Skip to main content

Full text of "Parametric study of the lubrication of thrust loaded 120-mm bore ball bearings to 3 million DN"

See other formats


N73 26479 



NASA TECHNICAL NASA TM X- 68264 

MEMORANDUM 



oo 

vO 
I 

X 



< 

CO 

< 






PARAMETRIC STUDY OF THE LUBRICATION OF THRUST 
LOADED 120-MM BORE BALL BEARINGS TO 3 MILLION DN 



by H. Signer, E. N. Bamberger, and E. V. Zaretsky 
Lewis Research Center 
Cleveland, Ohio 44135 

TECHNICAL PAPER proposed for presentation at 

Joint Lubrication Conference cosponsored by the American 

Society of Lubrication Engineers and the American 

Society of Mechanical Engineers 

Atlanta, Georgia, October 16-18, 1973 



1 



PARAMETRIC STUDY OF THE LUBRICATION OF THRUST LOADED 120-MM 
BORE BALL BEARINGS TO 3 MILLION DN 
by H. Signer , E. N. Bamberger , and E. V. Zaretsky' 
Lewis Research Center 
ABSTRACT 
A parametric study was performed with 120-mm bore angular-contact 
ball bearings under varying thrust loads, bearing and lubricant tem- 
peratures, and cooling and lubricant flow rates. Contact angles were 
nominally 20° and 24° with bearing speeds to 3 million DN. Endurance 
tests were run at 3 million DN and a temperature of 492 K (425 F) 
with 10 bearings having a nominal 24 contact angle at a thrust load 
of 22241 N (5000 lb). Bearing operating temperature, differences in 
temperatures between the inner and outer races, and bearing power con- 
sumption can be tuned to any desirable operating requirement by vary- 
ing 4 parameters. These parameters are outer-race cooling, inner-race 
cooling, lubricant flow to the inner race and oil inlet temperature. 
Preliminary endurance tests at 3 million DN and 492 K (425 F) indi- 
cate that long term bearing operation can be achieved with a high de- 
gree of reliability. 



* 

Industrial Tectonics, Inc. , Compton, California 

General Electric Co. , Cincinnati, Ohio 
"iNASA Lewis Research Center, Cleveland, Ohio - Member ASME 



INTRODUCTION 

Advanced air breathing engines for high—speed aircraft for the 
1980 's are expected to operate with bearing temperatures near 492 K 
(425 F) and at speeds approaching 3 million DN. (DN is a bearing 
speed parameter and is equal to the product of the bearing bore in 
millimeter and the shaft speed in rpm.) In support of these engines, 
as well as for similar high performance oriented bearing applications, 
a reliable bearing-lubricant system is required. Such a system re- 
quires essentially three key items. These are, a suitable lubricant, 
a reliable bearing structural material, and an optimized bearing de- 
sign coupled with the proper operational parameters needed to sustain 
ultrahigh speeds. 

Over the past decade several new classes of lubricants were de- 
veloped and evaluated, which extended the upper temperature range of 
lubricating fluids [1-4] . Of these, the poly-ester and tetra-ester 
fluids have proven to be most useful and applicable in typical air- 
breathing environments, and consequently have been widely accepted in 
current commercial and military applications [5]. These fluids have 
good thermal stability at temperatures to 505 K (450° F) . Bearing 
life at 492 K (425 F) with the tetra-esters exceeded AFBMA-predicted 
(catalogue) life by a factor in excess of four [3] . When test oil in 
[3] was replaced at a rate approximating the replenishment rate in 



Numbers in brackets designate References at end of paper. 



actual commercial engine usage, no significant increase in lubricant 
viscosity or acidity with time was observed at 492 K (425 F) . 

Research reported in [6] indicated that AISI M-50 steel can pro- 
duce the most favorable life results at elevated temperatures when com- 

i;^ ■: <. ' 
pared with other high-speed or high-temperature steels such asi'M-l, a.' I 

M-10, WB-49, etc. Rolling- element bearing tests to 589 K (600° F) 
with the AISI M-50 steel and a synthetic paraf finic oil produced lives 
in excess of 13 times the AFBMA predicted life. 

In conventional rolling-element bearings, both metallic and non- 
metallic cages have found widespread use. Precision bearings, such as 
those used for aircraft applications, are usually equipped with cages 
machined from iron or copper base alloys. Where marginal lubrication 
is suspected during operation, silver plating of the cage material has 
been used to attenuate the detrimental effects of intermittent or mom- 
entary metal to metal contact. The bearing tests reported in [1-4] 
successfully utilized a cage material made of a nickel base alloy (AMS 
4892). In most critical high-temperature, high-load and high-speed 

aircraft applications, the cages are a carbon steel (AISI 4340 or sim- 

«. Bcckweil C hotdness of apptoxim-atelji 
ilar) hardened to ^ 33 and silver plated for extra protection. 

AIST 
With the tetra-ester lubricant and the/^M-50 steel, two of the 

three key elements essential to successful large-diameter, high-load 
and ultrahigh-speed bearing operation are now specified. However, high- 
speed bearing operation in the range of 3 million DN requires more than 
the proper lubricant and bearing material. Heat generation within the 
bearing itself is extremely critical, as is component loading due to 



centrifugal effects. Jones [7-9] first considered speed effects on 
bearing life and dynamics without considering the effect of the lubri- 
cant. Subsequently, Harris [10, 11] expanded these bearing analyses 
including lubricant effects. 

At high speed, the effect of centrifugal loading of the rolling 
elements against the outer race of the bearing becomes extremely im- 
portant. Theoretical life calculations for a 150-mm bore angular- 
contact ball bearing operating at 3 million DN (20 000 rpm) predict 
that this bearing has approximately 20% AFBMA-calculated life [12]. 
This decrease in predicted life is due to the increased stress in the 
outer race caused by centrifugal effects. The expected final result 
is extremely short bearing life at speeds much above 2 million DN both 
in actual running time (hr) and in total bearing inner-race revolutions. 

Another problem of operating bearings at high speed is the need to 
adequately cool the bearing components because of excessive heat gen- 
eration. A method which has been used successfully to 3 million DN is 
cooling lubricant applied under the race [13] . In this method lubri- 
cant is centrifugally injected through the split inner-race and shoulders 
of an angular-contract ball bearing by means of a plurality of radial 
holes. As a result, both the cooling and lubricant function is 
accomplished. 

The research reported herein, which is based on the work reported 
in [14, 15], was undertaken to investigate the performance of optimally 
designed 120-mm bore angular -contact ball bearings at speeds to 3 million 
DN. The primary objectives were to (a) determine the operating 



characteristics under variable lubricant flow conditions at 3 million DN, 

(b) determine the effect of speed and load on bearing performance and 

(c) conduct preliminary bearing endurance tests at 3 million DN and at a 
temperature of 492 K (425° F) . 

