Skip to main content

Full text of "DTIC ADA392707: Anti-Icing Endurance Time Tests of Two Certified SAE Type 1 Aurcraft Deicing Fluids"

See other formats


DOT/FAA/AR-OI/13 

Office of Aviation Research 
Washington, D.C. 20591 


Anti-Icing Endurance Time Tests 
of Two Certified SAE Type I 
Aircraft Deicing Fluids 


April 2001 


distribution statement a 

Approved for Public Release 
Distribution Unlimited 


Final Report 


This document is available to the U.S. public 
through the National Technical Information 
Service (NTIS), Springfield, Virginia 22161. 


® 20010802 045 

U.S. Department of Transportation 

Federal Aviation Administration 



NOTICE 


This document is disseminated under the sponsorship of the US 
Department of Transportation in the interest of information exchange The 
United States Government assumes no liability for the contents or use 
thereof. The United States Government does not endorse products or 
manufacturers. Trade or manufacturer's names appear herein solely 
because they are considered essential to the objective of this report. This 
document does not /constitute FAA certification policy. Consult your local 

FAA aircraft certification office as to its use. 


This report is available at the Federal Aviation Administration William J 
H ughes Technical Center's Full-Text Technical Reports page! 
actlibrary.tc.faa.gov in Adobe Acrobat portable document format (PDF). 



2. Government Accession No. 


DOT/FAA/AR-01/13 


4. Title and Subtitle 


ANTI-ICING ENDURANCE TIME TESTS OF TWO CERTIFIED SAE TYPE I 
AIRCRAFT DEICING FLUIDS 


Technical Report Documentation Paqe 


3. Recipient's Catalog No. 


5. Report Date 


April 2001 


6. Performing Organization Code 


7. Author(s) 

Kathy Bouchard, Jean-Louis Laforte, and Arlene Beisswenger 


9. Performing Organization Name and Address 

Anti-icing Materials International Laboratory 
Universite du Quebec a Chicoutimi 
555, boulevard de V Universite 
Chicoutimi, Quebec G7H 2B1 


12. Sponsoring Agency Name and Address 

U.S. Department of Transportation 
Federal Aviation Administration 
Office of Aviation Research 

Washington, DC 20591__ 


15.' Supplementary Notes 

The FAA William J. Hughes Technical Center COTR was Paul Boris. 


8. Performing Organization Report No. 


10. Work Unit No. (TRAIS) 


11. Contract or Grant No. 


13. Type of Report and Period Covered 

Final Report 


14. Sponsoring Agency Code 

AFS-200 


This report presents the results of Anti-Icing Endurance Time (AET) tests performed with unsheared samples of two certified 
SAE Type I aircraft deicing fluids from September 5 to October 15, 1999, at the Anti-Icing Materials International Laboratory 
(AMIL). Over 100 tests, including 25 calibration and 50 fluid tests, were conducted at various temperatures and icing intensities, 
under the six environmental conditions addressed in the holdover time (HOT) guidelines published by the SAE as part of the 
ARP 4737: frost (3), freezing fog (6), snow (6), freezing drizzle (4), light freezing rain (4) and rain on a cold-soaked wing (2). 

The results obtained demonstrate the feasibility of performing the six AET testing procedures within the prescribed accuracy and 
repeatability. Indeed, environmental parameters in AET calibration and fluid tests were kept within the target values with 
variations within the allowable drifts. Moreover, AET results showed an expected inverse relationship between endurance times 
and precipitation rate; the shortest and longest failure times being obtained respectively under the highest and lowest icing rates. 

The AET test results were also compared and discussed with HOT data obtained in a parallel test set performed in July 1999 by 
APS Aviation (APS) at the Canadian National Research Council (NRC) facility. These tests include freezing fog, snow, freezing 
drizzle, and light freezing rain tests at -10°C, and rain on a cold-soaked wing at +1°C. However, their testing methods are 
somewhat different. AMIL failure times are systematically found to be 1 to 2 minutes shorter than APS’s measured values for 
mean variation up to 30%, depending on test conditions. These lower failure times can be partially attributed to differences in 
procedures used during the test performance. This includes the effect of delay between fluid application and start of precipitation 
and the effect of volume of fluid applied on the anti-icing endurance time were checked. 

The results obtained will be useful to discuss the differences between test methods and, ultimately, obtain a single set of 
standardized procedures which reflect reality, allowing any testing facility to perform AET tests, and thus obtain the failure time 
intervals to be used to fill cells of the Type I holdover time table of the ARP 4737 guidelines. 


17. Keywords 18. Distribution Statement 

Frost, Freezing fog, Freezing drizzle, Freezing rain, Snow, This document is available to the public through the National 
Anti-icing endurance time, Precipitation simulation, Holdover Technical Information Service (NTIS) Springfield, Virginia 
time, HOT table 22161. 


19. Security Classif. (of this report) | 20. Security Class! (of this page) 21. No. of Pages 


Unclassified _ 

Form DOT FI 700.7 < 8 - 72 ) 


