the bolts, and although it is a rare occurrence to find 
such a thing happen in a vertical engine, very serious accidents have fre- 
quently been caused by the junk-ring bolts getting loose in a horizontal 
cylinder. To prevent the possibility of a casualty the piston bolts of all 
engines should have a light wrought-iron ring secured to the junk-ring by 
studs, having square bodies and nuts secured with split-pins ; this ring 
(r. fig. 88) fits close to the heads of the bolts, and prevents them then from 
turning. When studs are used their bodies should be square or heart-shaped 
with a projecting side, the holes in the junk-ring corresponding in size and 
form to them, so that, when on, it prevents them from unscrewing ; the nuts- 
may be prevented from slackening by a ring, or each stud may have a split- 
pin through its end. 

There are some other methods, but none of them are either so efficient or 
so inexpensive as the above. 

Solid Packings. — In order that the weight of the piston of a horizontal 
engine may be taken by the broad packing-ring, instead of by the compara- 
tively narrow flange and junk-ring, it was customary and advisable to fit a. 



270 



MANUAL OF MARINE ENGINEERING. 



cast-iron packing between the body of the piston and the packing-ring for 
about one-third of the circumference in lieu of springs. The pistons of 
■diagonal and oscillating cylinders are also better if fitted in this way, and 




gjCg 





i 









i 


J. 


1 

i 

i 


L 



o 




Fig. 93. 
Piston-rod Ends and Guide Blocks, &c. 

heavy pistons of vertical engines may be also dealt with in this way to provide 
for the heavy rolling of the ship. 



PISTON-ROD. 271 

Piston-rod. — It is usual to have only one rod to each piston of a direct- 
acting engine, but some manufacturers, to suit a particular style of crosshead 
and connecting-rod, fit two. The single rod is preferable from practical 
considerations, even for large engines, because it requires very considerable 
care on the part of the' workman to bore the two holes in the piston, cylinder 
bottom, and crosshead so exactly that the rods will fit into their place without 
adjustment ; the friction of the two stuffing-boxes will be very considerably 
more than that of the one larger one ; the cost of labour will also be nearly 
double that for the single rod, and there are two stuffing-boxes, which require 
packing, and two glands demanding attention, instead of one. 

Return connecting-rod and large steeple engines of necessity required 
two rods to each piston, and Messrs. Humphreys fitted four rods in the case 
of the very large pistons of H.M.S. " Monarch," the better to distribute the 
load over the piston face, and to admit of a better form of crosshead. 

Diameter of Piston-rod. — Since the piston-rod is secured in the piston, 
and usually well guided at the other end, so that it cannot bend without 
meeting with considerable resistance, it may be treated as a strut or column, 
secured at both ends ; but when the outer end fits into a crosshead, which 
would offer little or no resistance to bending, as in a paddle-wheel diagonal 
engine, then the rod must be treated as a column loose at one end and secured 
at the other. 

From Mr. Hodgkinson's experiments, and Mr. L. Gordon's investigations 
the following are the formulae for computing the strength of columns : — 

/ S 

(1) For a column fixed at both ends, P = — ~~w- 

/ S 

(2) For a column loose at both ends, P = '- — „. 

2 / S 

(3) For a column fixed at one end only, P = '■ — p. 

2+5^ 

P is the load, I the length of the column in inches, d the diameter in inches. 
a for solid wrought iron and mild steel fI / OT , and / 36,000 lbs. per square 
inch, S being the area of section of the rod in square inches. Taking this 
value of / in the above formula?, P is the breaking load ; since it is usual to 
have a factor of safety for all important parts of a marine engine, of at 
least 6, the value of / for this reason should not exceed 6,000 lbs. for 
mere strength ; but as the piston-rod is liable to great shock, is always 
working with alternating stresses, and always receives its load suddenly, 
and must be rigid and without quiver, 3,000 lbs. should be taken as the 
value of / to calculate the diameter. These formula?, however, are too com- 
plicated for general use, but the size of a piston-rod may be checked by them 
easily after having been calculated by an empirical formula. 

Since, however, an approximate value may be safely taken for the relation 



272 



MANUAL OF MARINE ENGINEERING. 



between / and d of ordinary marine engines, the formulae may be reduced to 
a verv workable form. Hence, since 



P = — -xp; 



and / S = -r- X /. 



D being the diameter of the piston, and p the effective pressure on it in 
pounds per square inch. 

p.D 2 = /d 2 -l-f JL 



Let 



/ 



-7 be represented by r, 



Then 



taking / = 3,025. 



= D V? 



(1+ar 2 ) 



D 



Rule : — Diameter of piston-rod = rr vp (1 -f- « r 2 ) 



00 



(1) 



>/p 



The following are the values of r : — 

. r = 25. t- \/p 

. r = 30 
. r = 35 -5- 
. r = 40 4- Jp 

. r = 50 -5- -Jp_ 
„ „ medium ,, and compound, r = 37 -s- Vp 

The following rules, however, will give results sufficiently accurate for 

all practical purposes : — 

_ , ~. . . , , diameter of cylinder 
Rule : — Diameter of piston-rod = 



Short stroke direct-acting engines. 
Long 
Very long 

Oscillating engines of short stroke 

long „ 



F 



J¥. 



The following are the values of F 



Naval engines, direct-acting, . 

,, return connecting-rod, 2 rods, 

Mercantile ordinary stroke, direct-acting, . 



long 
medium 



oscillating. 



F = 50 
F = 70 
F = 45 
F = 42 

F = 40 



Note. — Long and very long, as compared with the stroke usual for the 
power of engine or size of cylinder. 

D is the diameter of the low-pressure and d that of the high-pressure 
cylinder of a compound system ; d v rf.>, etc., the diameters of the intermediate 
cylinders ; R is the ratio of low-pressure to high-pr ssure cylinder volumes ; 
/•j the ratio of volumes of the first intermediate and high-pressure cylinders ; 
r 2 that of the second intermediate and high-pressure cylinders ; if the stroke 
be the same in each cylinder 

D 2 dj dj 

d 2 ' Tl ~ d-' % ~ d 2 ' 



PISTON-ROD ENDS. 273 

The maximum pressure repeatedly applied on the pistons when working 
full speed are approximately as given by the following formulas : — Where 
p„ is the absolute pressure at or near the engines, and may be taken as the 
load on safety-valve plus 15 lbs. for purposes of calculation of sizes of piston- 
rods, connecting-rods, columns, etc., and their bolts and fittings. 

Values of p or maximum effective working pressure»in 

High-pressure cylinder of a compound engine, . .' . p a X 0*75 

Low-pressure ,, ,, ,, p a + (R X 1*2) 

High-pressure ,, triple-compound engine, . . p a X - 63 

Medium-pressure ,, ,, ,. . . p a -*- (r x X 1*5) 

Low-pressure ,, ,, ,, . . p a -i- (R X 1 - 5) 

High-pressure ,, quadruple-compound engine, . p a X 0*55 

1st medium-pressure cylinder of a quadruple-compound engine, p a -f- (r : X 1*65) 

2nd ,, ,. ,, ,, „ p a -f- (r. 2 X 1 '55) 

Low-pressure „ „ „ „ p a -r- (R x 1-65) 

The piston-rods in a compound engine are all of one size, except in the 
case of the triple, with two low-pressure cylinders, whose rods may be and 
usually are smaller than those of the high-pressure and medium-pressure 
cylinders. It is at all times desirable that the piston-rod shall move through 
the stuffing-box without vibration, but especially is this so when metallic 
packing is used. It is, therefore, a good thing to have the piston-rods larger, 
rather than smaller, than given by the above rules. 

Piston-rod Ends. — It is absolutely necessary that the rod should fit perfectly 
steam-tight into the piston, and also be of such a taper as not to " draw " 
in the least when subject to shock. If the rod end were made cylindrical or 
1 parallel," as it is technically called, and fitted in to satisfy the above con- 
ditions, it would be very tedious and difficult to get it out again. For this 
reason principally it is usual to turn the part fitting steam-tight into the 
piston " taper " or conical. If the taper is very slight the rod can be easily 
made a tight fit, but unless formed with a shoulder at the end of the taper, 
it would in time become so tightly held by the piston as to withstand all 
attempts at withdrawal ; moreover, there would be at all times a great 
danger of splitting the piston by the wedging action. If the taper extends 
the whole depth of the piston, it should be at the rate of f inch to the foot ; 
that is, the diameter of the rod at back is less than that at the front by one- 
sixteenth of the length of taper. Even with so liberal an allowance as this, 
great difficulty is often experienced in withdrawing the rod after a few months' 
work ; for this reason, and to obtain a larger screwed end, some engineers 
do not extend the taper the full depth of the piston. The most convenient, 
and at the same time reliable, practice is to turn the piston-rod end with a 
shoulder of -3- inch for small engines, and | inch for large ones, make the 
taper 3 inches to the foot (fig. 94) until the section of the rod is three-fourtls 
of that of the body, then turn the remaining part parallel ; the rod should 
then fit into the piston so as to leave £ inch between it and the shoulder 
for large pistons, and y T inch when small. The shoulder prevents the rod 
from splitting the piston, and allows of the rod being turned true after long 
wear without encroaching on the taper. 

18 



274 



MANUAL OF MARINE ENGINEERING. 



It was usual to prolong the piston-rods of vertical engines to admit of 
the " tail" end passing through a stuffing-box in the cylinder-cover, and so 
help to guide the piston, and prevent its unduly wearing the cylinder. Since 
gravity prevents moisture getting to the packing of this stuffing-box, and the 
lubricant applied externally soon gets carried through to the cylinder, some 
trouble is experienced in keeping it steam-tight ; the rolling of the ship also 
causes the piston to exert pressure sideways on the gland and packing, and 
further aggravates the evil : in other words, it is an unsatisfactory guide. 
For these reasons it is preferable to simply fit a brass or white metal bush 
in the cover for the " tail " end to work in, and case it with a dome or sheath 
fitted steam-tight and true on the cover ; a couple of spiral grooves in the 
side of the bush will admit and release the steam. But, on the whole, it 
is very doubtful if these tail-rods are efficacious, and certainly they cannot 
be so beneficial to the good working of the cylinder as a piston with broad 
bearing surfaces. It should be noted that when the packing-ring is pressed 
out by springs acting independently of the body of the piston, it is advisable 
to form the piston with greater depth of flange and junk-ring. 

The piston is secured to the rod by a nut, and the size of the rod at the 




Fig. 94. — Piston-rod Crosshead. 

nut should be such that the stress on the section at the bottom of the thread 
does not exceed 7,000 lbs. The depth of this nut need not exceed the diameter 
which would be determined by allowing this stress. To avoid the large cavity 
which is necessary in the cylinder-cover for the piston-rod nut, some engine 
builders recess it into the piston ; this recess does not materially affect the 
strength of the piston, and the plan may be followed with advantage. Al- 
though piston-rod nuts seldom work loose, and those of vertical engines are 
less liable to this than are others, still as a measure of safety in all cases a 
taper-split pin should be fitted to the rod behind the nut, and in the case of 
large engines it is usual to fit a " lock" plate to the nut itself, or to adopt 
some other means of preventing it from moving at all when at work. 

Cast-steel Piston-rod Crosshead. — Fig. 95 represents the modern form of 
crosshead made of cast steel and designed in such a way as a casting permits 
of. It is, of course, much lighter and cheaper than that of forged material, 
shown in fig. 94, and is especially suited to fast-running engines which require 
a large bearing surface. These crossheads can now be cast quite sound and 
of excellent material so as to be practically as good as a steel forging. 



PISTON-ROD GUIDES. 



275 



The stresses on the various sections are, as a rule, light, as the governin° 
requirements of surface naturally causes it to be much larger than would be 
the case for mere strength. The gudgeons can be cast solid, or lightened 
out as shown. The example shown in fig. 95 has white metal fitted to both 
the head-going and stern-going sides, and is without a loose slipper. Such 
however, can always be fitted to this type of crosshead if desired. 




95. — Cast-steel Crosshead. 

When cast steel is objected to as material for the gudgeons, a composite 
design has been adopted whereby the crosshead is made of hard-wrought 
steel and fitted to the cast-steel slipper, etc., and secured by studs and nuts. 

Piston-rod Guides. — The pressure on the piston is transmitted through 
the piston-rod to the connecting-rod, and the reaction of the latter rod acts 
in the direction of its length ; consequently, when the connecting-rod is not 
in line with the piston-rod, the force of its reaction can be resolved into two 
component forces, one in the direction of the piston-rod, and the other per- 
pendicular to it. This latter force is usually called the " thrust of the 
connecting-rod," and unless specially prevented, would tend to bend the 
piston-rod. To prevent such an occurrence, and to preserve the piston-rod 
in its true course, a guide is provided, and the piston-rod end fitted with 
blocks or slippers to work in it. This thrust varies from at the end of the 
stroke to its maximum point, which is towards the point when the crank is 

B 




Fig. 96. 

rit a right angle to the centre line through the cylinder, and depending on 
the cut-off point; when steam is cut off past half-stroke, then, neglecting 
inertia effects, this is exactly the point of maximum thrust. To determine 
the magnitude of the thrust when P is the total effective load on the piston, 
S the stroke, and L the length of connecting-rod, represented by A B, fig. 96 ; 
completing the parallelogram by the dotted lines A E, BE, A C, then the 



276 MANUAL OF MARINE ENGINEERING. 

reaction, R, of the connecting-rod, represented in direction and magnitude 
by A B, will be resolved into two forces, P and Q, represented by the lines 
E B, A E, both in direction and magnitude. P must equal the load on the 
piston, and Q the thrust on the guides, may be obtained by measuring B C 
on a graphically constructed diagram, or by geometry calculated as- 
follows : — 



A C 2 = A B 2 - B C 2 = L 2 



-(!)' 



Now, 








P 


:Q: 


:AC:B 


Therefore, 












S 


Q = 


p 


BC 


p 






2 




* AC " 








-- /S\2 



-p s 



^/4L 2 - S 2 

V " V2/ 
Or, by Trigonometry, 

Q = R sine B A C, 

p 

P = R cosine B A C, or R = — : ~ . _, = P secant BAG. 

cosine B A C 

Therefore, 

sineBAC ptanBAC 

cos B A C 

The angle B A C is found by knowing its sine to be half the stroke- 
-r length of connecting-rod. 

Example. — To find the thrust taken on the piston-rod guide of an engine 
whose piston load is 100,000 lbs.; the length of stroke is 60 inches, and the 
connecting-rod is 120 inches long. 

Thrust = 100,000 x 6 ° = 25,819 lbs. 

^4 x (120) 2 - 60 2 

Surface of Guide-block. — The area of the guide-block, or slipper-surface 
on which the thrust is taken, when going ahead should be sufficiently large 
to prevent the maximum pressure exceeding 100 lbs. per square inch. When 
the surfaces are kept well lubricated this allowance may be exceeded, but 
the reduction in surface should be effected by making shallow grooves and 
recesses in the face of the slipper, in which *he lubricant can lodge and impart 
itself to the guide as it is carried along. A good method of carrying this 
into effect is to provide a surface calculated on the allowance of 100 lbs. per 
square inch, and by cross planing so as to leave shallow recesses about -^ inch 
deep, reduce the actual surface which touches the guide to about f of the 
original area ; there will then be strips across the slipper 1-^ inches wide, 
with depressions between them f inch wide, filled with grease. 

When, however, the piston speeds exceed 300 feet per minute the area 
of guide-block should be increased to ensure continuous safe running, and 
follow this rule : — 

Working pressure per square inch = 1,760 -=- vS + 100. 

For example, an engine running at a piston speed of 900 feet per minute 



PISTON-ROD CROSSHEADS AND GUDGEONS. 277 

should have guide-blocks such that the maximum pressure per square incb 

does not exceed Jm + & , or o5 lbs. 

Guide-blocks designed on these lines for ahead motion have always ample, 
provision for astern-going. Then in any engine 

r> , r, t A M 1 V. T X J (S + 100) 

itw/e : — Gross area of guide- block shoe = 7ft fT~ sc l' ms ' 

Cast iron, hard and close-grained, is really good material for the guide 
plates ; its surface, after a few days' work, becomes exceedingly hard and 
highly polished, and offers very little resistance to the slipper or guide-block. 
So long as this hard skin remains intact, no trouble will be experienced, but 
if abrasion takes place from heating or other cause, it rarely works well after 
and should b? at once planed afresh, or, better still, ground smooth. 

The slippers or facing plates fitted to the piston-rod or crosshead were 
sometimes made of bronze ; but bronze never gets that smooth hard skin so 
essential to good and efficient working, and when once the surface is grooved 
and scratched, it will wear away very rapidly. Fine grain cast iron is, after 
all, the best metal for this purpose, if care is taken at the first working of the 
engine to run for a few hours at easy speed, so as to rub down and polish 
the surfaces ; after this is once thoroughly done, cast-iron surfaces will 
continue to work well with very slight attention. White metal is, however, 
now generally used for the facing of slippers, and works very well, and for 
high speed is reliable for good and safe working. The best way of using 
white metal for this purpose is to fit strips of this material into grooves planed 
across a cast-iron slipper, and leave them standing from ^ to \ inch above 
the cast iron. The strips should be about 2 inches wide, and the space 
between them from § to \\ inches, into which the lubricant can collect and 
lodge, as before described. A slipper fitted in this way is shown in fig. 92. 
It is cheaper, however, and more general to cast the white metal practically 
over the whole surface of the slipper, where it is retained in place by " tinning " 
and underlying the edges of the recess provided for it. It is usual to cut 
•oil- ways in the face of the guide, which distribute the oil across it, and metal 
combs secured to the slipper dip into the oil receiver at the end of the guide, 
and smear the face on the return stroke. 

The guide plates are sometimes planed across, instead of the slippers, for 
the same purpose of retaining the lubricant, or have circular grooved rings 
dotted about their surface. 

Piston-rod Crossheads and Gudgeons. — When there are two piston-rods, 
as in the case of the return connecting-rod engine, they are united to a 
common crosshead, having a turned journal in the middle for the connecting- 
rod to work on ; or else a bearing is fitted to the crosshead, in which a gudgeon 
on the connecting-rod works, ^he former (fig. 97) is the better and usual 
plan under ordinary circumstances. This crosshead is of wrought-iron or 
■steel, and made of a form suitable to the circumstances, and arranged to 
work in guides. The diameter at the middle must, of course, be sufficient 
to withstand the bending action, and generally from this cause ample surface 
is provided for good working ; but in any case the area, calculated by multi- 
plying the diameter of the journal by its length, should be such that the 



278 



MANl'AL OF MARINE ENGINEERING. 



pressure does not exceed 1,200 ibs. per square inch, taking the maximum 
load on the piston as the total pressure on it. 

Let L be the distance of the centres of the piston-rods in inches, and P 
the maximum load on the piston in pounds, then for strength 



Diameter of crosshead should not be less than 
For good wearing, / being the length of the journal. 

Diameter of crosshead should not be less than 



V P X L 
18 

P 

1.200 x r 



With fast-running engines the following rule may be followed where P 
is the maximum load on the piston, d is the diameter and / the total length 
of gudgeon or crosshead arms in inches, and R the revolutions per minute. 



Then for any engine 

Diameter of crosshead = 



P X * /R + 100 
12,500 

P x ^R + 100 
12,500 x / " 




Fig. 97. — Crosshead and Guide-block for double Piston-rods. 

Of course, the maximum load on the crosshead is really the reaction of 
the connecting-rod, but, to avoid any complication of the calculation, it is 
sufficient to take the load on the piston. 

When the diameter of the crosshead journal is calculated by the first 
rule, the length is usually made equal to it. 

Direct-acting engines have sometimes a gudgeon secured to the connecting- 
rod (fig. 98), which works in a bearing in the piston-rod end (figs. 91 and 
92) ; or have the gudgeon at the piston-rod end (figs. 93, 94. and 95). and 
connecting-rod (figs. 99 and 100) swung on it by brasses, etc., on either side. 
By the latter plan larger bearing surfaces are obtainable, and the brasses, 
being on the outside of the rods, are much easier watched and adjusted ; 
on the other hand, there are two sets of bolts, brasses, etc., to lubricate 
and keep in order, and there is the liability by careless adjustment to put the 



PISTON-ROD CROSSHEADS AND GUDGEONS. 



279 



whole of the load on one side only. In the main, however, this plan is a 
preferable one for large and heavy rods, and it is one which admits of the 
piston-rod being fitted into its end (figs. 94 and 95) instead of forged with 
it. This latter advantage is well worthy of consideration, for it is highly 
important that the piston-rod shall be quite free from flaws and reeds on its 




I 
— — i — 

I 







^ 





Fig. 98. 



Fig. 99. — Connecting-rods. 



surface, otherwise the packing soon gets damaged, and it is then found im- 
possible to keep the glands from leaking. A steel rod rarely has them, but 
if the rod simply fits into the crosshead or rod end, this latter may with 
advantage be made of cast steel, while the rod may be of forged steel of any 
desired quality, easily returned or ground in the lathe and cheaply replaced. 
Smaller engines are better with the gudgeon shrunk into the jaws of the 



280 .MANUAL OF MARINE ENGINEERING. 

connecting-rod, and working in brasses fitted into a recess in the piston-rod 
end (fig. 91), and secured by a wrought cap and two bolts. 

The diameter of the gudgeon = T125 X diameter of piston-rod. 
., length „ ,, = T4 X ,, ,, 

The area obtained by multiplying these is exactly double the area of the 
piston-rod section, and so if the maximum stress per square inch on the rod 
does not exceed 2,400 lbs., the above rule holds good ; if this allowance of 
stress is exceeded in calculating the rod, then the length of the gudgeon should 
be increased until the pressure on the section, as calculated by multiplying 
length by diameter is in accordance with the following : — 

12 500 
Rule. — Pressure per square inch on gudgeon = _^!_. — — lbs. for revo 

lutions R per minute. v R + 100 

When the gudgeon is fixed in the piston-rod (fig. 93), or formed with the 
guide ends as in fig. 94, the length of each end should be not less than - 75 the 
diameter of the piston-rod. 

The brasses when fitted into the piston-rod should be square-bottomed 
and of hard gun-metal, with good oil-ways, as owing to the motion being 
through a comparatively small angle only, the lubricant is not so easily 
spread. Some white metal does not work well on gudgeons and crossheads ; 
it has a tendency to abrade, and to wear the journal oval. This is generally 
observed whenever white metal is used on a bearing subject only to a small 
angular motion. White metal, however, can be used with mild steel gudgeons, 
especially in fast-running engines with advantage. 

Gudgeons, when fitted to the connecting-rods, should be made of hard 
iteel or mild steel case-hardened ; great care is required, if of the latter, 
that they are carefully ground true after hardening. 

The bolts securing the brasses should be of the best tempered mild steel ; 
they should be of such a size that the stress per square inch of section at 
the bottom of the thread does not exceed 6,500. It is true that where extreme 
lightness of machinery is a sine qud non, these stresses have been exceeded 
by some engineers ; but even under these circumstances it is unwise to exceed 
them by more than 10 per cent., as the saving of weight effected is exceedingly 
small, especially when compared with the risk run (v. Table lxxix.). 

When the bolts are less than 2 inches in diameter, owing to the uncertainty 
of the depth of thread, etc., the allowance should be not more than 6,000 lbs. 

Rule. — Diameter, of piston-rod bolts at the bottom of thread 



-»/ 



r 

D is the diameter of piston, p, the effective pressure on it per square inch. 

Where there are two bolts of steel/ = 13,000 lbs. ; under 2", 12,000. 
four „ J = 24,500 lbs. ; under 2", 22,000. 

The cap is of the same width as the piston-rod end — viz., 1*15 X diameter 
of the rod, and its thickness equal to the diameter of the bolts at the bottom 
of the thread. 

Connecting-rods. — The length of the connecting-rod measured from the 
centre of the gudgeon to the centre of the crank-pin should, if possible, be not 



CONNECTING-RODS 281 

less than twice the stroke. Quick-running vertical engines have almost 
invariably the connecting-rods twice their s'troke in length ; other vertical 
engines from twice to three times, but generally two and a quarter times is 
not exceeded. 

A connecting-rod may be viewed as a strut loose at both ends, the 

f S 
formula for which, as given on p. 271, is R = yr- . 

1 + 4«-jT 

a 1 
The value of R will be found by multiplying the load on the piston by the 
secant of the angle of obliquity of the connecting-rod (vide p. 276)- Or by 
geometry 

B.l. 2L , 

J4: L 2 - S* 

P being the load on the piston, S the stroke, and L the length of the 

connecting-rod as before. 

Simplifying the above formula by assuming a value r for the ratio of 

<ir d 2 
I to d, and substituting — - for S ; and taking the value of /at 3000 lbs. 

Diameter of connecting-rod in the middle = — — ^= — . 

48*5 

Example. — To find the diameter at the middle of the connecting-rod of 

an engine, 60 inches stroke, whose length is 120 inches, and the load on the 

piston 100,000 lbs., r being taken at 15. 

100,000 x 2 x 1 20 inQQ . Q „ 
R = — , = 103,358 lbs. 

^/4 x 120 2 - 602 



J 103,358 (l + ^~) 

Diameter at middle = — — ttt-= = 7-6 inches. 

48-5 

The following are the values of r in practice : — 

Naval engines — Direct-acting • . r = 9 to 11. 
Mercantile engines, ,, ordinary r — 12. 

,,, ,, long stroke r = 13 to 16. 

Taking 10 as the average value of r for naval engines, and 13 for mercantile; 

then, 

For a naval engine, 

Diameter of connecting-rod at middle = . / ~ ft( ,», 

For mercantile engines, 



R 

Diameter of connecting-rod at middle = A / ttttttv- 



/: 

V 1! 



The sizes given by these rules, although large enough for strength, are 
somewhat smaller than found in actual practice in the mercantile marine 
generally. 



282 MANUAL OF MARINE ENGINEERING. 

The following empirical formula will be found a very useful one, and the 
results given by it agree very closely with good modern practice : — 

Diameter of connecting-rod at middle = — • 

L is the length of the rod in inches, and 



K = 0-03 x ^Effective load on the piston in lbs. 

Example. — To find the diameter of the connecting-rod, 100 inches long, 
for an engine having a load of 55,000 lbs. 



K = 0-03 x ^55,000 = 70, 



Diameter =-^- — = 6*6 inches. 

* 

The diameter of the connecting-rod at the ends may be 0*875 of its 
diameter in the middle. The tapering of rods, or making them barrel- 
shaped, is usual in the case of those having single brasses at both ends ; 
then the diameter of the crank-pin end is 0"925 of the diameter at middle. 
Direct-acting engines have usually the connecting-rods tapering from the 
gudgeon end to the middle, and then parallel, or nearly so, to the crank-pin 
end. 

It is, however, simpler and sufficiently accurate to follow, in the case of 
connecting-rods, a similar rule to that laid down on p. 272 for piston-rods, 
viz. : — 

-r.. , , ,. , Diameter of cylinder / — 
Diameter ot connecting-rod = = — sjp. 

Here p is as calculated by the formulae on p. 273. F may be taken at 55 for 
the crosshead end of fast-running light engines, and 52 of mercantile engines 
of ordinary type. The diameter of the rods at the middle may be got by 
dividing by F — L, where L is the length of connecting-rod in feet. 

Connecting-rod Bolts. — The diameter of the bolts may be calculated by 
allowing the same stress per square inch as that given for piston-rod bolts. 
It is usual now, from practical considerations, to make the bolts of both 
piston and connecting-rod of the same size ; the bolts therefore should be 
calculated from the load on the connecting-rod. In order that the whole of 
the stretch shall not come on one section, as at the bottom of the last thread 
of an ordinary bolt, it is better to turn part of the body of connecting- and 
piston-rod bolts to the same diameter as at the bottom of the thread, leaving 
it a little larger than the diameter over the thread close to the head, and in 
way of any joint — that is, the bolt is made with a phis thread, and bearing 
collars where required. 

Connecting-rod Brasses. — The crank-pin brasses are more severely tried 
than any others about an engine, and, therefore, should be most carefully 
designed, and made of the very best material. Some engineers make the 
brasses to form the end of the rod (fig. 99), and retained to it by bolts and a 
steel cap ; others prefer that they shall only act as bushes or liners to the 
connecting-rod, sometimes fitting them into a square or octagonal recess in 
the rod end,' and held in place by a flat cap and bolts, just as is generally 
done to piston-rod ends; .but more generally they are fitted in duplicate 



CONNECTING-ROD BRASSES. 283 

halves, as shown in fig. 98. The former plan is an expensive one when they 
are ol large size and made of brass, on account of the great weight required, 
and consequently are also costly to renew when worn, besides which they 
are very liable to get out of shape when heated, and to crack through the 
crowns. The latter plan avoids the use of so much brass, gives a good solid 
bed to the brasses, and leaves the bolts free of all stress except tension. When 
rods are made in this way it is usual to forge the head of the rod solid, and 
turn it and the cap at the same operation ; the hole for the brasses is bored 
or slotted out (the latter when the hole is 9 inches and upwards in diameter) 
roughly ; the head is then slotted through or parted in the lathe so as to cut 
off the cap, the space left by the tool being equal to twice the difference in 
thickness of the brass at the crown and sides ; the cap is then bolted close 
to the rod, and the hole bored out to the diameter of the brasses measured 
across the rod. The brasses are kept from turning by a brass distance piece 
secured between the cap and rod and projecting between the brasses, and in 
the case of large brasses a short feather is fitted close to each flange in the 
crown. All brasses have a tendency to close on the pin or journal after 
having been hot, because the inner surface becomes warm first, and the 
metal in expanding tends to straighten the curved part ; this is resisted by 
the other part of the brass and the bed in which it is fitted, and in conse- 
quence this inner ourface gets compressed permanently, so that on cooling 
down it contracts, and tries then to give the brass more curvature, and so 
presses hard on the journal. It is for this reason that some bearings will 
never work cool but always a trifle warm ; this slight amount of heat causes 
the brass to expand so as truly to fit the journal. It is now not at all un- 
common to make these fittings of good cast or malleable cast iron when lined 
with white metal ; indeed, cast iron carries white metal better than brass 
does, and when of really good mixture is quite as strong as gun-metal, and 
stronger than the common brass so often used with white metal. 

Fig. 100 is an example of a modern connecting-rod with the jaws made 
as part of a cone, instead of as in fig. 99, thereby providing a better connec- 
tion between each jaw and the body of the rod, as well as permitting of its- 
being machined much more cheaply. This rod is fitted with cast-iron bearings- 
at each end, lined with white metal. This form of rod and " brasses " is now 
very generally used in the mercantile marine and Navy for engines of all- 
sizes. 