HIGH-SPEED BEARING TESTER 
A schematic of the high-speed, high-temperature bearing tester used 
in these tests is shown in ? x 1. This tester is described in detail 
in [1, 2] and has been subsequently modified to operate at speeds of 
25 000 rpm. The tester consists of a shaft to which two test bearings 
are attached. Loading is supplied through a system of ten springs which 
apply a thrust load to the bearings. Dual flat belts drive the test 
spindle from a 75 kW (100 hp) fixed speed electric motor. The drive 
motor is mounted to an adjustable base, so that drive pulleys for 12 000 
to 25 000 rpm can be used with the same drive belts. The drive motor is 
controlled by a reduced voltage auto-transformer starter which permits 
a selection of the motor acceleration rate during start-up. This con- 
trol protects the bearings efficiently from undesirable acceleration 

q 

during start-up. The lubrication system of the test rig delivers up to 

-2 
2.8x10 cubic meters (7.5 gallons). There are three lubricant loops in 

the system. The oil flow in each loop is metered by adjustable flow con- 
trol valves and can be individually measured by a flow rate indicator 
without interruption to the machine operation. Two of these loops are 
shown in "' s 2. The first of these loops supplies cooling oil to the 
test bearing outer race and is designated C . The second loop is divided 
by a lubricant manifold which feeds individual annular grooves or channels 



at the shaft internal diameter proportioning the amount of oil which is 
to lubricate and/or cool each bearing inner race. L. designates the 
oil flow to the bearing through a plurality of radial holes in the 
center of the split inner race. C. designates the lubricant utilized 
to cool the bearing inner race and lubricate the contact of the cage 
with the race land through a plurality of radial holes in the inner-race 
shoulder. The lubricant system permits a selection of various lubricant 
schemes, including bearing lubrication through the inner-race hplii:, lub- 
rication of the cage-race shoulder contact region, the application of 
inner- and/or outer-race cooling, and a selection of any desired flow 
ratio for cooling and lubrication as well as the conventional lubrica- 
tion through jets. The third lubricant loop is fed into the slave bear- 
ing which supports the shaft (not shown in Figs, 1 and 2) . By the sys- 
tem of valves and manifolds previously discussed an unlimited number of 
combinations of oil flows can be achieved to evaluate various conditions. 
Consequently, values of L., C, and C can be independent of each other. 

The machine instrumentation includes the standard protective cir- 
cuits which shut down a test when a bearing failure occurs, or if any of 
the test parameters deviate from the programmed conditions. Measurements 
were made of bearing inner-race speed, bearing cage speed, test spindle 
excursion, oil flow, test bearing inner- and outer-race and lubniani 
temperatures, and machine vibration level. The speed and spindle excur- 
sion measurements were made with proximity probes and displayed by numer- 
ical read-out and oscilloscope, respectively. The oil flow was estab- 
lished by a flow meter, and bearing outer-i&cc and lubricant inlet and 



outlet temperatures were measured by thermocouples and continuously re- 
corded by a strip chart recorder. The inner— race temperature of the 
front test bearing was measured with an infrared pyrometer. 

TEST BEARINGS 

The test bearings were ABEC-5 grade, split inner-race 120-mm bore 
ball bearings. The inner and outer races, as well as the balls were 
manufactured from one heat of double vacuum-^nelted (vacuum-induction 
melted consumable electrode vacuum remelted) AISI M-50 steel. The 
chemical analysis of the particular heat is shown in Tablft 1 The nom- 
inal hardness of the balls and races was Rockwell C-63 at room temper- 
ature. Each bearing contained 15 balls, 2.0638 cm (13/16 in.) in diam- 
eter. The cage was a one piece inner- land riding type, made out of an 

RockweM C hardness 
iron base alloy (AMS 6415) heat-treated to a/|range of 28 to i5 and hav- 
ing a 0.005 cm (0.002 in.) maximum thickness of silver plate (AMS 2410). 
The cage balance was 3 gm-cm (0.042 oz-in.). 

The retained austenite content of the ball and race material was 
less than 3 percent. The Inner- and outer-race curvatures were 54 and 
52 percent, respectively. All components with the exception of the cage 
were matched within + one Rockwell-C point. This matching assured a 
nominal differential hardness in all bearings (i.eo, the ball hardness 
minus the race hardness, commonly called AH) of zero [16], Surface 
finish of the balls was 2.5 ycm (1 microinch) AA and the inner and ouier 
raceways were held to a 5 ucm (2 microinch) AA maximum surface finish. 

A photograph of the test bearing is shown in Fl^. 3, Thf fce^: - 
ing design permitted under-race lubrication by virtue of radial slots 



A 



machined into the halves of the split inner races. It had been shown 
in [13] that this was the most reliable technique for lubricating high- 
speed bearings. Provision was also made for inner-race land to cage 
lubrication, by the incorporation of several small diameter holes ra- 
diating from the bore of the inner ace to the center of the inner 
:a e shoulder. 

LUBRICANT 
The oil used for the parametric studies as well as for the subse- 
quent long-time high-speed (3 million DN) endur^rsce tests, wtis a 5 -cent i- 
stoke neopentylpolyol (tetra) ester. This is a Type II oil, qualified 
to MIL-L-23699 as well as to the internal oil specifications of most 
major aircraft- engine producers. The major properties of subject oil 
are presented in T^^hle 2 and a temperature-viscosity curve is shown in 

Fig, 4. 

TEST PROCEDURE 
The test procedure was adjusted according to the test conditions 
to be evaluated. Generally, a program cycle was defined which would 
allow the evaluation of a number of conditions without a major inter- 
ruption. With the exception of speed, all test parameters such as load, 
lubricant flow rate and oil temperature could be adjusted while the 
tester was in operation. During operation, the tester was allowed to 
reach equilibrium condition before the data were recorded. For the 
long term endurance test at 3 million DN the procedure varied only to 
the extent that once the preset test parameters had been achieved no 
further adjustments were made. 



Power loss per bearing was determined by measuring line to line 
voltage and line current to the test-rig drive motor. Motor drive 
power was then calculated as a function of line current, reflecting 

bearing power usage at the various operating speeds. 

-3 

Approximately 1.9x10 cubic meters (0.5 gallon) of oil was re- 
plenished in every twenty-four hours of operation^ This amount of oil 
was removed from the sump and replaced with an equal amount of fresh 
oil. The rate of replenishment was approximately equal to 0.3 percent 
per hour of the entire sump capacity. This operation was performed 
without interruption of the test. By closely monitoring the oil re- 
plenished, it has been shown that no significant increases in either 
viscosity or acid number occur, even in tests on the order of 1000 
hours duration. 