Unclassified _ 

Reproduction of completed page authorized 



















TABLE OF CONTENTS 


Page 


EXECUTIVE SUMMARY ix 

1. INTRODUCTION 1 

1.1 Purpose ' ■ 1 

1.2 Background 1 

2. SET OF TESTS AND CONDITIONS 2 

2.1 Determination of the Set of Tests 2 

2.2 Anti-Icing Endurance Time Test Condition Selection 5 

3. EQUIPMENT, PRECIPITATION SIMULATION, AND CALIBRATION 5 

3.1 Climatic Chambers 5 

3.2 Frost Generation 5 

3.3 Supercooled Precipitation Simulation 6 

3.3.1 Freezing Fog Water Spray 7 

3.3.2 Freezing Drizzle, Light Freezing Rain, and Rain on a Cold-Soaked 

Wing Water Spray 8 

3.4 Snow Making, Storage, and Distribution System 9 

3.4.1 Snow Making 9 

3.4.2 Snow Storage 10 

3.4.3 Snow Distribution System 10 

3.5 Plate Setups 14 

3.5.1 Frost Tests 14 

3.5.2 Freezing Fog, Freezing Drizzle, and Light Freezing Rain 15 

3.5.3 Rain on a Cold-Soaked Wing Test 18 

3.6 Measured Parameters 18 

3.7 Calibration 18 

3.7.1 Frost Calibration 19 


3.7.2 Freezing Fog, Freezing Drizzle, Light Freezing Rain, and Rain on a 
Cold-Soaked Wing Calibration 


iii 


20 



3.7.3 Snow Calibration 


21 


4. RESULTS 21 

4 .1 Water Spray and High Humidity Endurance Tests (WSET and HHET) 2 1 

4.2 Anti-icing Endurance Time Test Results 24 

4.2 .1 Methodology 24 

' i , 

4.2. 1.1 Sample Dilution Preparation 24 

4.2. 1.2 Failure Criterion and Type 25 

4.2. 1 .3 Measurements and Failure Recordings 26 

4.2.2 Frost Tests 26 

4.2.3 Freezing Fog Tests 29 

4.2.4 Snow 32 

4.2.5 Freezing Drizzle 35 

4.2.6 Light Freezing Rain 38 

4.2.7 Rain on a Cold-Soaked Wing 41 

5. COMPARISON BETWEEN AMIL AND APS RESULTS 45 

5.1 Scope 45 

5.2 Comparison of AMIL and APS Data 45 

5.3 Comparison of AMIL and APS Testing Procedure 47 

5.3.1 Sample Dilution 49 

5.3.2 Amount of Fluid Applied and 5-Minute Delay 50 

5.3.2 .1 Five-Minute Delay Effect 53 

5.3.2.2 One Thousand mL vs Five Hundred mL Applied Fluid Effect 53 

5.3.3 Comparison of Endurance Time Under Similar Conditions 54 

6 . CONCLUSIONS 54 

7. REFERENCES 55 

8 . ADDITIONAL INFORMATION 56 


IV 



LIST OF FIGURES 

Figure Page 

1 Humidity Generator (Bird’s Eye View of Bath) 6 

2 Freezing Fog Water Droplet Size Distribution 7 

3 Freezing Drizzle Water Droplet Size Distribution 8 

4 Light Freezing Rain Water Droplet Size Distribution 9 

5 Snow Box Mounted in its Support Above Test Plate 11 

6 Snow Box on Track of Support 11 

7 Cross Section of Snow Box 12 

8 Side Views and Cross Section of Acetal Cylinder With Six Rows of Three Cavities 

Which Transfer Snow From the Box to the Test Plate 13 

9 Frost Setup 14 

10 Frost Test Plate Arrangement 14 

11 Test Plate Supports Used for Freezing Drizzle and Light Freezing Rain Tests 15 

12 Plate Arrangement in Freezing Drizzle and Light Freezing Rain Tests 16 

13 Plate Arrangement in Freezing Fog Test 17 

14 Cold Soak Box Plate Setup 18 

15 Ice Catch Pan Arrangement in Freezing Drizzle and Light Freezing Rain 

Calibration Tests 19 

16 Anti-Icing Endurance Time as Measured in WSET and HHET Tests 23 

17 Frost Anti-Icing Endurance Times 28 

18 Ice Front at-10°C, Typical of a Frost Failure 29 

19 Ice Front at-10°C, Typical of a Failure in a Freezing Fog Test 31 

20 Freezing Fog Endurance Times 32 

21 Snow Anti-Icing Endurance Times 34 

22 Slush at -10°C, Typical Snow Test Failure 35 


v 




23 Freezing Drizzle Endurance Times 37 

24 Ice Front at - 10°C, Typical of a Failure in a Freezing Drizzle Test 38 

25 Light Freezing Rain Endurance Times 40 

26 Speckled Ice Front at -10°C, Typical of a Failure in a Light Freezing Rain Test 41 

27 Rain on a Cold-Soaked Wing Endurance Times 43 

( , 

28 Failure in Rain on a Cold-Soaked Wing at+1°C 44 

29 Air and Plate Temperature Recordings (a) Test With 5-Minute Delay and (b) Test 

Without Delay 5 ] 

30 Air and Plate Temperature Recordings (a) Test With 500 mL of Fluid Applied and 

(b) Test With 1000 mL of Fluid Applied 52 


LIST OF TABLES 

Table « Page 

1 Society of Automotive Engineers Type I Fluid Holdover Table 3 

2 Icing.Intensity Corresponding to Time Values Given in the SAE HOT Type I Table 4 

-k 

3 Selected AET Test Conditions With Their Allowable Fluctuations 5 

4 Allowable Variation in Temperature and Icing Intensity for a Calibration Test 20 

5 Water Spray and High Humidity Endurance Tests Results for OCTAFLO and ADF 

Concentrate Deicing Fluids 22 

6 Fluid Dilution Selection and Identification 25 

7 Frost Tests Results 27 

8 Freezing Fog Results 30 

9 Snow Test Results 33 

10 Freezing Drizzle Test Results 36 

11 Light Freezing Rain Test Results 39 


vi 




12 Rain on a Cold Soak Box Test Results 42 

13 Comparison of AMIL and APS Results 46 

14 Anti-Icing Materials International Laboratory vs APS/NRC Test Conditions 48 

15 Comparison of Differences Between OCTAFLO and ADF Values 49 

16 Five-Minute Delay and Volume of Applied Fluid Effects 53 

17 Light Freezing Rain vs Freezing Drizzle at 13 g/dm 2 h 54 


vii 




LIST OF SYMBOLS AND ACRONYMS 


(am 

micrometer (micron) 

AET 

Anti-Icing Endurance Time 

AMIL 

Anti-Icing Materials International Laboratory 

APS 

APS Aviation 

CSW 

Rain on a Cold-Soaked Wing 

FAA 

Federal Aviation Administration 

FIE 

First Ice Event 

FOG 

Freezing Fog 

FP 

Freezing Point 

FRST 

Frost 

HHET 

High Humidity Endurance Test 

HOT 

Holdover Time 

kPa 

Kilo Pascal 

LZR 

Light Freezing Rain 

MIT 

Mean Icing Time 

MVD 

Median Volume Diameter 

NRC 

National Research Council 

OAT 

Outside Air Temperature 

PIL 

Plate Icing Length 

Rh 

Relative Humidity 

SAE 

Society of Automotive Engineers 

SNW 

Snow 

Tair 

Air Temperature 

Tplate 

Plate Temperature 

WSET 

Water Spray Endurance Test 

ZL 

Freezing Drizzle 


viii 





EXECUTIVE SUMMARY 


This report presents the results of 50 anti-icing endurance time (AET) tests performed at the 
Anti-Icing Materials International Laboratory (AMIL) with unsheared samples of two certified 
Society of Automotive Engineers (SAE) Type I aircraft deicing fluids: OCTAFLO of Octagon 
Process Inc. (propylene glycol-based) and ADF Concentrate of Union Carbide (ethylene glycol- 
based). The two candidate fluids were subjected to six different types of icing precipitation 
under various conditions of temperature and icing rate: frost (3 conditions), freezing fog and 
snow (6 conditions each), freezing drizzle and light freezing rain (4 conditions each), rain on a 
cold-soaked wing (2 conditions). Each type of precipitation with its specific condition are 
addressed in the holdover time (HOT) guidelines published by SAE as part of the ARP 4737 
document to help the pilot and transport management assess the protection times of SAE Type I 
deicing fluids. The SAE G-12 Holdover Time Subcommittee is charged with establishing and 
updating these guidelines. 

The first objective of this laboratory work was the determination of the anti-icing endurance 
times of two certified SAE Type I aircraft deicing fluids, under frost, freezing fog, snow, 
freezing drizzle, light freezing rain, and rain on a cold-soaked wing conditions at various 
temperatures and icing intensities. The second objective was the establishment of a 
comprehensive basis to analyze, compare, and discuss HOT data obtained by APS Aviation 
(APS) at the Canadian National Research Council (NRC) facility and, ultimately, the finalization 
of AET standardized procedures. 