White metal is better than bronze for the rubbing surface of the crank- 
pin " brasses," it is important, therefore, that the white metal shall project 
beyond the " brass," so that it alone shall bear on the pin. For this purpose 
strips of white metal should be fitted into grooves planed in the " brass " 
and be well hammered, so as to thoroughly fill the spaces, after which it 
should be smoothly bored and fitted to the pin. Brasses which have not 
been originally designed for white metal may be fitted in this way, or by 
boring some shallow holes, whose diameter at the bottom is more than at the 
surface, casting into them buttons of white metal, which, after hammering 
down, are bored out so that the white metal stands out beyond the original 
wearing surface. 

A very good plan, but somewhat more expensive and not more efficient 
than the one above described, is to run the white metal into recesses cast 
with the brass, hammer it well in place, bore out, and then plane out the brass- 



284 



-MANUAL OF MARINE ENGINEERING. 



intervening between the white metal patches, leaving only slight ridges 
surrounding the latter, which prevent it from being spread out. The most 






Fig. 100.— Connecting-rod. Modern Form. 

general plan, however, is to have a recess in the " brass" or shell, which is 
now often of cast iron instead of bronze, tin the surface and run the metal 



GUDGEON END OF ROD. 



285 



in place while the shell is warm. When cold bore out as usual, trim the 
edges, etc., and cut grooves for the oil to circulate in the surface of the metal 
(v. fig. 100). 

Caps of Connecting-rod Brasses. — The width of the connecting-rod end 
should be such as to efficiently support the brasses ; its thickness (in direc- 
tion of the length) = 0"6 x diameter at middle of rod. 

The thickness of cap at middle = 0*8 X diameter of body of bolts -j- 
0*1 X pitch of bolts. 

Thickness of cap at ends = diameter of bolts at bottom of thread. 

For ease of manufacture caps are generally made straight and of thickness 
given by the first rule. 

Gudgeon End of Rod. — The jaws of every connecting-rod are subject to 
forces which tend to open them on the down stroke and close them on the 
up. When fitted with a gudgeon-pin secured in the eyes, as in fig. 98, the 
jaws are subject to bending moments at the various sections, which become 
maxima at a point just below the gudgeon and also at an angle of about 40° 
to the vertical through the centre of curvature of the inner portion. When 
the rod is fitted with "brasses" working on a crosshead, as in fig. 99, the 
maximum bending moment is at an angle of about 40° to the vertical or axis 
of the rod, and is greater than when with a gudgeon secured to the rod ends. 
When the diameter of the gudgeon or crosshead ends is 



then 



Diameter of cylinder #— 
~I5~ " X V? ' 



Let P be the maximum re-current load on the piston. 

L the width from centre to centre of gudgeon-rings or of crosshead 
brasses when so fitted. 



F 



= IT 

V 3,000' 



then 



Diameter of gudgeon ring . 
Thickness ,, ,, 

Width of jaw . 
Thickness of jaw in line at 40 c 



. = diameter of gudgeon + D25 X F. 
. = 0-85 x F. 

. = 1*35 X F, or 1-1 X diameter of rod. 
. = '48 F X L when with gudgeon. 
. = *52 F X L when with brasses. 




Fig. 100a. — Connecting-rod End (Michel). 



286 MANUAL OF MARINE ENGINEERING. 



CHAPTER XII. 

SHAFTING — CRANKS AND CRANK-SHAFTS, ETC. 

The Shafting of a Modern Marine Engine is made wholly of mild steel, 
produced from a Siemens furnace, and generally having a tensile ultimate 
strength of about 30 tons. In section it is, of course, circular, and in the 
mercantile marine generally solid ; in Naval ships it is always made hollow 
by boring out the centre so as to remove the core of doubtful metal and 
provide the strongest shaft with the least weight, inasmuch as, with a large 
reduction in material, there is only a small reduction in strength by this 
treatment. For example, if a shaft have a borehole in diameter half that 
of the outside, the reduction in weight is 25 per cent., while the strength is 
diminished by 12J per cent. only. The cost of hollow shafts is, of course, 
greatly in excess of that of solid ones ; notwithstanding, it is found worth 
while to fit them to many of the high-speed express steamers crossing channels 
and entering shallow harbours. 

In Practice every Shaft in a Ship is subject to torsion, and, therefore, to 
the shearing stresses caused by torque. It has also to resist the additional 
shearing stress due to its own weight, but this addition is, as a rule, not great 
by comparison. But the weight causes a bending moment to act on every 
part of the shaft, which, if the bearings are wide apart, will cause a stress 
and deflection which must be taken account of. 

Throughout the Line of Shafting from the motor to the propeller is trans- 
mitted to the latter the torque generated in the former, and when the screw 
is heavy and racing badly in a sea-way, there mav be, and often is, a reverse 
action by the inertia stored in the latter transferring back a torque to the 
motor. 

The Propeller-shaft of every Ship is subject to a more severe load from 
the overhung heavy instrument setting up severe bending moments, due to 
gravity in smooth water, and in rough water to the inertia effects and the 
uneven action of the blades on the water. When a ship is pitching in a heavy 
sea the stern drops with great rapidity, so that the vertical velocity of a heavy 
screw is serious ; this velocity is checked with sufficient abruptness to put 
a very heavy bending moment and shearing force on the shaft end, where 
it ceases to be supported by the stern bush, as is equally the case when the 
wave motion causes a rapid rise of the stern. The side throw or lurch at the 
stern from wave action sets up horizontal inertia forces, also severe and 
trying, although not so intense as the vertical ones. The differential pres- 
sures on opposite blades of the screw also produce bending moments with 
the corresponding shearing forces. Hence the stern shaft of a cargo steamer, 
which is sometimes deeply laden and sometimes light, and always somewhat 
lively at the stern, is tried very severely, and should always be of ample size ; 



TUNNEL SHAFTING. 



287 



in fact, larger than that of the finer-lined express steamer of the same power, 
but with a screw lighter, because of smaller diameter and often of bronze. 
Crank-shafts of reciprocators are subject to complex loads at their various 
parts, and always have bending moments of kinds as well as torque to resist. 
The shafts of turbines also have to bear considerable bending moments, 
due to the weight of the rotor, but in their case both it and the torque are. 
constant and uniform ; there is no variation in torque through the revolution 
as with the reciprocator. It does, however, have to resist inertia forces of 
a kind when the ship is in a sea-way, and when the turbine is well aft, as in 
fig. 40, the stresses due to them will be heavy in bad weather. 

The Crank-shaft of the ordinary Oil Engine is subject to severe bending 
moments, and to shocks which produce heavy shearing forces. It is claimed 
for the Diesel engine that, owing to the very great amount of compression, 
there is little or no shock at explosion, and the diagram (fig. 101) would 
seem to prove this. The combined torque diagram from an oil engine on the 

IN0ICATOR DIAGRAM OF DIESEL 
2 STROKE CYCLE ENGINE-, 




Fig. 101. —Diesel Engine Diagram. 

four-cycle system, however,' shows a very large ratio of maximum to mean 
torque as well as the big bending moment. 

The Tunnel or Intermediate Shafting of a ship is subject theoretically 
to torque only ; it is, of course, in compression axially from the screw shaft to 
the thrust block, but this affects the torque so little as to be negligible (about 
1 \ per cent.). If the tunnel bearings are close together, and the rate of revolu- 
tion is not high, then torsion alone is the governing factor in determining its 
size. But, since the rate of revolution has increased so largely, especially 
where turbines are employed, it is necessary now to cousider the bending 
moments that may or do come on these shafts. However close, in reason, 
the bearings may be to one another, the weight of the shaft must cause a 
certain amount of sag ; if the centre of gravity of the shaft does not coincide 
with its axis of rotation, it may then whirl — that is, the C.G. will tend to 
move in a circle, whose centre is on the axis of rotation. At low speeds 
the whirling action is comparatively small, and not sufficient to further 
bend the shaft, but at higher speeds the centrifugal force may be so great 
as to be harmful. 



288 



MANUAL OF MARINE ENGINEERING. 



Ill order that a shaft may revolve freely without any danger of whirling 
due to the " sag " or deflection caused by its own weight, the bearings should 
never be further apart than the distance given by the following rule, where 
d is the diameter in feet and z is a multiplier depending on the rate of revo- 
lution : — 

Rule. — Pitch of plummer blocks should not exceed z Jd. 

When the revolutions do not exceed 200 per minute, z = 40. 



5 »» 

» >5 

5 >♦ 

» J> 

> >J 

> )J 
J >> 
J J> 



99 


250 


> ? 


2 = 36. 


»> 


300 


j> 


2 = 315. 


>} 


400 


j» 


2 = 29-0. 


>> 


500 


5J 


2 = 25-5. 


>» 


600 


JJ 


2 = 23-5. 


»> 


700 


»; 


2 = 21-5. 


>> 


800 


>f 


z = 20-0. 


>> 


900 


5> 


2 = 19-0. 


>> 


1,000 


JJ 


2 = 18-0. 



Considering that in a screw ship the lengths of tunnel shafts near the 
stern will be subject to considerable inertia forces, the pitch of bearings should 
be less than the limit given above for still water, and about 60 per cent, should 
not be exceeded. For example, a shaft 12 inches diameter at 200 revolutions 
should be supported every 24 feet, while at 400 revolutions it should be 
17*4 feet, and never exceed in smooth water 29 feet. 

Alternating Stresses. — If a shaft having a heavy body, such as a screw or 
paddle-wheel, at its overhung end is revolving under the action of torque, 
the material from the weight to the supporting bearing is subject to tension 
and compression, and shearing due to that weight and to shearing at every 
section throughout its length due to the torque. But the bending moment 
due to the overhung weight sets up tension on all particles above the neutral 
axis and compression on all below it, and inasmuch as the shaft is turning, 
so that what was at the top at one moment is at the bottom the next, hence 
the particles, from being subject to tearing apart when above, are quickly in 
a state of being crushed together below. The magnitude of the forces ranges 
from steadily to the maximum at top and bottom, so that the load and 
change of load is gradual ; nevertheless, seeing that at 120 revolutions the 
change from plus to minus is no less than four times in a second, while with 
turbine-driven shafts there is often as many as 24 alternations per second, 
the endurance of the material is severely tried. Such stresses as these which 
it has to resist are called Alternating. 

Piston-rods, Connecting-rods, Valve-rods, and Columns are all subject 
to these reversals of load and stress, but in their case, the application and 
release are sudden and made with more or less shock, but modified somewhat 
by the action of inertia forces and the effect of cushioning. 

The Arm of a Lever vibrating through an angle when transmitting energy 
is subject to alternating stresses also, more or less abruptly applied and 
released. 



SAFE WORKING STRESS. 289 

But the Arm of a Crank-shaft, which is a lever moving always angularly 
in one direction, is subject to no such changes of stress. On the one side 
on which the force is applied the molecules are subject to tension only, and 
thos-e on the opposite side to compression only ; but in the case of an engine 
crank the Load and stress, as measured by the turning moment, are not so 
abrupt in their application, etc., and they also are modified by inertia forces 
and cushioning. With the reversal of the engine into " astern gear " these 
stresses are, of course, reversed. 

The Bolts of Connecting-rods, Main Bearings, Cylinder Feet, and 
Cylinder Covers are all subject to a load of one kind, but it is applied 
and released, then ceases for a moment, to be again applied more or 
less suddenly. Such loads and the stresses caused by them are called 
Intermittent. 

Iron, Steel, and all Materials suffer more or less severely from these inter- 
mittent and alternating stresses, but much more so from the latter, as the 
effect on their structures is to gradually destroy them. The higher the stress 
and the greater the number of alternations or intermissions per minute, the 
quicker is the strength or virtue of the material annihilated. Of steels some 
varieties have a longer life than others, and of the bronzes it is even more 
marked. It does not follow of necessity that because a steel or bronze has 
a very high elastic limit that it can be used with greater longevity in a rapidly 
active structure than one whose elastic limit is less. Professor Arnold has 
shown that bar steel, by rolling and drawing cold, might have its yield 
point raised by as much as 8 to 12 tons, while the endurance was reduced 
by 38 per cent. ; in the case of aluminium copper alloys the endurance 
of one was more than double that of others not differing largely from it in 
composition. 

The Influence of Numbers of Reversals of Stress and their magnitudes 
has been demonstrated also by Prof. Arnold and others, and the results of 
experiments on steel and iron is startling. 

Wohler's experiments showed that the iron used for axles then (1870) 
required 56,430 repetitions with a plus and minus load of 15*3 tons per square 
inch to produce fracture, while with 8 - 6 tons each it required over 19 millions. 
Similar experiments with Krupp's steel showed that with plus and minus loads 
of 20 - l tons per square inch 55,100 repetitions produced fracture at 15 - 3 tons; 
it was done with a little over 3 millions, while some bars with 143 tons took 
over 45 millions. It is clear, then, that since some definite number of alter- 
nations will break down the tenacity of a metal, and that the greater the 
stress the fewer they will be, that for long life a low working stress is 
necessary. 

Safe Working Stress on the material must, therefore, vary with the con- 
ditions under which it performs its duty. In a general way the highest 
stress allowable under any circumstances is one-half that of the elastic limit, 
and for margins of safety to cover inaccuracies and small hidden defects 
the highest working stress should not exceed 40 per cent, of the elastic limit. 
If the material is subject to intermittent stress, 90 per cent, of this maximum 
or 36 per cent, of the elastic limit should be observed in the design. If, 
however, the stresses are alternating, 66 of the maximum or 26*4 per cent, 
of the elastic limit should not be exceeded. 

The elastic limit of mild steel, as found in any workshop, is about 

19 



290 MANUAL OF MARINE ENGINEERING. 

30,000 lbs. The working stresses on it with moderate speed of revolution 
should be 12,000, 10,800, and 7,920 lbs. per square inch in tension. 

For high rate of revolution even these allowances are too great, and 
over 500 revolutions per minute the reduction in stress should be quite 
10 per cent., so that for screw shafts of fast-running engines the stress 
allowed in calculating the sizes should not exceed 7,000 lbs. for continuous 
working. 

For work where the alternations or intermissions are effected with con- 
siderable shock the working stress should be at least 10 per cent, less than 
when effected gently. 

It should be noted, however, that certain qualities of steel and bronze 
stand shock and alternation of stress much better than others. 

Twisting Moment. — If a force is acting on a shaft so as to turn it, or tend 
to turn it, round on its axis, it is called a twisting force, and the effort of this 
force is measured by multiplying it by its distance from the axis, and called 
the twisting moment or torque. Suppose P is the thrust along the connecting- 
rod when at right angles to the crank, and L is the distance of the centre 
of the crank-pin from the centre of the shaft, P X Lis the twisting moment 
on the shaft. 

When one force is acting on the shaft as above described, the second 
force, which completes the couple, is the reaction of the bearing, which is 
equal to P, but acts in the opposite direction. If the force P and the reaction 
R act in a plane perpendicular to the axis of the shaft, they will cause no 
bending action on the shaft, but there will be a force R tending to shear 
the shaft across. But in actual practice it is almost impossible that P and 
R shall act in such a plane, and they usually act in planes parallel to one 
another, and perpendicular to the axis ; hence, the shaft is also subject 
to a bending action. But if a shaft is turned by means of two equal forces 
acting in opposite directions, one on either side of the shaft and equidistant 
from the axis and in the same plane, then the shaft is balanced, these forces 
will cause no pressure on the bearings, and it is subject, therefore, to twisting 
strains only. If one shaft is coupled to another shaft, from which it is to 
transmit power by two coupling bolts equidistant from the centre, it will 
only receive a twisting strain. Such is the state of the shafting from the 
crank-shaft to the propeller-shaft of a screw steamship. 

Resistance to Twisting. — Let T be the twisting moment on a shaft in 
inch pounds, d the diameter of the shaft in inches, and / the stress per 
square inch on the transverse section of the shaft. Then (Rankine, Applied 
Mechanics, p. 355) for solid shafts of diameter d, 



T = '^-f = 0-1964/d 3 ; or d = M x 51. 



For hollow shafts with a bore diameter d lt 
T = 0-1964/(— ^j-^V 
Example. — To find the diameter of a shaft subject to twisting only, the 



DIAMETEK OF A SHAFT SUBJECT TO TORSION. 291 

force being 100,000 lbs. acting at a distance of 24 inches, stress to be 8,000 lbs. 
T = 100,000 x 24 = 2,400,000 inch-lbs. 



d- 3 A40 
V 8 



400,000 _ . .... , 

x 5-1 = 11 *5 inches. 



8000 

Diameter of a Shaft subject to Torsion. — If a constant force P were applied 
to the crank-pin tangentially to its path, then the work done per revolution 

will be P x ■— l — ; L being the length of the crank in inches ; then if R be 

the number of revolutions per minute, 

2 it L 
Work done per minute = P x '* x R. - - (1) 

But this work is equal to I.H.P. x 33,000 ; and the twisting moment is 
P x L constantly. Then 

(PxL)x^xR = I.H.P. x 33,000, 

and 

„ T I.H.P. x 33,000 x 12 
P x L = ^ • 

2 ir x R 

That is, 

(2) 



.3) 



Diameter of shaft = a/-^^ X 42-84. - ... (4) 

But as shafts must be strong enough to resist the maximum twisting stress, 
it is necessary always to base calculations on it instead of on the mean twisting 
moment. The factor 42 "84 must, therefore, be multiplied by the ratio of 
maximum to mean moment, as given in Table xxxv. 

Professor Rankine directs (Rules and Tables, p. 250), in order to find the 
greatest twisting moment from the mean : if a shaft is driven by a single 
engine, multiply by 1*6 ; if by a pair of engines with cranks at right angles, 
by ri ; if by three engines with cranks at angles of one-third of a revolution, 
by 1-05. 

These values are, however, very much lower than usually obtained in 
modern practice if the effect of inertia forces are neglected. 

For a three-crank engine, cranks at 120°, cutting off at half to two-thirds 
stroke, multiply by 1*15. 

For a two- and four-crank engine having cranks at right angles, cutting 
off steam at half to three-quarter stroke, multiply by 1*3. 

For a single-cylinder engine cutting off steam at half stroke, by 2"0. 



290 MANUAL OF MARINE ENGINEERING. 

30,000 lbs. The working stresses on it with moderate speed of revolution 
should be 12,000, 10,800, and 7,920 lbs. per square inch in tension. 

For high rate of revolution even these allowances are too great, and 
over 500 revolutions per minute the reduction in stress should be quite 
10 per cent., so that for screw shafts of fast-running engines the stress 
allowed in calculating the sizes should not exceed 7,000 lbs. for continuous 
working. 

For work where the alternations or intermissions are effected with con- 
siderable shock the working stress should be at least 10 per cent, less than 
when effected gently. 

It should be noted, however, that certain qualities of steel and bronze 
stand shock and alternation of stress much better than others. 

Twisting Moment. — If a force is acting on a shaft so as to turn it, or tend 
to turn it, round on its axis, it is called a twisting force, and the effort of this 
force is measured by multiplying it by its distance from the axis, and called 
the twisting moment or torque. Suppose P is the thrust along the connecting- 
rod when at right angles to the crank, and L is the distance of the centre 
of the crank-pin from the centre of the shaft, P X Lis the twisting moment 
on the shaft. 

When one force is acting on the shaft as above described, the second 
force, which completes the couple, is the reaction of the bearing, which is 
equal to P, but acts in the opposite direction. If the force P and the reaction 

R. ant in a. r»la.np v\prneru]if>iilsir t.n fhp mris ni thp shaft t.ViPV will pause no 



The reference on page '291 to Table WW. should be 
Table XXXIV., page 298. 



transmit power Dy two coupling uvivo cijumiciuau UU u. ^„~ , 

only receive a twisting strain. Such is the state of the shafting from the 
crank-shaft to the propeller-shaft of a screw steamship. 

Resistance to Twisting. — Let T be the twisting moment on a shaft in 
inch pounds, d the diameter of the shaft in inches, and / the stress per 
square inch on the transverse section of the shaft. Then (Rankine, Applied 
Mechanics, p. 355) for solid shafts of diameter d, 



- d 3 It 

T = —/ = 0-1964/d3; ovd = ^ x 5-x. 



For hollow shafts with a bore diameter d u 



T = 0-1964/(-- d -^ 4 ). 
Example. — To find the diameter of a shaft subject to twisting only, the 



DIAMETER OF A SHAFT SUBJECT TO TORSION. 2\)l 

iorce being 100,000 lbs. acting at a distance of 24 inches, stress to be 8,000 lbs. 
T = 100,000 x 24 = 2,400,000 inch-lbs. 



'J 



2,400,000 _ . .... , 

x 5-1 = 11 '5 inches. 



8000 

Diameter of a Shaft sutject to Torsion. — If a constant force P were applied 

to the crank-pin tangentially to its path, then the work done per revolution 

2 t L 
will be P x — — — ; L being the length of the crank in inches ; then if R be 

the number of revolutions per minute, 

2tL 

Work done per minute = P x x R. - - (1) 

But this work is equal to I.H.P. x 33,000; and the twisting moment is 
P x L constantly. Then 

(PxL)x^xR = I.H.P. x 33,000, 

and 

n T I.H.P. x 33,000 x 12 

P x L ^ 

J T X R 

That is, 

T H P 

Mean twisting moment = -^-^ — ■ x 63,000. - - - - (2) 

And as before 

d _ 3 /I.H.P. x 63,000 x 5-1 



R xf 



I.H.P. 321,300 



"V""R~ x ' / 

If/ be taken at 7,500 for mild steel 



(3) 



Diameter of shaft = */ •■■' ~ x 42'84 



V R-X 42-84. - - . . ( 4 ) 

But as shafts must be strong enough to resist the maximum twisting stress, 
it is necessary always to base calculations on it instead of on the mean twisting 
moment. The factor 42 "84 must, therefore, be multiplied by the ratio of 
maximum to mean moment, as given in Table xxxv. 

Professor Rankine directs (Rules and Tables, p. 250), in order to find the 
greatest twisting moment from the mean : if a shaft is driven by a single 
engine, multiply by 1 -6 ; if by a pair of engines with cranks at right angles, 
by 1*1 ; if by three engines with cranks at angles of one-third of a revolution, 
by 1-05. 

These values are, however, very much lower than usually obtained in 
modern practice if the effect of inertia forces are neglected. 

For a three-crank engine, cranks at 120°, cutting off at half to two-thirds 
stroke, multiply by 1*15. 

For a two- and four-crank engine having cranks at right angles, cutting 
off steam at half to three-quarter stroke, multiply by 1/3. 

For a single-cylinder engine cutting off steam at half stroke, by 2*0. 



292 MANUAL OF MARINE ENGINEERING. 

The following rule holds good for the ordinary engines, as found in general 
use iu the merchant service : — 

/r u p 

Diameter of the tunnel shafts = ?/ - * ' : X F. • • (5) 

Two-stage compound engines, cranks at right angles — 

F = 8 Jp, 

where p is the absolute actual pressure, or boiler pressure + 15 lbs. 
Triple-compound, three-cranks at 120 c — 

F = 4Wp. 

Quadruple-compound engines, two cranks, right angles, F = 5*5 Jv. 
„ „ four cranks, F = 5-C Jp. 

Expansive engines, cranks at 90°, and the rate of expansion 5, F = 9 - 5 J p. 
Turbines, F = 54. 
Single-crank compound engines, pressure 80 lbs., F = 10 s '« 

The shafts of torpedo-boats, destroyers, and fast craft which are run at 
full speed only occasionally, and for short periods, may be designed by taking 
F at about a half of the above values. 

The Torsional Stiffness of a Shaft is of more importance to the marine 
engineer to-day than it has ever been before, inasmuch as it is employed as 
the measure of the torque of the turbine, and from it the horse-power trans- 
mitted is measured (v. p. 150). 

The torsion meter employed on shipboard to obtain the measure of the 
power of a turbine is really in essence only an instrument for indicating the 
angle of twist of a definite portion of one of the shafts that previous to fitting 
in the ship was tested by torque to ascertain the angular displacements 
corresponding to the magnitude of each application. The following formula 
is a means for computing what that angular displacement should be in degrees, 
with a shaft whose external diameter is d and the diameter of bore d x ; the 
twisting moment or torque is T l in inch-pounds, and C is the modulus of 
stiffness, which for solid shafts of best mild steel is 11,750,000, and for hollow 
shafts 12,150,000. is the angle in degrees; L is the length of the shaft 
under observation in inches, R the revolutions per minute. 



(«) 



584 X T x XL 

C X./ 1 
T,xL 



for solid shafts. 



20, 120 d* 



584 x T, X L 

' CX («**-(*!*) 



(I>) 



- for hollow shafts. 
T lX L 

- &A J 



21.147 {d*-d,*) 



BENDING MOMENT. 293 

nT , , , ,, S.H.P. x 33,000 , , „ 
(c) I he torque on a shaft = — = — - foot-lbs. 

= ^5^ X 63,000 inch-lbs. 

17 *-- / x rp ^X 20,120 d* S.H.P. . Q/vvn 
Prom equation (o) Tj = ' = —= — x 63,000. 

(b) T l = dx21Ml( d —=^) = ^^ X 63,000. 
Then S.H.P. = f x *L f or so lid shafts, 

K rf 4 — (7 * 

and S.H.P. = f- X nQ i for hollow shafte. 

Li Ji'VO 

Example. — Find the value of 6 for a shaft 9 inches diameter and 6 inches 
"bore when transmitting 5,000 S.H.P. at 350 revolutions per minute. The 
length under observation is 40 inches. 

5 000 
Here the torque = -^ X 63,000, or 900,000 inch-lbs. 

u< *■ ,m /, 900.000 x 40 ft _ oq , , 

Equation (6), fl = 21,147 (6,581 - 1,296) = ° -323 of a de ^- 

Bending Moment — If a force is acting on a shaft tending to bend 
it only, its effort is called the bending moment, and is measured by- 
multiplying the force by the distance at which it acts from the support 
of the shaft. 

If the shaft is overhung like a cantilever, and a force P is applied at a 
•distance L from the point of support, 

The bending moment = PxL- - -(1) 

If supported on two bearings, whose distance apart is L, and a force P is 
■applied at a point midway between these two bearings, 

P X L 

The bending moment = — j — • " - - (2) 

If the bearings are long — that is, exceeding the diameter of the shaft in 
length, and are also strong and rigid, so that the shaft is held by them suffi- 
ciently to prevent flexure taking place in the bearing, 

P X L 

The bending moment = — « — -. - - - (3) 

o 



294 MANUAL OF MARINE ENGINEERING. 

Since, however, few shafts are so secured as to comply with these conditions 
exactly, any shaft supported in strong hearings not less than 1 diameter long, 
and whose distance apart does not exceed 10 diameters, and which has to 
work freely in its bearings, may be treated as partly complying with these 
conditions, and 

The bending moment = — - — . - - (4) 

When the bearings are not rigid enough to prevent flexure of the shaft 
within them, L must be measured from the centres of the cap bolts, so that 
where each cap is held down by a pair of bolts L is measured to the centres 
of bearings. If the caps and bearings are strong and rigid enough to resist 
any tendency to bend by the action of the shaft, L may be measured from 
the edge of the bearing or cap. If the bearing is fitted with brasses, which 
project beyond the cap and bed so much as to receive little or no support 
from them, L must still be measured from the edge of cap. In a few words, 
the distance must be measured from what would be the actual points of sup- 
port if it is bent by severe pressure. 

Resistance to Bending. — The strength of a circular section shaft to resist 
bending is only half of that to resist twisting. If M is the bending moment 
in inch pounds, and d the diameter of the shaft in inches, 



M = ^x/;andrf = *fi&~ W -2 



f is a factor which depends on the material of which it is composed, and 
the value may be based on the fact that it is in tension and compression. 
The only shafts in a marine engine which are subject to bending only are 
some weigh-shafts having double-ended levers, similar to the side levers of 
paddle-wheel engines, and their diameter is determined from other considera- 
tions than that of mere strength ; but with them, as with the crossheads of 
return connecting-rod engines, care should always be taken that the size suit- 
able for good working in the bearings should be sufficient for strength. 

Equivalent Twisting Moment. — When a shaft is subject to both twisting 
and bending simultaneously, the combined stress on any section of it may be 
measured by calculating what is called the equivalent twisting moment — that 
is, the two stresses are so combined as to be treated as a twisting stress only 
of the same magnitude, and the size of shaft calculated accordingly. Pro- 
fessor Rankine gave the following solution of the combined action of the two 
stresses (vide Rankine, Rules and Tables, p. 227) : — Let T be the twisting 
moment on a shaft when M is the bending moment on a section, then taking 
T T as the equivalent twisting moment, 

Tj = M + JM? + T 2 . 

Example. — To find the diameter of a section of a shaft at which the bend- 
ing moment is 40,000 inch-pounds, when the twisting moment is 250,000 inch- 
pounds. The shaft of steel / = 7500 lbs. 

Here 

T x = 40,000 + X A0,000 2 + 250,000 2 

= 40,000 + 10,000 Vi* 2 + 25* 
= 293,170 inch-pounds. 



i . s/T 5 .! _ 3/ 293,170 x 5- 1 = 
V/ V T.50U 



CURVE OF TWISTING MOMENTS. 



295 



Crank Shafts. — These shafts are subject always to twisting, bending, and 
shearing stresses ; the latter are so small compared with the former that 
they are usually neglected directly, but allowed for indirectly by means of 
the factor f, as already stated. 

The two principal stresses vary throughout the revolution, and the maxi- 
mum equivalent twisting moment can only be obtained accurately by a series 
of calculations of bending and twisting moments taken at fixed intervals, 
and from them construct a curve of strains. 

Curve of Twisting Moments. — The twisting moment at any position of the 
crank is equal to the pressure on the piston multiplied by the distance inter- 
cepted by a line through the connecting-rod on a line at right angles to centre 
line through centre of cylinder. 




Fig. 102. 

Let AB (fig. 102) be the centre line of the engine through the cylinder and 
shaft centres, A C the position of the crank, B C the connecting-rod, and A D 
a line at right angles to A B. Produce B C to cut the line A D, and drop 
from A a line A E perpendicular to B C. P is the load on the piston, and R 
is the thrust on the connecting-rod. It will easily be proved that the angle 
D A E is equal to the angle A B D, called for convenience a. Then 

P = R cos a ; and AE = AD cos a. 
The twisting moment = RxAE = RxAD cos a = P x A D. 

Let the twisting moment be calculated at equal intervals of say 10° of 
angular movement of the crank, so that in the whole revolution there will be 
36 observations, or 18 in the half revolution. Draw a line AB (fig. 103), and 
divide it into 18 equal parts, Aa 15 a x a 2 , &c; erect at these points perpen- 
diculars, and cut off parts a 1 b v a 2 b 2 , &c, to represent the value of the twisting 
moments at each corresponding position of the crank to a suitable scale. 
Through the points b x b 9 , &c, draw a curve, which represents the curve of 
strain on the shaft during the forward movement of the piston ; by producing 
A B, and going through a similar operation for the second half of the revolu- 
tion, the curve of strain during the backward movement of the piston can be 
obtained. 