The rationale for this replenishment is that it serves to maintain 
the lubricant properties, and specifically, viscosity and acid number, 
at a relatively constant level. If no oil changes were made, the lubri- 
cant would present a continuously changing parameter with indeterminate 
effects on bearing performance. The selected replenishment rate is 
based on actual oil consumption rates in turbojet engines. 

RESULTS AND DISCUSSION 
Effect of Speed 

The effect of speed on the operating characteristics of the 120-mm 
bore angular-contact ball bearings are shown in i'lgs, 5 r..nd 6 for 
three thrust loads. The 20 and 24 contact angle bearings were run at 
nominal speeds of 12 000, 16 000, 20 000, and 25 000 rpm. Test 



10 



conditions included an oil— inlet temperature of 428 K (310 F) ; a lubri- 

-3 
cant flow rate through the inner race, L., of 1.2x10 cubic meters per 

_3 
minute (0.313 gpm) ; an inner-race cooling rate, C. , of 3.6x10 cubic 

meters per minute (0.94 gpm) or 3 L.; and an outer-race cooling flow 

-3 
rate, C , of 1,9x10 cubic meters per minute (0.5 gpm). Generally, 

throughout the range of speeds for the 6672- and 13345-N (1500- and 3000- 
Ib) thrust loads, the bearing inner-race temperatures were 3 to 8 K (5 
to 15 F) higher than the outer— race temperatures (Figs, 5(a) and (b)). 
For the 22241-N (5000— lb) thrust load, the temperatures of the inner 
and outer races were generally within 3 K (5 F) of each other. Over 
the range of speeds, the temperature of the bearing races increased from 
the range of 441 to 453 K (335° to 355° F) , to temperatures in the range 
of 478 to 494 K (400 to 430 F) or an increase of approximately 36 to 
42 K (65 to 75 F) . In general, the 24 contact angles bearings ran 
approximately 3 to 8 K (5 to 10 F) cooler than the 20 contact angle 
bearings over the entire speed range. This difference in temperature 
was not considered significant. 

The power loss for the bearings over the speed range is shown in 
figure 6. As would be expected, bearing power consumption increases 
linearly with speed for the three loads shown. At the thrust load of 
22241 N (5000 lb) and a speed of 25 000 rpm, the power loss is approx- 
imately 15 kW (20 horsepower) per bearing for the two contact angles. 
At the speed of 12 000 rpm and a thrust load of 6672 N (1500 lb) , the 
power loss is approximately 3 kW (4 horsepower) or 20 percent of the 
power loss at the high-speed, high-load condition. 



11 



Comparing the power consumption of the two contact-angle bearings at 
the 22241N (5000 lb) load and at low speeds, the 24° contact-angle bearing 
had a power loss approximately 25-percent greater than the 20° contact- 
angle bearing. At the higher speeds and for all three loads, the dif- 
ference In power loss between the two contact angles was essentially 
insignificant. 

Another effect of speed on bearing operation is the centrifugal 
effect on outer-race load or stress and dynamic contact angle. The ef- 
fect of speed on contact stress and operating contact angle using the 
methods of Harris [10, 11] are shown in Table 3. For these calcula- 
tions, a bearing operating temperature of 478 K (400 F) was assumed. 
These calculated data show that there is a more marked increase in 
stress at the outer-race ball contact at the lower thrust load of 
6672 N (1500 lb) than at the 22241-N (5000- lb) thrust load. This dif- 
ference in the relative increase in stress at the outer-race ball con- 
tact, probably accounts for the slightly higher rate of increase in 
power consumption with speed with the lower thrust- loaded bearing tests. 
Contact- angle measurements were made on bearings having an initial con- 
tact angle of 24° and tested at 12 000, 16 000, and 25 000 rpm at a load 
of 22241 N (4000 lb). The actual measurements were in excellent agree- 
ment with the prediction made in Table 3, with the exception of the 
inner- race contact angle at the maximum speed condition. Here the meas- 
ured angle was approximately 25 versus a predicted operating angle of 
34 . The reason for this discrepancy is not understood at the present 
time. 



12 



Effect of Load 
The effect of load on the bearing temperature is shown in Fig, 7 
Bearing temperatures increased nearly linearly with load for most of the 
speeds. However, a temperature rise of only 6 to 11 K (10 to 30 F) 
occurs when the load was increased from 6672 to 22241 N (1500 to 5000 lb). 
The difference in temperature between the inner and outer race for the 

24 contact-angle bearings was insignificant and generally was within 

±3 K (±5 F) . For the 20 contact-angle bearings, the variation in tem- 
perature between the inner and outer races was slightly greater at the 
lower loads. This difference in temperature, however, did not exceed 
8 K (15 F) . Bearing power loss as a function of load for the two con- 
tact angles is shown in figure 8. The greatest increase in power con- 
sumption at any speed was approximately 3 kW (4 horsepower) from the 
6672-N (1500'-lb) thrust load to the 22241-N (5000-lb) thrust load. The 
increase in power consumption with thrust is considered negligible for 
most practical applications. 

Effect of lubricant flow . - The effect of lubricant flow into the 
bearing, and that used for cooling of the inner and outer races was de- 
termined. A test condition was selected which incorporated a speed of 

25 000 rpm (3 million DN) and a thrust load of 22241 N (5000 lb). This 

condition was chosen on the basis of approaching maximum Hertzian 

6 2 
Stresses [2068x10 N/m (-300 000 psi)] and speeds which can reasonably 

be anticipated in advanced state-of-the-art turbojet engines. Outer- 
race cooling flow (C ) was 9.5xlO~^, 1.9x10""^, 3.8xl0~^, and 5.7xlO~^ 
cubic meters per minute (0.25, 0.5, 1.0, and 1.5 gpm) , Inner-race 



13 



-4 
lubricant flow (L.) ranged from 7.6x10 cubic meters per minute (0.2 

_3 
gpm) to approximately 5.7x10 cubic meters per minute (1.5 gpm) , Inner- 
race cooling (C.) was varied as a function of L.. Tbe result; ::z ihese 
tests are shown in Figs, 9 and 10. 