The laboratory tests were conducted from September 5 to October 15, 1999, under a Federal 
Aviation Administration (FAA) award following a recommendation made by the SAE G-12 
Fluid Subcommittee meeting in Toronto on May 19, 1999, in order to investigate the difference 
between HOT testing methods and facilities. The AET testing procedures are based on the 
laboratory testing protocol established at the Montreal Fluid Subcommittee meeting held in 
March 1999 and revised accordingly in two subsequent meetings between APS/NRC and AMIL, 
the first held at Montreal on July 30, 1999, and the second at Chicoutimi on October 6, 1999. 

The air temperature conditions in the AET tests are -3° and -10°C for freezing drizzle and light 
freezing rain; -3°, -10°, and -25°C for freezing fog and snow; 0°, -10°, and -25°C for frost; and 
+1°C for rain on a cold-soaked wing. For frost, fluids are tested at only one icing intensity at 
each temperature, i.e., -3° -10°, and -25°C. For all other types of precipitation, fluids are tested 
at two precipitation rates corresponding to light and moderate icing intensities. All tests were 
conducted in climatic chambers with specialized equipment and a valid calibration test. To 
determine the distribution of icing intensity, an individual calibration was conducted before 
testing a fluid under each AET condition. 

Frost requires a humidity generating system consisted of a water bath and a “frosticator” which 
cools the test panels to a temperature 3°C below that of the air during a test. Freezing fog, 
freezing drizzle, and light freezing rain tests are conducted similarly, with the height and the size 
of the supercooled water droplets different for each type of precipitation. The snow tests were 
conducted in a two step process: the first of which consists of making artificial snow in the form 
of agglomerates of tiny frozen droplets of about 20-40 pm in diameter. The second step consists 
of distributing the snow in an even manner over the test plate by means of an automated system. 


IX 



For the rain on a cold-soaked wing simulation, a cold soak box is used to cool the test plate 
below the air temperature. 

For all tests but frost, the fluid test panels were 500 mm long, 300 mm wide and 3.2 mm thick. 
For frost, the panels are 300 mm long, 100 mm wide and 1.6 mm thick. They were at the air 
temperature at the beginning of the test but are free to vary during the test. For frost, test plates 
are maintained at a prescribed temperature throughout the test by means of a special cooling 
system. For rain on a cold-soaked wing, the test plate is at -10°C at the beginning of the test and 
is free to vary during the test. In all AET tests with the exception of frost, failure is called when 
30% of the plate is covered with frozen contamination. In frost tests, failure is called when there 
is a 50% ice-covering of the plate because of the smaller plates. The following measurements 
are performed for each AET tests: icing intensity, anti-icing endurance times, photographs of the 
ice front at failure as well as continuous recordings of air and plate temperature and humidity. 

Sheared and unsheared samples of the two candidate fluids, were first subjected to two standard 
laboratory tests: the Water Spray Endurance Test (WSET) and the High Humidity Endurance 
Test (HHET), to ensure that anti-icing endurance times exceed the minimum prescribed values of 
3 and 20 minutes respectively, confirming they are SAE Type I approved fluids. For both fluids, 
the First Ice Event (FIE) or AET exceeded the minimum values specified for WSET and HHET 
for an SAE Type I fluid. 

In the AET tests of OCTAFLO and ADF Concentrate, when results of the two tested samples are 
compared, time variations of 1 minute or less are generally observed between the endurance 
times measured with the same sample. These variations do not appear dependent on the fluid nor 
the testing temperature. The 1-minute variation is considered to be within the experimental error 
of measurement. 

The AET results were also compared and discussed with HOT data obtained by APS in a parallel 
test set conducted in July 1999 at the NRC facility using a somewhat different test method. The 
compared tests include freezing fog, freezing drizzle, light freezing rain, and snow tests 
performed at -10°C, and for rain on a cold-soaked wing at +1°C. AMIL failure times were 
generally 1 to 2 minutes shorter than APS’s measured values, resulting in an average difference 
of 30% (depending on tests). As the failure times obtained with Type I fluids are shorter, when 
compared to those of Type II and IV fluids, the relatively short times have the effect to overvalue 
these differences when expressed as percentages. Considering a 1-minute time variation could 
be within the expected acceptable experimental error of measurement. 

An examination of APS and AMIL testing procedures allows for the identification of thirteen 
differences, among which the following six can be judged more significant: the plate working 
area, the sample dilution, the failure call, the precipitation rate measurement method, the amount 
of fluid applied and finally, the 5-minute delay prior to the start of precipitation. The last two 
factors may partially explain the 1 to 2 minutes lower failure times observed. This interpretation 
is supported by the results obtained by two tests in which effects of these two factors were 
compared. 


x 



These results will be useful in evaluating differences between APS/NRC and AMIL procedures 
in order to ultimately finalize a single set of procedures to be approved and published. As a 
consequence, any testing facility, with the appropriate capability, could perform AET tests 
according to an approved procedure and thus obtain the HOT values which can be used by the 
SAE committees responsible of substantiating the HOT tables of the current ARP 4737 
guidelines. On the basis of the test results obtained with the two Type I fluids, individual cells of 
the Type I HOT table substantiated using the APS and AMIL data, would show lower time 
intervals by comparison to numbers of the generic Type I HOT table actually in use. 

In the process of reducing the number of parameters which are not the same in the APS/NRC and 
AMIL procedures, it is recommended that each parameter for which a difference is identified in 
this report be analyzed and discussed. In order to realize this, real conditions and actual practices 
of using fluids in airports during deicing and anti-icing operations should be taken into 
consideration, as well as the feasibility of performing reproducible tests in a laboratory. 


xi/xii 




1. INTRODUCTION. 


1.1 PURPOSE . 

The first objective of this work is the determination of anti-icing endurance times (AET) of two 
certified Society of Automotive Engineers (SAE) Type I aircraft deicing fluids under the 
following six environmental conditions: frost, freezing fog, snow, freezing drizzle, light freezing 
rain, and rain on a cold-soaked wing. Each of these conditions are addressed in the holdover 
time (HOT) tables published by SAE as part of the ARP 4737 standard as guidelines to help 
pilots and transport management assess the protection time of certified deicing and anti-icing 
fluids. Results obtained by these AET tests with two Type I deicing fluids are to be compared 
with HOT values of the current Type I table published by SAE and cells not in agreement with 
the table will be identified. The Holdover Time SAE G-12 Subcommittee is in charge of 
establishing and updating these guidelines. 

The second objective is the establishment of a comprehensive basis to compare and discuss data 
obtained by APS Aviation (APS) in the National Research Council (NRC) laboratory and 
ultimately to finalize standardized Anti-icing Endurance Time test procedures. The discussion 
and the finalization of AET procedures are to be done among representatives of Anti-Icing 
Materials International Laboratory (AMIL) and APS with the presence of the Federal Aviation 
Administration (FAA), Transportation Development Center (TDC) and the SAE G-12 Holdover 
Time subcommittee. 