296 



MANUAL OF MARINE ENGINEERING. 



Divide the area enclosed between this curve and the line A B by the 
length of A B, and the quotient is the mean twisting moment, and repre- 
sented by A M in fig. 103, so that the rectangle A M N B is equal in area to 
the figure A B C. 

The value of A M may be calculated by taking a mean of the values oi 
Ojfej, a 2 6o. etc. 

When there are two engines — that is, two pistons operating on one shaft 
— the combined twisting moment is found by drawing the curve of twisting 




Fig. 103. — Curve of Twisting Moments. 



moments of each crank separately, transposing that of one on that of the 
other in a position corresponding to the relative position of the cranks. In 
fig. 103a A C B is the curve of strain on one crank, and A 1 C 1 B l that on the 
other, which is at an angle with it of degrees represented by AA X . The 
combined twisting moment at any period a is represented by a d, which is 
equal to a b 4- a c, and the dotted curve CdC lt etc., represents the curve of 
combined twisting moments. 

The maximum twisting moment will be at the point where the curve is 




A A. ;a B B 

Fig. 103a. — Curve of Combined Twisting Moments. 

highest, and the ordinate may be measured and its value found by referring 
to the scale to which the curve is drawn. The mean twisting moment may be 
found by measuring the area included between the dotted curve and the base 
line and terminal ordinates, and dividing by the base line, or by taking a 
mean value from the ordinates as before. 

If there are three engines, a similar operation will indicate the maximum 
twisting moment. 



CURVE OF TWISTING MOMENTS. 



297 



There is another, and perhaps a better, method of showing the curve of 
torque whereby the magnitude of the twisting moment at any angle through- 
out the revolution is found by taking the length of the radial line intercepted 
between the crank-pin circle and the curves, which are constructed by using 
that circle as a reference instead of a base line. Fig. 104 is a good example 
of this method, as prepared by Professor Jamieson, for a triple-compound 
three-crank engine working under ordinary conditions as given below, and 
will be found instructive. 




Fig. 104. — Crank Effort Diagram of a Triple- 'Expansion Engine. 

Ratio of expansion, 10-4. 
Length of connecting-rod = 9 feet. Stroke = 4-5 feet. 
Mean efficiency of steam, or ratio of area of work in cylinder to full theoretical diagram 

55 per cent. 



Cylinder's diameter, . 
Area, ..... 
Ratio, ..... 
Mean pressures, lbs. per sq. in., . 
Range of temperatures, Fah., 


HP. I.P. L.P. 

. 28" 46" 77" 
. 615-8 1661-9 4656-6 
1 2-80 7-56 
67-6 28-2 9-7 
64-3° 74-9° 80-6° 


Steam, 164 lbs ; Vacuum, 26£ ins. ; Receivers, 52 and 5 lbs. ; 
Revolutions, 62i per minute. 


Cut-off H.P. =- 33A ins. 

„ LP- = "„ 
,, L.P. — „ 


I.H.P. = 710 L.P. 
„ = 799 I.P. 
„ = 764 H.P. 




2,273 total I.H.P. 




ssss 



298 



MANUAL OF MARINE ENGINEERING. 



The pressures at the different points may be taken from actual indicator 
diagrams, or by constructing steam diagrams from the conditions under 
which the engine is to work. 

The bending moment on a section of the shaft will vary exactly with 
the pressure on the crank-pin, and to find the maximum equivalent twisting 
moment on a section, it is only necessary to construct a secondary curve from 

the formula Tj = M + -/M 2 -f- T 2 between the point of maximum twisting 
and that at which the pressure on the piston is greatest. 

When steam is not cut off in the cylinder before 0"4 of the stroke, the 
maximum load on the piston may be used to calculate the bending moment, 
which is to be combined with the maximum twisting moment to find the 
maximum equivalent twisting moment. Only when steam is cut off earlier 
than this does the point of maximum equivalent twisting moment differ 
much from the point of maximum twisting. 

Momentum of Moving Parts. — In making these calculations it has been 
assumed that the moving parts, such as the piston and rods, have no effect 
on the force exerted on the shaft ; but this is never strictly true, for since 
these parts are of considerable weight, a part of the energy of the steam is- 
absorbed at the commencement of the stroke in overcoming their inertia, and 
consequently the load on the crank-pin is less then than is represented on 
the curves. Again, towards the end of the stroke the momentum, or energy 
thus stored in these moving parts, is given out on the crank-pin, and causes 
larger loads on it than shown by the curve. The further consideration of 
the effect of the inertia and momentum of the moving parts will be found 
in Chap. xxix. It is sufficient to say here that the general tendency is to 
modify the stresses on the crank shaft, as also those of the connecting-rods, 
piston-rods, etc., so that if any of these are strong enough to withstand the 
stresses due to external forces, they are sufficiently strong for the engines 
when moving. Indeed, it is only at very high speeds that momentum need 
be taken seriously into account. 

The following Table (xxxiv.) gives the relation between the maximum 
and mean twisting moments of engines working under various conditions, 
the momentum of the moving parts being neglected : — 



TABLE 


XXXIV. 




Description of Engine. 


Steam cut-off at 


Max. Twist. 
Mean Twist. 




0-2 


2-625 


ii 






0-4 


2 125 


ii 






06 


1-835 








08 


1-698 ! 


Two-cylinder, cranks at 90*, - 




. 


o-i 


1-872 


n »» • ■ 




. 


02 


1-616 


i» i» 




. 


3 


1-415 


»» >• 




• 


4 


1-298 


n i» • ■ 




. 


5 


1-256 


ii »» 




- 


6 


1-270 


ii ii 




- 


07 


1-329 


ii ii 




- 


0-8 


1-357 


Three-cylinder compound, three ci 


•anks at 120°, 


H.P. 5, L.P. 6G 


1-40 


„ triple, „ 


06 


115 


Four-cylinder quadruple, four cranks at 90°, 


6 


1-26 



OVERHUNG CRANK. 



299 



Overhung Crank. — The simplest form of crank is that known as the over- 
hung crank, such as is usually fitted in mill engines, but hardly found now 
in marine engines. The shaft projects beyond the bearing, and has keyed to 
its end a lever or disc, in which is secured the crank-pin. 

The pin is subject to bending and shearing forces, due to the thrust on 
the connecting-rod. The maximum bending moment on the part of the pin 
close to the crank is found by multiplying the greatest thrust of connecting- 
rod by the distance to the centre of the connecting-rod. 

If R is the thrust of the connecting-rod, and I the length of the pin, then 

, . R x l 
* Bending moment on crank-pm = 



2 



and diameter of pin 



R x I 10-2 
■ x 



y^rr; 



5-1. 



Example. — To find the diameter of the crank-pin whose length is 14 inches 
1 




Fig. 105. — Cranks of Paddle-wheel Engine. 

and the thrust of connecting-rod is 125,000 lbs., /being of steel and taken at 
9,000 lbs. 



Diameter 



-v/- 



,000 X u 



9,000 



X 51 = 9*97 inches. 



The crank-arm (fig. 105) is to be treated as a lever, so that if a is the 
thickness in direction parallel to the shaft axis, and b its breadth at a section 
x inches from the crank-pin centre, then 

Bending moment M at that section = R x x, 

a x 62 M 
6 -/' 

6M 



and 



or 



a = 



62 x/* 

If a crank-arm were constructed so that b varied as v x (as given by the 
above rule), it would be of such a form as to be inconvenient of manufacture, 
and consequently it is customary in practice to find the maximum value of 

* For other conditions as to size of pins, vide p. 309. 



300 MANUAL OF MARINE ENGINEERING. 

b, and draw tangent lines to the curves at the points ; these lines are gener- 
ally, for the same reason, tangential to the boss of the crank-arm at the shaft. 
The bending moment decreases as the distance from the crank-pin 
•decreases, while the shearing stress is the same throughout the crank-arm ; 
•consequently this latter stress is large compared with the bending stress 
close to the crank-pin, and so it is not sufficient to provide there only for 
bending stresses. The section at this point should be such that, in addition 
to what is given by the calculation from the bending moment, there is an 
extra square inch for every 8,000 lbs. of thrust, on the connecting-rod. 
Moreover, the crank-arm is subject to twisting from the action of the pin ; 
strictly speaking, therefore, it should be calculated from the formula 

Tj = M + VM 2 + T*. 

The length of the boss h, into which the shaft is fitted, is from 0"75 to 
1 "0 of the diameter of the shaft, and its thickness e must be calculated from 
the twisting stress R X L. 

The crank turns the shaft (fig. 105) by exerting a force S on the key, 
whose centre of effort is on the circumference, and therefore at a distance 
of half the diameter from the axis of the shaft, so that 

SxB=RxL; orS = 2Rx|. 

If the crank is loose, the area of the section of the key parallel to the 
shaft must therefore not be less than S + 10,000 lbs. And the load on 
the section of the crank-boss opposite the key is 

2L 



Q -S- R=R( 2 ^ - l) 



The stress on *,he section of the boss crossways is T, so that 

T — "t-? = R x L ; or T = 2 R _ L -. 
2 D + e 

The stress on this section should not exceed 9,000 lbs. 

To avoid a complicated expression, it is convenient to assume a relation 

between h and e, and to substitute the value of e thus found in the above 

expression. The value of — in practice varies from 2, when there is not 

much space for the crank, to 3, when there is ample room. 

Example. — To find the section of the boss of a wrought-steel crank 8 inches 
long ; the pressure on the crank-pin is 54,000 lbs., the diameter of the shaft 

10 inches, and — assumed at 2 - 2. Stroke of piston 60 inches. 

Here assume 

8 
e = — = 3 - 67 inches. 

30 
T = 2 x 54,000 iq-tT3^7 = 237,015 lbs. 

Area = 237,015 -v- 9,000 = 26 '33 square inches. 

And since // = 8 inches, e = ^— = 3 3 inches. 

o 



PADDLE-SHAFTS. 301 

The cranks of marine engines are always of steel, wrought or cast, and 
generally of the same materials of which the shaft is made, so that the length 
and thickness of boss may bear a constant relation to the diameter of the shaft. 

When h = D, then e = 032 D. 
„ h = 0-9 D, „ e = 0-34 D. 
„ h = 0-8 D, „ e - 0-35 D. 
„ h = 07 D. .. e = 0-36 D. 

The crank-eye or boss into which the pin is fitted should bear the same 
relation to the pin that the boss does to the shaft. 

Cranks are always shrunk on to both shaft and pin, and when this opera- 
tion is carefully and well done, a key to the latter is almost unnecessary, and 
some engineers have latterly omitted to fit one to even very large pins ; some 
engineers simply drill a hole half into the shaft and half into the crank, and 
drive into it a steel pin so as to answer the purpose of a key. 

The diameter of the shaft end on to which the crank is fitted should be 
1*1 X diameter of the journal. Overhung cranks are never fitted now to 
screw engines, as they often proved to be very unsatisfactory, from the fact 
of the whole of the pressure coming on one bearing, and the whole of the 
bending and twisting stresses being taken by one crank and journal. 

Paddle-shafts. — The cranks of a paddle-wheel engine (fig. 105) are still 
often overhung, and in the case of double engines, the arm to which the pin 
is secured is the one fitted to the intermediate shaft ; the pin fits loosely into> 
an eye on a crank or disc secured to the paddle-shaft, and so drives this latter 
shaft. The effect of this arrangement is to give a very equable strain to the 
paddle-shaft, for the pressure of the pin is always at right angles to the crank 
on the paddle-shaft ; and in smooth water the power of each engine is very 
nearly equally divided between the two wheels, and the bending action on 
the paddle-shait never exceeds half that due to its own cylinder, for when 
near the dead points the bending moment is at its maximum, and is whollv 
taken on the crank-arm to which the pin is secured. For these reasons the 
shaft to which the arm having the crank-pin secured is fitted must be stronger 
than the outer shafts, especially when the ship is intended to work in rough 
water, as it is liable then to have to transmit the whole twisting force of one 
engine, and always takes, during certain periods of the revolution, the whole 
bending force from that engine. Hence, if T be the maximum twisting 
moment from one piston of a double paddle-wheel engine, and M the maxi- 
mum bending moment from that piston, the 

Maximum equivalent twisting moment on the intermediate shaft 

= M + JW + T 2 , 

And maximum equivalent twisting moment on the paddle-shaft 



4+vW^ 



Exception may be taken to the latter, since at times when one wheel is out 
of water the whole of the twisting force of both engines is transmitted 
through the shaft of the wheel which is deeply immersed ; but when the 
maximum combined effect of twisting is on this one shaft, the bending 

M 

moment on the crank- journal is probably less than — -, and is that due to the 



302 MANUAL OF MARINE ENGINEERING. 

force found by dividing the maximum twisting moment by the length of the 

T x J2 
crank, which is approximately T ; the distance at which this force acts 

is measured from the face of the crank-arm to the edge of the casting, into 
which the journal brass is fitted. 

Example. — To find the diameter of intermediate and paddle-shafts of 
a double paddle-wheel engine, having cylinders 80 inches diameter and 
60 inches stroke, using steam of 45 lbs. pressure absolute, cutting off at 
0-6 the stroke ; the distance between the bearing beds being 50 inches. 

Maximum effective pressure in the cylinder will be about 40 lbs. per 
square inch. Hence 

Load on piston - - = 5026 x 40, or 201,040 lbs. 

Maximum twisting moment = 201,040 x 30 = 6,031,200 inch-lbs. 

Maximum bending moment = - — - — -. — = 2,513,000 inch-lbs. 

(1) Maximum equivalent twisting moment on intermediate shaft 

= 2,513,000 + ^2,513,0002 + 6,031,200 2 = 9,056,000 inch-lbs. 
Diameter of shaft 

(2) Maximum equivalent twisting moment on paddle-shaft 

_ 2,513,000 = J(*fim*f + 6>03 ^ _ 7i417>800 inoh . lb . 

Diameter of shaft 



,eyM£^ + 5 . 1 = 16 . 8inohes . 



8000 

The outer part of a paddle-wheel shaft is subject to twisting and bending 
from the reaction of the water on the floats, and from bending due to the 
weight of the wheel itself. The pressure on the float can be found by divid- 
ing the twisting moment by the distance to the centre of the pressure of the 
float from the shaft axis in inches. It is practically sufficiently accurate to 
measure to the centre of fixed floats, and to gudgeons of feathering floats. 

For example, the reaction of the water on the floats of the engine in the 
preceding example, whose radius to float centres is 140 inches, will be found 

.p _ , 6,031,200 .oAoniu 

Pressure on floats = , ' — , or 43,080 lbs. 

140 

The twisting moment on the shaft is the same at the outer bearing as at 
the inner, and is 6,031,200 inch-lbs. The weight of the wheel is 20 tons, or 
44,800 lbs., and the distance of its centre from the bearing is 40 inches. 

* Maximum bending moment 

= (44,800 + 43,080) x 40 inches = 3,515,200 inch-lbs. 

* In smooth water the bending force is really the resultant of the weight and reaction 
on the floats, and may be taken = J weight 2 + reaction 2 . 



CRANK-SHAFT OF SCREW ENGINES. 303 

Maximum equivalent twisting moment 

= 3,515,200 + V3,515,200 2 + 6,031, 200 2 = 10,515,200. 
Diameter of shaft 

. lySflgS „ 5 -l , !M8 inches. 

The outer end of a paddle shaft is subject to alternating stresses due to 
the weight of wheel, and the inner part to intermittent ones (v. Table lxxxiii.). 

The crank-shafts of paddle engines are now made in the same way as those 
of screw engines. Sometimes they are made in one piece from a " solid " 
forging, and sometimes in one piece " built up " (v. fig. 19). Those in very 
large engines have a separate shaft for each cylinder, coupled as in screw 
engines. 

Crank-shaft of Screw Engines. — In case of 'the forward crank of a double 
or treble engine, and the crank of a single engine having two arms, there is 
the action of one engine only on it. On the forward journal and crank-arm 
there is a twisting action sufficient to overcome the friction, and to drive the 
eccentrics if fixed in this part, and half of the whole bending moment due to 
the thrust on the crank-pin. On the aftward journal, the other half of the 
bending moment, and the whole of the twisting moment, except the small 
portion required as above ; this portion is at certain periods of the revolu- 
tions so small, that in calculations for the journals it may be neglected. 

Then equivalent twisting moment on aftward journal 



"♦y®^ 



2 



M 

Strain on forward journal = — . 

In multiple-crank engines the aftward crank has not only to resist the 
action of its own piston, but also to transmit the twisting strains of the 
forward engines. There will be strains from its own piston, which may be 
calculated in the same way as those on the forward cranks, and to these must 
be added the twisting strain of the forward engine. 

Let T 2 be the maximum twisting moment on the after engine from its own 
piston, and M 2 the corresponding bending moment, T x the twisting moment on 
the forward engines at the same period. 

Then on the forward journal of the after crank, the twisting moment is T,. 
M 9 v 

and the bending strain -5—, so that — 

Equivalent twisting moment on forward journal of after crank 

On the after journal of the aftward crank, the twisting moment is 
T 2 + T p and the bending moment -3, so that — 

Equivalent twisting moment on after journal of aftward crank 



f ♦ Ar) * v. 



304 MANUAL OF MARINE ENGINEERING. 

The bending moment on the after-arm of the aftward crank will be found 
by calculating the maximum force on the crank-pin tending to twist the shaft. 

Let T„ be the maximum combined twisting moment, as found by the 
methods indicated before, L the length of the crank or half-stroke of piston. 
Then the maximum twisting force at the crank-pin is T„ ■•- L. 

The maximum bending moment at any section of the after crank-arm of 
the aftward crank, whose distance from the centre of the crank-pin is x 

inches, is—-? x x. 

The maximum bending moment on a section of the forward arm of the 

same crank is — l x x. 

Example. — To find the sizes of the parts of a crank-shaft of a double 
expansive engine of 1000 I.H.P., the length of stroke is 40 inches, the cut- 
off 0-6, and the stroke and the cranks at right angles. Revolutions 60 per 

minute. Mean twisting moment of one engine = -— x 63,000. Since the 

cut-off is 0-6, the ratio of maximum to mean twisting moment is 1-835 
(Table xxxiv. ) ; therefore 

Maximum twisting moment of one engine 

= 1-835 x — x 63,000 = 963,375 inch-lbs. 
60 

Mean twisting moment of both engines 

= l ™ x 63,000. 
60 

Ratio of maximum to mean twisting moments is 1-27 (Table xxxiv.); 
therefore 

Maximum twisting moment of both engines 

= 1-27 x 1229 x 63,000 = 1,333,500 inch-lbs. 
60 

Maximum turning force on forward pin 

= 963 > 375 ^ 48,168 lbs. 
20 

Maximum turning force on aftward pin 

= 1 > 333 > 5QQ = 66,675 lbs 
20 

Assuming the distance between the bearings on which the brasses are 
bedded to be 30 inches, 

The maximum bending moment on each of the two forward journals 

= 48 ' 168 u X 3 ° = 180,630 inch-lbs. 

8 

That on the two journals of aftward crank 
_ 66,675 x 30 = 250)0()() . nch lba 
8 



TWISTING MOMENT. 305 

Then diameter of foremost journal 



80 ' 0Q x 10-2 = 6-13 inches. 



\r- 



The maximum equivalent twisting moment on after journal of forward 
crank 



= 180,630 + 7l80,630 2 + 963,375 2 = 1,160,630 inch-lbs. 
Diameter of journal 

/l,160,630 



8000 



x 5-1 = 9*04 inches. 



The maximum equivalent twisting moment on fore journal of aftward 
crank 



- 250,000 + 7250,0002 + 963,375 2 = 1,245,000 inch-lbs. 
Diameter of journal 



3 /1,245,000 _ . n __ . , 
V 80Q0 - 5-1 = 9-25 inches. 



Maximum equivalent twisting moment on aftermost journal 

= 250,000 + s/250,000 2 + 1,333,500* = 1,606,700 inch-lbs. 
Diameter of journal 

3 A,606,700 



8000 



x 5*1 = 10*1 inches. 



The aftermost crank-arm will be 1 1 inches across the face ; to find its 
thickness 18 inches from the pin. 

Bending moment at that section = 66,675 x 18 = 1,000,000 inch-lbs. 

Thickness = , 19 '" or L n = 7*44 inches. 
11* x oOOO 

In actual practice the crank-shaft would not be made with the four 
journals all of different diameter, but some engineers make the shafts partly 
in accordance with theory, by arranging the two forward journals of the same 
diameter, and the two aftward journals of the same diameter ; that is, for 
the example given above, the journals of the forward crank would be each 
9 04 inches diameter, and those of the aftward one 10*1 inches diameter. 

Example. — To find the dimensions of the crank-shaft of a single engine, 
whose cylinder is 30 inches diameter and stroke 50 inches, the steam used is 
65 lbs. per square inch absolute pressure, and the cut-off at 0*3 the stroke. 
The distance between foundation facings for shaft brasses is 40 inches. The 
connecting-rod is 100 inches long. Back pressure and loss at piston are 
5 lbs. 

The maximum pressure on the piston is 60 x 706 = 42,360 lbs. 

The maximum twisting moment occurs just at the cut-off in this case, and 
is 42,360 x 24, or 1,016,640 inch-lbs. 

The bending moment on each journal at that period — — tt or 

211,800 inch-lbs. 

20 



306 MANUAL OF MARINE ENGINEERING. 

The bending moment on the after arm, at a distance of 22 inches from the 
centre of crank-pin, is 42,360 x 22, or 931,920 inch-lbs. 

Diameter of fore journal 



*/ 



211,800 .... 

x 10-2 = 6-46 inches. 



8000 
Maximum equivalent twisting moment on after journal 

- 211,800 + N /211,800 2 + 1,016,640 2 - 1,250,200 inch-lbs. 
Diameter of after journal 



V 



f/lZgjF. «M =9-27 inches. 



If the crank-arm is 10 inches wide at the face, then thickness of crank- 

™ . ^ , • 6 x 931,920 _ . . 
arm at 22 inches from pin = -y™ 8000 ~ ' incnes - 

The bending moment at the centre of the pin of a solid or rigidly built 
op crank-shaft is n — . 

Note. — A solid shaft, or one whose continuity of strength is unbroken from end to 
end, is treated, so far as bending stresses are concerned, as a girder or beam secured at 
its points of support ; or as a continuous girder supported at several points when there 
are more than two journals. Hence, the bending moment in the middle between two 

"R. v T Ft x L 

journals is — - — ; and at the points ofsx^pport also — ^ — , since change of flexure takes 

T 

place at a distance — from the supports. Marine crank-shafts whose arms are at least 

0*7 of the diameter thick, and whose bearings thoroughly support the shaft close to the 
crank -arms, are really subject to little or no bending action at the journals. 

At and near the end of the stroke the crank-arms are subject to a bending 
moment applied very suddenly when the steam enters the cylinder, on opening 
to lead after only slight compression. The force should be taken at twice 
the load on the piston (2 P), and if L x is the length of the pin -f- the thickness 
of the crank-arm as close to the shaft, then 

„ ,, , 2PxL x PxL, 

Bending moment = - L =s — - — -. 

4 Ji 

Crank-shafts are subject to intermittent loads, and therefore to inter- 
mittent stresses (v. Table lxxxiiL). 

And the bending moment on the section of each arm caused bv this force 

PxL 

is — -r — -, and acts at right angles to the bending force, due to the force on 

the crank tending to twist the shaft. Hence, 

. JPXL,) 3 P x Lj 



BUILT CRANK-SHAFTS. 



307 



In the previous example L x may be supposed to be 23 inches, and P is 
42,360 lbs., a is 7 lbs. ; then 



b = 



3x 42,360 x 23 
2 x 1- X 8,000 



= 3 - 6 inches, 



so that the forward crank-Arm must not be less than this thickness at 
any part. 

Crank-shafts for mercantile screw engines, when above 10 ins. diameter, are 
generally made in duplicate pieces, so that in case of damage to one only a 
part of the shaft is condemned, and a spare piece can be easily carried on 
foreign voyages. And also by this plan there is less labour in replacing the 
damaged part, than if the whole shaft is moved. 

Built Crank-shafts. — Shafts above 10 inches diameter are better built up 
than in one forging, and they can then be made of steel at much less cost 
than a solid one ; indeed, many engineers now make crank-shafts of all 
sizes on this principle. The crank-arms are usually of the same thickness 
at the pin as at the shaft, and equal to 0*7 to - 8 of the diameter of shaft 

I 





— 




j 





• — 






i 


1 






1 

i 

-r~- 

i 
i 


j 













r 






"\ 




Fig. 106.— Built up Crank-shaft. 

journals ; the end view, as in fig. 106, shows the usual shape for large 
cranks — smaller ones are often straight on the sides. Great care is 
required properly to construct such a shaft so as to be perfectly true 
when finished, and to have the arms shrunk on sufficiently tight without 
leaving the metal around the pins, and shaft-ends in such a state of tension 
as to be dangerous. 

The thickness of the metal around the shaft, etc., can be calculated, as 
before stated for the overhung crank. 

The crank-arms are sometimes forged with the shaft-ends, and the pins 
shrunk into eyes in the arms. This method has advantages, but it is very 
unsightly, and misses one of the chief merits of a built crank-shaft. Another 
arrangement is to make the crank-pin and arms in one piece, generally a 
steel casting, and shrunk it on the shaft-ends or shanks ; and sometimes 
it is secured to the latter by flanges, bolts, etc. 

There are a number of other patented forms of crank-shafts, some having 
the crank-arms of cast steel, and some of forged steel and iron, so arranged 
a? to couple the shafts at the cranks instead of between them. 



308 



MANUAL OF MARINE ENGINEERING. 



Fig. 107 shows a piece of crank-shaft as generally made for fast-running 
engines, and indicate the modifications made when further weight is to be 
saved, as in naval and other high-speed ships. 

Couplings. — It is usual now to have the coupling forged with the shaft 
instead of keyed on as formerly. The tube shafts of twin-screw engines, 
however, generally have one coupling keyed on. As a rule, the only stress 
to which a coupling is subject is due to twisting ; hence, if t be the 
thickness of the flange, and r the distance of any part of it from the centre 
of the shaft which is subject to a twisting moment T : the section of metal 
resisting the force is 2 n r t ; and if / be the stress per square inch on this 
section, acting at the distance r . then 

T 

T = 2 cr r t f x r = 2 t r- . f . t, that is, thickness of flange = .-: =-> 

J J ° 2 t r-f 

If r is the radius of the shaft subject to twisting only, so that -q-/is 
equal to T. Then 



Thickness of flange = 



irr' 6 f 



3*i*f, or T 



Ci 





/ Section, ( 




"7 


P®®»r 


I 


< 




— 








• 




r 





Fig. 107.— Naval Crank-shaft. 



From practical considerations the thickness of the flange should not be 
less than the diameter of the coupling bolts, and since the strength of a 
coupling is somewhat impaired by the holes drilled for the bolts, it should 
be about '27 the diameter of the shaft subject to twisting only. 

Coupling Bolts. — When shafts are close coupled, and the bolts are a good 
fit in the holes, they are subject to a shearing force only, caused by the 
torque on the shaft ; hence, if d be the diameter of the bolts, whose 
number is n, K is the distance from centre of bolts to centre of shaft, T the 
twisting moment, and D the diameter of the shaft subject to twisting only. 
Then 



T = n—r—/ x K ; or d 



V o- 



T 



7854/. K.r* 



but 



Hence, 



tD 3 



16 



Diameter of bolts = -^ . / ^. 

2 \ n \ K. 



SURFACE OF CRANK-PINS AND SHAFT-JOURNALS. 309 

If K is always taken at 0-8 x D. Then 

Diameter of bolts = — . / zr— . 

'2 V 0-8 x n 

Then when there are 5 bolts, 

Diameter of bolts — -~ */ = j—z = t • 

2 v x - o 4 



If 3 bolts, diameter of bolts = diameter of shaft 



„ 5 

., 6 
„ 7 
„ 8 
., 9 
„10 


»i 
>> 




»> 


»t 


»» 


>i 


it 


i» 



of shj 


ift -=- 3-10. 


»» 


3 58. 


>> 


4 00. 


ii 


4 38. 


»> 


4 73. 


>» 


5 06 


>» 


5-37. 


11 


5 67. 



The number of bolts in a coupling depends sometimes on circumstances, 
but usually there should be two, and an allowance of one more for each 
2 inches of diameter of shaft, and the above proportions are based on this 
allowance ; but when it is necessary to have the couplings as small as possible 
the number may be increased, and with the consequent decrease in diameter, 
the centres of bolts may be nearer to the centre of shaft. 

The couplings of a two-crank engine, whose shaft is in duplicate halves 
at right angles, should have four or a multiple of four bolts ; and for those 
of a three-crank engine, whose shaft is in three duplicate pieces, the number 
of bolts must be a multiple of three. 

With shafts of mild steel, the bolts should be of a harder kind ; indeed, 
as they are subject to shearing stresses only, they may be made with 
advantage of a steel 40 to 45 tons tensile strength. 

Surface of Crank-pins and Shaft-journals. — Measuring, as in the case of 
gudgeons and crossheads, the effective bearing surface as the diameter multi- 
plied by the length of the bearing, the bearing surface of crank-pins was 
such that the pressure per sq. in. did not exceed 500 lbs. ; this, however, is 
"hardly sufficient for the pins of high-speed engines, which should be such 

, 6,500 
■that the pressure per square inch does not exceed ,- _ , _ = lbs. — that is, 
r r n v n + 100 

if d is the diameter of the crank-pin, and I its length, L the maximum 
recurrent load on piston, R the revolutions per minute, then 

. . T 6,500 WlTflOO 
(ZXJ = L "-v/R + 100 ; ° r ~^500 . 

"When the brass is recessed, so that it bears only in parts on the shaft, the 
actual bearing surface should not be exposed to more than 600 lbs. pressure 
per square inch. 

The pins of paddle-wheel engines, owing to the comparatively slow speed 
of shaft, may be designed, if necessary, to take a pressure of 800 to 900 lbs. 
per square inch. 

The main bearings in which the crank-shaft runs should be such that the 
pressure never exceeds 600 lbs. per sq. in. in paddle engines, and in screw 
•engines it should not exceed 400 lbs. The main bearings of screw engines, 



310 MANUAL OF MARINE ENGINEER TNG. 

when room admits, should be such that the nominal pressure in lbs. per 

■ , , 4 > 300 

sq. in. does not exceed -, p ■ /, measuring the whole of the bearing. 