Referring to Fig. 9(a), the test results are shown for the 20 
contact-angle bearing with no inner-race cooling flow (C-) except for 
that lubricant entering the inner race through the slots in the mating 
surfaces of the split inner race. The temperatures of the inner race 
varied generally from 6 to 11 K (10 to 20 F) for a particular lubri- 
cant flow (L.) over the range of outer— race cooling rates (C ) . At a 

_3 
lubricant flow (L.) of approximately 2.3x10 cubic meters per minute 

(0.6 gpm) to the inner race, the temperature of this component ranged 

from approximately 486 to 497 K (415 to 435° F) at an oil inlet reropet- 

perature of 394 K (250 F) » The actual temperature depends upon the outer- 
race cooling flow (C ) . At an increased lubricant flow rate to the 

_3 
inner race (L.) of approximately 4.5x10 cubic meters per minute (1.2 

gpm), the temperature ranged from 472 to 475 K (390° to 395° F) . 

-4 
At an outer-race cooling flow (C ) of 9.5x10 cubic meters per 

minute (0.25 gpm) (fig. 9(a)), the temperature of the outer race was 

nearly equal to that of the inner race. As the flow rate to the outer 

race (C ) was increased, the outer-race temperature decreased. At an 

-3 
outer— race flow (C ) of 5.7x10 cubic meters per minute (1.5 gpm) the 

temperature of the outer race was approximately 17 K (30 F) lower than 
the Irxner- race temperature. The amount of decrease in outer-race tem- 
peratures, with increasing inner race flow (L.) for all values of C , 



14 



generally paralleled those of the inner race. What is significant is 
that the internal clearances of the bearing will be affected with the 
changes in the outer-race cooling flow (C ) . 

Referring to Fig. 9(b) wherein the inner-race cooling flow rate 
(C.) was 1.33 L , the temperature of the inner race ranged from 478 K 
(400 F) at an inner-race lubricant flow rate (L.) of 1.1x10 cubic 

meters per minute (0.3 gpm) to approximately 461 K (370 F) when the 

-3 
inner-race flow rate was doubled to 2.3x10 cubic meters per minute 

(0.6 gpm). Beyond this value of L., the temperature of the inner race 
increased again. 

In general, the outer-race temperature paralleled the inner-race 
temperature for the various outer-race cooling flow rates (C ). How- 
ever, the inner-race temperatures were not significantly affected by 

the outer-race flow rates (C ) . At an outer-race cooling flow (C ) of 

o ° o 

-3 
1.9x10 cubic meters per minute (0.5 gpm), the temperature of the 

inner and outer race was approximately equal (fig. 9(b)). However, at 

-3 
an L. value of 2.3x10 cubic meters per minute (0.6 gpm) and with 

inner-race cooling (C.) , a decrease of as much as 36 K (55 F) can be 
achieved over the same bearing having no lubricant cooling flow to the 
inner-race lands (Fig. 9(a)). It may be concluded, that inner-land 
cooling (C.) can play a significant role in reducing the detrimental 
thermal effects on the bearing and specifically under marginal lubri- 
cation conditions. 

The data obtained with the 24 contact-angle bearing and a value 
of C. = are shown in Fig. 9(c). The results under this operating 



15 



condition were generally similar to those obtained for the 20 contact 
angle bearing shown in Fig. 9(a). However, the inner-race tempera- 
tures of the 24 contact-angle bearing varied over a greater range. At 

_3 
a lubricant flow to the inner race (L.) of approximately 1.9x10 cubic 

meters per minute (0.5 gpm) , the temperature ranged from approximately 
478 to 494 K (400 to 430 F) with various levels of outer-race cooling 
flow rate (C ) . This temperature of the inner race decreased with in- 
creasing inner-race lubricant flow rate (L.)' At a value of L. equal 

-3 
to 3.8x10 cubic meters per minute (1 gpm), operating temperatures were 

at a minimum ranging between 464 to 479 K (375 to 390 F) . Beyond this 

flow rate temperatures began to increase as lubricant flow (L.) increased. 

This rise in temperature was probably due to the increased quantity of 

lubricant within the bearing cavity and to the resultant churning effects. 

Again, the outer-race temperature closely paralleled the inner-race tem- 

outer- 
perature and decreased with increased! race cooling flow (C ). The mini- 
mum outer-race temperature, achieved with an outer-race cooling flow (C ) 
of 5.7x10 cubic meters (1.5 gpm), was approximately 455 K (360 F) . 

At an inner-race cooling flow of C. equal 1.33 L^ shown in Fig. 9(d), 
the minimum inner-race temperature was approximately 458 K (365 F) . The 

minimum temperature at the outer race was obtained at an outer-race cool- 

-3 
ing flow (C ) of 5.7x10 cubic meters per minute (1.5 gpm). This tem- 
perature was approximately 447 K (345 F) . At a value of L. in excess 

-3 
of 1.5x10 cubic meters per minute (0.4 gpm), temperature at the inner 

race began to increase. Likewise, the temperatures at the outer race 

which parallel those of the inner race, except at an outer-race flow rate 



16 

-4 9 

of 9.5x10 cubic meters per minute (0.25 gpm) , began to increase. For 
a lubricant inner-race flow C. equal to 3.0 L. (Fig. 9(e)) the range 

of component temperatures was very narrow, decreasing from 460 K (370 F) 

-4 
at a flow rate of L. equal to 7.6x10 cubic meters per minute (0.2 gpm) 

to 350° F at L. equal to 1.5x10 cubic meters per minute (0.4 gpm). 

Beyond this point, temperature rapidly increased at the inner race. The 

optimum matching of inner- to outer-race temperatures was obtained with 

-3 

outer-race lubricant cooling rates (C ) ranging between 3.8x10 to 

_3 
5.7x10 cubic meters per minute (1 to 1.5 gpm). 

A summary of the inner-race temperatures is shown in Fig. 9(f). ,, 
From these data it may be concluded that the 24 contact-angle bearing 
ran slightly cooler than its 20 contact— angle equivalent. However, 
this difference in temperature was only about 8 K (15 F) over most of 
the comparable operating conditions. As the inner— race cooling flow 
(C.) increased, temperatures decreased significantly. The maximum dif- 
ference in temperature 50 K (90 F) was observed between C. equal to 
3.0 L. and C. equals zero. In the case where no cooling was attempted 
of the inner race except for that provided by the inner-race lubricant 
flow (L.), most of this difference can be accounted for by the fact that 
a major cause of the heat generation within the bearing is likely due to 
churning of the lubricant within the bearing cavity. 