The laboratory tests were conducted from September 5 to October 15, 1999, under a Federal 
Aviation Administration (FAA) award following a recommendation made by the SAE G-12 
Fluid Subcommittee meeting in Toronto on May 19, 1999, [1] in order to investigate on the 
difference between HOT testing methods and facilities. The AET testing procedures are based 
on the laboratory testing protocol established at the Montreal Fluid Subcommittee meeting held 
in March 1999 [2] and revised accordingly in two subsequent meetings between APS/NRC and 
AMIL, the first held at Montreal on July 30, 1999, [3] and the second at Chicoutimi on October 
6, 1999 [4]. The procedures are detailed in reference 5. 

Appendices referenced in this report contain details pertaining to various aspects of the test 
program. Due to the combined length of all the appendices, they are not appended to this report. 
The details noted in these appendices are referenced in the body of the text and are not necessary 
for the overall comprehension of the tests and results described. 

1.2 BACKGROUND . 

Deicing and anti-icing fluids are commonly used during the winter to remove and prevent 
aircraft contamination by any frozen deposit while on the ground. Anti-icing fluids are able to 
protect the aircraft for a time period that depends on environmental conditions including the 
nature of precipitation, the outside air temperature (OAT), and the precipitation intensity. 

The FAA’s William J. Hughes Technical Center continues to support research and related efforts 
directed toward the improvement of aircraft deicing methods and practices. One such effort is 
the standardization of HOT test procedures for deicing fluids. In the past, HOT testing has 


1 


largely been performed by APS of Montreal, Canada. In general the international aviation 
community has accepted the APS results. AMIL an accredited anti-icing laboratory, was tasked 
by the SAE G-12 Holdover Time Subcommittee meeting in Zurich on May 21, 1996, [6] to 
prepare HOT test procedures that could be performed in a laboratory environment. In the review 
of these procedures, discrepancies were noted between HOT values obtained by the two 
facilities, i.e., APS/NRC and AMIL. This points out the necessity to standardize the testing 
method, procedures, environmental test conditions, and interpretation of fluid failure in order to 
eliminate disagreements among testing facilities. Such standardization would allow any testing 
facility, with the appropriate capability, to perform HOT testing by adhering to an approved 
published procedure. 

HOT tables for SAE Type I fluids have virtually remained unchanged since they were initially 
published in 1992. Some of the fluids that were used to establish this table are no longer in 
production; and new fluids that have been introduced since are assumed to meet the holdover 
time guidelines of this table. As a first step in the standardization of holdover time testing, 
following the May 1999 recommendation of the SAE G-12 Fluid Subcommittee [7], the FAA 
proposed to subject two currently approved Type I fluids, one ethylene glycol-(EG) and one 
propylene glycol-(PG) based, to the anti-icing endurance time methods and procedures at AMIL. 
Prior to testing, the procedures were discussed and coordinated between AMIL and APS with the 
concurrence of the FAA and TDC in two meetings; the first held at Montreal on July 30, 1999, 
[3] and the second at Chicoutimi on October 6, 1999 [4], This is a draft version to be discussed 
further in subsequent SAE Fluids, and Holdover Time Subcommittee meetings. 

The intent of this work is, ultimately, to determine the variation between the two methods and 
facilities and to compare, reconcile, and ideally, establish a single set of procedures to be 
published after concurrence by the SAE G-12 Fluids and Holdover Time subcommittees. One of 
the expected outcomes of such testing will be the substantiation of the current Type I HOT table. 
Once substantiated, new fluids will have to be tested according to the AET set of tests and the 
generic table adjusted accordingly prior to their use. 

2. SET OF TESTS AND CONDITIONS . 

2.1 DETERMINATION OF THE SET OF TESTS . 

The set of tests to be conducted was determined in accordance with the first objective, that is, the 
determination of the anti-icing endurance times of two certified SAE Type I aircraft deicing 
fluids under the six different types of iced precipitation shown in the SAE Type I holdover time 
(HOT) table: frost, freezing fog, snow, freezing drizzle, light freezing rain, and rain on a cold- 
soaked wing. The most recent version of the SAE Type I fluid HOT table published in August 
1999 is presented in table 1. This is a generic table which applies to all certified SAE Type I 
fluids. 

Each column of table 1 corresponds to one type of icing precipitation, and is divided into one to 
three individual cells. With the exception of the frost, each cell is dependent on the OAT and 
comprises two numbers. These numbers correspond to time values expressed in minutes which 
delimit the interval of the protection times which can be expected for a SAE Type I fluid at that 
OAT. The larger number corresponds to HOT values expected under light icing conditions at the 


2 


TABLE 1. SOCIETY OF AUTOMOTIVE ENGINEERS TYPE I FLUID HOLDOVER TABLE 


< 

o 

“O 

c 

CO 

CO 

c 

.2 

B 

c 

o 

O 

i— 

CD 


CO 

© 


c 

2 

o 

c 

3 

LL 


III 


CD 
< 

H 

M. W 

£ 2 

oi 

§1 


a: 

ui 

co 

3 

Ui 


x 

t 

5 

CO 

z 

< 

s 

LU 

a 

<c 

a 

ui 

(0 

ui 


9 - 2 


_ ® 

-I a- 

UI UJ 
n <t 

> w 

g £ 
w £ 

CO 

g_ 

‘o 


< 

O 

-I 

a. 

CL 

< 

UI 


a: 

o 

u. 

>- 


< t 


CO 

CD 

e 


2 

o 

22 

o 

I 


CD 

C 

"© 

■D 

*3 

o 


CD 

CO 

z 

o 

x 

CO 

UI 

a: 

ui 

x 

h 


cc 

UJ 

I 

H 

o 


co 

c 

o 

♦3 

■5 

c 

o 

a 

k. 

CD 

X' 

4- 

CO 

0) 

«5 

CO 


> 

© i 

I E 

= jo 

CO 3 
* O 

E jz 


a> 

> 

o 

2 

o 

X 

0) 

CO 


X 

o 

k. 

o. 

Q. 

< 


0 
z 

OQ^ 
_ J n 
S O ui 
<U^ 

a: < 
o 
</> 


o 

hS z 

X N i 

-i ui a 
x 

IL 


(JJ CM 

— UJ 

SB 

UJ = 


UI 


< 

o 


< 5 
£| 
CO 


a 

Sg 

uj O 
UJ u. 
X 


CO 

o 

X 

LL 


O 


eS 

■• 5 X 
—IZ L o 

2 s « 

5 &■§ # 

6 < o ® 
. O -e 2 

g 

© 


in 

o 


CM 

O 


in 

o 


in 

T“ 

6 

s 

© 


CO 

d 

I 

CM 

T- 

© 


in 


CM 

co 

a> 

> 

o 

-Q 

CO 


o 

CD 

> 

O 

£X 

CO 


in 

o 


CM 


in 

o 


in 

o 

g 

o 


in 


CM 

co 


in 


CD 

o 


in 


in 


3 

o 

© 


3 

o 

© 

-Q 


< 

o 

1 

© 

43 


© co 
3 -c 

CO T3 
CO C 
O CO 
CL . 
— C 

Q co 


O 

o 

o 

C E r- 


£ 

5 

(0 

k. 