If, then, d be the diameter of the journals and / the length of each — 

dxl - 2 X 4,300 * 

The length of the crank-pin is from 1 to 1*5 of the diameter, and that of 
each journal from 1*2 to 1*5 the diameter of the journal. Vertical engines 
have usually sufficient space for a crank-pin 1*25 the diameter, and each 
journal 1*5 the diameter. The foremost journal of a compound engine is 
often made much shorter than the others, to allow the eccentric sheaves to- 
be nearer the centre, so as to come in line with the valve-rod. 

Owing to the comparatively small pressure on the crank-pins and journals 
of three-crank compound, and triple- and quadruple-compound engines, the 
intermediate journals may be generally somewhat shorter ; but the first and 
last journals should not be materially less than was usual in a two-crank 
engine. 

Drivers. — In order to avoid any of the thrust of the propeller coming on 
the crank-shaft and its bearings, the coupling-bolts connecting it to the 
thrust-shaft are sometimes made without heads, so that they are free to move 
in and out of the holes in the coupling flange of one of these shafts while held 
firmly in that of the other. When this is so they should be of larger diameter 
than the ordinary coupling-bolts, and the part projecting from the face 
of the flange into which they are secured, should be larger still, so as to form 
a shoulder, against which they may be tightened up, and give the necessary 
strength to resist bending. The faces of the flanges when thus loosely coupled 
should be from j to J inch apart. It was found necessary, generally, to pro- 
vide means for lubricating these drivers, especially in heavily armoured and 
high-powered warships with single screws. Drivers are still necessary in all 
ships which are liable to change shape from varied loading, as in cargo steamers 
in heavy seas, and in long steamers of light build. As a rule, however, it 
is better to connect the crank and thrust-shafts in the ordinary way with 
bolts, etc., and to leave such clearance in the bearings and brasses that the 
crank-shaft may have some longitudinal " play." 

Taper Bolts are often used in lieu of the ordinary parallel ones, especially 
for the shaft next the propeller-shaft, to facilitate their withdrawal. Taper 
bolts can be used with advantage when the flange is by necessity small, for 
the screwed end is much smaller than the part at the junction of the two- 
shafts subject to shearing. These bolts also are necessarily a tighter fit in 
the hole, since the tightening of the nuts draws them farther into it. 

Cross-keys are sometimes fitted to couplings. Half the key is bedded 
into a recess in the face of each flange, and so it takes the shearing stress 
from the bolts. 

Since with a number of bolts or drivers it is possible, from wear or bad 
workmanship, that the load is taken only by a part of them, it is usual to- 
provide an excess of strength. This provision can be conveniently effected 
by proportioning them to the diameter of the crank-shaft as if it were subject 
to twisting only. 



PROPELLER-SHAFTS. 311 

Propeller-Shafts,* sometimes called "screw" shafts, and sometimes "tail- 
end " shafts. The propeller shaft is subject always to the torque of the 
engine, and to bending due to the weight of the propeller. In rough weather, 
when the ship is pitching, the strains are increased and often become 
very severe ; for when the screw is partially immersed, in addition to the 
twisting moment by the reaction of the water acting on one side only, there 
is a bending stress as on a paddle-shaft ; besides which the momentum of 
the screw when pitching also adds severe bending stresses, all of which are 
moreover alternating. 

In still water the bending moment on the shaft is the weight of the screw 
multiplied by the distance of its centre from the stern bush. To provide 
for the strains in rough weather, the bending moment should be taken at 
twice this value ; for ships which may cross the Atlantic at any time in 
ballast trim, even this is insufficient to produce a shaft sufficiently large to 
last a satisfactory length of time, inasmuch as the material of a shaft working 
under such conditions is subject to alternating stresses of considerable intensity, 
which tend to degrade it and render it unfit to resist such stresses as come 
on it. The diameter of the screw is also an important factor in determining 
the size of screw shafts of ships subject to rough weather. 

Hence, if T is the maximum twisting moment on the crank-shaft. \V the 
weight of the propeller in pounds, and L the distance of its centre from the 
stern bush, 

Maximum bending moment = 2 W x L ; 
and 

Maximum equivalent twisting moment T\ 

= 2 W x L + n/(2W x Lp 7~T2; 

and as before, 

/t 

Diameter of screw-shaft = I ' — ~ x 5'1. 

V / 

Example. — To find the diameter of the screw-shaft for an engine whose 
maximum twisting moment is 1,333,500 inch-lbs. The weight of the screw 
is 6000 lbs., and the distance of its centre from stern bush is 20 inches. 

The max. bending moment 

= 2 x 6000 x 20 = 210,000 inch-lbs. 
The max. equivalent twisting moment 

= 240,000 + n/240,000 2 + 1,333,500 2 = 1,594,000 inch-lbs. 
Diameter of shaft 



=y- 



1,594,000 



x 5*1 = 11*2 inches. 



6000 

It is such a very serious matter when the screw-shaft breaks, that it 
should always be of ample size, and for ships" in the Atlantic trade it should 
be specially strong. It was usual to make it the same diameter as the crank- 
shaft, but in some ships even this is not sufficient, and it is now not at ail 
an unusual thing to make them 20 per cent, stronger than the crank-shaft. 

Where the screw is fitted in a " banjo " frame for lifting above water when 
the ship is under sail, the shaft is, of course, nearly wholly free from bending 
stresses. 

* The outboard shafts of naval ships and large express steamers are generally hollow, and for stilfncis 
are of large diameter, with the hole as much as 0-7 x the external diameter. 



312 MANUAL OF MARINE ENGINEERING. 

Outer Bearing. — It was formerly customary to provide an outer bearing 
on or in the rudder-post, for the extreme end of the screw-shaft to rest upon ; 
but since the rudder-post practically gives no support sideways, and a very 
precarious one in any direction, the practice is now obsolete. Also, it was 
found that when ships so fitted touched the ground with the heel, the screw- 
shaft was often bent, and sometimes dangerously so. The strongest argu- 
ment in favour of this outer bearing is, that it prevents the loss of the screw 
when the shaft is broken ; but if the shaft is broken, and the ship has to 
depend on the sails, it is better perhaps to be without the screw ; and if the 
shaft is broken diagonally, as is often the case, and the screw is caused to 
revolve from the motion of the ship, there is great risk of splitting the stern 
tube. If there is no outer bearing, and the fracture is well within the tube, 
the screw will not be lost, but will go back until the shaft-end butts against 
the rudder-post, and revolves then without danger. If the shaft breaks close 
to the propeller, and there is an outer bearing, the danger of damage to the 
rudder-post and rudder is very great indeed, from the wrenching of the bearing 
on the propeller falling out. 

Screw-shaft End. — The shaft end fitted into the screw boss should be 
turned to a taper of f inch to the foot ; if the taper is less than this, as was 
sometimes the case, extreme difficulty is experienced in getting the screw off. 
The screw should be secured by a key extending the whole length of the 
boss, and driven into place after the screw is thoroughly well driven on. The 
screw is retained in place by a nut, the screw-thread of which is the reverse 
hand of that of the screw itself. A tail key through the shaft end was pre- 
ferred by some engineers as a means of retaining the screw in place ; but 
although it is a very safe plan, it is not so convenient as the nut. When a 
nut is employed a safety key or pin is fitted in rear of it (v. fig. 173), or else 
a set-screw or other simple means of locking it is used. These nuts are now. 
often of cast steel or bronze, and made so as to cover the end of the shaft. 
When the propeller is of bronze a " cap " nut of the same material is 
necessary. Such nuts are secured by set-screws in the end " out of centre," 
or set-screws in the boss-fitting in recesses in the side of the nut (v. fig. 174). 

Screw-shafts are encased with bronze from the propeller to the inner end 
of the stern tube in H.M. Navy ; in merchant ships this is now frequently 
the case, although objected to partly on account of the expense, and partly 
because it prevents examination of the shaft and the detection of flaws, which 
may extend unobserved until rupture takes place. Bronze casings, on the 
bearing parts, are used not so much to protect the shaft from corrosion, as 
to provide a better wearing surface when running on lignum vita?, and to admit 
of wear without weakening the shaft thereby. Lloyd's Register now require 
all casings to be the full length of the stern tube. 

When working in sandy water lignum vita? wears very quickly and grinds 
away the bronze casing. Ships which are often exposed to this are better 
without the bronze casing and the lignum vita?, and should be fitted in lieu 
with Fenton's white metal, and the shaft either without casing or cased 
with a hard steel liner, which may be renewed when worn. 

The Stern Bush should be of such length that the pressure per square 

inch (measured as stated for bearings) does not exceed — . lbs. This 

v & v/R 4- 100 

bush has to sustain the weight of the propeller and that of a considerable 



THE STERN BUSH. 313 

portion ol its shaft ; in a seaway it has also added to its load that due to 
inertia. The above allowance, however, for smooth-water conditions provides 
a large enough bush for sea-going ones. If d is the diameter of the shaft 
over the bronze casing, d 1 = - 9 d that of the shaft under it ; I the length 
of the bearing part of the bush, all in inches, and W the weight of the propeller 
in pounds, and the length of shaft, whose weight is borne by the bush, as 
15 diameters, or 15 d t = 13*5 d. 

Weight of shaft = ^ x 15 d x x 0-28 = 3*3 tf x 3 , or 2-4 dK 

Then total weight on bush = W + 2*4 r/ 3 pounds. 

Then lx dx . 5Q ° = W + 2*4 d 3 , 
s/R + 100 

and I = {-j + 2*4 d*) X >-^— • 

In a general way this total weight on the bush = 1*3 W for solid shafts, 
and 1*225 W for hollow ones. 



TV, i a *i. t. -Wv'E+100 

Then, length of bush = ^ 1 , 

° ¥ X d 

where for solid shafts F = 385, and for hollow shafts F = 408. 

Example (1). — What should be the length of the bearing strips of a stern 
bush for a ship whose screw weighs 25,000 lbs., revolves at 100 per minute, 
and the shaft is 17 inches over the liner. 

T . /25,000 . -, , 1Q .\ v'200 „ nQ . , 
Length = f — ^ (- 2*4 X 289 ) -^r- = 60-a inches. 

Example (2). — For a turbine-driven ship the screw is 6,600 lbs. in weight, 
the shaft 12 inches over the liner, revolutions are 500. 

T . 6,600 x s'600 

Length = 409 x 1 J~ = 33 mches - 

The stern bush in practice is of a length equal to three to four times the 
diameter of bore. The stern-shaft should be supported on a bearing in the 
tunnel when possible ; when this is either not possible or inconvenient, it 
should rest on a bush in the stern tube just abaft the stuffing-box. 

In the mercantile marine the screw-shaft, when partly cased with bronze, 
and running on bushes fitted with lignum vita?, the brass casings should 
extend from the screw boss to an inch or two beyond the inner end of bush, 
and should do so also where the shaft passes through the stuffing-box. and 
inner bush when there is one. Of late years there has been a tendency to 
follow the Admiralty method of casing in the whole length covered by the 
stern tube. In order to make the process as easy and inexpensive as possible, 
the stern tube is made very short, so that the casing is very little longer 
than the sum of the lengths of the two liners as formerly fitted. 

The Thickness of Brass Casing at bearing parts is 0-3 inch + 0*035 x 
diameter. In order to easily withdraw the shaft, the diameter of the inner 
casing should be £ inch larger than the outer. 

* Lloyd's Register requires it to be at least 4 diameters in length. 



314 MANUAL OP MARINE ENGINEERING. 

Stern Tube. — In the Navy the stern tube is always of bronze, and is within 
another steel tube secured to the framing of the ship. The lignum vita? 
strips are fitted into the tube, either in separate grooves for each strip, or the 
strips fit in side by side with a brass strip at the top secured by screws to 
the tube which keys them so as to form a bush. A similar but shorter set 
of strips is fitted next the stuffing-box. The brass stern tube fits into 
the stern-post or spectacle piece accurately and tightly, and is secured to 
the bulkhead by a flange, etc.* 

In the mercantile marine the stern tube is nearly always of cast iron, 
whoso thickness is 0*5 inch + # 08 the diameter. The stern bush of bronze 
fits accurately into the tube, and the stuffing-box neck ring and gland are 
lined with brass. There are various ways of fitting the stern tube in place. 

The common plan adopted by most engineers is to turn the outer end so 
as to fit accurately and tightly into the hole bored in the stern forgings or 
castings, and secure it by a nut screwed on its end. The inner end has a 
flange, and next it a projecting rim, which is turned to fit in the hole bored 
in the bulkhead ; the flange is bolted to the bulkhead after a liner is fitted 
between them (v. fig. 108). 

The tube was sometimes bolted to the stern-post by two lugs cast with it, 
one above and one below it. 

Another plan of fixing the stern tube is to fit ifcs outer end into a recess 
bored in the stern-post, and secure it by bolts to the bulkhead as before 
described, and by two strong draw-bolts passing through the flange to a 
partial bulkhead two or three frame spaces nearer the stern. In this case 
the stern bush is partly in the tube and partly in the stern-post. 

Stern Bushes. — When made of white metal they should be of a thick- 
ness = 05 inch -(- 0"03 x diameter. Those fitted with lignum vitae are of 
bronze, and formed with a flange, which is secured to the stern tube by screws, 
which prevent it from turning or coming out. Stern bushes are often made 
of strong cast iron, with white metal strips driven in, or with it run in as in 
a main bearing. Such bushes have flanges like the bronze ones. The lignum 
vita? is sometimes fitted in strips, as in stern tubes, and sometimes into 
square holes, the bush being cast as a skeleton to hold the wooden blocks. 
This latter plan is very convenient for small ships, but not a good one for 
large ones, as the wood by the continued concussion gets impressed on the 
cast-iron tube. Lignum vitse wears best in end grain, especially when it is 
of inferior quality ; when cut from a good tree of large size it wears equally 
well either way. 

Lignum vita? and white metal strips should be from f-inch to f-inch 
thick, and about three -to four times their thicknesses in breadth; they 
must be bevelled so as to leave free watercourses between them. The brass 
behind the strips should be 0'04 X diameter in thickness, and the metal 
ridges between the wooden strips of the same thickness. Sometimes the 
bronze bush is dispensed with, and the strips fitted into the cast-iron tube 
as in the brass tube, but this is not good practice. 

A pipe is fitted leading from the top of the stern tube to the bulkhead, 
through which the water may run from the tube so as to cause a fresh supply 
to enter from the sea, and thereby prevent heating (v. fig. 108). 

* It is usual, however, to fit a separate bush crntaining the lignum vitae strips in all 
but small ships for convenience of repair. 



f .V-'- ' ' ' * * ssss s 







315- 



5.^25351 



e 

o 



5 



- 



u 

«2 

■a 
a 

*- 
a 
a, 



u 

4> 

6 



e 

s 
E* 



c 



O 
CQ 



CO 

o 



OK 



•316 MANUAL OF MARINE ENGINEERING. 

From time to time attempts have been made to exclude sea water from 
the stern tube, and lubricate the shaft bearings with oil or soapy water. 
Ordinary stuffing-box, gland, etc., have been fitted to the tube end, and the 
lubricant fed through a pipe from the upper deck, so as to give a " head " 
superior to that of the sea water, and so prevent any leakage inward. No 
great degree of success has followed this plan, as no provision is made for the 
wear of the bush. But if the bush did not wear the violent action of the 
screw in a seaway would inevitably cause leakage. This difficulty has been 
got over by Mr. Cedervall, who prevents water from entering by the means 
shown in fig. 108, which have been generally successful in practice. 

Thrust-shaft. — Although the crank-shaft was sometimes made with a 
■collar or collars on it, to take the thrust, it is not good practice, especially 
for large engines. The crank-shaft should be required to take only such 
loads and motions as are due to the direct action of the pistons, and be free 
to move around in its bearings without end pressure ; and since any longi- 
tudinal displacement of the crank-shaft tends to throw abnormal strains 
•on the working parts, it is better to remove all causes of such a derangement. 
To this end the thrust collars should be on one of the intermediate shafts, 
and for convenience on that one next the crank-shaft. If possible the thrust 
bearing should bean the engine-room, and it is for this purpose chiefly that 
the collars were sometimes on the crank-shaft. 

Thrust. — To find the thrust along the shaft of a screw engine, it is neces- 
sary to know the speed of the ship and the effective horse-power. The 
•effective horse-power is in this case the power actually employed in pro- 
ducing thrust, and, of course, its relation to the indicated horse-power depends 
•on the combined efficiency of the engines and propeller. This ratio may be 
taken as 0*77 with the best high-speed engines, and - 68 with the ordinary 
merchant steamers' machinery. With turbines it is probably 08. If P be 
the pressure in pounds exerted by the propeller against the thrust bearing, 
.and S the speed of the ship in feet per minute, then 

Work done in moving the ship = P x S, 
and therefore Effective H.P. = PxSt 33,000, 

P X S 



or I.H.P. x/ = 



33,000' 



+X. T, T TT T, 33,000 - 

then P = I.H.P. x -~ — k/. 



Now, if K be the speed in knots, 
Then 



„ v 6080 



p = i.h.p. x 326 K X £ . 

J? is called the mean normal thrust. 

Thrust may be ascertained fairly accurately from the formula — 

i t> D* x VA x V 2 * 

Thrust in pounds P = 5 — X tr. 



THRUST. 317 

D is the diameter in feet, A the acting surface in square feet, V is = pitch X 
revolutions per second, G = 0*4 to 0"5, and P r is the pitch ratio. 

Example. — To find the thrust on the shafting of an engine whose I.H.P: 
is 2,000, and the speed of 'the ship 12 knots, the efficiency 0*66 

P = 2,000 X 326 * °' 66 = 30,166 lbs. 

Now, it will be seen that P varies with the I.H.P., and inversely as the 
speed, so that the thrust of a particular screw may vary very considerably ; 
for if from some cause the speed is decreased, without a corresponding decrease 
in the power, the thrust must of necessity increase. This actually occurs 
in practice, and must be provided for always. The times when the actual 
thrust exceeds the normal thrust are when the engine first moves, and its 
power is employed in overcoming the inertia of the ship, also when the ship 
is towing, and when driving against a head wind or sea. It is also to be 
noted that the speed of ship means speed through the water ; for it is on this- 
account that so little strain comes on the moorings of a ship whose engines 
are working at full speed when in a dock or confined piece of water. In this 
case it is only at first starting that any great tension is thrown on the moor- 
ings, for as soon as the water is set into motion so as to flow past the ship in 
a steady stream, the power is absorbed in facing the stream and really pro- 
pelling the ship through the water. 

Another cause of variation in the thrust is the variation in the twisting 
moment, which is, as before shown, very great in certain classes of engines. 

The surface exposed to thrust may, however, be calculated from the mean 
normal thrust, and allowance made for all emergencies. This surface should 
be such that the pressure per square inch from the mean normal thrust does 
not exceed 70 lbs. ; and for tugboats or ships especially exposed to severe 
weather, or service analogous to either of these, it should not exceed 50 lbs. 
Ordinary merchant ships have usually such surface that the nominal thrust 
does not exceed 60 lbs., and naval ships 45 lbs., as they may have to tow 
or do analogous work at full power. 

Friction Loss at the Thrust Block is not so great as generally supposed 
by those who devise and patent special means for its reduction. As a matter 
of fact, with carefully turned steel shafts with multiple collars running 
against good white metal well lubricated the loss is never more than 1 '5 per 
cent, of the I.H.P. transmitted by the shaft, and generally is as follows : — 

The ordinary merchant steamer, . . . 04 per cent. 

Express steamers and large naval ships. . . 0'5 ,, 

High-revolution steamers and naval ships. . 0'65 ,, 

Turbine steamers of large size, . . . 0"70 ,, 

,, high revolution, . . 1*0 to 1 "3 per cent. 

The following rule holds good for all : — 

Loss per cent. = 0*05 vrevs. per min. 

♦Pressure per Sq. In. on a Thrust Bearing should not exceed 2,700 + VRrf+100, 
R being the revs, per min and d the diameter of thrust-shaft in inches. 

Diameter ot Thrust Collars. — Let P be the mean normal thrust, d the 

* With Michel bearings the pressure may be as much as 17,500 + y/Hd + 100. 



318 MANUAL OF MARINE ENGINEERING. 

diameter of the shaft, and D the diameter of the thrust collars, whose number 
is n, and the pressure 60 lbs. Then 

P _ 60 (li^! - I*) n = 47 (B 2 - d 1 ) 

and 



n 



°= y 



2 + 



47 n 
The thickness of each collar for mere strength 



»s» 



= --=- (rrd x 1000) = 



n v ' 3142 dn 

In practice the thickness of each collar = 0*4 (D — d). 

(1) Space between the collars, if rings are of solid brass = 0*4 (D — d). 

(2) Space between the collars, if rings are of cast iron faced with bronze or 
white metal = 0*75 (D — d). 

(3) Space between the collars, if rings are of hollow bronze for water to 
circulate through = D — d. 

The number of collars depends very much on the size of the engine and 
the prejudice of the designer. If there are many collars, they are of necessity 
somewhat small, and although the chances are in favour of the majority of 
them acting efficiently, provision must be made for the contingency of the 
whole thrust coming on only one of them, and the larger the number of 
collars, the less able is each one separately to resist the whole thrust. The 
chief objection to a few collars is, that they are of necessity of comparatively 
large diameter, and have, therefore, a greater resistance to turning owing to 
the longer radius, besides a higher speed of rubbing surface ; there is also 
the consideration of cost of forging against large collars. On the other 
hand, when there are a few large collars a better design of thrust-block is 
possible, and the rings can be made adjustable without removal. 

The number of collars should vary with the size of the shaft, and a very 
good rule is, that there should be one collar for shafts up to 5 inches diameter, 
and then an additional collar for every 1*8 inches of diameter beyond this 
when the collars are large, and the engines slow running ; fast-running naval 
engines require more than this to get sufficient surface, say an additional 
collar for each additional inch and a quarter. That is, 

A 5 

Number of collars for naval and express = 1 -f- 



mercantile engines = 1 -f- 



1-25 
d-o 

1-8 



Michel's Thrust-block, shown in fig. 109, is based on the principle followed 
by this engineer in all rubbing surfaces, whereby perfect lubrication is obtained 
and freedom from metallic contact of surfaces ensured. This arrangement 
requires one shaft collar only ; on each side in the block is a horse-shoe-shaped 
inverted collar, each containing six segments of suitable 'metal — bronze or 
white metal — loosely held and pivoted out of centre so as to tip slightly but 
sufficiently to admit the passage of oil freely as the shaft revolves and main- 
tain oil films between the surfaces, which are sheared continuously. The 
inventor claims that this resistance to shear is constant and about J lb. per 



THRUST-BLOCKS. 



319 



square inch, whatever the thrust pressure may be ; the heat generated is, 
therefore, 

B.T.U. = number of square inches X speed -fr- 2 X 778. 





Fig. 109. — Self-contained Michel Marine Thrust Bearing. 

In practice the coefficient of friction is only 0-0015, and a pressure of 500 lbs 
per square inch is not excessive under ordinary working conditions. With highly 
viscous oil us much as 5 tons per square inch has been maintained satisfactorily. 
Fig. 110 is the form of block generally used. Here the horse-shoes fit 




L_ 




Fig. 110— Thrust-block with Adjustable Collars. 

over two screwed bars, one on either side of the block; nuts are fitted to 
these bars, so that each collar may be adjusted by its own nuts, or the whole 
of them by the nuts at the end. A simpler and cheaper method is to turn 



320 MANUAL OF MARINE ENGINEERING. 

these side bars so that there is a collar between each pair of shoes instead 
of a pair of nuts — that is, the bars have a set of collars corresponding to those 
of the shaft — and take the thrust of the shoes as before. 

Both these plans are most successful in practice, in great measure 
due to the fact that the collars are open and exposed at the top, so 
as to be easily lubricated and cooled by the air, and to their running 
in oil. or in a mixture of oil and soapy water contained in the trough 
below them. 

It is most important that a bearing be placed close to the thrust, 
so that the shaft cannot vibrate and cause uneven pressure over the 
surface of the collars. The function of the thrust bearing? is to take 
only end pressure. This is particularly the case when designed with 
horse-shoe rings. 

The Length of the Bearings of Tunnel Shafting will depend on the size of 
the shaft and their distance apart. If d is the diameter and L the distance 
in inches, the weight of the shaft which each bearing has to support is 

W = jXLx 0-282, or 0-22 rf 2 xL 

The bearing surface is taken at I X d square inches, and the pressure 
should be about 60 lbs. per square inch in smooth water when the rate of 
revolution does not exceed 100 per minute ; taking 

* 

Pressure per square inch = 800 -s- vrevs. + 100. 

800 
Then 0-22 d* x L = d X I X 



</R + 100' 

, „ . , , , . rfxLx Jr + loo . , 

and Length of bearing =-. inches. 

Example. — What should be the length of tunnel shaft bearings in a turbine 
ship having shafts 6 inches diameter running in bearings 10 feet apart at 
800 revolutions per minute ? 

t -T, 1 1 • 6 X 120 X x/800 + 100 _ ... . . 
Length of bearing = ■ = 0*93 inches. 

Diameter of Shafts, Rules for, in Practice. — As almost every ship is classed 
with one or other of the shipping registers, the shafting must in no case be 
smaller than provided for in the rules of the corporation with which she is 
classed, and from whom a machinery certificate is necessary. All British 
ships carrying more than 12 passengers must have a certificate from the' 
Board of Trade, whose surveyor has to report that, inter alia, the shafting is 
fit and proper, etc. ; in his case it means that the diameters are not less than 
given by the Board's Rules. 

The Admiralty now generally specify the sizes of the various shafts, 
which are almost invariably hollow steel. 



BOARD OF TRADE RULES. 



321 



Board of Trade Rules for Shafts. 

For compound condensing engines with two or more cylinders, when the 
cranks are not overhung : — 

~T5* 



s 



3 / C x P x PS 



p = 



/ X S« 



D«\ 



8" ( n D»\ 



Where S = diameter of shaft in inches. 

d 2 = square of diameter of high-pressure cylinder in inches or sum 
of squares of diameters when there are two or more high 
pressure cylinders. 
D 2 = square of diameter of low-pressure cylinder in inches or sum 
of squares of diameters when there are two or more low- 
pressure cylinders. 

P = absolute pressure in lbs. per square inch — that is, boiler 
pressure plus 15 lbs. 

C = length of crank in inches. 

f = constant from following table. 

Note — Intermediate pressure cylinders do not appear in the formulae. 

For ordinary condensing engines with one, two, or more cylinders, when 
the cranks are not overhung : — 



S 



-V E 



x P x D 8 



P = 



3 x / x S 3 
C x D 2 ' 



3 x/ 

Where D 2 = square of diameter of cylinder in inches ; or sum of squares 
of diameters where there are two or more cylinders. Other symbols as above 

TABLE XXXV. — Board of Trade Factors for Shafts. 



For Two Cranks- 
Angle 
between Cranks. 


For Crank and Thrust Shafts. 


For Tunnel Shaft. 
/ 


For Propeller Shaft. # 
/ 


90° 
100° 
110° 
120° 
130° 
140° 
150° 
160° 
170° 
180° 


1047 

966 

904 
For paddle engines of ""' 

ordinary type multiply „ . _ 

constantin this column 7fi8 

suitable for angle of ^..^ 

cranks by 1*4. ij.i 

743 

740 


1221 
1128 
1055 
997 
953 
919 
894 
877 
867 
864 


890 
821 
768 
727 
694 
670 
651 
638 
631 
629 


For Three Cranks. 
120° 


1110 


1295 


943 



Note. — When there is only one crank the constants applicable are those in 

the Table opposite 180°. 

* The portion of the propeller shaft which is forward of the stern gland, and all 
the thrust shaft, with the exception of the part enclosed in the thrust bearing, may be 
the same diameter as the intermediate tunnel shafting. 

21 



322 



MANUAL OF MARINE ENGINEERING. 



Lloyd's Rules with regard to Shafting. 

The diameters of crank and straight shafts are to be not less than those 
given by the following formulae : — 

TABLE XXXVI.— Diameters of Shafts— Lloyd's Rules. 



Description of Engine. 



Compound — Two cranks 

at right angles, 
Triple — Three cranks at ' 

equal angles, 
Quadruple — Two cranks 

at right angles, 
Quadruple— Three cranks, 

Do. — Four cranks, 



Diameter of Intermediate Shaft in 


Inches. 




(•04 A+ 006 D+ -02 S) x 3 JF. 
(038 A + 009 B + -002 D + 0165 S) x 

(034 A + -011 B + '004 C + "0014 D + 

(•028 A + -014 B + 006 C + "0017 D + 
(•033 A + -01 B + -004 C + "0013 D + 


•016 S) x 

•015 S) x 

•0155 S) x 


VpT 



Where A is diameter of H.P. cylinder in inches. 
B „ first LP. „ 

C „ second LP. „ 

D „ L.P. „ 

S is stroke of pistons in inches. 
P is boiler pressure above atmosphere in lbs. per square inch. 



), where P is the diameter of the propeller 



The diameter of the crank-shaft and the thrust-shaft between the collars to 
be at least f £ of that of the intermediate shafts. The diameter of thrust- 
shaft may be tapered off at each end to the same size as that of the inter- 
mediate shafts. 

The diameter of the screw-shaft to be equal to the diameter of intermediate 

shaft multiplied by ( - 63 -\ — ^ 

.and T that of the intermediate shaft, both in inches. In no case, however, 
must the diameter of the screw-shaft be less than TOT T. 

Note. — This size of screw-shaft is intended to apply when continuous brass liners 
are fitted the whole length of stern tube. If no liners are used, or if two separate ones, 
then the shaft must be ? J of that given by the Rule. 

Lloyd's Register Rules for Oil Engine Shafts. 

The diameter of the shafts of oil engines for driving screw propellers may 
be determined in the following manner when made of ordinary mild steel and 
the maximum pressure in cylinders does not exceed 500 lbs. per square inch : — 

(1) For petrol or paraffin engines for smooth-water service, diameter of 
crank-shajt in inches = C ^D 2 X S. D is the diameter, and S the stroke 
in inches. C = 0-34 ; for engines with six cylinders C = 0-36 ; if for open 
sea work add 0-02 to each. 

(2) Single-acting Diesel type engines : — 

Diameter of crank shaft = N^D 2 X (A S + B L). 
Here L is the distance apart of bearings (inner edges) ; the values of (A S+B L) 
are as in this table. 



BRITISH CORPORATION RULES. 



323 





TABLE XXXVII.— Crank Coefficients. 




Four-cycle Single-acting 
Engine. 


Two-cycle Single-acting 
Engine. 