Power loss as a function of inner-race cooling (C.) and lubrication 
(L.) is presented in Fig. 10. The power loss increases linearly with 
increased flow to the inner race. For outer-race cooling flows from 
9.5xl0~ to 3.8x10 cubic meters per minute (0.25 to 1.0 gpm) there was 



17 



essentially no significant difference in power loss. At an outer-race 

-3 
cooling flow (C ) of 5.7x10 cubic meters per minute (1.5 gpm) the 

power loss was approximately 0.75 to 3 kW (1 to 4 horsepower) greater 

than at the lower values of C . This difference in power consumption 

can be attributed to the reduction in bearing Internal clearances due 

to temperature differences between the inner and outer races. The rate 

of increase in power loss with L, is due to both the amount of oil 

used for the inner-race cooling flow (C.) and that used for the primary 

lubricant flow, L.. 
1 

The rapid increase in power consumption with Increased C. can be 
attributed to a larger quantity of lubricant being entrapped within the 
bearing cavity at the higher lubricant flow rates. For all values of 
C , the extrapolated value of power loss where L. was zero was in the 
range of 9 to 12 kW (12 to 16 horsepower) . 

Referring to Fig. 10(f) which is a summary of the range of power 
loss for the 20 and 24 contact angles, there appears to be no signifi- 
cant difference between the power loss with either of these contact 
angles for a particular inner- race cooling flow (C). This is expected 
since power loss increases when the amount and viscosity of oil within 
the bearing cavity is increased. It must be recognized, however, that 
all power loss data presented herein are based solely on shaft horse- 
power measurements. As such, they do not necessarily encompass the 
power requirement for pumping the lubricant, and, more specifically, 
for circulating the oil used to cool the outer race. 



18 

The above results indicate that the bearing can be temperature and 
power tuned to any specific operating condition depending upon the lubri- 
cant characteristics. The concept of Bearing Thermal Management proposed 
herein, is believed to be the proper technological approach to high-speed 
bearing operation. The basis of this is the recognition that total and 
flexible thermal control over all of the bearing components is essential 
to achieve a reliable high-speed, highly-loaded bearing. This in turn re- 
quires a lubrication scheme of sufficient sophistication to achieve the 
thermal controls and still permit its practical use in actual flight 
hardware. 

Bearing endurance at 3 million DN . - Preliminary bearing endurance 
tests were conducted with the 24 contact— angle bearing at a speed of 
25 000 rpm (3 million DN) and a thrust load of 22241 N (5000 lb) . Under 
these conditions the maximum Hertz stresses in the inner and outer races 
are 1965x10 and 2096x10 N/m (285 000 and 304 000 psi) , respectively. 

For these long-time, high-speed bearing tests the cooling-flow rate per 

-3 

bearing to the outer race (C ) was 2.8x10 cubic meters per minute 

_3 
(0.75 gpm) . Lubricant flow to the inner race (L.) was 1.3x10 cubic 

meters per minute (Oc.35 gpm) and inner— race cooling flow (C.) was approx- 

~3 
imately 3-9x10 «:ubli: meters per mirxx^ (1 gpii;. 

The standard AFBMA (catalog)-life calculation predicts a bearing 
ten-percent (B,„) life under these operating conditions of approximately 
16 hours. (The bearing lO-percent (B,^) life is the operating time 
(life) at which 90% of a group of bearings will survive.) However, us- 
ing the material, lubricant and speed factors given in [17] a B^_ life 



19 

of about 175 hours would be a more reasonable prediction. Of the ten 
bearings initially tested all ran for 1000 hours without failure. These 
results show that long-term bearing operation at 3 million DN can be 
achieved with a high degree of reliability using sophisticated but cur- 
rently available state-of-the-art bearing materials and designs, lubri- 
cants, and lubrication techniques. 

SUMMARY 
A parametric study was performed with 120-mm bore angular-contact 
ball bearings having nominal 20 and 24 contact angles under varying 
thrust load, bearing and lubricant temperature, and cooling and lubri- 
cant flow rates at speeds to 3 million DN. Endurance tests were run 
at 3 million DN and a temperature of 492 K (425 F) with 10 bearings 

having a nominal 24 contact angle at a thrust load of 22241 N (5000 lb) 

f\ ft 7 

producing a maximum Hertz stress of 1965x10 and 2096x10 N/m (285 000 and 

304 000 psi) on the bearing inner and outer races, respectively. The 

following results were obtained: 

1. Bearing inner- and outer-race temperatures and power consump- 
tion were found to vary with load, speed, lubricant flow rate into the 
bearing, and lubricant cooling to the inner race. Lubricant cooling 
flow to the outer race was found to affect outer-race temperatures sig- 
nificantly, but had only a small effect on the measured inner-race tem- 
perature. Power loss due to change in lubricant cooling flow to the 
outer race was relatively insignificant. 

2. Bearing operating temperature, differences in temperatures be- 
tween the inner and outer races, and bearing power consumption can be 



20 

tuned to any desirable operating requirement by varying 4 parameters. 
These parameters are outer-race cooling, inner-race cooling, lubricant 
flow to the inner race and oil inlet temperature, 

3. All ten bearings which were endurance tested ran for times 
in excess of 1000 hours without failure. These results indicate that 
long— term bearing operation at 3 million DN can be achieved with a high 
degree of reliability using sophisticated but currently available state- 
of-the-art bearing materials, designs, lubricants, and lubrication 
techniques. 



21 



REFERENCES 

1. Bamberger, E. N. , Zaretsky, E. V., and Anderson, W. J., "Fatigue Life 

of 120-mm Bore Ball Bearings at 600 F with Fluorocarbon, Polyphenyl 
Ether and Synthetic Paraffinic Base Lubricants", NASA TN D-4850, 
1968. 

2. Bamberger, E. N. , Zaretsky, E. V., and Anderson, W. J., "Effect of 

Three Advanced Lubricants on High Temperature Bearing Life", Journal 
of Lubrication Technology, Trans. ASME, Series F, Vol. 92, No. 1, 
1970, pp. 23-33. 

3. Zaretsky, E. V., and Bamberger, E. N. , "Advanced Airbreathing Engine 

Lubricants Study with a Tetraester Fluid and a Synthetic Paraffinic 
Oil at 492 K (425° F)", NASA TN D-6771, 1972. 

4. Parker, R. J., and Zaretsky, E. V., "Effect of Oxygen Concentration on 

an Advanced Ester Lubricant in Bearing Tests at 400 and 450 F", 
NASA TN D-5269, 1969. 

5. D'Orazio, A. J., "Development and Utilization of Specification Mil-L- 

23699 - Synthetic Lubricating Oils for Aircraft Gas Turbine Engines", 
U.S. Navy NAPTC-AED-1868, 1968. 

6. Parker, R. J., and Zaretsky, E. V., "Rolling Element Fatigue Lives of 

Through-Hardened Bearing Materials", Journal of Lubrication Technol- 
ogy, Trans. ASME, Series F, Vol. 94, No. 2, 1972, pp. 155-173. 