© 

X 

1-1 
w O 

< ™ 
m O) 

2-S 
£ © 
= £ 
O UL 

II II 


§x 
O XL 


CO D) 

® R 

- (D 


M 


o 

x 


= 2 
. c « 

t- n ® 
W ® ■= 
2 -a 
c 

— CO 

“* ° ® 
ui c -S 

>.2 2 

P ® -§ 

O .2 O 

<1 E 

o © co 
c — _9> 
.2 §15 
o -I cl 
< g '© © 

O O o 

I 8| 

£ E O) 


© 


3 

o 

c 

© 


3 C 

.3 © 
i2 £ 
© © 
o x 

© © 
© © 

£ £ 
CD CD 
© © 
D D 
ll II 


a u. 


o © 

>* o 
q. 2 w 

«•§ 8 . 
£ 2 > 
w o) § 
2 .2 c 
.2 N « 

£ © - 
T3 © 

C * 


© 

© 

© 

4-4 

© 

© 

8 

I 

1 

© 

£ 

o 

x 

UL 

© 

-C 


o 

© 

"O 

* 

© 

© 

© 

© 


<o C 


3 
o 

■c © © 

3 © © 

a 3 x 

CO 

UI 

H 

o 

Z t** C\J 00 


X 111 

i! 


^ UJ 

co5 
uj 2 

CC UI 


H (/) 
< UJ 


a. 

a. 


_< > 

cl o O O 
O-JZQ 

Q.f hi 

w.>0CUI 
> > UJ -J 

$p>o 

< j 2 < 

uj uj O q 
x m -i 

► uj £2 

z!^< 

2 ^ 3 UJ 

h X L. T 

Zogz 
2 Q a x 

o^t 

ttxlff ^ 

UJ ^ 

jES|_ 

ai«j: 


UJ 

o 

ui 

-i 

m 

< 

i- 

UJ 


f" z 
£ = 


UJ 


; S 


i-5cn 5 

■ it — 


> rf ^ 

XH< w 

2 <ZUJ 
— —i UJ x 

z sy Q UJ 

X O D H 
Q< 


UI 
Ui I- 
CQ £ 

o a 


>» Q 
H UJ 
X rr 


£ 

© 

5 


© 

a 

>* 

H 

UJ 

< 

co 


UI 

O 

X 

X 


a 


a 

z 

o 

o 

UI 

H 

o 

X 

X 

UI 

o 

UJ 

o 

> 

o 

X 

X 


E o 


o 

Q 

a 

z 

< 

X 

o 

X 

a 

Ui 

o 

z 

LU 


c/> o 

uj o 

CD — 1 

-J > (0 2 
d D -J § 

2 > < O 
o x s u; 
h 2 uj Q 
O X 2 UJ 

ill r— —I 


I- BD 
^tE< 

Seg 


11.5 50 

Oo^ 

2p2uiz 
h (/) O O 

D LiJ r Z UJ 

<( I o ^ X 

O h 5 X K, 


UJ 

X 

< 

0 

z 

a 

UJ 

o 

a 

z 

3 

O 

X 

0 

0 

z 

X 

3 

o 

a 

UJ 

co 

3 

0 

a 


3 


TRANSPORT CANADA. AUGUST 1999 





































cell temperature while the smaller one is the protection time expected under moderate icing 
conditions. Light, moderate, and heavy or severe are relative terms and could be confusing. 
Although these terms appear in various meteorological charts, frequently with associated 
intensity criteria, it, nevertheless, can be a source of dispute. Therefore, for this report, the 
extremes of icing intensity for each precipitation test condition will be referred to as the lower 
and the higher icing intensity. 

The values of anti-icing endurance times obtained in AET tests performed in the present work 
are to be used to. fill individual cells of the HOT table shown in table 1. Intensities regarding the 
shortest and longest times to consider are noted in table 2. They have been tentatively 
established at the November 97 SAE G-12 Subcommittee meeting held in Montreal [2] with the 
latest revision issued after a meeting held in Chicoutimi on October 6, 1999 [4], 

• The shortest time will represent the holdover time obtained with a fluid tested at the 
highest icing intensity for this cell; 

• The longest time will represent the holdover time obtained with a fluid tested at the 
lowest icing intensity for this cell; 

• The high and low icing intensity are to be determined by meteorological standards 
modified to take into account the likelihood of an icing intensity at a given temperature. 

The set of icing intensity intervals was selected on this basis for each cell of the table 1. The 
conditions which are specifically applicable for an SAE Type I fluid are in table 2. For a Type I 
fluid, three temperature intervals are currently given in the HOT table depending on the type of 
precipitation: above 0°C, 0° to -10°C, and below -10°C. In each cell of this table, the high and 
low icing intensity values are shown in bold and italics, respectively. For frost, only one value is 
presented in each cell and the temperature corresponds to the air temperature (except for -3°C) 
while the plate temperature is 3°C below this value. 


TABLE 2. ICING INTENSITY CORRESPONDING TO TIME VALUES 
GIVEN IN THE SAE HOT TYPE I TABLE 


Test 

Temp. 

(°C) 

Icing Intensities Under Various Weather Conditions, g/dm 2 h 

Frost 

Freezing 

Fog 

Snow 

Freezing 

Drizzle 

Light Freezing 
Rain 

Rain on a Cold- 
Soaked Wing* 

-3 (l) 

0.2 

5-2 

25 -10 

13-5 

25-13 

75-5 

-10 (2) 

0.15 


25-10 

13-5 



-25 (3) 

0.06 

■23 

25-10' 





For Frost, "T*, = 0°C T plates = -3°C, a %„ = - 10°C T plates = -13°C, <3) T air = -25°C T plales = -28°C 
*For rain on a cold-soaked wing, T air = +1°C 


4 















2.2 ANTI-ICING ENDURANCE TIME TEST CONDITION SELECTION. 


Table 3 presents the particular environmental conditions of temperature and icing intensities with 
their allowable variations that were retained for the different AET tests selected for the present 
testing program. For frost, the plate temperature is -3°C lower than the air temperature. For the 
five other tests, plate temperature is the same as the air temperature. 


TABLE 3. SELECTED AET TEST CONDITIONS WITH THEIR 
ALLOWABLE FLUCTUATIONS 


Test 

Icing Intensities Under Various Weather Conditions, g/dm 2 h j 

Temp. 

(°C) 

Frost 

Freezing 

Fog 

Snow 

Freezing 

Drizzle 

Light 

Freezing Rain 

WSmm 

-3 ±0.5 (1) 

0.2 ±0.02 

2 ±0.2 

10 ±0.5 

5 ±0.2 

13 ±0.5 

5 ±0.2 

5 ±0.3 

25 ±1.0 

13 ±0.5 

25 ±1.0 

75 ±3.0 

-10 ±0.5 (2) 

0.15 ±0.02 

2 ±0.2 

10 ±0.5 

5 ±0.2 

13 ±0.5 


5 ±0.3 

25 ±1.0 

13 ±0.5 

25 ±1.0 


-25 ±1.0 (3) 

0.06 ±0.01 

2 ±0.2 

5 ±0.2 

• Vr- / 77 — 



5 ±0.3 

10 ±0.5 

i;; V ;/' C/Vv/V 

77; ' ' V -' ■ 

l . 7, V • 


For frost, (,) T air = 0°C T plales = -3°C, (2) T* = -10°C T plat « = -13°C, (3) T* = -25°C T plates = -28°C 


*For rain on a cold-soaked wing, = +1°C 

There are three conditions for frost, six for freezing fog and snow, four for freezing drizzle and 
light freezing rain and, finally, two for rain on a cold-soaked wing. For each condition, the 
testing program consists of a single AET test using two plates, giving a total of 25 calibration 
tests without fluid and 50 tests with the two candidate Type I fluids, i.e., 25 tests with each fluid. 
Should the allowable variation be exceeded during the course of the test, the test was terminated 
and repeated. Only tests performed within the allowable variations are included in the report. 