Value of Coefficient, 
AS + BL. 


a 
b 
c 
d 


4 or 6 cylinders. 

8 
10 to 12 „ 
16 


2 to 3 cylinders. 

4 
5 to 6 

8 »> 


•089 S + -056 L 
•099 S + -054 L 
•111 S + -052 L 
•131 S + -050 L 



For auxiliary engines values 5 per cent. less. 

In solid forged shafts, breadth of webs 1-33 diameter, and thickness 
0-56 diameter of shaft or of equivalent strength. 
(3) When no flywheel is fitted — 

Diameter of intermediate shafts = coefficient X ^D 2 x S. 
The value of this coefficient is as follows : — 



TABLE XXXVIIa. 



Four-cycle Single-acting. 


Two-cycle Single-acting. 


Value of 
Coefficient. 


4 cylinders. 

6, 8, 10, or 12 cycles. 

16 cycles. 


2 cylinders. 

3, 4, 5, or 6 cylinders. 

8 cylinders. 


0-456 
0-436 . 
0-466 



N.B.— When the 
stroke is not less than 
12 nor more than 16 
diameter of cylinder, 
then 



Diameter of internal shaft = coefficient (735 D + "273 S). 
(4) When flywheels are fitted, then coefficient : — 

TABLE XXXVIIb. 



Four-cylinder 

Single-acting 

Engine. 


Two-cycle 

Single-acting 

Engine. 


Value 

of 

Coefficient. 


Four-cylinder 

Single.acting 

Engine. 


Two-cycle 

Single-acting 

Engine. 


Value 
>i 

Coefficient. 


4 cylinders. 
6 " „ 
8 „ 


2 cylinders. 

3 „ 
4 


0-405 
0-400 
0-409 


10 cylinders. 
12 „ 
16 


5 cylinders. 

6 „ 

8 „ 


0-420 
0-427 
0-461 



(5) Diameter of screw shaft 



= t(o- 



63 + 



0-03 P 



)• 



-), and never less than 



107T; T being diameter of intermediate shaft and D that of propeller in inches. 
The Board of Trade Rules for Shafting of Oil Engines is the same as the 
above rules of Lloyd's Register. 

The British Corporation Rules for Shafts. 

Diameter of Shafting. — The minimum diameters of crank, thrust, pro- 
peller, and intermediate shafts may be found from the following formulae, 
except where the ratio of length of stroke to distance between main bearings 
is unusual, when they will receive special consideration : — 



»-y 



P X L 2 x S 



C 



X B. 



Where D 
P 
S 
L 
B 
B 
B 



diameter of shaft. 

absolute pressure — i.e., boiler pressure + 15 lbs. 

stroke of engine, in inches. 

diameter of low-pressure cylinder, in inches. 

1 "0 for crank- and thrust-shafts. 

- 95 for intermediate shafts. 

for propeller shafts to be taken from the following Table 



324 MANUAL OF MARINE ENGINEERING. 

TABLE XXXVIII. — British Corporation Factors for Shafts. 



Coefficient of 


i 
Ratio of Diameter of Propeller to Diameter of Crank-shaft. 


Displacement of 








Vessel at four-fifths 














Moulded Depth. 


13 


14 


15 


16 


17 


18 


•6 


10 


101 


102 


103 


1-04 


105 


•62 


101 


102 


103 


1 


04 


105 


1 


06 


•64 


102 


103 


1-04 


1 


05 


1-06 


1 


07 


•66 


103 


104 


105 


1 


06 


107 


1 


08 


•68 


104 


105 


1-06 


1 


07 


1-08 


1 


09 


•70 


1-05 


106 


107 


1 


08 


1 09 


1 


10 


•72 


1 06 


1-07 


1-08 


1 


09 


110 


1 


11 


•74 


107 


1-08 


1-09 


1 


10 


1-11 


1 


12 


•76 


1-08 


1-09 


1-10 


1 


11 


1-12 


1 


13 


•78 


109 


1-10 


111 


1 


12 


113 


1 


14 


•80 


110 


111 


112 


113 


114 


115 



L2 



The value of the divisor C in the formula depends on the ratio t™, 

where L = diameter of low-pressure cylinder and H of high-pressure 
cylinder, in inches : — 





Two Cranks at 90°, Compound 






L 2 


or Quadruple, also Three 


Three Cranks at 120°, 


Four Cranks at 90°, 


BP 


Cranks at 120°, Quadruple 
Expansion. 


Triple Expansion. 


Quadruple Expansion. 


Ratio 3 


9,910 








i H 


10,160 


••• 


.. 






, H 


10,410 


••• 


,. 






, 3§ 


10,660 


• •• 


•.. 






, 3£ 


10,910 


•• . 


•• 






, 3f 


11,160 


••• 


• • 






. H 


11,410 


•• • 


a# 






, H 


11,660 


••• 


,. 






, 4 


11,910 


••> 


• • 






, H 


12,160 


• a • 


.. 






, 4£ 


12,410 


• •• 


.. 






. 4g 


12,660 


... 


.. 






, *i 


12,910 


13,650 


.. 


. 




, H 


13,375 


14,160 


.. 






, 5 


13,840 


14,670 


•• 






, 5i 


14,305 


15,180 


.. 






, 5* 


14,770 


15,690 


.. 






, 5f 


15,235 


16,200 


,, 






, 6 


15,700 


16,710 


.. 


i 




. 6i 


16,630 


17,730 


,, 






, 7 


17,560 


18,630 


,. 






. 7* 


18,410 


19,530 








> 8 


19,260 


20,430 


22,'66( 




, H 


20, 1 10 


21,330 


23,660 




, 9 


20,960 


22,200 


24,660 




, 9* 


21,750 


23,070 


25,660 




, 10 


22,540 


23,940 


26,580 




, ioi 


23,330 


24,810 


27,500 




, 11 


24,120 


25,660 


28,420 




, Hi 


24,900 


26,500 


29,340 




, 12 


25,680 


27,340 


30,260 



shafts for screw engines. 325 

Rules of the Bureau Veritas. 

Shafts for Screw Steamers. 

(a) Crank Shafts. — § 7. When the crank of a screw engine is not over- 
hung, the diameter of the shaft shall be determined by one of the following 
formula? : — 

For non-compound condensing engines : « 

, 3 / n P L D2 ... 

<*- V— C - - - (A) 

For double, triple, and quadruple-expansion engines : 

• d= y PLKD-f 0-lnD'X . (B) 

For shafts having a single overhung crank, the form under the radical 
sign is to be multiplied by 

8 + 7* 2 + 1. 

For two-cylinder single-crank tandem engines the formula will therefore 



be 



= V pL(D» + 0-1D«)(« + J«» + l) . (C) 



In those formulae : 

d = diameter of the after shaft bearing in inches. 
% = number of high-pressure cylinders. 

D 1 = diameter of each high-pressure cylinder in inches. If there are 
several high-pressure cylinders the diameters of which are not 
the same, n x D x 2 represents the sum of the squares of theii 
respective diameters. 
n = number of low-pressure cylinders. 

D = diameter of each low-pressure cylinder in inches. If there are 
several low-pressure cylinders the diameters of which are not the 
same, n D 2 represents the sum of the squares of their respective 
diameters. 

N. B. — For triple or quadruple-expansion engines the intermediate cylinders 
do not come into account in the formula?. 

L = length of stroke in inches, common to all pistons. 

P = boiler pressure above atmosphere in pounds per square inch. 

s = - (see fig. 110a). In order to determine a, B is supposed to be situated 
half-way the length of the bearing, unless the latter be longer 
than lh times the diameter ; in this case B C may be considered 
as being equal to f of the diameter. 
C = a constant, the values of which are given below for certain cases. 
The values given apply to navigation in a sea-way ; for smooth 
water (tugs excepted) the constants may be increased by 
30 per cent. 
If it is above 15 inches, it should be increased by an amount to be 
determined by the Administration ; for built-up shafts, however, this latter 
increase will not be required. 



326 



MANUAL OF MARINE ENGINEERING. 



For hollow shafts the diameter must be increased by 
1 per cent, if the diameter of the hole is 0*4 of the outside diameter. 



» 
»» 



>> 



10 



0-5 
0-6 
0-7 



eF 



l_ 



'0"'-'<0i 



-•!•—£- 






>> 










>» 










Q] 


ider 


0-4 


of 


the 





increase 


will 


be 



V 



— K 



Fig. 110a. 



If the hole is 
outside diameter, no 
required. 

The Administration mav allow a 
reduction on the diameter in certain 
special cases, for instance in well-balanced 
engines with light moving parts, or for 
very superior workmanship, etc. On the 
other hand, the Administration may re- 
quire an augmentation for engines which 
differ much from the average propor- 
tions found in practice, thus, for instance, 
for engines having a comparatively small stroke ; for compound engines, the 
low-pressure cylinder of which has a very large size compared with the high- 
pressure cylinder, etc. 

Values of Constant in Formula (A) is 6,230 ; in (B) for double-expansion 
two-cranks 90° it is 3,400 ; for triple three-cranks 120°, 3,900 ; for quadruples 
and triples four-cranks 90°, 4,000, and for special crank setting for minimum 
torsion add 100. 

Other Cases. 

(6) Propeller, Tunnel, and Thrust Shafts. — The diameter of the crank-shaft 

must be ( 1*7 -j — 15 J per cent, in excess of the diameter of the crank-shaft 

calculated from one of the formulae (A), (B), or (C) ; D being the diameter 
of the propeller in inches, and d the diameter of the crank-shaft in inches. 
It is recommended to fit the propeller shaft in such a way that it cannot 
move endways, if for some reason or other it has been uncoupled from the 
rest of the shafting. Liners fitted on propeller shafts to be tapered off at 
ends. 

For tunnel-shafts a reduction of 6 per cent, on the diameter of the crank- 
shaft will be allowed. 

The diameter of thrust shaft at the bottom of the collars, both between 
and immediately beyond these latter, to be equal to that of the crank- 
shaft, and tapered off at each end to the smaller diameter of the body 
of the shaft. The thrust of the screw propeller must be taken up by 
an efficient thrust block, so as to prevent any fore and aft strain on 
the crank-shaft. 



Shafts for Paddle Steamers. 

§ 8. In side wheel steamers having double, triple, or quadruple-expansion 
engines with an intermediate shaft, each end of which carries an overhung 
crank-pin fitting loosely into an eye of the paddle-shaft crank, the bearing of 



HORSE-POWER TRANSMISSIBLE THROUGH SHAFTS. 



327 



00 




K 




O 




(n 




O 




a 




O 




i—t 








o 


t~ 




o 


H 


CD 

M 


« 


e3 


o 


fa 
O 


CO 


fa 


W 


3 


a 


O 


h-« 


r^ 


pa 
05 


>> 


H 


00 




r^ 


>H 


t^ 


n 


6 




CO 


W r / 


^J 


£ w 


S3 


cS 


3 5 


fa 

o 


dqM 


ID 




fa 


gEH 


J3 


P fe 


>> 


o o 


.o 


Ph 


„ 


gpq 


6 


M 


>> 


H H 


-O 


►J M 

pa te 

CO 


>> 


&, 


NSMIS 
CCORD 




2 <j 


(3 


« 


3 


H fc 


O 
ft 


« 


ti 


H 


O 


£ 


U 


O 


J4 


PM 


e. 


■ 


s 


H 


o 


CO 




Ph 


o 


O 


is 


a 


H 



X 
X 
X 

fa 
w 













too-woioaocoo 




o 


«iO"035©OTl>t- 




o 










00 


<nt»<-*io©ooc5 — ico 




OOCO'MOOOOt^^TjIO 




o 


©"■*OOOiCO©-*<iO-H 




o 






c~* 


<^fcocoT^^'coI>c»05--< , 




e»t»^iooiQe33eoesi<NO 




o 


Tt"C0-*Olt^<MCOCOU3-? 




o 






© 


NMn^Toe t^»o>« 




HHioan-HOOMOo 




o 


G5CO-*<ai-*O5C0— 'Oi-^CO 




o 


t>lO00"*CO— < © -h Ol © CO 




K3 


PH -I 




«00OffiH«h9^ii0OO 




o 


FHt^O^OOCCWfflfflMOO 




iO 






<* 


*»■.».■•■..»■«■«■»* * * * * * ■ ' 




-HCNcMCOCOtJiiOOi>05<MO 






— — 




coiooaowiooooHooo 




o 


C0OJt^C5LOr^O100C0C0I>r~© 




o 


T(lON^-t-iWfflO00'*10'M *. * 




■>* 


■iNOINeo^l^lOfflX'H't* 




• 
*«HO)aM«h(St»iiOOO 




o 


i0t»O^rtCStMt»ONWiC«0Q 




«a 








a 


CO 


-iHrtNM«i<*iOt-0»iflffl 


p 




^* h^ ~h ^ 




loaot^cit^Ttioscat^co— iooo 


S 


g 


t^-ioaooiOrtehwowoo 












CO 


rt«HNMMM'*T|ifflct)OMO 


go 
S3 




»* HH 




oiooicsccioi— >oooo5ooo©-hoo 


t-l 


o 


fflO'MTtnOTjlrtlO'^'HOrtOOCSOt^ 


&< 


IQ 


OCri'idit--iiOOW-iOHrtno;oM 


p 


(M 




H) 


— < — I — CNCNCOCO'^lOC-a)— iCOt-O 


O 

P4 




HWHfl 




I'NceOiOt^^TjiaiOWWrtiMOtHW^t^ 




O 


HrtMOWM-H^i-HWMOfflOffli'lOn 




o 


fO-H^tHO^iXM^I^NH-HCBMMt-IO 




CM 






^_H-H-HCNcNCNeO'*»Oi>05-HCO'»05iricO 






pHfH rH-H NN 




Mioteeaffi'fffl^MxiHtt^ooNMO 




>o 


iMOQ«o»i-<xoe-t>»ao»ooiN 




t> 


CBMO)NCI> i-H "* © 00 © CO © I> ©_ OJ © ©_ N . . 




i— * 


HHiHHCM 




r-COCO©-*<t>T*llO©— < i— itO00lO©COO5'#C0©© 




O 


MMWOOlN-HlOXHOOtMCSOMOtOOTllO 




ia 


ot-xoffi>oa-H'*nM^i»M?)wiooeiioio 




^* 


rt rt iHiH£>Jcjfn^"io"ox"o"cw"i > '''o!'««d' 

^H f— I ^H i— I f— i CM CM 




ooxpHioxco-Hx^o^iooaoffl-HCffloo 




LO 


*t»rtI»ft>rtl>rH65X10000XO!MXOt» 




<N 


■<ii»r5i>oo©c>4ioi>©i>i-~u5©©ii';-H©'Mio©© 






r-i" i-h h i-T of of co •* io «o ao © of tjh" © © of 






X^OOMXfl!NC»t*t»TiiMt»OC»M(00500 




O 


o w o o a -i o ?i «3 o o i< ia a o ■* m- !s a o 




O 


O! ^ 13 M» O »[* ffl f) M 3 ffl a M rt 9 M«) ffl o 






,-T ph h h of of co' ■* iq so oo" ess' — ' co" 10" t-" 




050!>-*|>^t^t^-*©©COTtlOO©(ML0 05<M©© 






ofiNN^cBoteiiisonrtao-isnist-ii 




WM*OOl > "0)OMCOi-t>^rt-H-«N100)t»N 










HrtHM«M^l3C8t«00a- iCO 






1— H — H 


« 


r- 


O 13 O W O 13 O 13 


2S 


S^ 


O'<ot»i»xxe»c s-HNcorcocotHxao-Hfi 




n 



328 



MANUAL OF MARINE ENGINEERING. 



the latter (see A of the following sketch) must have its diameter calculated 
from the formula : 



3 / P L {n~Df + 0-1 n D 2 ) (s + Js 2 + 1) 



(D) 



where the letters have the same meaning as before (§ 7), except that a, for 

determining s = -, is to be measured as shown in the sketch below, the point 
r 

B being the middle of the bearing. 



Fig. 1106. 

For two-cylinder compound receiver engines with two cranks at 90°. 

C = 13,000 for navigation in smooth water. 
7,100 for coasting vessels. 
5,700 for sea-going vessels. 

For triple-expansion engines with three cylinders and cranks at 120°. 

C = 14,900 for navigation in smooth water. 
8,150 for coasting vessels. 
6,540 for sea-going vessels. 

The diameter of the outer bearing of the paddle shaft and of the inter- 
mediate shaft to be submitted to the Administration or the Surveyors for 
approval. The same applies to other cases not dealt with in this paragraph. 



Turbine Shafting (Bureau Veritas) 



Diameter of tunnel-shafts 



XT p 

70x-^=(^ 



R 



H.P. is the shaft horse-power, R the corresponding revolutions per minute. 
The rotor shaft diameter = 1 "05 X dt at least. 

diameter of screw in inches 



Diameter of screw-shaft = dt + 



160 



SHAFTS OF OIL ENGINES. 



329 



Shafts of Oil Engines {Bureau Veritas). 

Diameter of crank shaft = C \/D 2 x S, 

where D is the diameter of cylinder, and S the stroke of piston in inches 

S 8 

C is a factor which varies inversely as =., and is 0*493 when ^ is 2*0. and O560 

when it is only - 85. — is frequently T5. Then C = 0-520. 

Steel Shafts (Bureau Veritas). 

§ 9. Shafts made of steel must have a tensile strength of 26 to 30J tons, 
and test pieces cut from the forging must satisfactorily withstand the pre- 
scribed tests. When it is desired to make use of steel of softer or harder 
quality, the corresponding alteration in the diameter must be submitted 
to the Administration (of the Bureau Veritas) for approval. 

Summary. — The following is a summary of certain parts of this Chapter 
which will be useful for reference : — 



Rule 1. — Diameter of shaft 



V ihr 

V revoiutio: 



oiutions 



X F. 



The various values of F can be ascertained from the following table, where 
p is the absolute initial pressure, or that at which the safety valves are loaded 
-+ 15 lbs., when there is no reducing valve : — 



TABLE XXXIX.— Values of Factor F in Formula for Shafts 

d 



-^ 



H.P. 



R 



x F. 









F for Crank 


F for Tunnel 








Shaft. 


Shaft. 


Single-crank expansive 


1 cyl. 


Cut-off, 0.2 stroke, 


170 + 3Vp 


135 + 3\ p 


)J M 


1 „ 


„ 0-G „ 


113+ 3Vp 


85+3 v'p 


„ compound 


2 „ 


„ 0-G „ 


100 + 3Vp 


75 + 3 V / 


Two-crank 90° expansive 


2 „ 


„ 0-2 „ 


86 + 3Vp 


70 + 3\> 


)» if 


2 


„ 0-6 „ 


68 + 3 Vp 


55 + 3\p 


„ compound 


2 „ 


„ 0-6 „ 


63 + 3 Vp 


50 + 3Vp 


„ quadruple 


4 » 


„ 0-6 „ 


58 + 3Vp 


45 + 3Vp 


Three-crank 120 c compound 


3 „ 


„ 0-5 „ 


51 + 3Vp 


40 + 3Vp 


„ triple 


3 „ 


„ 0-6 „ 


41 + 3Vp 


30 + 3 v'p 


Four-crank 90° triple 


4 „ 


„ 0-6 „ 


47 + 3Vp 


35 + 3 Vp 


„ quadruple 


4 „ 


. „ 0-6 „ 


55 + 3 Vp 


38 + 3Vp 


Three-crank triple-compound naval 


engines (hollow) . , 


72 


63 


Four-crank „ 


M 


n ( « ) • 


75 


61 


Turbines taking S.H.P., . 


• 


. . • • 


• ■ 


56 



330 MANUAL OF MARINE ENGINEERING. 

The Diameter of the Propeller Shaft of a screw ship can be obtained with 
a close approximation to the correct size by adding to the diameter of the 
tunnel-shaft an allowance based on the diameter of the screw and the con- 
ditions of working. If d be the diameter of the screw-shaft at the outer 
bush, d x the diameter of the tunnel-shaft or that to resist torque only, both 
in inches, and D the diameter of the screw in feet. Then 

d = (d x -f- x D) inches. 

The value of x for a single screw ship working in smooth water only is - 06. 

For ocean-going ships with single screws, . . . x = - 075. 

For twin-screw ships with outer brackets, . . . x = - 09. 

For Example. — An ocean-going single-screw ship has a propeller 18 feet 
diameter, and the tunnel-shafts are 13 inches diameter, what should be that 
of the propeller shaft. 

Diameter of propeller-shaft = 13 -f 0-075 X 18 = 14-35 inches. 

Example. — A twin-screw steamer having stern brackets and a screw 
12 feet diameter, has tunnel-shafts 11^ inches diameter, what size should be 
the outer propeller-shaft ? 

Diameter shaft = 1125 + 0-09 x 12 = 12-33. 

The crank- and screw-shafts of ships which are run at full speed only 
occasionally, and for short periods, may be calculated by taking F at half the 
above values. 

Example. — To find the diameter of the crank-shaft for a three-crank 
triple-compound engine having cylinders 30 inches, 45 inches, and 75 inches 
diameter, and a stroke of 50 inches. The load on safety valve is 165 lbs., 
the revolutions 90, and the I.H.P. 3,300. 

(1) By above Rough Eule. 

Diameter of crank-shaft = ^/^° X 41 +3 N /l80 = 14-39 ins. 

(2) By Board of Trade Rule. 

/ 25 x 180 x 75 2 
Diameter of crank-shaft = 8 / lUQ U + 75*\ = 14-12 ins. 

(3) By Lloyd's Rules. 
Diameter of crank-shaft 

= \* (-038 x 30 + -009 x 45 + -002 x 75 + -0165 x 50) Vl65 = 14-5 ins. 

(4) By British Corporation Rule. 



/Tu 

Diameter of crank-shaft = \ / — — ■,„' nn „" — = 14-3 ins. 



[80 x 752 x 50 
17,300 

(5) By the Rules of the Bureau Veritas. 

' , , .. 3 / 165 x 50 (30 2 + 0-1 x 752) , A .,. 
Diameter of shaft = \/ vJOO ™ im ' 



DETAILS OF CRANK-SHAFTS. 



331 



TABLE XXXIXa. — Shafts foe Paddle Engines. 



Description of Engine. 


Value of F 
Intermediate 
Shaft Journal. 


Value of F 

Paddle-shaft 

Inner Journal. 


Value of F 

Paddle-shaft 

Outer Journal. 


Single-crank single-cylinder, 

Two - crank two - cylinder ; cranks 1 
virtually at right angles, and con- > 
nected by link, - - - - ) 

Two-crank two-cylinder, with inter- j 
mediate shaft ; cranks at right > 
angles, ) 

Two-crank two-cylinder ; solid crank- ) 
shaft ; cranks at right angles, - ) 


••• 

58 

• •• 


80 
53 

50 
55 


100 
65 

65 

65 

1 



For paddle steamers working only in smooth water the above values of F 
may be reduced 20 per cent. 

The above values of F were for iron shafts, but it is better not to 
reduce them when employing mild steel, unless the forgings are from ingots 
of the highest quality. 

Details of Crank-shafts. 

Crank-arms if forged solid with the shaft — 

*Breadth - - »- - = 1*1 x diameter of shaft. 
Thickness - - - = 0'75 
Diameter of coupling - = 2*0 
Thickness „ - - = 027 
Number of coupling bolts = 2 



x 

X 
X 



)) 
>» 



Diameter 



diameter of shaft in inches 
4« + 19 



= diameter of shaft -?- 



10 



Diameter of crank-pins - = 1 to 11 x diameter of shaft. 
Length „ - = 1 to H the diameter of shaft. 

Length of journals - - = 1 to 1£ „ 

The following convenient rules give sizes closely approximating to those found 
in practice, and may be used to obtain the diameter of the shafts, preliminary 
to making a more elaborate calculation. They are, therefore, very useful in 
the initial stages of a marine engine design to enable the designer to get on 
with the work ; but being purely empirical they should be used with some 
caution : — 

d is the diameter of the cylinder in inches. 
H MP 

D ,, ,, L.P. ,, 

S the stroke also in inches. 

F a factor, which for the crank-shaft of ordinary compound engines 
with cranks at 90° is 12 ; and for the tunnel-shafts is 13 ; for 
three-crank triple-compound engines F is 14 for the crank shaft 
and 15-2 for the tunnel-shafts ; four cranks, 13 and 14'2. 



* Breadth* x thickness = 0-9 X 4*. 



332 MANUAL OF MARINE ENGINEERING. 

(1) Ordinary compound engines 



Diameter of shaft = 



d + D+S 



{2) Triple-compound three-crank engines 

d+^+D+S 



Diameter of shaft = 
Taking the same example as before 



F 



Diameter of crank-shaft •= r-j = 14*3 inches. 

14 

Rule for Determining quickly the Diameter of Crank-shaft. — Taking D 
as the diameter of each L.P. cylinder of a compound system, and S the stroke 
both in inches, and p the initial pressure absolute. Then the diameter of 
the crank-shaft may be found rapidly and with close approximation to correct- 
ness by the following : — 

TABLE XXXIX6. — Crank-shafts of Screw Engines. 



Description of Engine. 


Diameter. 
Crank-shaft. 


Two cranks at 90° compound, two cylinders, .... 
Three cranks at 120° compound, three cylinders, . . 
Three cranks at 120° triple-compound, three cylinders, . . 
Four cranks at 90° triple-compound, four cylinders, . . . 
Four cranks at 90° quadruple-compound, four cylinders, . 


1-6 D + S r 
no vp 

2-85 D H- S ,- 
150 N P 

2 D + S i- 

185 ^ 

3 D + S ,- 

205 Vp 
2.5D + S V - 
225 7 



The tunnel -shafts should be 0*95 X diameter of crank- shaft. 

Board of Trade Rule for Shafts of Turbine Ships is as follows : — 



Diameter 



3 /I.H.P. .__ _. 
= ^/ -jj— X 40-2 x C, 



where R is the number of revolutions per minute. 

For the intermediate or tunnel-shafts, . . - . C = T500. 

,, propeller-shafts. . . . . . C = 1*667. 

Hollow Shafts, now so extensively used in special ships, are especially 
suited for the propeller-shafts of multiple-screw ships, inasmuch as they are 
less liable to sag under their own weight, and, therefore, freer from the ten- 
dency to whirl than solid ones would be under the same circumstances. 



DETAILS OF CRANK-SHAFTS. 333 

If the diameter of the hollow shaft is d, and that of its bore d v then 

— A — L * 
The relation of the weights of these shafts will be -™ — "TT 
In practice d x is about a half of d ; assuming this to be so, 

j 2 

Ratio of weight of solid to hollow = ■ * ,j . 

d, = Vd* (1 - 0-0625) = 0-979 d. 

Substituting this value of d s , 

958 
Ratio of weight of solid to hollow = ==» or 1-278. 

That is, the solid shaft of the same strength is nearly 28 per cent, heavier. 

The Outboard Shafts of large naval ships with twin screws and ordinary 
brackets are of large diameter (as much as 28 inches), with a bore-hole 0-7 of 
the external diameter. Under these circumstances — 

(a) d s = ^/ i '- = 0-9W. 

, , . , . , (0-91rf) 2 -828 

(6) The ratio of weights = ^ _ ^ t = -^ Q - 

That is, the solid shaft is 1-6 times the weight of the hollow one ; but in salt 
water the difference is greater. Taking the specific gravity of steel as 7-86 
and salt water as 1-03, then — 

(c) Hollow shaft = d\\ - 049) X 7-86 - l-08eP = 2-98. 

Solid shaft = (0-91cZ) 2 X (7-86 - 1-03) = 5-66. 
The ratio is now 1-9. 



334 MANUAL OF MARINE ENGINEERING. 



CHAPTER XIII. 

FOUNDATIONS, BED-PLATES, COLUMNS, GUIDES, AND FRAMING. 

For the good working of an engine it is essential that the fixed parts, such as 
bed-plates, framing, etc., shall not only be strong enough to resist the strains 
to which they are subject, but rigid and stiff enough to prevent any tendency 
to rack or change of form which would throw abnormal stresses on the working 
parts. With this object in view it is usual to construct such parts of cast iron 
or cast steel ; and from their form and general construction thus enables the 
engineer to manufacture a cheap structure having the necessary qualities. 
But since ordinary cast-iron has not a high tensile strength, and is compara- 
tively unsuited to withstand sudden shocks, Structures made of it cannot 
be so light as if made of wrought-iron or steel ; so that when extreme lightness 
of machinery is aimed at, the framing is usually made of wrought steel, and 
rigidity given to it by cross bracing, etc. This latter system is, of course, 
an expensive one in most engines, and only adopted when economy of weight 
is of more importance than economy of manufacture. Although cast-iron 
framing and bed-plates are undoubtedly cheaper and better for engines 
generally, and continue to be used for all sizes in the mercantile marine, a 
system of construction with wrought iron or steel for very large engines may 
be adopted with advantage, and taking into account cost of patterns and 
risk in casting may not be less economical. 

Steel manufacturers can now, however, produce large and moderately 
complicated castings in steel, and the foundations and frames of express 
engines of large sizes are being made wholly of that material. Increased 
experience in the making of steel castings has no doubt given greater faith in 
their soundness as well as permitted of a lower price, and with improvement 
will be more freely used for columns and foundations (fig. 32). At present 
they can only be used to advantage in large engines owing to the considerable 
thickness still demanded by the steel moulders. 

Bed-plates and Foundations. — Vertical engines are usually built on a 
structure called by these names, but are sometimes known as the sole-plate. 
It contains the main bearings for the crank-shaft, and on it are the facings 
for the columns, condenser, etc., and it used sometimes to contain the water- 
ways leading from the condenser to the pumps. The foundation generally 
contains only the main bearings, and has facings for the front columns only 
when it is bolted to feet on the front of the condenser, so that with the latter 
it forms the engine base. The condenser is in this case lower down, and the 
weight and cost of half the sole-plate is saved. The part of the foundation 
fitting to the feet on the condenser front should be of good depth, the flanges 
strong, notched one into the other, and strongly bolted at top and bottom. 
The transverse parts of the bed-plate, into which the main bearing brasses 
are fitted, are sometimes formed like inverted bowstring girders, and are 



MAIN BEARINGS. 335 

unsupported by the bed built in the ship, but span the space between the fore 
and aft bearers. This is convenient sometimes, especially when the shaft 
must be low down in ships having a good rise of floor, and also for very small 
engines ; but it must be of strong form, and as it depends for strength only 
on the connection to the longitudinal parts of the foundation, it is a most 
convenient form when steel castings are used, as also for very light engines ; 
additional support may be, and often is, given in the case of large engines by 
bolting to the ship at the middle (fig. 114). When this particular style is 
adopted great care should be exercised in designing the foundation, so that 
the transverse portions have a good extended connection to the longitudinal 
ones, especially in the direction of the column bases. If the longitudinal 
parts are flat and straight on the bottom, so as to be in the same plane as 
the rest of the foundation, they may be formed with flanges and bolted to the 
steel seating in the ship, and from it receive support and strength. 