7. Jones, A. B., "The Life of High Speed Ball Bearings", Trans. ASME, 

Vol. 74, No. 5, 1952, pp. 695-703. 

8. Jones, A. B. , "Ball Motion and Sliding Friction on Ball Bearings", 

Journal of Basic Engineering, Trans. ASME, Series D, Vol. 81, No. 1, 
1959, pp. 1-12. 



22 

9. Jones, A. B. , "A General Theory for Elastically Constrained Ball and 
Roller Bearings under Arbitrary Load and Speed Conditions", Journal 
of Basic Engineering, Trans. ASME, Series D, Vol. 82, No. 2, 1960, 
pp. 309-320. 

10. Harris, T. A., "An Analytical Method to Predict Skidding in Thrust 

Loaded, Angular Contact Ball Bearings", Journal of Lubrication 
Technology, Trans. ASME, Series F, Vol. 93, No. 1, 1971, pp. 17-24. 

11. Harris, T. A., "An Analytical Method to Predict Skidding in High Speed 

Roller Bearings", ASLE Transactions, Vol. 9, No. 3, 1966, pp. 229-241. 

12. Scibbe, H. W. , and Zaretsky, E. V., "Advanced Design Concepts for High 

Speed Bearings", ASME Paper 71-DE-50, 1971. 

13. Holmes, P. W. , "Evaluation of Drilled Ball Bearings at DN Values to 

Three Million", NASA CR-2004, NASA CR-2005, 1972. 

14. Zaretsky, E. V., Bamberger, E. N. , and Signer, H. , "Operating Charac- 

teristics of 120-mm Bore Ball Bearings at 3x10 DN", Proposed NASA 
Technical Note, 1974. 

15. Bamberger, E. N. , Zaretsky, E. V., and Signer, H. , "Effect of Speed 

and Load on Ultra High Speed Ball Bearings", Proposed NASA Technical 
Note, 1974. 

16. Zaretsky, E. V., Parker, R. J., Anderson, W. J., and Reichard, D. W., 

"Bearing Life and Failure Distribution as Affected by Actual Compo- 
nent Differential Hardness", NASA TN D-3101, 1965. 

17. Bamberger, E. N. , et al., "Life Adjustment Factors for Ball and 

Roller Bearings, An Engineering Design Guide", ASME, 1971. 



23 



TABLE 1 - CHEMICAL ANALYSIS OY VACUUM INDUCTION, CONSUMABLE- 
ELECTRODE VACUUM REMELTED AISI M-50 BEARING STEEL 



Element 


Composition, wt. % 






Races and Balls 


Carbon 


0.83 


Manganese 


.29 


Phosphorus 


.007 


Sulfur 


.005 


Silicon 


.25 


Chromium 


4.11 


Molybdenum 


4.32 


Vanadium 


.98 


Iron 


Balance 



24 



TABLE 2 - PROPERTIES OF TETRAESTER LUBRICANT 



Additives 


Antiwear 

Oxidation Inhibitor 

Antifoam 


Kinematic viscosity, cS, at - 
311 K (100° F) 
372 K (210° F) 
477 K (400° F) 


28.5 
5.22 
1.31 


Flash point, K (°F) 


533 (500) 


Fire point, K (°F) 


Unknown 


Autoignition temperature, K ( F) 


694 (800) 


Pour point, K (°F) 


214 (-75) 


Volatility (6.5 hr at 477 K 
(400° F)), wt. % 


3.2 


Specific heat at 47 7 K (400° F) , 
J/(kg)(K) (Btu/(lb)(°F)) 


2340 (0.54) 


Thermal conductivity at 477 K 
(400° F), J/(m)(sec)(K) 
(Btu/(hr)(ft)(OF)) 


0.13 (0.075) 


Specific gravity at 477 K (400° F) 


0.850 



25 

TABLE 3 - CALCULATED OPERATING CONTACT ANGLES AND STRESSES AS A 
FUNCTION OF INITIAL CONTACT ANGLE, SPEED AND LOAD 



Speed, 
rpm 


Thrust 

load, 

N (lb) 




Contact angle. 


deg 


Maximum Hertz 
stress, N/m (ksi) 


Unloaded 


Operating 




Outer 
race 


Inner 
race 


Outer race 


Inner race 


12 000 


6672 
(1500) 


20 


13 


26 


1365x10^ 
(198) 


1420x10^ 
(206) 




13345 
(3000) 




18 


26 


1572 
(228) 


1806 
(262) 




22241 
(5000) 




21 


27 


1779 
(258) 


2130 
(309) 




6672 
(1500) 


24 


14 


30 


1344 
(195) 


1351 
(196) 




3345 
(3000) 




19 


30 


1531 
(222) 


1731 
(251) 




22241 
(5000) 




23 


31 


1731 
(251) 


2048 
(297) 


16 000 


6672 
(1500) 


20 


9 


27 


1517x10^ 
(220) 


1400x10^ 
(203) 




3345 
(3000) 




14 


28 


1682 
(244) 


1779 
(258) 




22241 
(5000) 




18 


28 


1855 
(269) 


2096 
(304) 




6672 
(1500) 


24 


9 


32 


1503 
(218) 


1330 
(193) 




3345 
(3000) 




15 


32 


1648 
(239) 


1703 
(247) 




22241 
(5000) 




19 


32 


1813 
(263) 


2020 
(293) 


20 000 


6672 
(1500) 


20 


6 


28 


1682x10^ 
(244) 


1386x10^ 
(201) 




3345 
(3000) 




11 


29 


1813 
(263) 


1758 
(255) 




22241 
(5000) 




15 


29 


1958 
(284) 


2068 
(300) 




6672 
(1500) 


24 


6 


33 


1682 
(244) 


1324 
(192) 




3345 
(3000) 




12 


33 


1923 
(260) 


1682 
(244) 




22241 
(5000) 




16 


33 


1924 
(279) 


1993 
(289) 


25 000 


6672 
(1500) 


20 


5 


29 


1889x10^ 
(274) 


1365x10^ 
(198) 




3345 
(3000) 




8 


30 


1993 
(289) 


1731 
(251) 




22241 
(5000) 




12 


30 


2110 
(306) 


2041 
(296) 




6672 
(1500) 


24 


5 


34 


1862 
(270) 


1310 
(190) 




3345 
(3000) 




8 


34 


1993 
(289) 


1662 
(241) 




22241 
(5000) 




12 


34 


2096 
(304) 


1965 
(285) 



FIRST TEST BEARING 
LOAD PLATE 



FORWARD 
HOUSING 
ASSEMBLY- 

SIGHTTUBE 

SAPPHIRE 
WINDOWS- 



CONNECTOR 
FRONT PLATE- 



RETAINING 
PLATE -" 




r OUTER-RACE COOLING 
/ ,- INNER-RACE /-THERMOCOUPLE 
/ / COOLING^/ ^ 

ii^\ /"TrBELLOWS / ^^ REAR HOUSING 

ASSEMBLY 
-INNER-RACE 
LUBRICATION 

DRIVE SHAFT7 



OILOUT^-— ^ 



♦ 



OIL 

DRAIN Kx.^ 
T UBE ^ 

v///My//^// 



Fig. 1 High-speed, higii-temperature bearing test apparatus. 