3. EQUIPMENT. PRECIPITATION SIMULATION. AND CALIBRATION . 

3.1 CLIMATIC CHAMBERS . 

The AET tests were performed in the AMIL climatic chambers at the Universite du Quebec h 
Chicoutimi (UQAC). The choice of the chamber for each type of precipitation was determined 
by its volume and height. The freezing fog and frost tests were conducted in the smaller 
chambers whereas freezing drizzle, light freezing rain, and rain on a cold-soaked wing tests were 
performed in the 9-meter-high climatic chamber. For convenience, the artificial snow was made 
in a small environmental chamber that was different from the one in which the fluid and snow 
calibration tests were performed. 

3.2 FROST GENERATION . 

Frost is generated when a mass of humid air comes in contact with a surface colder than the air. 
The quantity of frost accumulated depends on the level of humidity in the air and the temperature 


5 




































differential between the air and the surface of deposition. To obtain the high level of water 
moisture required in frost tests, a humidity generator, consisting of a 90-cm-long, 60-cm-wide, 
and 30-cm-deep bath of water which is maintained at a temperature warmer than air, is used. 
Forced air circulates throughout the bath to increase surface area and promote evaporation of the 
water. The humidity generator is shown in figure 1. 


mechanical float switch for water level 



Notes: The bath is about 30 cm deep 

Compressed air is injected into the water by means of a copper pipe with small holes on its 
underside, which blows air bubbles into the water 


FIGURE 1. HUMIDITY GENERATOR (BIRD’S EYE VIEW OF BATH) 

3.3 SUPERCOOLED PRECIPITATION SIMULATION . 

For supercooled liquid precipitation, two different types of water spray systems were used: the 
first is a pneumatic spray system for the freezing fog tests, and the second consists of different 
hydraulic nozzles for the freezing drizzle, light freezing rain, and rain on a cold-soaked wing 
tests. All systems use ASTM D1193 Type IV water. 


6 




The system used for the freezing fog tests consists of a pneumatic water spray nozzle oscillating 
over the test area. The nozzle, supplied with water and compressed air at 270 kPa pressure, 
located 1.45 m above the test plate support, allows for the continuous production of a water spray 
of very fine droplets presenting a 23 ±5 pm median volume diameter (MVD) (see histograms 
and droplet size cumulative frequency shown in figure 2). 



5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 

DROPLET DIAMETER (pm) 


(a) Droplet Size Histogram 



DROPLET DIAMETER (pm) 

(b) Droplet Size Cumulative Frequency 


FIGURE 2. FREEZING FOG WATER DROPLET SIZE DISTRIBUTION 








The system used for generating the freezing drizzle, light freezing rain, and rain on a cold-soaked 
wing tests consisted of one or two hydraulic water spray nozzles oscillating over the test area. 
The nozzles, located about 7.0 m above the test plate, allow for the production of droplets 
presenting MVD between 150 and 1400 pm, depending on the orifice diameter of the nozzle 
selected for the test. The water spray intensity for a given nozzle is controlled by varying the 
time sequence of “on/off” pulses. 

For the freezing drizzle tests performed, the measured droplet MVD is 237 ±20 pm (see the 
histogram and droplet size cumulative frequency in figure 3). For the light freezing rain, the 
measured droplet MVD is 970 pm (see the histogram and droplet size cumulative frequency in 
figure 4). 



50 70 90 110 130 150 170 190 210 230 250 270 290 310 

DROPLET DIAMETER (pm) 


(a) Droplet Size Histogram 



0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 
DROPLET DIAMETER (pm) 


(b) Droplet Size Cumulative Frequency 

FIGURE 3. FREEZING DRIZZLE WATER DROPLET SIZE DISTRIBUTION 


8 










400 500 600 700 800 900 1000 1100 1200 1300 1400 1500 


DROPLET DIAMETER (ion) 


(a) Droplet Size Histogram 



DROPLET DIAMETER (pm) 

(b) Droplet Cumulative Size Frequency 


FIGURE 4. LIGHT FREEZING RAIN WATER DROPLET SIZE DISTRIBUTION 
3.4 SNOW MAKING. STORAGE. AND DISTRIBUTION SYSTEM . 

3.4.1 Snow Making . 

The artificial snow was made in a cold chamber by means of a pneumatic water spray nozzle 
supplied with water and compressed air. The nozzle produces a spray of very fine water droplets 
which becomes supercooled in cold air and freezes to form solid ice crystals on contact with a 


9 











collection plate on the chamber floor. Water flow and air pressure are adjusted to ensure the ice 
crystals conform to the requirements of the laboratory-made snow. Typical parameters are: 

• Air temperature: -20° ±5°C 

• Water droplet size: 22 ±3 pm MVD 

• Water flow rate to nozzle: 70 mL/minute 

• Air pressure to nozzle: 260 kPa 

• Artificial snow density 0.1 g/cm 3 

3.4.2 Snow Storage . 

The laboratory-made snow is placed in an insulated heat container, which is stored in a cooler 
kept at a temperature below -10°C. The snow quality is verified prior to each test by means of a 
density measurement. Furthermore, if the artificial snow shows any evidence of sintering, 
agglomeration, or crystallization, it is not used for the snow tests. 

3.4.3 Snow Distribution System . 

For the snow tests, the snow is distributed as ice particles in the form of clusters in the range of 
intensities specified in table 3. The snow distribution system was designed so that the mass of 
each cluster is 0.03 g or less. The snow is placed in a U-shaped aluminum box, 320 mm long, 
253 mm high, and 132 mm wide at the top, with a 65-mm high drawer at the top with a sliding 
base which allows the addition of snow in between tests, above the test plate (figure 5). The box 
is suspended from a track around 760 mm above the center of the test plate (figure 6). The track 
is attached to a motor which provides the lateral movement of the snow box. The lateral 
displacement speed depends on the desired snow intensity. The snow is continually stirred 
inside the box by a rotating system consisting of three blades, aligned at 120° angles from each 
other (figure 7). Each blade measures 50 x 300 mm and consists of a frame housing a wire 
mesh. The continued rotation of the blades prevents clumping of the snow prior to dispensing. 
The box contains an opening at the base, 10 mm wide along the length of the box. This opening 
houses a 32-mm diameter Acetal cylinder, which contains 18 cavities arranged in six rows of 
three cavities each at 60° spacing (figure 8). Each cavity has a diameter of 11 mm and is drilled 
to a U-shape. The cavities on each row are spaced at 87-mm intervals and out of phase with each 
other row. The cylinder, turns after a given time interval thus dispensing snow clusters onto the 
test plate. The rotation speed of the cylinder is predefined to accommodate the desired snow 
intensity. 