Main Bearings. — The bearings in which the shaft journals run should 
approximate, as far as possible, to a hole through a solid support. If it were 
practicable a hole with a bush of suitable metal in it would form the best 
possible bearing for a shaft ; but since the bearing, however well designed 
and made, is liable, in course of time, to wear somewhat, it becomes a neces- 
sity that there shall be some means of adjusting the brasses, so as to prevent 
the shaft from having too much movement when they are worn.* In the case 
of the vertical engine, the weight of the shaft, and the pressure from the 
piston, act very nearly in the same direction, so that the wear is only verti- 
cally above and below the shaft ; consequently the adjustment is necessary 
only in a vertical direction. The greatest loads on the bearings, however, 
are during the first half of the stroke, and consequently the position of mean 
pressure on the journals is not exactly vertical ; this is also somewhat modi- 
fied on the upstroke by the tendency of the shaft to roll on the surface of the 
brasses, and on the downstroke it is aggravated from the same cause. In 
fitting the brasses for a vertical engine, this should be borne in mind, and 
every allowance made for taking the wear due to these causes. It is of the 
utmost importance for the good working and endurance of a crank-shaft 
that the bearings be rigid in themselves, and that the framework containing 
them shall be rigid enough to sustain them perfectly in line one with another. 
Crank-shafts are more severely tried by the giving or springing of the bearings 
than any other cause, and they were more often broken from want of rigidity 
in the bed-plates and seatings, than from the normal strains from the pistons,! 
so that a shaft may be of ample size to bear the twisting and bending moments 
if properly supported in its bearings, and yet give way after a few weeks' 
work because it is in a light foundation in a weak ship. 

The brasses were usually formed as shown in fig. Ill, and carefully bedded 
into the recesses provided for them in the foundation. At one time it was 
usual to design them with projecting facings, called chipping strips, to avoid 
the labour of chipping and filing the whole of the surface ; this was, however, 
found to be highly objectionable as engines increased in size ; with the in- 
crease of boiler pressure, and consequent increased percussive action due 
to the high initial loads, such an effect was produced on these strips and 
the cast-iron surface on which they were borne, that engineers have gradually 
abandoned the practice ; the planing machines also have rendered such a 
device unnecessary, as it is practically as cheap to fit brasses now to bear 

• With forced lubrication and the bearings shut off from dust, etc., tiicre is in practice no wear. 



336 



MANUAL OF MARINE ENGINEERING. 



over the whole surface as to do so only on strips. The square bottom brass 
is objectionable on two grounds ; one being that it is impossible to remove 
it in most engines without lifting the shaft, and the other that when it becomes 
hot it is invariably distorted, from the variation in thickness of metal, with 
the ultimate result that it is broken through the middle longitudinally. 





Fig. 111. — Crank-shaft Bearing. 



The first of these evils is avoided by making the bottom brass round and 
of even thickness, so that it can be got out when relieved of the weight of 




Fig. 112. — Improved Form of Crank-shaft Bearing. 

the shaft, by being moved around until it is on the top of the journal. The 
second evil is also partly avoided by making it of an even thickness ; but 
this form of brass is often found cracked, and is liable to heat from its want 
of stiffness. Both these brasses, when first heated by. abnormal friction, 



MA.IX-BEARIXG BOLTS. 337 

tend to expand along the inner surface, which is in contact with the shaft ; 
this would open the brass, and make the bore of larger diameter, if not resisted 
by the cooler part near the cast iron, and by the bed-plate itself. If the 
brass has become hot quickly and excessively, the resistance to expansion 
produces permanent set on the layers of metal near the journal, consequently 
on cooling this contracts, the brass closes and tends to grip the shaft ; it will 
then set up sufficient friction to become again hot, and expand sufficiently 
to ease itself from the shaft, when so long as that temperature is maintained 
the shaft runs easily in the bearing. This is the reason why some bearings 
always are a trifle warm, and will not work cool. A continuance of heating 
and cooling will set up a mechanical action at the structure of the brass, 
which must end in rupturing it, just as a piece of sheet metal is broken by 
continually bending backwards and forwards about a certain line. 

This pernicious action of the brass can be prevented by securing it to the 
bed-plate, along its two longitudinal edges, as shown in fig. 112, by an H- 
shaped strip, which holds both top and bottom brasses, so that they cannot 
move in their beds. This method is a very simple one, and has been found 
most successful in engines of all sizes. 

It is also essential that the bearing to be efficient should be rigid through- 
out its whole length ; this is not the case when the brasses have long over- 
hanging ends, which afford little or no support to the shaft. To this end 
it is better, when possible, to extend the bed for the brasses, so as to support 
them over the whole of their length, as shown in fig. 112. 

Caps or Keeps for Main Bearings are very generally made of wrought iron, 
but as stiffness is as necessary as strength, cast steel, as in figs. 113 and 114, or 
even cast iron, as in fig. 112, may be used with advantage in their construction. 
A wrought-steel cap, which may be amply strong, is often far from stiff enough, 
while a cast-iron cap, which is stiff enough for good working, is generally 
amply strong. 

Let d be the diameter of the main-bearing bolts (when there are only two 
to each cap), t the thickness of the cap, and b its breadth, I is the pitch of the 
bolts, all in inches ; / a factor, which for wrought iron and forged steel is 1, 
for cast steel 1 % and for cast iron 2. 

Thickness of cap - - = d . / - x f. 

Thickness of brass at middle = — . / — . 

3 V b 

Main-bearing Bolts. — Each cap is usually held by two bolts, but very 

large bearings have four bolts, two on each side, so as to avoid large bolts 

and heavy nuts, and to distribute the load over the cap. When everything 

is in good order and properly adjusted, the load from the piston should be 

equally divided between the bolts ; but since, from a very slight difference in 

setting of the nuts, the load may come on three, and sometimes even on two 

bolts only, due allowance must be made for this. To meet this it should be 

assumed that each bolt is capable of sustaining one-third the load on the 

piston. If P is the maximum load on the piston in lbs., 

P 
Area of each bolt at bottom of thread = .,-,• 

<V 

22 



338 



MANUAL OF MARINE ENGINEERING. 



For mild steel/ = 5,000 lbs. for small and 7,000 lbs. for large bolts (r. Table 
lxxxiii.). /— - 

Also, Diameter of main-bearing bolt = diameter of cylinder X ^J JL,. 

p is the maximum pressure per square inch, and is, as stated on p. 273, and 
may be taken generally at 0'75 X absolute boiler pressure for the high-pres- 
sure cylinder of a triple engine, to equal the load on the low-pressure piston. 




Fig. 113.— Solid Steel Main Bearing Frame. 




Fig. 114. — Cast-steel Main Bearing Frame for Naval Engines. 

Brasses, so called because formerly always made of brass. — They should 
be made of a metal which itself will withstand wear without wearing the 
shaft journals, and whose surface is such that the shaft runs on it with 
minimum amount of friction. The metal must also possess sufficient strength 
in itself or its shell not to fracture under the percussive loads of the piston, 
and be free from brittleness, so as not to crack when quickly cooled. Good 
gun-metal or bronze possessed all the qualities essential for brasses when the 
journal was. of wrought iron ; but the soft steel as used for shafts will not 
run satisfactorily on bronze bearings. There are, however, other metals 
which have certain of these qualities in a higher degree without having them 



BRASSES. 339 

all. Cast iron is harder than ordinary bearing bronze, and when once worn 
to a smooth surface gives equally good results ; but it is liable to fracture 
from continued shocks and to crack when cooled suddenly. White metals 
offer least resistance, or produce least friction, but most of them are too 
soft to be used alone. Of the patent bronzes there are few which are suitable 
for heavy bearings, and none of them have so far been shown to be much 
superior to good gun-metal. 

White metal carried in a bronze or cast-iron shell is beyond doubt the 
best bearing for all qualities of steel journals, and if kept well lubricated 
there is practically no wear on it or the journals. 

When a bearing is of ample size, properly designed and constructed, well 
lubricated and looked after, it may be of almost any of the white metals. If, 
however, the bearing surface is limited there is a considerable difference in 
the behaviour of the different metals ; but if badly designed and constructed 
even the best metal will give trouble ; moreover, if not properly looked after 
by the engineer, the best metal and the most careful design are of no avail. 

Certain of the white metals have given the best results as a bearing surface, 
and there is every reason for this, inasmuch as they do not cause abrasion 
of the shaft, and if their own surface is injured it will not, as a rule, form 
into tine, hard grit, which grinds both the surfaces, as all the hard bronzes do 
more or less. When white metal is used it is highly important that the 
shaft shall bear wholly on it, and not partly on it and partly on the metal 
containing it. and also that efficient courses for the distribution of the lubricant 
are provided. For heavy loads the tin compounds are the best. 

There are three common methods of fitting the white metal into a setting 
of other metal — (1) By casting it into oblong recesses ; (2) by casting it into 
a large number of small circular recesses ; and (3) by driving strips into 
longitudinal grooves, in the same way as the lignum vita? in a stern bush. 
The last plan is, on the whole, the most satisfactory with large bearings, for 
the strips are well secured, and extend over the whole length of the bearing, 
leaving several oil courses longitudinally, and the shaft bears on the white 
metal only; this also possesses the advantage that a damaged strip may be 
taken out, and a new one re-fitted with ease, and without heating the brass 
and running the risk of distorting it. Cast iron is now very often used as 
a setting for the white metal, and answers the purpose very well indeed, 
being much harder than brass, and thereby affording the softer metal a better 
support. When cast iron is used the shell is generally made thicker than 
when of brass ; and sometimes advantage is taken of this to cast the shell 
hollow, so as to admit of water circulating through it. 

The thickness of " brasses " in the crown depends principally on the 
diameter of journal. 

When of bronze .... =0*10 X diam. of journal + 0*10 m. 

,, cast iron .... =0*15 x ,, +0-15 in. 

When fitted with white metal thickness = O20 X diameter -f (W5 in. 

When fitted with white metal strips as follows : — 

Thickness of strips . . . = O04 X diam. of journal -f- £ in. 

Breadth „ . . . = 0*16 ,, + \ ,, 

Space between strips . . . = thickness of strip. 

Thickness of metal beyond . . = - 065xdiam. of journal if bronze. 

„ 0-12 X „ „ iron. 



310 MANUAL OF MARINE ENGINEERING. 

Columns. — The columns which support the cylinders of a vertical engine 
are subject to alternate tensile and compressive stresses from the steam 
pressure acting on the ends and cover of the cylinder ; to a steady compressive 
stress from the weight of the cylinders, etc. ; and to cross-breaking stresses 
when the ship is rolling and pitching. As a rule, columns, if designed from 
considerations of mere strength only, would not be stiff enough for good 
working. The same reasons, therefore, which decide the using of cast iron 
for foundations strongly influence most engineers to choose this metal for 
columns. The front columns are often made, however, of wrought iron or 
steel, turned smooth (vide fig. 114) ; and all the columns for exceedingly 
light engines are made of wrought steel, well braced together to prevent 
vibration (vide fig. 34). Since cast iron is so superior to wrought iron 
or steel for resisting compression, and so inferior to either for resisting 
tension, a good composite column is formed by fitting a steel tie-bar or bars 
through a hollow cast-iron column, the latter supporting the cylinder while 
the former holds it down. The chief objection, however to both this com- 
posite column and those of wrought steel for large engines is the difficulty of 
getting good attachment to the cylinder ; and since it must be outside the 
cylinder the columns are necessarily far apart, and away from the centre 
line of compound engines. To avoid this difficulty wrought-steel columns 
are formed with a flange at each end like a shaft-coupling ; but even then 
the load on the cylinder is very much concentrated. Columns when of 
cast iron or of cast steel (v. fig. 32) are made of various shapes, and no rule 
can be laid down in favour of any particular form. 

The columns should be so arranged as to support the cylinders and resist 
the reaction on the foundation. Some engineers, in thoroughly effecting the 
latter, completely hide from view the working parts, and make all the bearings, 
etc., very inaccessible (v. fig. 38). They should be so placed at the cylinder 
bottoms that the piston-rod axis is within the lines drawn through the 
extreme points of back and front columns ; and when there are only two 
columns to each cylinder, the front ones are better to be spread out some- 
what, so as to act as struts to resist any tendency to motion of the cylinders 
when the ship is rolling and pitching, and thereby leave the working parts 
more open to view. When there are guides on both back and front columns, 
or when the front columns only have the guides, then this is not possible. 

Back columns are generallv of different form from the front ones, to suit 
the guides and bearings for the pump weigh-shaft of the levers when so 
fitted. 

Some engineers have utilised the back columns as exhaust pipes to the 
condenser, but this is not good practice, inasmuch as the heat of the steam 
causes them to expand, and when the guides for the piston-rods are on them 
the heat is conducted to them with prejudicial results ; this latter difficulty, 
however, was sometimes overcome by placing the guides on the front columns. 
It is also bad practice to expose any important part of the engine structure 
subject to heavy strain, to unnecessary wear, such as wasting of the inner 
surfaces of the casting by corrosion. 

Guide-plates. — In order to have a sound and hard surface for the piston- 
rod slides or shoes to work on, the guide-plates should be separate from but 
secured to the columns ; when so fitted they also admit of adjustment, and 
may be cast hollow, so as to permit of a flow of cooling water through them 



FRAMING. 341 

(v. fig. 32). This is especially needful for large quick-running engines, where 
the speed of piston is very high, and any casual want of lubrication would 
soon cause most serious damage. Cast iron when once worn smooth pro- 
vides a splendid surface for a slide, but if by any mischance this surface 
suffers a little abrasion, it is most difficult to get in good working order 
again, and cannot be relied on to work well until replaned or ground 
smooth. 

The face of the guide plates should have good oil courses cut on it, so 
that the lubricant is well distributed, and they should be cut deep enough 
to retain it and also prevent them being choked with the soapy deposit from 
some kinds of oil. The piston-rod slide should always be provided with 
a comb, which will carry the lubricant from the drip-boxes, and spread it 
over the face of the guide at every stroke ; the plate may with advantage 
have recesses in the upper part to catch the oil so spread, where it is 
retained and stored to trickle down at leisure. 

Framing. — Horizontal engines required a different arrangement of bed- 
plate and framing from that of the vertical type. Trunk and return con- 
necting-rod engines had no sole-plate proper, as the cylinders were connected 
to the condenser casting (fig. 31) by A frames, which contain the main bearings ; 
the trunk engine required no guides, and those for the crossheads of the 
return connecting-rod engine were on each side of the condenser. These 
frames had to be sufficiently strong to take the whole load from the pistons, 
and stiff enough to be rigid under them, otherwise the crank-shaft was 
liable to distortion. The usual form approximated to the letter A, the two 
feet being connected to the cylinder front, one at top and one at bottom, 
in line with the brackets on which the cylinder sat, and by which it 
was secured to the seatings in the ship. Projecting feet were cast on the 
cylinder front to meet those of the frames, so that the connection was 
made with driven bolts. 

Usually there were only three frames to a two-cylinder engine, and four 
frames to a three-cylinder engine, the middle ones being very much stronger 
than the other two, as they were required to take part of the strain of both 
engines. 

Horizontal direct-acting engines have a sole-plate, which connects the 
cylinders to the condense casting, and contains the guides for the piston- 
rods, and the brackets for the main-bearings ; these latter are usually stayed 
to the cylinder tops by tie-bars through cast-iron struts or by steel tie-rods 
only. 

The framing for the older diagonal paddle-wheel engines was made some- 
what on the same principle as that for the horizontal screw-engine, with 
modifications (v. fig. 19) to suit the altered conditions. The main part of these 
frames must extend from the cylinder to shaft and down again, so as to form 
a support for the latter, and having guides for the piston-rod crossheads. 
Intermediate supports or columns connect this main frame to the foundation. 
It is now usual to make these frames of steel bars (vide fig. 20), and for some 
very light draught steamers of large power frames made of steel sections 
and plates have been found considerably lighter than the ordinary frames, 
and can be designed to add materially to the stiffness and strength of the 
ship in the neighbourhood of the machinery (v. fig. 22). A modified form of 
this kind of frame is imperative, when exceedingly light draught is a necessity, 



342 MANUAL OF MARINE ENGINEERING. 

as the hull is so light, to comply with the requirements, as to be unable by 
itself to stand the racking strains from the engines. 

Even when weight is of secondary consideration, the wrought-steel frame 
is preferable to the cast iron, and when the cost of patterns is taken into 
account, it is no more expensive. The bearings for the shaft, and the guide- 
plates for crossheads, are, of course, of cast iron or cast steel fitted to the 
wrought-steel work direct. 

Side stiffness and stability are obtained by connecting the four frames 
by wrought-iron tie-bars through the top (v. fig. 22), and by the main beam 
before the shaft. The main-bearings for the shaft of a diagonal engine are 
sometimes so set that the shaft can be lifted vertically ; but a better plan is 
to set them at a slight angle, so that their centre line is in the direction of 
the resultant of the iveight of shaft, etc., and mean pressure on the journals 
due to the thrust on the connecting-rod. 

Entablature of Oscillating and Steeple Engines. — This is usually of cast 
iron, but may with advantage be made of wrought-steel plates and angles, 
or, better still, of cast steel, as the strains on it from the overhung cranks 
are severe and concentrated owing to its being supported at so few points. 
It was no uncommon thing formerly to find it broken and patched after 
only a few months' use, and not many work without a certain amount of 
give-and-take, which must tend to produce rupture in course of time. 

It is usually supported (fig. 27) by four columns to each crank, and addi- 
tional stiffness and stability are imparted by diagonal cross braces to the 
foundation at each end. It is seldom possible to place the supporting 
columns in line with the main girders of the entablature, but when possible 
this should always be done, so as to avoid the canting action which is caused 
by the centre of support not being in the same plane with the centre of force 
on the journals. When this is not possible, the sides of the entablature 
connecting the main girders or rockers should be of extra stiffness, and well 
connected to them by spreading out webs or fillets. Special advantage 
should also be taken of the main beams worked into the ship, to form a 
powerful tie to the entablature girders, and to prevent their tendency to 
canting. This can be done generally by multiplying the number of the bolts, 
and fitting cast-iron filling pieces in lieu of hardwood ones only. The bottoms 
of the cross-bearers should, when possible, be tied together by bars parallel 
with the crank-shaft to keep them from " giving " or twisting. In the case 
of steeple engines (fig. 28) the thrust of the connecting-rod on the guides 
caused great stress on the entablature and framing, and necessitated diagonal 
ties between it and the foundation. The horizontal force in an oscillating 
engine also puts cross or horizontal strains ou the entablature which the 
ordinary columns are not fitted to resist, hence the diagonal cross frame 
which Penn generally fitted. 

To resist, as far as possible, the tendency to spring, the supporting columns 
should be of extra size, with strong and broad Manges. 

When there are four supporting columns of wrought steel, their diameter 
in the body should be - 7 the diameter of the piston-rod. and at each end 
0-55. 

The collars or shoulders on which the entablature is supported, should be 
equal in diameter to that of the piston-rod, and 02 the diameter of the 
piston-rod in thickness. 



ENTABLATURE OF OSCILLATING AND STEEPLE ENGINES. 343 

If the columns happen to come in line, or nearly so, with the centre of the 
shaft journal, they may be 10 per cent, less in diameter than given above. 
The breadth of the rockers should be not less than the diameter of the shaft 
journals, and the depth at the centre should be calculated as for a box-girder, 
subject to sudden loads applied at the middle of its length, which is measured 
from column to column. 

Roughly speaking, the depth of the rocker under the bearing brass should 
not be less than four times the diameter of the piston-rod for engines of 
ordinary dimensions. 

The thickness of metal of sides of rockers 



= 0-4 \/diameter of piston-rod. 

Thickness of top and bottom = 0*6 ^diameter of piston-rod. 

The bottom brasses of the main bearings should be round, so that the 
recesses for them may tend to strengthen the rockers rather than weaken 
them, as would be the case if square-bottomed. 



344 MANUAL OF MARINE ENGINEERING. 



CHAPTEK XIV. 

.THE CONDENSER. 

The function of the condenser is to cool the steam on leaving the engine, so 
as to reduce its pressure to a minimum ; in doing this the steam is converted 
back into water, 'ine very early engines could only work by the aid of 
condensation, as the steam with which they were supplied was generally 
of a little lower pressure than the atmosphere ; it is, in fact, owing to this 
that the steam-engine owes its birth, for steam was preferred by the early 
mechanicians because it was so readily changed from vapour to liquid, and 
so produced that vacuum which Nature was supposed to abhor, and to fill 
which she would perform the work of horses. The proper relation of the 
condenser to the engine is better understood by following the early history 
of the steam-engine from the day when the cooling water was admitted to 
the cylinder containing steam, and then allowed to run freely away from the 
bottom on the descent of the piston, to the time when Watt, having per- 
ceived the waste of energy in always forcing the piston up against the atmo- 
spheric pressure, and in admitting the hot steam into the cold cylinder, made 
the engine double acting, and effected the condensation in a separate chamber. 
The jet or spray of water continued long after Watt's time as the means of 
cooling the steam, and gave in later days the distinguishing name to the 
condenser, which is now nearly entirely superseded by a more perfect 
apparatus. 

' The Common or Jet Condenser, now really uncommon on shipboard, con- 
sists essentially of an air-tight chamber, into which the steam flows from the 
cylinder after having been exhausted of its available energy ; at its entry 
the steam is intercepted by a spray of water, caused by the inrush from the 
sea through small holes or narrow slits in a pipe placed across the steam- 
way. If the spray is fine, like a shower of rain, it mixes mechanically with 
the steam, as well as cools it by surface contact ; should the pipe have slits, 
so as to cause the water to flow in thin broad streams like ribbons, the cooling 
is effected principally by surface contact. The result in either case is the 
turning of the steam into water, which falls to the bottom, and is pumped away 
by the air-pump. It might be supposed that the mere turning of the steam 
into water, thereby causing it to occupy far less space, will produce the 
vacuum in the condenser ; in practice it does, but to so slight ah extent 
and of such an evanescent nature, that unless some other means were at 
hand, this condenser would be nearly useless. Water readily absorbs air 
when freely exposed to the atmosphere, and gives it up again on being heated, 
or relieved of atmospheric pressure. Fresh feed-water contains air, which 
becomes mechanically mixed with the steam in the boiler, and passes with it 
through its various passages, until it enters the condenser, when it parts com- 
pany with it, and remains as cooled rarified air after the steam is converted 



AMOUNT OF INJECTION WATER. 345 

to water. The cooling water also contains air, and readily gives most of it 
up on becoming heated by the exhaust steam, especially under the diminished 
pressure of the condenser. After a few strokes of the engine a sufficient 
amount of air will be accumulated to raise the pressure to that of the atmo- 
sphere, and although the condenser may be kept quite cool there will be 
no vacuum, but the reverse. The pump, therefore, which exhausts the 
condenser draws away the air as well as the water, but since the latter could 
run away by gravity, it is only the former which is of necessity pumped ; 
hence this pump is rightly named the air-pump. 

The shape of a jet condenser is immaterial so long as the inlet for the 
steam is high enough to prevent the water running back into the cylinder, 
and the bottom so shaped that the water will all drain into the air-pump 
bottom. It was generally formed to suit the ship and the working parts of 
the engine, and was often a part of the engine framing ; the back columns 
of vertical engines were often utilised for the purpose, and did extremely 
well, except that occasionally rapid corrosion so weakened them as to become 
dangerous. The frames of the horizontal engines were arranged by :;ome 
engineers to do duty for condenser, until the Admiralty finally forbade the 
practice. 

The capacity of the jet condenser should not be less than one-fourth that 
of the cylinder or cylinders exhausting into it, and need not be more than 
one-half of it, unless the engine is a very quick running one ; one-third the 
capacity is generally, however, sufficient. The objection to a large condenser, 
beyond its cost and weight, is that a longer time is necessary to form a 
good vacuum in it ; and the objection to a small condenser is its liability 
to flooding and overflowing to the cylinders, unless the engineer is most 
attentive. 

The amount of injection water depends on the weight of steam to be con- 
densed and its temperature ; the exact quantity of water per pound of steam 
depends on the temperatures of the steam, of the cooling water, and of the 
" hot-well," or receptacle into which the air-pump delivers the products of 
the condenser. As the supply of water to the boilers (called the feed-water) 
is taken from the hot-well, and it is an obvious advantage for it to be as 
warm as possible, the cooling water used is only such as sufficient to produce 
& good vacuum. With the jet condenser a vacuum of 24 inches was con- 
sidered fairly good, and 25 inches as much as was possible with most con- 
densers ; the temperature corresponding to 24 inches vacuum, or 3 lbs. pressure 
absolute, is 140°. In actual practice the temperature in the hot-well varied 
from 110° to 120°, and occasionally as much as 130° was maintained by a 
careful engineer. To find the quantity of injection water per lb. of steam 
to be condensed : 

Let T x be the temperature of the steam whose latent heat is L ; T the 
temperature of the cooling water, whose quantity in lbs. is Q ; T 2 the tem- 
perature after condensation, or that of the hot-well. 

The total heat of the steam = T, -f L. 

The heat absorbed by the cooling water will be (T x -f- L) — T 2 . 

But the heat absorbed by the cooling water is also represented by 
Q (T 2 — T c ). Hence 

(T 2 + L) - T 2 = Q (T - T ). 
Or, Q = (T 1 + L)-T 2 ^(T 2 -T ). 



346 MANUAL OF MARINE ENGINEERING. 

Now (Tj + L) — T., is equivalent to the total heat of evaporation from 
T 2 and at T a , and is therefore equal to 966 c + 0-7 X 212° + 0-3 xT ; - T 2 . 

Or, Q(T 2 -T ) = 1,114° + 0-3 x T x - To. 

Therefore 

Q 1,114° + 0-3 x T x -T 9 

y " To - T 

Example. — To find the amount of injection water required for an engine, 
the steam at exhaust being at a pressure of 10 lbs. absolute ; the temperature 
of the sea is 60°, and it is required to keep the hot-well at 120°. 

The temperature corresponding to 10 lbs. is 193°. 

n 1.114° +0-3 X 193° -120° ,--«« 

Q= 120° - 60° =17-53 lbs. ^ 

That is, the amount of injection water is 17 '53 times the weight of steam 
for this particular case. 

The allowance made for the injection water of engines working in the 
temperate zone was usually 27 to 30 times the weight of steam, and for the 
tropics 30 to 35 times, the summer temperatures being about 60° and 80° F. 

The Area of injection orifice and size of pipe is governed by the head 
of water, vacuum, and length of piping, or, in other words, by the equivalent 
head at the condenser. 

Neglecting the resistance to flow at the orifice, and in the pipes and 
passages, the velocity at the condenser may be found as follows : — 

Let h be the head of water above the valve on the condenser in feet, p 
the pressure in the condenser, and h x the equivalent head n for gravity, 

ftj = h + (15 - p) 2-3, 

and velocity in feet per second = V2 g h x = 8-025 Jh x . 

Example. — To find the theoretical velocity of flow into a condenser in 
which the vacuum is 26 inches, and the orifice 12 feet below the water-line 
of the ship. Here 

h x = 12 + (15 - 2) 2-3 = 42 feet. 
Then 

velocity = 8-025 \/42 = 52 feet per second. 

In practice owing to loss of head due to resistance at valves in the pipes, 
etc., the actual velocity was only about half that given by the above rule. 
Hence, in designing it was usual to calculate on a velocity of only 25 feet 
per second for shallow draught steamers, and 30 for deeper ones. 

From these rules, and with such allowances, the following holds good : — 

Area of injection orifice in square inches = number of cubic feet of 
injection water per minute -3- 10 - 4 to 12-5 according to circumstances, or = 
weight of injection water in pounds per minute -f- 650 to 780. 

A rough rule sometimes used was — 

Allow one-fifteenth of a square inch for every cubic foot of water con- 
densed per hour. And another 

Area of injection orifice = area of piston + 250. 

The injection valve was usually a simple slide or sluice valve, which 



SURFACE-CONDENSER, 347 

readily opened or shut, and regulated the amount of water. The handle 
or lever for working the injection valve must be very near the starting gear, 
so that the water may be shut off as soon as the engine stops. 

Snifting Valve. — It is usual to fit. at the bottom of a jet condenser, a 
non-return valve, through which the water, etc., may run or be blown out 
by steam ; it shuts by its own weight, and is held on its seat by the pressure 
of the atmosphere. This is called the sniiting valve, and it allowed of the 
condenser being emptied of water and air by steam before starting the engine ; 
it likewise prevents the pressure in the condenser exceeding that of the 
atmosphere to such an extent as to be dangerous in case of mishap with the 
injection water or steam, and on that account is a useful and necessary 
adjunct to the condensers of turbine engines, as there is no hindrance to 
the full and direct flow of steam into it should the rotor stick. The valve 
through which the steam is admitted is called the blow-through valve, and 
was a simple mushroom valve, raised by means of a lever, and closed by the 
steam pressure on the lever being released. 

The snifting valve was usually exposed without a casing, so as to be 
easily inspected or removed in case of being gagged with dirt, etc. ; to prevent 
the water from being spurted about the engine-room, the valve was formed 
with a curved rim, which completely covered and overhung the seat like an 
inverted saucer. 

The internal injection pipe or rose should be placed below the flow of 
steam, so that the cooling water may pass in a shower twice through the 
steam. 