FLOW PATH > 



OUTER / 
RACE^, 



INNER 
RACE 



SHAFT 



BEARING 
HOUSING 



/ ^TEST 
/ BEARING 




FLOW PATH> 



Fig. 2 Lubricant system for test bearings. 



RADIAL 
LUBRICATION 
HOLE IN RACE 




RADIAL LUBRICATION I 
HOLE IN RACE SHOULDER -1 



Fig. 3 Unfalledl20-mm bore angular-contact high-speed test ball bearing. Running time, 
1000 hours; speed, 25 000 rpm (3 million DM); temperature, 492 K (425° F); thrust load 
22 241 N (5000 lb). 



KINEMATIC 

VISCOSITY, 

cS 



50 000 
10 000 
2 




300 350 400 450 500 
TEMPERATURE, K 



100 200 300 400 500 

TEMPERATURE, °F 

Fig. 4 Viscosity as function of 
temperature for tetra-ester 
(Type ID lubricant. 



vO 

in 
I 



440 


r- 500 


420 


490 




480 


400 


- 




470 


380 


- 




460 


360 


- 




450 


340 


- 




440 



THRUST LOAD, 
N(LB) 





440 


p 500 


*- 


420 


- ^490 


LU 

^3 




Z3 


< 

s 


400 


!<480 
_ cc 

•s. 




380 


5470 

- < 


z 

a 

CD 


360 


1 

BEARING- 




340 


L 

440 



440 



420 



400 



380 



360 



340 



1- 500,- 



490 



480 



470 - 



460 



450 



440 




-i_ 



(a) CONTACT ANGL£, 20°. 

THRUST LOAD, 
N(LB) 




(b) CONTACT ANGl£, 24° 



CONTACT 
ANGLE, 20° 




10 



_L 



_L 



26x10^ 



14 18 22 

INNER-RACE SPEED, RPM 

(cl SUMMARY OF INNER-RACE TEMPERATURE RANGE. 

Fig. 5 Range of bearing race temperature as a function of 
speed for various thrust loads. Bearing type, 120-mm bore 
angular-contact ball bearing; lubricant flow to inner race, 
L], 1. 2x10'^ m^minlO. 313 gpml; inner-race cooling flow, 
C|, 3.6x10'^ m^/min (0.94 gpm); outer-race cooling flow, 
Cp, 1.9x10"^ m^/min (O.Sgpm); oil inlet temperature, 
428 K (310° Fl. 




(a) CONTACT ANGLE, 20°. 



20 



,.- 16 




(b) CONTACT ANGLE, 24°. 



QL 




26x10^ 



14 18 

INNER-RACE SPEED, 

(cl SUMMARY. 



Fig. 6 Range of power loss as a function of speed for 
varying thrust loads. Bearing type, 120-mm bore 
angular-contact ball bearing; lubricant flow to inner 
race, Lj, 1.2x10-3 m3/min (0.313 gpm); inner-race 
cooling flow, C|, 3.6x10'^ m^/min (0.94gpm); outer- 
race cooling flow, C., 1.9xl0"'/min (0.5gpm); oil inlet 
temperature, 428 K (310° F). 



440 



420 



400- 



380 



M- 360 



i 340 



? 440 

o 

z 

Of 

< 

« 420 



400 



380 



360 



340 



INNER RACE 
OUTER RACE 



SPEED, 
RPM 

25 000 



20 000 




16 000 



^^^^ 12 000 



(a) CONTACT ANGLE. 20°. 



^ 500 

I 

o 

§ 490 

CO 



480 
470 
460 

450 
440 




16 000 



12 000 



J_ 



50 100 150 200 

BEARING THRUST LOAD, N 



250x10^ 



± 



I 



10 20 30 40 

BEARING THRUST LOAD, LB 

(b) COI^ACT ANGL£, 24°. 



50x10^ 



Fig. 7 Range of bearing race temperature as a function of 
bearing thrust load for various speeds. Bearing type, 120-mm 
bore angular -contact ball bearing; lubricant flow to inner 
race, Lj, 1. 2x10"^ m^/min (0.313 gpm); inner-race cooling 

^ nrlmln (0.94gpm); outer-race cooling flow, 
'/min (0.5 gpm); oil inlet temperature, 428 K 



flow, Cj. 3.6xiq- 
Cn, 1.9xlO'^m^; 



(3°10° F), 



440 



r 500f- 



un 
I 



420 



400 



380 



o 360- 






< 

Qi. 

\ 

O 



< 



340 



SPEED, 
RPM 




^ (c) COMPARISON OF INNER-RACE TEMPERATURES. 

UJ 



440n^ 500 



L§ 490 



420 



400- 



380 



360- 



340 



480- 



470 



460 



450- 



440 



20° CONTACT ANGLE 

24° CONTACT ANGLE 




12 000 



J_ 



_L 



J 



50 100 150 200 

BEARING THRUST LOAD, N 



250x10^ 



± 



10 20 30 40 50x10-^ 

BEARING THRUST LOAD, LB 

(d) COMPARISON OF OUTER-RACE TEMPERATURES. 
Fig. 7 Concluded. 



a. 

:r 

<S 

z 

< 

OQ 

Q£ 
LU 
O. 

1/5 

Q 



o 
a. 



< 
I— 
o 



20 



16 



12- 



8 - 



4- 



SPEED, 
RPM 




24° CONTACT ANGLE 
20° CONTACT ANGLf 

_J I 



50 



L_ 
10 



100 150 200 

BEARING THRUST LOAD, N 

J I I I , 

20 30 40 50x10^ 

BEARING THRUST LOAD, LB 



250x10^ 



Fig. 8 Bearing power loss as a function of bearing 

thrust load for various speeds. Bearing type, 120-mni 

bore angular -contact ball bearing; lubricant flow to 

inner race, L, 1. 2x10^ m^/min (0.313 gpm); inner- 

■3„3, 



m^/min (0.94gpm); 
3„3 



race cooling flow, Cj, 3.6x10 

outer-race cooling flow, Cq, 1.9x10'^ m^/min (0.5 gpm); 

oil inlet temperature, 428 K (310° F). 