10 





FIGURE 5. SNOW BOX MOUNTED IN ITS SUPPORT ABOVE TEST PLATE 



FIGURE 6. SNOW BOX ON TRACK OF SUPPORT 


11 














FIGURE 7. CROSS SECTION OF SNOW BOX 


12 












FIGURE 8. SIDE VIEWS AND CROSS SECTION OF ACETAL CYLINDER WITH 
SIX ROWS OF THREE CAVITIES WHICH TRANSFER SNOW FROM THE 

BOX TO THE TEST PLATE 













3.5 PLATE SETUPS. 


3.5.1 Frost Tests . 

Frost tests are conducted using the frosticator shown in figure 9. It consists of a support, the top 
of which is inclined at an angle of 10°. On the support is placed a set of six plates 
300 long by 100 wide by 1.6 mm thick. The support is maintained throughout the test at -3°C 
below the prescribed air temperature by a cooling system. 



FIGURE 9. FROST SETUP 

For frost tests, the frosticator is covered with three polished AMS 4037 aluminum alloy plates 
coated with the candidate fluid and nine small 100-by 100-by 1.6-mm bare aluminum plates 
placed adjacent to the fluid-coated plates to measure ice catch (figure 10). The mirror polished 
surface of test plates corresponds to a roughness between 0.1 and 0.2 pm. For ice calibration 
tests, eighteen 100- by 100-mm bare aluminum ice catch plates are used to measure the frost 
intensity. 



ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 

ICE 


(b) Ice Catch Calibration Tests 
FIGURE 10. FROST TEST PLATE ARRANGEMENT 


14 
















3.5.2 Freezing Fog. Freezing Drizzle, and Light Freezing Rain . 


For freezing fog, freezing drizzle, and light freezing rain tests, movable stands were used which 
could accommodate a removable 500- by 300- by 3.2-mm-thick AMS 4037 aluminum alloy 
mirror-polished panels at a 10° inclination. Each test plate stand is designed to minimize the 
contact between the test surface and the support (see figure 11). As with the plates used in the 
frost test, the measured average roughness of the panel surface is between 0.1 and 0.2 pm. For 
fluid testing, the two panels are coated with the candidate fluid and they are surrounded with at 
least eight 100-by 100-mm small aluminum pans or plates used to measure icing intensity and 
distribution (see figure 12). 

The ice catch pans are made of 0.8-mm-thick aluminum foil surrounded by a 15-mm-high rim, 
while the ice catch plates are 1.6-mm-thick aluminum without a rim. Pans are used for freezing 
drizzle, light freezing rain, or rain on a cold-soaked wing where the water droplets do not freeze 
immediately on impact and run down slope. However, when the temperature is sufficiently low, 
droplets freeze on impact; for example, freezing drizzle calibration tests performed at -10°C 
showed that there was no significant difference between the ice catch on plates or in pans. For 
this reason, in freezing fog and frost tests, where the ice is formed on contact with the plate, ice 
catch plates are used instead of pans. 

Eight pans are positioned around each test plate for the freezing drizzle and light freezing rain 
tests (figures 11 and 12) and eight plates for the freezing fog tests (figure 13). For the ice catch 
calibration tests, each panel is covered with twelve 100- by 100-mm ice catch plates or pans 
(figures 12b and 13b). For convenience, the term “plate” will be used in this section to refer to 
both plates and pans used for ice catch measurements. 



FIGURE 11. TEST PLATE SUPPORTS USED FOR FREEZING DRIZZLE AND LIGHT 

FREEZING RAIN TESTS 


15 



















ICE 


ICE 






ICE 


ICE 

FLUID 








• 


ICE 


ICE 





* 

ICE 


ICE 


(a) Fluid Test 



FLUID 




ICE 

ICE 

ICE 

1 1 

ICE 

ICE 

ICE 

1 1 

ICE 

ICE 

ICE 

1 

ICE 

ICE 

ICE 




ICE 

ICE 

ICE 

1 1 

ICE 

ICE 

ICE 

1 1 

ICE 

ICE 

ICE 

1 

ICE 

ICE 

ICE 



(b) Ice Catch Calibration Tests 

FIGURE 12. PLATE ARRANGEMENT IN FREEZING DRIZZLE AND 
LIGHT FREEZING RAIN TESTS 


16 































































(b) Ice Catch Calibration Tests 


FIGURE 13. PLATE ARRANGEMENT IN FREEZING FOG TEST 


17 

















































































3.5.3 Rain on a Cold-Soaked Wing Test . 


For the rain on a cold-soaked wing test, the plate setup consists of a cold soak box filled with a 
65/35 propylene glycol/water volume ratio mixture on to which the test panel is placed (see 
figure 14). The box is contained within a 25-mm-thick polystyrene insulating jacket equivalent 
to a thermal resistance RSI equal to 1.3. During a calibration test, the panel is covered with 
twelve 100- by 100-mm pans. 



FIGURE 14. COLD SOAK BOX PLATE SETUP 

3.6 MEASURED PARAMETERS . 

The validity of the tests depends essentially on the three following environmental parameters: 
icing intensity, air, and plate temperatures. Recordings of the air and plate temperatures ensure 
that they are maintained during the test at target values within the prescribed allowable 
variations. Air temperature and humidity sensors are located within 1.5 meters from the test 
plates. The plate temperature sensors consist of thermocouples and platinum resistance 
temperature detectors (RTD) fixed to the underside of the test plate 150 mm from the top and the 
side edges. All these sensors are linked to a data acquisition system computer which records and 
logs test data in real time throughout the course of a test at the sampling rate of two data per 
second. 

3.7 CALIBRATION . 

Calibration tests are conducted for each condition of AET tests in order to establish that even and 
reproducible ice formation occurs over the surface of the test plates, i.e., 

• The target icing intensity for the test is within an acceptable range, and 

• The icing intensity over the surface of the panels exhibits a good distribution. 


18 



The allowable variations in temperature and icing rates are shown in table 4. Calibration tests 
consist of measuring the icing intensity by means of small 100- by 100-mm ice catch plates or 
pans placed on test panels. Figure 15 shows the arrangement of 20 pans used for calibration tests 
of freezing drizzle and light freezing rain. These pans or plates are weighed prior to and on 
completion of each test and the difference in the recorded weights is the ice catch for that plate. 
The icing intensity 7 for each plate is then calculated using the following relation: 

/ (g/dm 2 h) = _ Ice weight (g) _ 

[ice catch plate area (dm 2 ) x calibration test duration (h)] 



FIGURE 15. ICE CATCH PAN ARRANGEMENT IN FREEZING DRIZZLE AND 
LIGHT FREEZING RAIN CALIBRATION TESTS 


For a calibration test to be considered valid, the average icing intensity over the whole test panel 
surface must correspond to the value prescribed for that particular test condition and variations 
shall be within the limits specified in table 4. If not, calibration tests are repeated varying 
parameters until the required ice catch and distribution is obtained. Therefore, before conducting 
a fluid test under a given condition, a valid calibration test is conducted in that particular 
condition. 