Surface-Condenser. — It has been seen that with jet condensation the con- 
tents of the hot-well consist of a mixture of condensing and condensed water 
in the proportion of about 30 to 1, so that the water available for feeding the 
boiler was nearly as salt as sea water. If the cooling water is kept separate 
from the condensed steam, the latter, which is pure water, may be used as 
feed- water. The idea was by no means new, since so far back as 1794 Cart- 
wright took out a patent for an engine, in which the steam was condensed 
on the cold surfaces of two metal cylinders placed one within the other, and 
having cold water through the inner one and about the outer. An engine 
was made on this patent, and is said to have given great satisfaction. 
Brunei, in 1822, took out a patent for the same object ; his invention was 
designed more especially for ships, and consisted of groups of small tubes. 
In 1833, L. Herbert and J. Don patented an arrangement whereby " the 
air and steam from the eduction passage is drawn by a fan through or among 
small tubes, so as to be condensed. The tubes may be below the water." 
In 1835, W. Symington patented a plan " for condensing the steam from the 
cylinder, and cooling the surplus water from the air-pump, by tubes laid 
along the keel exposed to water outside a steam vessel." In 1838, J. B. 
Humphreys took out a patent for " surface- condensation by leading the 
steam through tubes in a vessel kept cold by a flow of water." The practical 
success, however, of this mode of condensation is chiefly due to Mr. Samuel 
Hall, with whose name the surface-condenser is generally and properly 
associated. He took out a patent in 1831 for a system of surface-condensa- 
tion, and in his specification claims, among other things, to condense the 
waste steam from the safety-valves, and to distil fresh water, to make up 
loss, by an apparatus, in principle similar to the distillers of to-day. One 



348 MANUAL OF MARINE ENGINEERING. 

of the first ships fitted w : th Hall's condenser and appurtenances was the 
*' Siriua," which made the first voyage under steam from England to America 
in 1838. In 1837 engines, made for the " Wilberforce," were fitted with 
Hall's condenser, and it is interesting to note that these engines had piston 
valves, etc., pretty much as now used. Ignorance and prejudice drove 
Hall's invention off the market for many years, so that it was not till 1860 
that the surface-condenser came into general use, and then only slowly ; 
indeed, it was not until mineral oil was used exclusively for internal lubrication 
that objections to surface-condensation ceased. In Hall's condensers the 
steam passed through the tubes and the circulating water outside them ; 
had the cooling water gone through the tubes it is highly probable they would 
not have " choked with mud and sand " and been condemned, as they were 
in the case of the " Wilberforce " in 1841. 

Condenser Tubes. — It is essential that the material on the surface of 
which steam i3 to be condensed should be metallic, thin, and a good 
conductor of heat, strong enough to resist the pressure of the water on 
it, amounting to at least 15 lbs. per square inch, and capable of withstanding 
sudden changes of temperature without fracture and the corrosive action of 
salt water and distilled water. 

The circular section being best suited to resist both internal and external 
pressures, tubes were naturally choser as the means of separating the steam 
from the water, and small tubes admit of a Very large amount of surface in 
a small space. 

Copper, being highly ductile and one of the best conductors of heat, was 
at first chosen as the material from which to make the tubes. But it was 
soon found that the acids derived from the fatty matter from the cylinders 
dissolved some of the copper, and produced soluble salts of that metal, which 
were pumped into the boiler with the feed-water, and there caused great 
injury to the iron surfaces. This, for a time, gave the surface»-condenser 
an ill repute, as it was found that the value of fuel saved by them was ex- 
ceeded by that representing wear and tear of the boilers. This was eventually 
obviated by having the copper tubes coated with tin, and by discontinuing 
the use of tallow in the cylinders. But, notwithstanding this, the boilers 
in H.M. Navy continued to show signs of premature decay, such as was 
not customary with those receiving water from a jet-condenser. It was 
attributed to the highly corrosive power of redistilled water on the bare 
surface of the iron, together with the impossibility of keeping a protective 
6cale on the surfaces when such water was used ; later research has shown also 
that some gases which entered the boiler with the original water chemically 
combined with bases, which kept them comparatively innocuous, were freed, 
and came back mechanically mixed or in solution with the feed-water from 
the hot- well, capable of highly destructive action on iron surfaces, carbonic 
acid being a common and very corrosive one. 

Copper tubes were, of course, expensive, and the tinning added to their 
cost very considerably, thereby rendering the first cost of a surface-con- 
denser an appreciable addition to that of the engine. In 1860 brass tubes 
had long been used for boilers, and as this material was very ductile, and 
its galvanic action on iron feeble, it was tried as a substitute for copper in 
the manufacture of condenser tubes with at first mixed success, the want 
of complete success being partly due to want of care in manufacture, and" 



CONDENSER TUBES. 340 

partly to prejudice. The partial has since become a complete success, arid all 
Condenser-tubes are now made of brass, but somewhat richer in copper. 

The Admiralty, and consequently all foreign governments, required brass 
tubes to be tinned when the condenser was of iron, and some of the large 
steamship companies also adopted this practice, but, as a rule, now the tubes 
are untinned. The tinning adds ljd. per pound extra to the cost of the 
tubes, and amounts to an increase of 16 per cent. ; it is not necessary, but 
is an additional safeguard against the formation of copper salts on the one 
side, and to the corrosive action of sea-water on the other. Brass tubes, 
untinned, after twelve years' constant use, have been found, on being cleaned, 
to be nearly as good as when new ; on the other hand, brass tubes have 
pitted badly, and in places been perforated in a few months from causes 
practically unknown* It was conjectured that galvanic action had set up 
from iron filings, carried on to the tube surface by the steam or circulating 
water, causing a separation of the copper and zinc and the dissolving away 
of the latter. The investigations Prof. Bengough made for the Institute 
of Metals has thrown much light on the matter, and will eventually clear 
up what was somewhat of a mystery. As the condensers of warships are 
now always made of brass or steel, the tubes are untinned, but the Admiralty 
require the addition of 1 per cent, of tin to the mixture of copper and zinc, 
as this composition does not usually pit or corrode so readily as the ordinary 
brass mixtures have so often done. 

The loss from blowing off the boilers to prevent dangerous incrustation 
when fed from the hot-well of a jet-condenser, amounted to as much as 
25 per cent., and was seldom less in general practice than 12 per cent. This 
loss is almost wholly avoided by the use of a surface-condenser, and an addi- 
tional saving of no mean importance is effected by cessation of necessity 
to scale and clean out the boilers as was the case when jet-condensers were 
used. The net saving of fuel by the use of a surface-condenser averages 15 per 
cent. ; and in the hands of a careful engineer, the economy has been found 
to extend to even 20 per cent. 

When sea-water is raised to a temperature of 280° F., which corresponds 
to a pressure of 50 lbs. absolute, or 35 lbs. above the atmosphere, what are 
called its insoluble salts are wholly precipitated, and these form a hard scale 
on the hot surfaces. The principal insoluble sait in sea-water (v. Chapter xxx. } 
is sulphate of lime ; it is called insoluble, because it does not dissolve in 
water under ordinary circumstances, and consequently when deposited on 
the surface of the tubes, etc., will not easily redissolve and wash off again. 
The carbonate of lime, and the salts of soda and magnesia, are comparatively 
harmless, for although the former is precipitated, it is only in a soft muddy 
state, and mixed with the brine products of the latter can be blown off, and 
easily removed from the boiler when in port. It is for this reason that a 
surface-condenser is an imperative necessity for engines using steam above 
35 lbs. pressure, or 50 lbs. absolute. 

For the same reason it is advisable to fit a surface-condenser to steamers 
running in muddy fresh water, as otherwise the boiler soon fills with deposit, 
which, unless removed and the boiler thoroughly cleaned, will cause serious 
damage, and be a source of danger and expense, as well as a constant cause 
of priming. 

It is seen, then, that by the use of a surface-condenser steam of higher 

* Vide Reports of the Corrosion Committee of the Inst, of Metals, 



350 MANUAL OF MARINE ENGINEERING. 

pressure than 50 lbs. absolute may be employed with safety ; a considerable 
saving of fuel and time is effected ; and there is considerably less risk of 
burning and otherwise damaging the boiler. A better vacuum is also obtained 
in it, as a rule, than was possible with the jet-condenser. On the other hand, 
a surface-condenser is heavier, more costly, and occupies more space than 
the jet-condenser; a second pump for the cooling water is necessary, and 
although the air-pump need not be so large as for a jet-condenser, it was 
often made so in case the jet was used or the tubes leaked. ■ It was urged that 
the wear and tear and the store account are increased somewhat when there 
is a surface-condenser, and more care and responsibility are laid on the 
engineers ; but these are mere trifles compared with the benefits, and are 
only mentioned now as examples of the arguments used by the opponents 
of Mr. Hall's system in the bitter controversy that raged in the forties of 
last century, and prevented its general adoption. For sea-going vessels the 
surface-condenser is indispensable, and now always fitted, notwithstanding 
its weight and other drawbacks. 

Condenser efficiency was never, as a matter of fact, seriously considered 
until quite recent times, for the old school of engineers had other distractions, 
and were quite content with what is now considered to be only a moderate 
or even a low vacuum, especially were they complacent if with "it was a hot- 
well temperature of 120° F., without enquiring very closely as to how it was 
obtained. Moreover, so long as it was accepted as an axiom that any higher 
vacuum could be of no service to the marine engine, there was no incentive 
to enquire into, much less to attempt better things. The advent of the 
turbine as a marine motor, however, together with the demand for the highest 
speed with some ships at any reasonable cost, has changed all this, and turned 
the apathy of the past into the activity of the present with regard to the 
condenser. Moreover, engineers have been impressed with the apprehension 
that while little remained to be gained at the upper part of the expansion 
diagram, at the lower the field was large and fallow. Thoughtful minds have 
turned, therefore, to it, and made a careful study of the theory and practice 
involved in the working of a surface condenser. To Mr. James Weir, Mr. 
Morison, Prof. Weighton, and others, we are much indebted for the better 
knowledge we possess, and although these gentlemen do not always agree 
they are all equally frank and communicative of what they discover when 
experimenting, so that to maintain 29 inches vacuum in a modern condenser 
in the temperate zone is easy. To the turbine such a small back pressure is 
of the utmost value, whatever it may be to a reciprocator, and its efficiency 
with it very high. But it must not be taken now for granted that no advantage 
can accrue to a reciprocating engine by the rediiction in back pressure. No 
doubt, owing to restriction in the valve and passage sections existing in 
most of these engines, the high volume steam has to flow at such excessive 
speeds that the difference in pressure between the cylinder and condenser 
is necessarily greater than required between a turbine and its condenser, 
or even what obtains when the back pressure is greater. At 29 inches of 
vacuum, or a half-pound pressure, a pound of steam has a volume of 640 
cubic feet, while at 26 inches it is only 172 ; for the same weight of steam 
the velocity of flow to the condenser will, therefore, have to be nearly four 
times at the high vacuum if the pressure difference of it is to be the 
same as at 26 inches. But it is quite unlikely that, with a decrease in the 



CONDENSER EFFICIENCY. 



351 



pressure in the condenser, there will be none in the cylinder, whatever its 
ports and passages be. That there is some decrease is certain, and such 
decrease will enable a greater power to be developed ; and that the decrease 
may be considerable may be seen on examining the diagram shown in 
fig. 1186. Each pound of back pressure means 2\ to 3 per cent, difference 
in the power developed by a triple- or quadruple-compound engine, and 
such an increment as this is quite possible in a properly designed engine by 




Fig. 115.— " Weir.Uniflux " Condenser for a Turbine of 11,000 S.H.P. 

merely increasing the vacuum from 26 to 29 inches. Moreover, if marine 
reciprocating engines are designed with a multiplicity of cylinders that 
each may be kept of small size, as they are on the lines of the oil engine, so 
that instead of one or. at most, two L.P. cylinders per engine there are 
three or four, then the port and passage ratios may be so much larger that 
*iigh vacuum would be of advantage to such engines. 



352 



MANUAL OF MARINE ENGINEERING. 



The effect on economy of consumption is another matter, and has to be 
considered from quite a different standpoint. More energy will be required 
by the air-pumps to maintain such higher vacuum ; and as the temperature 
due to 29 inches is 80°, against the 126-8° of 26 inches, the feed water must 
leave the condenser so much cooler ; but whether the reheating will be 
more costly than the gain is rather a matter of experience and trial than of 
calculation ; for so great a difference probably it is so ; but as 28 inches 
can be much more cheaply obtained than 29 inches, and the temperature 
difference is then only 25° against 46-8° as above, the loss, if really anything, 
cannot be great in these days of feed heaters. 

Mr. Weir attaches no importance to the question of water deposit on the 
tubes in a surface-condenser, and thinks there is nothing to be gained, while 
there is something lost, by fitting, collecting, and training plates among the 
tubes, whereby it is conducted away as soon as formed without further 




Kinetic Air Pump 
Suction 

-Water Pump Suction. 

Fig. 116. — Cylindrical Condenser on Morison's System 

contact ; whereas Mr. Morison advocates and fits them to all his condensers, 
and claims to get thereby a distinct benefit. The former authority is, hew- 
ever, an equally strong advocate for having warm hot-well water, and he 
should, therefore, not object to any reasonable natural means of attaining 
that very desirable end. With that end in view, the exposure of the water 
when once formed to a further cooling action cannot be helpful ; on the 
contrary, the already condensed water must lose some of its own heat, and 
add to that of the cooling water, thereby reducing its efficiency. Further, 
by means of such plates as Mr. Morison and others fit, the steam and air are 
dispersed over the whole of the cooling surface without scattering the water 
when condensed. But, on examining the design of these gentlemen's con- 
densers, one is struck by the necessity for such arrangements in the one and 
its absence of it in the other. Mr. Weir's condenser is of triangular cross- 



SURFACE CONDENSER. 353 

section or heart-shaped, as an approximation to it (fig. 115). Steam enters 
at the butt end through such an enormous orifice as to enable it to spread 
at once over the whole breadth and length (which is generally limited) of 
the top of the condenser, and gravitate to the contracted bottom as it cools 
and shrinks in volume. In this case the water, as it forms, can, and does, 
drop on to the inclined sides, on which it courses to the bottom, which is 
not very far distant. Here there is little chance of any of the water dripping 
on to the colder lower rows of tubes from the majority of the upper ones. 
But the shape of the condenser is not economical of space or suitable to 
resist unaided external pressure. On the other hand, the cylindrical 
condenser of Mr. Morison (fig. 116) only requires inexpensive guide plates, as 
does ako that of Mr. Morcom, which shall cause a perfect circulation of the 
steam among the tubes and the leading away of the water when formed 
is accomplished by the same means surely with some advantage. In the 
case of the compact rectangular section condenser forming part of the engine, 
and economic of room and cost, as in common mercantile practice, the Con- 
traflo system (fig. 117), with the similar drainage scheme, is also advantageous 
— in other words, circumstances alter cases. Fig. 117a is a diagrammatic 
form of a Contraflo, and shows how the actual one, as in Fig. 117, is 
virtually the same as the wedge-shaped or triangular section one. 

For efficiency, a complete and rapid circulation of the steam over the 
whole of the cooling surface is essential, as is also the concentrating of the 
air and water into a natural flow towards the passage to the air-pump ; the 
simplest means of effecting these two things are probably the best. 

Equally important is the question of cooling-water distribution. Its 
quantity depends on the difference in temperature between it as it enters 
and that of the condenser where it leaves it, the weight of steam to be con- 
densed, and the rapidity of flow. In practice condenser tubes are usually 
| inch in external diameter in the mercantde marine, and f inch on naval 
ships. Now, a tube f inch diameter, 18 gauge thick, and 10 feet long, has 
a surface of 1*95 square feet, and inasmuch as 30 lbs. of steam per square 
foot per hour may be condensed on it when clean, it will be necessary for a 
vacuum of 28 inches that about 1,200 lbs. of cooling water should pass through 
it per hour in winter time in the temperate zone, and 3,600 lbs. in the tropics ; 
the flow then will be at the rate of 400 feet per minute in the latter case, 
which is somewhat excessive, and would require considerable power to obtain, 
as the friction head would be about 3| feet per tube length. On the other 
hand, in the temperate zone in winter the flow would be too moderate. 

This means that if a good vacuum is to be maintained in the tropics 
the tubes must not be very long with so small a diameter, while, on the other 
hand, if a ship is never to be where the cooling water is above 60°, the tubes 
may be long with advantage. As, however, the rate of flow will vary as the 
square of the diameter of tube, while its surface is directly as the diameter, 
a small increase in diameter will admit of considerable increase in length. 

In practice, the cooling water may have sufficient heat imparted to it 
to raise it to a temperature very little below that of the condenser top before 
leaving it. As this warming up, however, is usually done in stages — that is, 
the water passes and repasses through the tubes on its journey from the 
entering in to the leaving them — three stages is the most common, so that 
in a general way the increment added to the temperature of the circulating 

23 





Fig. 117. — Ordinary Marine Engine with Contmflo Condenser. 





Fig. 117a— Diagram showing the Flow of Steam in the Contraflo System. 



SURFACE CONDENSER. 355 

water at each pass should be only one-third of the total increase. To do this 
the number of tubes in the lower or first group should be much less than 
that in the second, and less there than in the third, as the efficiency owing 
to difference in temperature between it and the condenser is decreasing at 
each stage. This circumstance accentuates the reason for shorter tubes 
or of larger diameter when the water is warm and a high vacuum desired. 

Surface Condenser Efficiency has, however, become a matter of first-rate 
importance ever since the introduction of the turbine on shipboard, for 
it was there discovered how seriously the efficiency of that instrument was 
augmented when working with the very low back pressures obtainable there 
at comparatively cheap rates, owing to the unlimited supply of coolingi 
water obtainable with little sacrifice of power due to the small resistance 
" head." Mr. Parsons, and those interested in the success of the turbine, 
turned their attention to improving the means whereby high vacuum could 
be obtained and maintained ; Mr. Morison, ably assisted by Prof. Weight on, 
of Durham, followed on with a series of most interesting and instructive 
experiments, and their investigations have been the means of throwing 
quite a flood of light on the subject, and clearly demonstrating exactly what 
takes place in a condenser and its pumps. 

The effect of Air mixed with Steam or water vapour had been noted by 
Prof. Osborne. Keynolds, and others years before, and it was well known 
that the presence of air was always a cause of retardation of condensation, 
and the reduction of amount of water deposited on a cold surface in a fixed 
time. If steam alone entered a surface-condenser, it would be wholly trans- 
formed into water, and a vacuum corresponding to the temperature would 
be maintained in it. The amount of steam condensed per square foot of 
surface would be very high, so long as the tube remained clean inside and 
outside and the cooling water supply plentiful with the flow through the 
tubes rapid. The only pump necessary under these circumstances would be 
one to withdraw the condensed water as it falls to the bottom. But, as a 
matter of fact, it is impossible in practice to work with a circuit so com- 
pletely closed that no air gets into the system when once the water put into 
the boilers has been deprived of the air it originally held in solution, for 
every condenser of a marine engine contains more or less air always, and, 
therefore, an air-pump is necessary to it, in order to maintain any sort of 
steady vacuum ; and if the vacuum is to be high, the air-pump must be efficient 
as well as sufficiently large ; for, however large it may be, if it is not efficient 
no high vacuum at all is possible. 

Air Leaks to the Condenser may have originated at the glands of the 
main and auxiliary engines when the pressure inside them is less than that 
of the atmosphere. Much air comes from the feed-water, which, if taken 
from storage tanks, will contain from 2J to 4 per cent, by volume of air. Even 
the water of the hot-well is charged with a considerable amount, and this 
Mr. Weir endeavours to eliminate in his feed-heaters before it can enter the 
boiler. Auxiliary feed-water is subject to the same action in the feed-heaters. 
Leakage may, and does, arise at times from faulty or damaged jointings of 
the condenser and connections ; these leaks, however, should be found 
out and stopped ; for this purpose a periodical examination is made by 
some careful engineers in charge of turbines ; those with reciprocators should 
do the same. The leakage from auxiliary machinery, however, is often a 



356 



MANUAL OP MARINE ENGINEERING*. 



most serious cause of loss, and is said to amount to as much as the cost of 
the whole power developed in them. No doubt the Admiralty are wise to 
have separate condensers for the auxiliary machinery, and in large merchant 
steamships the same practice might be followed with considerable advantage ; 
even if such condensers were simplified so that the water circulation was 
" natural," and the condensed water allowed to drain into a tank by gravity, 
it is better to have them rather than to run the risk of spoiling the main 
condenser efficiency by admitting aerated auxiliary exhaust to it. 




Beveporismg Chei.iber 



Feed Temperature 
Regulating Chamber 



Air Pump 
* Suction 



Feed Temperature Regulating Valve 



Fig. 118. — Gontraflo ©ondenser with Feed Temperature Regulatof. 



Air is heavier than Water Vapour, so that, if the exhaust steam contains 
but a very small quantity, it will accumulate at the bottom of the condenser, 
and unless removed it practically blankets the cooling surface there, and 
virtually reduces the size of the condenser. Even when the air-pump draws 
it away, if only at the same rate as it flows in, there may be a portion of the 
surface constantly surrounded by air and effecting no condensation. It is- 



MORISON S DIAGRAM. 



357 



necessary, therefore, that the air-pump shall be large enough to keep the 
inside of the condenser free from air lodgment. 

It follows, then, that, in addition to the large efficient pump, means 
shall be provided that the flow to it will be as direct as possible ; and at the 























29" 














































\ 

■ 




























1 


















1 


t 












































i 


I 
« 






























u 

3 
















j 


1 

l 






























e J* 
















i 


i 


28 


5' 
























C 05 




















1 
\ 








































j 




/ 


| 














































i 
i 


/ 


1 

1 




























N | 


2400 
















\ 


/ 


i 






NO 


AIF 


v.e 


S/ 


TU 


*AT 


lO 


BTE, 


\M 




i i 


Z V 




RA 


10 I 


)F A 


IR * 


O V 


APO 


JRl 








\* 


■s" 






/ 


• 














a 
i h 
1 co 




2200 






V 


s 








v 


i 










• 


/ 
















a w 
f- 

I . ^/x « 
















1 \ 




1 




7\ 




/ 




































\l ' 




1 


/ 




l / 








































\ 1 


i 


1 

\ i 




I 




















y oq 








►s 




. 


33 '? 


1800 




















u 




\ 


k 


















o 


















/ \ 


p 


X 


| V 
/ \ 






j \ 


















L iJ 




1400 
1200 
1000 
800 
600 
40O 
200 














\ 




\ 




» 
\ 




(\ 


















R-300 ® 


60 












j 




1 1 


\ 




\ 

\ 


{ 




V 
V 

\ 
















P w 


5x • 












,/ 




1 X 


jf\ 


\ 


J 




I 


»7-5 
















5 i 












i 




/ 


\ 1 
* 1 


\ 




( \ 




J 


\ 
\ 
















□ 
j- C 












r 


\ 




\l 




\ / 


! / 






27" 














c « 










1 




V. J 






\ 


\J 








b > 


\ 
\ 












as 












kf 






\ 




V 


t / 


\ 


J 


\ ' 






\ 
\ 










u » 


o o 

5 "* 










V 




\ 


\ 


\ 




f\ 


\ 


J 


\f 


i 






tf- 










B g 






\ 




\ 
\ < 


V 


\ 
\ 






s 




\ 

\ 


1 \ 

1 \ 


X / 


\o 


i 


'"" 


J s 




IS* 






* .2 

O O 




\ 

\ 




\ 




\ 
\ 


i 






\ ■ 






S 
\ 


7\ 


o 


\ 
X j 






.^i 


X 
X 


■, 2 * 






\ 
\ 




\ 




\ 
> 


> 




,•**• 


X 
+ 


**,_ 






r>f 




. / 








— ■ ' 


j 




J 1 


\ 




X 






x 


h> 


II 
x 


X 


J 


?t 


x 


/k 


v 


X 


> 


r"v 
















s x. 


<^ 


r^ 


" 








^-x 


jV 


i 


[/* 








X, J 


«, 


-/ 










^ 




<"x 


; . 




^ 


^ 


■■>. 


■m,^ 




*^"C 






J0T 




.*• 


*» 




























~-^ 


^ 


CD 


Js— 




~^. 


^- , 


















KM3 






XI 


rrs 


JUL 


I "'I 


TTg 






-BTT 


m 


nnr 


ni 




■L-n 


r* 31 


W] 


— 







4-5 50* 60° 70 SO 90 100 HO 120 130 140 

Temperature in Degrees Fahr. 

fig. 118a. — Morieon's Diagiam. Ai» Saturated with Water Vapou*. 
Carves showing relation between vacuum, temperature, volume, and ratio of air to vapour. 

same time that every part of the cold surface should be active ; hence there 
must be no pockets or eddies anywhere, and if the design of condenser doea 



358 



MANUAL OF MARINE ENGINEERING. 



not ensure thi« constant and general flow, the vapour must be directed by* 
guide plates and baffles. Mr. Morison has given great attention to this, 
and all else that pertains to the surface-condenser, and been good enough to 
impart his knowledge in the papers he has read at the meetings of the Insti- 
tution of Naval Architects, and of the N.E. Coast Institution of Engineers 
and Shipbuilders, the transactions of which may be studied with advantage. 
The Flow of Vapour and Water in its passage through the condenser 
from the exhaust to the air-pump suction must be at all times positive, but 
separate. When water is once formed, it should not be kept in contact 
with the cooling surface, or allowed even to touch it again, but at once take 
its passage to the drain. The vapour should be made to pass or repas3 
over the whole cooling surface, so that no part of it is inactive. In modern 
condensers the continuous contact is done by repassing largely in order 
to economise the space taken by the condenser, which makes it virtually 
the same as a long one of wedge shape, transverse section, like Mr. Weir's 
(fig. 117), having the butt end next the cylinder, and the drain to the air-pump 
at the thin end. In this way the air is concentrated before entering the 
pump, and is moreover cooled as much as is possible, so that the air-pump 



Top 



K 



A 



4~ 






-J 



s. 



Mean Press 3-5 Lbs. 
Scale fjf 



>C 



\ 



Steam, .. 

Vacuum, 

Revolutions, 



176 His. per square inch. 
28} inches. 
66 pur minute. 



High-pressure engine, 
Intermediate-pressure engine, 
Low-pressure engine 



6S0 I. If. P. 
6!)7 ,, 

7o5 ,, 



Total, .. 2,035 „ 

Fig. 1186. — L P. Indicator Diagram showing High Vacuum. S.S. " Xigaristan." 

(Richardson and Westgarth.) 



can be kept cool and fit to produce in its chamber a very high degree of 
vacuum. For this purpose the lowest part of the condenser is often arranged 
so that the lowest rows of tubes through which the coldest water passes are 
" drowned, "so that the water which surrounds them is cooled down below 
the temperature of the bulk of the condensed water. 

Fig. 118a, Morison's diagram showing the relation between vacuum, 
temperature, volume, and ratio of air to vapour, is a most interesting and 
instructive one, and may be referred to with advantage when considering 
condenser problems. 

Cooling Surface. — Mr. Weir has shown that, with efficient air-pumps, 
as much as 16 lbs. of water can be condensed on a square foot of cooling 



COOLING SURFACE. 359 

surface per hour in a vacuum of 26*75 inches when the sea-water is as hot 
as 80° F., and the hot-well temperature as high as 106°. With sea-water 
at 54° F., and the vacuum 28 inches, as much as 28 - 6 lbs. were condensed per 
square foot per hour. He also demonstrates how as much as 35 lbs. can be 
condensed with the vacuum at 27*3 inches, with the cooling water at the 
same temperature, but with the hot- well temperature reduced to 101°, or 
nearly that in the condenser. 

It would seem, then, from these facts that 1 square foot of cooling surface 
per I.H.P. is sufficient for any ship, so long as the condenser is fairly clean ; 
and, further, it is evident that h square foot per I.H.P. is ample allowance 
for such ships as work in temperate zones with cooling water under 60°. 
Moreover, as most ships on service run with less power than developed on 
trial, the allowance for trial conditions practically provides a margin for 
loss of efficiency of surface afterwards. Experience with turbine-driven 
ships, having a vacuum in their condensers of 28*5 to 29*0 inches, has estab- 
lished the fact that with cooling water at 50° F. and the hot-well kept at 
80° to 85° F., as much as 25 lbs. of steam can be condensed per square foot 
of surface per hour. 

The Allowances of Cooling Surface may be computed in the following 
way : — Ships trading to all parts of the world, and, therefore, sometimes in 
the tropics, should have 1 square foot of cooling surface for each 16 lbs. of 
steam condensed. Turbine steamers working under the same conditions 
should, to maintain high efficiency, have 1 square foot for every 12 lbs. 

Ships limited to service in temperate zones, or whose service in the 
tropics is short, or when there high efficiency is not of consequence, 25 lbs. 
for reciprocators and 20 lbs. for turbines is not too large an allowance. 
The temperature of the condenser with a vacmim of 29 inches is 80° F. ; 
it is obvious, therefore, that, with sea-water at this or a somewhat higher 
temperature, such a vacuum is impossible, whatever be the surface, and 
even with 28 inches the amount of cooling water required would be so great 
as to be almost prohibitive in commercial practice. * 

The temperature of the condenser is somewhat higher near the entering 
in of the steam than at the bottom when the engine is working at full power, 
but at reduced power the difference is very trifling ; at full speed, therefore, 
the circulating water may be nearer the temperature at bottom of condenser 
than at low powers. The water should, therefore, for this and other reasons, 
always flow in the opposite course to that of the steam — that is, the coldest 
water should be where the steam is coldest, and where the steam is entering 
the condenser, and, therefore, at its hottest the heated cooling water may still 
take up more heat from it, leaving it with heat still in it for abstraction below. 

The Cooling Surface per Horse-power may be as follows : — 

Triple-compound express steamers, 080 sq. ft. home waters, 125 sq. ft. tropics. 

„ „ economic „ 070 ,, „ 1"06 „ ,, 

Quadruple „ „ „ 65 „ „ 95 „ „ 

Turbine-driven ordinary „ 0'65 „ „ 110 „ „ 

,, express „ 0"80 „ „ 1'25 „ „ 

For ordinary cargo steamers going to all parts of the world 1 square 
foot of surface per trial trip I.H.P. is sufficient, as the temperature of sea- 
water does not exceed 85° F., and is not often over 80°. This allowance is 
sufficient to permit of good vadium being maintained on service conditions 

• At 80° F. the cooling water must be 63 times the weight of steam ; and at 85° no less than 153. 



360 MANUAL OF MARINE ENGINEERING. 

with the surface somewhat foul. Destroyers with reciprocating engines on 
trial in summer time have condensed 28 lbs. of steam per square foot, and 
the allowance per I.H.P. was only \ square foot, and the vacuum 25 to 26 
inches at full power. 

Professor Weighton, with a Contrarlo condenser, got 33 lbs. condensed 
per square foot from a triple-compound engine: and even 40 lbs. have been 
got by other experimenters. 

Cooling Surface. — Professor Rankine suggested the following as a means 
of ascertaining the amount of cooling surface : — Let t denote the temperature 
of a film of liquid at one side of a metal plate, S the extent of cooling surface ; 
let heat be communicated to the liquid at a temperature t by some such 
process as the condensation of steam, and let that be abstracted by the 
flow of a current of air, water, or other fluid, in contact with the metal plate ; 
the weight of fluid which flows past per second being W, its specific heat 
C, its initial temperature T T , being lower than t, but higher than T 2 . Then 
in all the equations t — T^ is to be substituted for T 1 — t, and t — T 2 for 

T 2 — t in the equation — ^ = a \ pp ™ f ; but he also added 

that there are not sufficient data to obtain the value of the constants. 