CO 

in 

I 

W 



440 



420 



400- 



380 



360 



340 



440,- 



420- 



500 r- 



490 



INNER- 
- RACE 
TEMPERA- 
TURE RANGE 



470 



4«0 



450 



440 



(Saumini 




\ ^ 3. 8x10'^ (1.0) 
^5. 7x10'' (1.51 



J_ 



_L 



_L. 



_L 



_L 



(a) CONTACT ANGIi, 20^; INNER-RACE COOLING 
FLOW, C|, 0. 

500r 



.400 



< 380 



5 360 



340 



440r 



420 



400 



380 



360 



340 



490 



, 470 



£ 460 



450 



440 



OUTER-RACE TEMPERATURE 



9. 5x10"^ (0.25) 

1.9)(10"3(0.5) 

INNER-RACE 
TEMPERATURE RANGE 




3. 8x10"' (1.0) 



■5. 7x10"' (1.5) 



J 



(b) CONTACT ANGLE. 20P; INNER-RACE COOLING 
FLOW, C|, 1.33Lj. 

500 r 



490 



480 



470 



460 



450 



440 



INNER-RACE 

|- TEMPERA- 
TURE RANGE 




3. 8x10"' (1.0)-^ 
-5.7x10"' (1.5) 



J_ 



12 3 4 5 6x10"' 

LUBRICANT FLOW TO INNER RACE, L|, m'/MIN 

I I J I I 

.4 .8 1.2 1.6 

LUBRICANT FLOW TO INNER RACE, L|, GAL/MIN 

(c) CONTACT ANGLE, 24°; INNER-RACE COOLING 
FLOW, C|, 0. 

Fig. 9 Bearing race temperature as a function of lubricant 
flow into bearing, L|, for varying inner (C|) and outer 
(Cq) race cooling rates. Bearing type, 120-nnm bore 
angular-contact ball bearing; bearing thrust load, 22 241 N 
(5000 lb); speed, 25 000 rpm (3x10^ DNl; oil inlet tempera- 
ture, 394 K (250° F). 



44U 


500 


420 


. 490 


400 


480 




470 


380 


- 




460 


360 


- 




450 


340 


- 




440 



OUTER-RACE 
COOLING FLOW, 

m'/MIN (SaUM IN) 




INNER-RACE TEM- 
PERATURE RANGE 



1.9x10"' (0.5) 



3. 8x10"' (1.0) 



5. 7x10"' (1.5) 



(d) CONTACT ANGLE, 24°; INNER-RACE COOLING 
FLOW, C|, 1.33L|. 



440 



420 



400 







< 

DC 


S 




s: 


£ 




Sj 








S 

«£ 


380 


"S 



E 360 



340 



500 



490 



470- 



460 



450 



440 



OUTER-RACE TEMPERATURE 




1.9x10"^ (0.5) 
^INNER-RACE TEM- 
PERATURE RANGE 
-3. 8x10"' (1.0) 

•5.7x10"' (1.5) 



J_ 



J 



440 



420 



400- 



380 



360 



340 



(e) CONTACT ANGLE, 24°; INNER-RACE COOLING 
FLOW, C|, 3 L|. 

500,- 



490- 



480 



470 



460 



450 



440 



0; CONTACT 
LE, 20° 




C| ■ 1. 33 L|; CONTACT 
ANGLE, 20° 
■ C| ■ 1. 33 L|; CONTACT 
ANGLE, 24° 
'- C| ' 1. 33 L|; CONTACT ANGLE, 24° 

I I I I \ I 



1 



5 



6x10" 



LUBRICANT FLOW TO INNER RACE, Lj, m^/MIN 

I I , I I I 

.4 .8 1.2 1.6 

LUBRICANT ROW TO INNER RACE. L|, GAUMIN 

(f) SUMMARY OF INNER-RACE TEMPERATURES. 

Fig. 9 Concluded. 



22 

20 

18 

16- 

14 

12 

10 



c- 24- 



OUTER-RACE COOLING FLOW, 
m'/MINI^AUMIN) 

5. 7x10"' (1.5) ->^ 
3.8xl0"'(1.0)-\\_ 
1. 9x10"' (0.5) - 




9. 5x10'^ (0.25) 



_L 



(a) CONTACT ANGLE, 20°; INNER-RACE COOLING 




(bl CONTACT ANGLE, 20<'-, INNER-RACE COOLING 
FLOW, C|, 1.33L|. 

22r 




6x10"- 
LUBRICANT FLOW TO INNER RACE, Lj, m%IN 

I I I I I 

.4 .8 1.2 1.6 

LUBRICANT FLOW TO INNER RACE, L|, GAUMIN 

(c) CONTACT ANGLE, 24°; INNER-RACE COOLING 
FLOW, Cj, 0. 

Fig. 10 Bearing power loss as a function of lubricant flow 
into bearing, Lj, for varying inner (Cj) and outer (C^) 
race cooling rates. Bearing type, 120-mm bore angular- 
contact ball bearing; bearing thrust load, 22 241 N 
(5000 lb); speed, 25 000 rpm (3x10^ ON); oil inlet temper- 
ature, 394 K (250° F). 



20 



g 12 



12- 



'U 




1 


20 


- 


5.7x10"' //, 


18 


- 


(1.5)^ /# OUIER-RACE 
y# COOLING FLOW, 

llll m'/MIN(GAL/MIN) 


16 


- 


i 


14 


- 


^^^3. 8x10"' (1.0) 
^-1.9x10"' (0.5) 


12 

10 




'^9. 5x10"" (0.25) 


8 


- 




6 




1 1 1 1 1 1 



(d) CONTACT ANGLE, 24°; INNER-RACE COOLING 
FLOW, Cj, 1.33 Lj. 
22 r 



20 



18 



16 



I/, 14 

s 

g 12 

o 
a. 

g 10 




^5. 7x10"' (1.5) 
^1.9x10"' (0.5) 



3. 8x10"' (1.0) 



(e) CONTACT ANGLE, 24°; INNER-RACE COOLING 
FLOW, Cj, 3 L|. 

^Ci-1.33Lj; 
CONTACT ANGLE, 24° 




6x10"- 



LUBRICANT FLOW TO INItR RACE, L,, m'/MIN 

I 1 \ I I 

.4 .8 1.2 1.6 

LUBRICANT FLOW TO INNER RACE, L|, GAUMIN 

(f) SUMMARY OF BEARING POWER LOSSES. 

Fig. 10 Concluded. 



NASA-Lewis