Another Basis for Calculating Cooling Surface is obtained by assuming 
that the maximum mean flow of cooling water permissible is 400 feet per 
minute, and the units of heat to be abstracted being that latent at the different 
pressures, and the cooling water in the tropics 80° F., and in home waters 
60°. The diameter of the condenser tubes is taken aa f inch for naval practice, 
| inch in general mercantile, and 1 inch in exceptional oases. 

The weight of water passing per minute through a f-inch tube at this 
rate will be 44 lbs., through a f-incb 60 lbs., and through a tube 1 inch in 
diameter 107 lbs. The temperature of the water at discharge from the 
condenser will be taken at about 2 per cent, below that of the hot-well water, 
or that of the condenser, so that for a vacuum of 28 inches it will be 0-98° X 
90 - 4 c = 88-6°. Under these circumstances, in the tropics the heat abstracted 
will be 88-6 — 80, or 8-6°. The latent heat with this vacuum will be 1,051° ; 

hence the weight of cooling water per lb. of steam = ' = 122 lbs. : a 

44 

tube f inch in external diameter will condense, therefore, in an hour 60 X ^ 

or 21-6 lbs. 122 

The greatest length of tube through which the water flows should not 
exceed 400 diameters, and, therefore, if the water passes three times, as is 
common practice, the f-inch tube should not be in the aggregate more than 
(f X W")' or 21 feet, and each tube is consequently only 7 feet long, or if 
only twice through 10*5 feet. 

Now, 21 feet of f-inch tube has a surface of 3-437 square feet. It follows, 
then, that the cooling surface in a condenser having f-inch diameter tubes 
for service in the tropics when 28 inches vacuum is required must be at the 
rate of 1 square foot for each, 21*6 -=- 3-437, or 6-3 lbs. of steam to be condensed. 
Taking the consumption of steam in a turbine steamer at 12*5 lbs. per 
S.H.P., then- 
Cooling surface of tubes f inch in diameter per S.H.P. for tropics 28 inches 
vacuum is 2 square feet. 



COOLING SURFACE. 361 

If sea-water is 60°, there will be then the following, viz. : — 
Heat abstracted 88-6 — 60, or 28-6°. 

Weight of cooling water = 1,051 ^ 28*6, or 36-7 lbs. per lb. of steam. 
Water condensed per hour = 60 X 44 -s- 36-7, or 72 lbs. 
Rate of condensation per square foot per hour = 72 -f- 3*437, or 21 lbs. 
For home service, therefore, 12-5 -4- 21, or 0*596 square foot per S.H.P., 
is sufficient for f -inch diameter tubes. 

For the Mercantile Marine with f -inch tubes and a 28 inches vacuum, the 
following will hold good, viz. : — 

Heat abstracted in the tropics, as before, 8*6 units. 

Water passed per hour, 60 X 60, or 3,600 lbs. 

Weight of cooling water per lb. of steam, 1,051 -f- 8*6 ; or 122 lbs. 

Weight of steam condensed per hour = 3,600 -=- 122, or 29-5 lbs. 

Maximum length of tubes, f X ^j^ , or 25 feet. 

Each tube being 8*33 feet long if three times through, or 12-5 if twice. 

The surface of 25 feet of -f-inch diameter tube is 4*91 square feet. 

Quantity of water condensed per square foot is then 29*5 -=- 4-91, or 6 lbs. 

If, however, the condenser were made the same length as that with f-inch 
tubes, the surface would be as before 3-437 square feet, and the quantity 
of water condensed per square foot 29*5 -s- 3*437, or 8*6 lbs. under these 
circumstances. 

The cooling surface per S.H.P. = 12*5 -*- 8*6, or 1*45 square feet. 

If the condenser tubes are made 1 inch in external diameter, the water 
passed per hour is 60 X 107, or 6,420 lbs. ; weight of cooling water as before, 
122 lbs. 

Steam condensed per hour. 6.420 -5- 122, or 52*6. 

Maximum length of tube = 1 X -^°-, or 33*3 feet. 

So that in this case each tube may be 11*0 feet long. 

The surface of 33 feet of 1-inch tube is 8*64 feet. 

52*6 
Quantity of steam condensed per square foot = ^i, or 6*1 lbs. 

If the cooling water has a temperature of 60°, the number of heat units 
abstracted by each pound will be as before 88*6 — 60, or 28*6°. 

The weight.of water for eaeh pound of steam, ? , or 36*7 lbs. 

28"6 

Steam condensed = 6,420 -f- 36*7, or 175 lbs. 

Steam condensed per square foot = 175 -f- 8-64, or 20*3 lbs. 

' From the above it will be seen that if the maximum combined length 
of 400 diameters of tube is followed as the rule for condensers, that 
a vacuum of 28 inches should be maintained with a rate of condensation 
of about 6 lbs. per square foot of surface when in the tropics, and 20 lbs. 
in the temperate zone. For the steam consumption per horse-power this 
would indicate that the cooling surface for tropical work must be at least 
three times that for cool climates when high vacua arc required and necessary 
as with turbines. It has been, however, pointed out that under tropical 
-conditions much more than 6 lbs. of steam per square foot of surface can 
be condensed. It follows, then, that the length of tube must be reduced, 
and the combined length be inversely as that quantity is to 6. 



362 



MANUAL OF MARINE ENGINEERING. 



That is, if Q be the quantity in pounds, then — 

fi tt 

Combined length = 400 X^X vq- 

d being diameter of tube in inches. 

If, however, a vacuum of 26 inches is sufficient, as is the case with 
reciprocating engines in the -tropics, then the temperature of condenser 
will be 120°, and the latent heat 1,030°. 

The number of units abstracted per lb. of cooling water will be 120 — 80, 
or 40. 

Weight of water for each pound of steam, 1,030 4- 40, or 25*75. 

Taking the tubes f inch in diameter — 

The steam condensed = 3,600 -*- 25-75, or 140 lbs. 

140 
The steam condensed per square foot surface = r^-r = 28 - 5. 

It is evident from the rates observed by Mr. Weir and others that the high 

vacua are obtained in the ordinary ship's condenser ; it follows, then, that 

either the tubes were short or the rate of flow through them much higher 

than 400 feet per minute or both. The difference in temperature has been 

taken here at only 8 - 6° F. for the tropics, which is, of course, very low. But 

if long tubes are fitted to a condenser the rate of flow must be very high to 

maintain high vacua, and the allowance of surface more than shown above. 

If, therefore, 28 inches is to be maintained with T25 square feet of cooling 

surface per S.H.P., the flow of water through the condenser, whose tubes 

400 x 2 
are 400 diameters in length, at a rate per minute = — =-^ — , or 640 feet, 

which is high for any tube, but especially so for those only f inch in diameter. 
The tl head " to overcome the mere resistance of 21 feet of such tube will be 
at a velocity of 640 feet per minute, probably as much as 45 feet, while at 
400 feet it will be no more than 16 4 feet. 

With tubes § inch in external diameter and 25 feet aggregate length the 
"head" to overcome resistance will be 15*7 with a velocity of 400 feet, and 
35-7 feet if the velocity is raised to 600, while with tubes of 1 inch external 
diameter, and an aggregate length of 33 feet, the resistance at 600 feet per 
minute flow will be only 14 feet " head." 



TABLE XL. — Effect of Vacuum on Steam Consumption in Lbs. per 
I.H.P. in a Turbine and Quadruple-expansion Engine 
(from Prof. Weighton's Observations), and 

in a Triple-expansion Engine (200 I.H.P.) and 1,000 K.W. Turbine. 

(from Sir C. Parsons' Observations.) 



Vacuum, . . Ins. 


20 


22 


24 


26 


27 


28 


29 


Turbine, Lbs. per H.P. 
Quadruple, ,, 


19-2 
16-7 


18-1 
16-1 


16-9 
15-5 


L5-6 

150 


14-8 
14-7 


13-9 
14-5 


130 
14-3 


Triple-comp. eng. „ 
Turbine „ 


14-8 
19-3 


14-35 
18-1 


14-0.") 
16-9 


13-90 
15-6 


13-8 
14-9 


13-77 
14-0 


13-76 
130 



CONDENSER TUBES. 



363 



TABLE XLI. — Temperature, Latent Heat, and Volume of 
Steam of very Low Pressure. 



Pressure. 








Pressure. 












Tempera- 
ture F ». 


Latent 


Volume of 






Tempera- 
ture F°. 


Latent 


Volume of 


Lbs. 




Heat 

F°. 


1 Lb. of 
Steam. 


Lbs. 




Heat 
F°. 


1 Lb. of 
Steam. 


Absolute 


Vacuum. 








Absolute 


Vacuum. 
















Cub. Ft. 










Cub. Ft. 


0-3, 


29-4 


67-5 


1,070 


1,067 


1-7, 


26-6 


120-3 


1,029 


200 


0-4, 


29-2 


740 


1,063 


800 


1-8, 


26-4 


122-4 


1,028 


190 


0-5, 


29-0 


80-0 


1,058 


640 


1-9, 


26-2 


124-6 


1,026 


181 


0-6, 


28-8 


85-5 


1,054 


535 


2-0, 


26-0 


126-7 


1,025 


172 


0-7, 


28-6 


90-4 


1,051 


461 


2-1, 


25-8 


128-6 


1,024 


165 


0-8, 


28-4 


94-5 


1,048 


410 


2-2, 


25-6 


130-4 


1,022 


158 


0-9, 


28-2 


98-5 


1,045 


367 


2-3, 


25-4 


132-2 


1,021 


152 


i-o, 


28-0 


102-0 


1,042 


333 


2-4, 


25-2 


1340 


1,020 


146 


11, 


27-8 


105-0 


1,040 


306 


2-5, 


25-0 


135-6 


1,019 


140 


1-2, 


27-6 


108-0 


1,038 


282 


2-6, 


24-8 


136-9 


1,018 


135 


1-3, 


27-4 


1110 


1,036 


260 


2-7, 


24-6 


138-2 


1,017 


130 


1-4, 


27-2 


113-7 


1,034 


240 


2-8, 


24-4 


139-6 


1,016 


125 


1-5, 


270 


1160 


1,033 


225 


2-9, 


24-2 


141-0 


1,015 


121 


1-6, 


26-8 


118-2 


1,031 


212 


3-0, 


240 


142-2 


1,014 


118 



Condenser Tubes. — They are. as a rule, made of brass, solid drawn, and 
tested both by hydraulic pressure and steam ; the latter test is a very useful 
one, as faults which escape detection under water pressure are often found 
out by steam ; these faults are due generally to minute particles of flux or 
slag in the original ingot, and sometimes the faults are in the form of cracks 
done in the process of drawing. It is by no means an uncommon thing to 
find a few tubes in a new condenser leaking through minute holes of various 
shapes ; these holes soon become enlarged if the tube is not at once stopped 
or withdrawn. These faults are not confined to the tubes of a few makers, 
but may be found in those of all makers at some time or other. Tinning is, 
as a rule, a preventive, as the defective places are in the process covered 
or filled with that metal, but it is seldom resorted to now. The Admiralty 
method of adding a small amount of tin to the alloy of copper and zinc has 
proved a good preventive of corrosion, and Mr. Philip, Admiralty Chemist, has 
shown by statistics its efficacy in a large number of condensers in H.M. Navy. 

Condenser tubes were usually made of a composition of 68 per cent, of best 
selected copper, and 32 per cent, of best Silesian spelter* The Admiralty, 
however, always specify the tubes to be made of 70 per cent, of best selected 
copper, and to have 1 per cent, of tin in the composition, and test them 
to a pressure of 300 lbs. per square inch. To prove that the tubes are of the 
70/30 alloy, a few pounds of them are melted in a closed crucible, and suffi- 
cient spelter added to bring the mixture to contain 62 per cent, of copper. 
The metal is then cast into an ingot, which when cold is rolled into a sheet, 
strips are cut from it and tested, and if satisfactory should have an ultimate 
tensile strength of 24 tons per square inch. In the mercantile marine the 
tubes are, as a rule, J inch diameter externally, and 18 L.S.G. thick (0-049 
inch) ; and 16 L.S.G. (0-065), under some exceptional circumstances. In 
H.M. Navy, the tubes used to be 18 to 19 L.S.G. thick, tinned on both sides ; 



* The tubes of the mercantile marine are usually 
cent, of lead has been tried with success. 



70/30 mixture without additions, but latterly 2 per 



364 MANUAL OF MARINE ENGINEERING. 

but now the Admiralty do not require the tubes to be tinned. On account 
of the economy of space and weight that is effected with small tubes all 
Naval engines are now fitted with condensers having tubes f inch diameter. 
The smaller the tubes, the larger is the surface which can be got in a 
■certain space. Since larger tubes are of necessity somewhat thicker than 
the smaller ones, a square foot of surface costs more when they are adopted, 
-and is not so efficient. Patent tubes made from sheet brass 22 B.W.G. 
thick, and joined at the seams like a tinsmith's joint and soft soldered, have 
been tried. The advantage claimed for them is the uniformity, whereby as 
little as 22 B.W.G. is sufficient .thickness, while it would not be safe to use 
tlrawn tubes so thin. 

The length of the tube depends on the arrangement of the condenser, but 
when they are not held tightly in the plates, but only packed, their unsup- 
ported length should not exceed 100 diameters ; when held with tight-fitting 
ierrules it may be 120 diameters. 

Tube-plates are now always made of brass, either cast or rolled into plates 
of suitable size ; the latter is preferable, as the rolled brass is very tough and 
close grained, and as strong as wrought iron. Formerly it was no uncommon 
thing to make the tube-plates of cast iron from If to 2h inches thick, and 
while some were converted into a substance resembling plumbago after two 
or three years' work, others have been found sound and good after twelve 
years' continuous work. 

Rolled brass tube-plates should be from T3 to 1*5 times the diameter of 
tubes in thickness, depending on the method of packing. When the packings 
go completely through the plates the latter, but when only partly through, 
the former is sufficient. Hence, for f-incb. tubes the plates are usually f to 
1 inch thick with glands and tape-packings, and 1 to 1£ inches thick with 
wooden ferrules. In the Navy the tube-plates are generally 1 to f inch 
thick, the tubes being £ inch diameter and 19 L.S.G. thick, but in the 
" Destroyers " the plates are only £ inch thick ; in their case, however, the 
plates are of small diameter and stayed in the middle ; and it may be added 
that leakage of tubes was no uncommon occurrence in such ships, so that 
this is as thin as they can safely be employed. 

The tube-plates should be secured to their seatings by brass studs and 
nuts, or brass screw-bolts ; in fact, there must be no wrought iron of any 
kind on the sea-water side of a condenser. When the tube-plates are of 
large area it is advisable to stay them by brass rods, to prevent them from 
bulging or collapsing. 

Tube Packings. — All attempts to drift or expand the tubes tightly into 
holes in a brass plate fail, owing to the softness of both plates and tubes ; and 
if it could be done it would be found impossible to draw the tubes for examina- 
tion and cleaning without damage. Fig. 119 shows a very simple plan, and 
one that proved effective under all circumstances, and essential with cast-iron 
plates. The ferrule is made of soft wood, such as pine or lime tree, very dry 
and well seasoned ; they were made nearly an eighth of an inch larger in 
diameter than the hole into which they had to fit, and a good fit on the tube. 
Before fitting them into place they were squeezed through a die in a press 
until they could be easily driven into their holes ; soon after being fitted into 
place they absorb moisture and expand circumferentially at each end, and 
become exceedingly tight on the tube and in the hole. After twelve years' 



TUBE PACKINGS AND FERRULES. 



365 



service they were found quite sound. It is urged against them that they are 
apt to shrink and drop out when the condenser is not in use, but this is not 
the case, as the swelled projecting ends form collars to prevent this, and- 
they do not shrink so much as is generally supposed, unless by unusual heat. 
This is one of the cheapest forms of tube-packing, and although not used 
now, was often employed in the mercantile marine of this and other countries. 
The plan adopted in H.M. Navy, and generally in the mercantile marine, 
is that shown in fig. 120. Each tube end passes through a stuffing-box fitted 





Fig. 119. 



Fig. 120. 




Fig. 121. 
Figs. 119-121. — Condenser Tube Packings and Ferrules. 

with a screwed gland, and kept tight by a tape washer, or some soft cordv 
as packing. This method is somewhat expensive, but it admits of the 
water being on either side of the tubes ; the packing is not affected by heat, 
and the condenser may remain unused for a very long time, and be quite 
tight at the end of it ; for these reasons it was chosen by the Admiralty. 
It is likewise the plan used by Hall in his early surface-condensers, and on. 



366 MANUAL OF MARINE ENGINEERING. 

the whole the most satisfactory one. Modern methods of drilling and tapping 
the holes, making the ferrules, etc., have very much reduced the cost, and 
so practically removed the only objection. Fig. 121 shows a modification 
of gland specially to suit vertical tubes ; the gland has an inside rim, which 
prevents the tube from slipping. It is now always used in condensers with 
horizontal tubes, and is especially necessary when they are long to prevent 
creeping. 

Steam Side and Water Side of Tubes. — This was somewhat of a vexed 
question, about which there is much to be said on both sides. The naval 
practice was formerly to circulate the water outside the tubes, so that the 
condenser shell may be kept cool and prevented from making the engine- 
room hotter than can be helped. The almost universal practice of the 
merchant service and Navy now is to circulate the water through the tubes. 
Independently of the particular reason for the choice of the Admiralty, the 
balance of argument is in favour of circulating the water through the tubes ; 
for when this is the case there is (i.) a larger surface of metal exposed to the 
hot steam ; (ii.) the grease that may be deposited on the tubes is easily 
removed by working a trifle warm, and using a solution of caustic soda or 
potash, and if this does not remove it, the deposit being soft does not prevent 
the tubes from being easily drawn, as is the case when scale from salt-water is 
deposited on their exterior surface ; (iii.) the scale from sea-water, which must 
be removed mechanically, can be so done without removing the tubes ; 
(iv.) a more extended and complete circulation of the cooling water is possible, 
and that without risk of air accumulation and without special and expensive 
diaphragms, etc. ; (v.) the condenser is more easily designed, and fits into the 
general arrangement of most engines, and is smaller, inasmuch as there is no 
need of an expansion chamber in front of the tubes, as is the case when 
steam passes through them ; (vi.) when it is necessary to examine the packings, 
or to plug a defective tube, only a water joint is broken ; and (vii.) the thin 
tubes are stronger to resist internal than external pressures, especially of brass. 
On the other hand, when the steam passes through the tubes the bulk of 
the fatty matter is deposited on the front tube-plate, and prevented from 
coating the tubes ; the large flat sides of the condenser are subject to the 
very slight pressure due to the head of water, and so may be made much 
lighter ; and there is less hot surface exposed in the engine-room. As little 
or no oil is now used for direct internal lubrication, and such as gets in with 
the piston and valve- rods is small, and a mineral oil, the argument as regards 
grease deposit is considerably modified from what it was when animal and 
vegetable oils were freely used, and even tallow on occasion, in both cylinders 
and boilers. 

Spacing of Tubes, etc. — The holes for ferrules or glands are usually \ inch 
larger in diameter than the tubes ; but when absolutely necessary the wood 
ferrules may be only J^ inch thick. 

The pitch of tubes when packed with wood ferrules is usually J inch more 
than the diameter of the ferrule hole. For example, the tube beinsf f inch 
external diameter, the ferrule hole will be 1 inch and the pitch of the tubes 
1 J inch. In the Navy, with tubes f inch diameter and the tube-plates | inch 
to 1 inch thick, the pitch of holes is 1] inch, and in "Destroyers" only 
}f inch. In the mercantile marine, with tubes f inch diameter and the tube- 
plates 1 inch thick, the pitch is 1| inch; for |-inch tubes, 1^- inch; and 



CONSTRUCTION OF SURFACE CONDENSER. 



367 



for 1-inch tubes, 1-^ inch. The tubes are generally arranged zigzag, and 
the number which may be fitted into a square foot of plate is as in Table xlii. 







TABLE XLII. 






Pitch of Tubes. 


No. in a 
square foot. 

1 


Pitch of Tubes. 


No. in a 
square foot. 


Pitch of Tubes. 


No. in a 
square foot. 


Inches. 




Inches. 




Inches. 




31 
15 


184 




... 


... 




1 


172 


1 : ' 


128 


h 


lib 


1A 


161 




... 


... 




1A 


152 


1A 


122 


1 ■• 


105 


1 3 


144 






... 




H 


136 


i£ 


lie 


1 » 


99 



The Body of the Condenser. — The surface-condenser was generally in the 
form of a cylinder or a rectangular parallelepiped, and sometimes a flattened 
cylinder ; the first and last forms are the best suited when weight is a great 
consideration, and the second and most convenient when space is of first 
importance ; the cylindrical form is by far the cheapest, as the patterns are 
very simple — the body, when not made of steel, copper, or brass sheets, 
being struck out ; the covers, tube-plates, and corresponding flanges can all 
be faced in a lathe ; this form also is by far the lightest, for the two reasons, 
that the circular plate is the form giving the minimum perimeter for a given 
area, and consequently a minimum barrel, and that the cylindrical form for 
. either internal or external pressure is the strongest, so that the thickness of 
metal will be the minimum. A modification of the cylindrical form possesses 
many of these advantages. 

The waterways, or chambers at each end of the condenser, are sometimes 
cast with it, and sometimes cast separately ; in the latter case there is an 
additional joint, but this is mitigated by its also forming the tube-plate joint ; 
in the former case there is only one joint less at each end through which aic 
can leak, but the plates are more troublesome to fit, and, except in the case 
of the cylindrical form, the tube plate flange is difficult to face. 

Great care should be taken in designing a condenser that free outlet is 
given to any air that may collect near the tubes, and all pockets, where it or 
dead water could lie, should be avoided, as a few of the tubes may get hot 
and leak from the above causes. 

The Construction of the Surface Condenser is to-day in all naval ships, in 
express steamers, and even in some cargo ships of steel sheets or plates from 
£ to | inch thick in the body, and of cast iron or brass in the waterways and 
parts exposed to sea-water in naval ships, and of cast iron in the mercantile 
marine where weight is not of paramount consideration. It is recognised 
now that there is no corrosive action on the body exposed to steam and 
condensed water, since the use of sea-water as the supplementary feed has 
been given up, and brass in the waterways can be protected by zinc slabs, or 
even by a few steel studs ; cast iron also is easily protected by zinc slabs. 
The condensers of such craft as destroyers, where the last ounce must be 
saved, should be of sheet brass, cylindrical in form, and of sufficient thickness 
to withstand atmospheric pressure and its own weight. With such condensers, 



368 



MANUAL OF MARINE ENGINEERING. 



as also with the brass waterways, the mudhole and peephole doors may be 
of cast iron or steel with advantage, as then they form the protectors against 
corrosion of the brass from sea-water, and can be easily and cheaply replaced. 
The shape of the modern condenser is either as an approximation to 
a triangle, or what is known as heart-shaped in transverse section, as 
shown in fig. 118. The cylindrical condenser is generally fitted with baffles 
of some kind, although with the modern entering in so much enlarged, 
especially as it is for turbines, the need to protect the upper rows of tubes 
from the pulsating impact of the steam has almost ceased to exist. With 
the ordinary rectangular or nearly rectangular condenser of the mercantile 
marine and the circular section exhaust pipe there should still be this screen, 
as well as means for draining the condensed water when formed without 
further contact with other tubes. 

TABLE XL III. — Trials of I.J. Battleship "Ibuki" — Curtis Turbines. 



Date. 



Kinds of 
Trial. 



Aug. 12, 1909, Full power 
Aug. 7, 1909, 
July 31, 1909, 
July 26, 1909, 
July 24, 1909, 



Baro- 
meter 

in 
Inches. 



29 995 
29 990 

29 900 

30 300 
30 320 



Temperature (Deg. Fahr.). 



Super Heat 

ifter through 

Regulating 

Valve. 



Star- 
board. 


21-8 

150 

9-8 

1-3 

31 



Port. 



22-9 

10-3 

3 9 

2 

4-3 



Main Condenser. 



Sea Water F" 



Inlet. 



Out- 
let. 



74 6 

71-2 

75 
73 2 

78 2 



977 
914 
93-4 

90-8 
92 3 



Con- 
densed 
Water, 

F°. 



90-9 
S6 

88-5 
86-7 
88-7 



Feed- 
Water 

after 
through 

Feed 
Heater. 



13111 

132-75 

151-6 

136-8 

145-2 



Mean 


Vacuum 


in Main 


Condensers 


at 


Top. 


Bottom 


26-33 


2S-42 


26 90 2S25 


27-10 ' 27 90 


27-80; 2S50 


28-03 


28-32 



Date. 



Boiler. 



8.H.P. 

per 
Sq. Foot 

of 
Grate. 



Aug. 12, 1909, 
Aug. 7, 1909. 
July 31, 1909. 
July 26, 1909, 
July 24, 1909, 



16 45 

12-80 

9-72 

9-23 

9 05 



Water Rate of Main 
Engine per S.H.P. 



Main Condenser. 



Not 
Corrected 

to 
Contract 
Condi- 
tions. 



Corrected *&£ 
Contact I Bfcfi* 

tions. (V ™- J per Hour 
(Per hour. )Cernom.; .„ Lbg- 



15 050 
15-686 

16 505 
18-652 
21074 



13-77 
14-76 
15-62 
17-75 
20-57 



1 1 -505 
9 099 
7 253 
5-038 
3-065 



Circulat- 
ing 
Sea \\ ater 

per Lb. 
of Steam 

in Lbs. 



44 -296 
39-685 
41113 
310§6 
30 946 



Main Engine. 



Bucket 

Speed 

in Feet 

per Sec. 



Steam 

Velocity 

in Feet 

per Sec 

(IstStage.) 



157 3 

147-7 

136-2 

118-4 

95-6 



2,142-5 
2,330 
2,379 
2,650-0 
2,943 



Efficiency 

of 
Turbine. 
Per cent 



52 45 
48-95 
46 30 
40 70 
3510 



The Stiffening of Flat Surfaces should be by means of ribs cast with it 

when of cast iron or cast brass, and 25 thicknesses apart for cast iron and 

40 for tough brass and sheet steel and brass, which should have angle or 

("-iron stitt'eners riveted on. Malleable flat surfaces can be stiffened by 



QUANTITY OF COOLING WATER. 369 

corrugations formed with it about 35 thicknesses pitch if fairly deep ; if 
the corrugating is light, the pitch should be much less. 

Quantity Of Cooling Water. — The necessary amount of circulating water 
may be calculated in the same way as that for injection water (see p. 345). 
on the principle that the exhaust steam has a certain quantity of heat which 
is to be expended in raising a mass of sea-water of a certain temperature to 
nearly that corresponding to that in the condenser. The quantity of sea- 
water will, therefore, depend on its initial temperature, which in actual 
practice may vary from 40° in the winter of temperate zones to 80° of the 
West Indies and other subtropical seas. In the latter case, with a vacuum 
of 28 inches, a pound of water requires only 20 thermal units to raise it to 
100°, while 60 are required in the former. From this it is seen that the 
quantity of circulating water required in the tropics is three times that of 
the North Atlantic in the spring of the year. 

As before, let T l be temperature of the steam on entering the condenser, 
and L the latent heat : T the temperature of the circulating water, and Q 
its quantity ; T the temperature of the water on leaving the condenser, and 
T 3 the temperature of the hot-well. 

The heat to be absorbed by the cooling water is now (Tj + L) — T 3 ; 
and this amount of heat must be equal to Q (T 2 — T ). Hence, 

Q = (T, + L) - T 3 ■*■ (T, - T„). 
Q= 1,114 + 03 T.-T, 

■*-2 x 

Example. — To find the amount of circulating water required by an engine 
whose steam exhausts at 8 lbs. pressure absolute, the temperature of the 
sea being 60°, and (2) the amount required when the temperature of the 
sea is 75°. The temperature of the hot- well to be 120°, and th'at of the 
water at the discharge 100°. Vacuum 26 inches. The temperature corre- 
sponding to 8 lbs. is 183 . 

(1) Q = UH + 0-3X183 -120 = 
W w 100—60 

That is, the water required is 26-22 times the weight of steam. 

With the jet condenser under similar conditions the quantity was only 
17-48 times {v. p. 346). 

(2) When the sea is at 75° 



n 1,114 + 0-3 x 183-120 . 

Q = loo— 75 = 41 9o tlmes - 

It is usual to provide pumping power sufficient to supply 30 times the weight 
of steam for general traders, and as much as 40 times for ships working in 
subtropical seas. As will be shown in another chapter, if the circulating 
pump is double-acting, its capacity may be -^ in the former, and -^ in the 
latter case of the capacity of the low-pressure cylinder. 

Table xliv. shows the least weight of cooling water required in practice 
per pound of steam entering the condenser at a pressure of 12 lbs. absolute. 
Modern air pumps can rffaintain a vacuum of 29 inches, but it will be seen 

24 



S70 



MANUAL OF MARINE ENGINEERING. 



by this table that the quantity of cooling water required for it at a tem- 
perature of 70° is enormous, and with water at 80° is impossible. 

TABLE XUV. — Ratio of Cooling Water to Steam Condense/). 



Vacuum, 


. Ins. 


250 


25-5 


26-0 


26-5 


27-0 


27-5 


28-0 


28-5 


29-0 


29-3 


Sea water, 


. 50 c F. 


13-9 14-7 


15-4 


16-4 


18-2 ! 20-1 


22-8 


28-0 


4(1-4 


65-0 


)> 


. 60° F. 


16-0 


17-1 


18-1 


19-5 


22-1 24-S 


29-0 


38-0 


04-3 


]3H 1 


»> 


. 70° F. 


18-9 


20-5 


21-8 


24-0 


27-9 32-3 


40-0 


66-3 


156 


. . 


j» 


. 80° F. 


231 


25-5 


27-6 


31-0 


37-9 ! 46-3 


63-0 


133 


" 





The highest possible vacuum with the temperature of cooling water at 
80 c is 28"8 ; and the ratio no less than 270. The highest vacuum when the 
water is 70° will be 29' 1 inches, and the ratio then 220. In the temperate 
zone with sea-water at 60° a vacuum of 29 - 4 can be maintained by passing 
280 times the weight of steam condensed. 

In every-day practice, and the condenser not too clean, the quantity 
of water required may easily be 10 to 20 per cent, more than that given in 
this table. 

Passage of Circulating Water. — The water must be caused to pass over 
a sufficient amount of surface to become duly heated, if the minimum quantity 
is to be used. In practice it should tra\el at least 20 feet lineally through 
the tubes before leaving the condenser ; if this cannot be arranged, then it 
must remain longer in contact with the surface. Hence, in small condensers, 
where the steam is outside the tubes, the water circulates only twice through 
them at a slow pace ; in larger condensers it may circulate twice through 
long tubes, or three or four times through shorter tubes at a higher ve