3.7.1 Frost Calibration . 

A total of eighteen 100- by 100-mm small ice catch plates are used in the frost calibration tests 
performed for each temperature condition. This includes nine ice catch plates replacing the three 
300- by 100-mm test plates and nine others used to control icing intensity during fluid test. The 
average ice catch over the test surface corresponds to the frost accumulation for that plate. The 
average ice catch over the whole test surface corresponds to frost accumulation for that particular 
test condition and variation must be within the limits specified in table 4 for the calibration test 
to be considered acceptable. 


19 







TABLE 4. ALLOWABLE VARIATION IN TEMPERATURE AND 
ICING INTENSITY FOR A CALIBRATION TEST 


Test 

Temperature 

(°C) 

mmmsm 



Freezing 

Fog 

Snow 

Freezing 

Drizzle 

Light 

Freezing Rain 

Rain on a Cold- 
Soaked Wing* 

-3 ±0.5 (1) 

0.2 ± 0.02 
(± 0 . 03 ) 

■SSI 

10 ±0.5 
(± 0 . 5 ) 

5 ±0.2 

(± 0 . 3 ) 

13 ±0.5 

(± 0 . 7 ) 

1 

5 ±0.3 
(± 0 . 3 ) 

25 ±1.0 

(± 1 . 5 ) 

13 ±0.5 

(±0.7) 

25 ±1.0 
(± 1 . 5 ) 

1 

^m£E !■ 

-10 ±0.5 (2) 

0.15 ±0.02 

(± 0 . 03 ) 

2 ± 0.2 

(±0.2) 

10 ±0.5 
(± 0 . 5 ) 

5 ±0.2 

(± 0 . 3 ) 

13 ±0.5 
(±0.7) 


5 ±0.3 
(± 0 . 3 ) 

25 ±1.0 
(± 1 . 5 ) 

13 ±0.5 

(± 0 . 7 ) 

25 ±1.0 
(± 1 . 5 ) 

. 

-25 ±1.0 (3) 

0.06 ±0.01 

(±0.01) 

WEmm 

5 ±0.2 
(±0.2) 


. ' 1 n is > |l 

5 ±0.3 

(± 0 . 3 ) 

10 ±0.5 
(± 0 . 5 ) 


Ug n ggn | 

—Sul USi 1 1SHIB til‘i: 1 S1 


For frost, <1) T air = 0° ±0.5°C T platts = -3° ±0.5°C, = -10°C T plates = -13° ±0.5°C, 

0) T air = -25 °C T plales = -28° ±1.0°C 


*For rain on cold-soaked wing, = +1° ±0.5°C 


3.7.2 Freezing Fog, Freezing Drizzle, Light Freezing Rain, and Rain on a Cold-Soaked Wing 
Calibration . 

For freezing fog, freezing drizzle, light freezing rain, and rain on a cold-soaked wing calibration 
tests, each test plate is replaced with twelve ice catch plates, which, in turn, are surrounded by at 
least eight additional reference ice catch plates (shaded on figures 12b and 13b), for a total of at 
least twenty 100- by 100-mm ice catch plates. Like in the frost tests, these plates are weighed 
prior to and on completion of each test and the recorded weight difference is the ice catch for that 
plate. The average ice catch is calculated on the ice catch plates placed over test plates as well as 
on the small reference plates surrounding them. It is the Ratio between these two calculated 
values that is used to estimate the icing intensity during a fluid test run, when only the reference 
pans are available, i.e., 


Ratio = Le!2!L 

Ipiate is the average ice catch on the pans over the test panel, 

7 re /is the average ice catch on the reference plates. 

During the course of a fluid test, the ice catch is measured on the reference surrounding plates 
and this value is then multiplied by the Ratio calculated above in previous calibration tests 


20 
































performed under the same conditions. The resulting value is the estimated icing intensity (7 p i ate ) 
over the test panel. 

For the fluid test run: 


Estimated I plale = I ref xRatio 

Where I re f is measured during a fluid test and Ratio has been determined in a previous valid 
calibration test. • 

3.7.3 Snow Calibration . 

The snow distribution box was built to accommodate only one test panel at a time without using 
any surrounding reference ice catch pans. Therefore, for a snow calibration test, ten snow catch 
150- by 100-mm pans with a 15-mm edge are used for each set of conditions, which are placed 
over the test panel surface. As with ice catch plates, the snow catch pans are weighed prior to 
and on completion of each test and the difference corresponds to the snow intensity for that 
particular test condition. The snow intensity is the average of the snow catch collected in the ten 
pans with a calculated standard deviation. This average value should be within the target values 
specified for each test condition of table 3, whereas the calculated standard deviation from the 
ten snow catch pans must fall within the variations given in table 4. The degree of 
reproducibility is checked by performing not less than two successive calibration tests before a 
test run in the same condition. 

4. RESULTS . 

4.1 WATER SPRAY AND HIGH HUMIDITY ENDURANCE TESTS (WSET AND HHET) . 

The first task was to perform WSET and HHET standard tests on a 50/50 dilution sample of the 
two candidate fluids to verify whether they can be considered as approved SAE Type I fluids. 
After receiving the two candidate fluids, ADF on June 23, 1999, and OCTAFLO on July 12, 
1999, dilutions were prepared according to the sample selection procedures and each fluid was 
tested sheared and unsheared using a set of three standard plates according to Annex A of 
AMS 1424 B. Full WSET and HHET test description and procedures are detailed in reference 8. 

WSET and HHET results are presented in table 5 and figure 16. For both tests, ice 
catch measured on the reference plates is within the prescribed target values, which are 
5.0 ±0.2 g/dm 2 h in the case of the WSET and 0.30 ±0.05 g/dm 2 h in the case of the HHET. For 
both fluids, the first ice event (FIE), the numbers in bold in the table 5, exceed the 3 and 20 
minutes minimum values specified for WSET and HHET respectively for an SAE Type I fluid to 
be approved. In WSET tests, sheared samples (solid points) show failure times varying by 
about ±30 seconds of values of fluids tested unsheared (open points). In HHET tests, failure 
times of sheared samples are 1 to 4 minutes lower than those observed with the fluid tested 
unsheared, EG times being a little shorter than PG times. However, these differences are within 
the range of the experimental error and thus are not considered to be significant. The 
experimental error in the WSET is estimated to be 1 min for Type I fluids. Even if the ice 


21 



TABLE 5. WATER SPRAY AND HIGH HUMIDITY ENDURANCE TESTS RESULTS FOR 
OCTAFLO AND ADF CONCENTRATE DEICING FLUIDS 


Fluid 

Shearing 

Test Code 

Date 

(y-m-d) 

ICE DATA 

FLUI 

D DATA I 

Plate 

WfflBSM 

Plate 

FIE 1 
(min: sec) 

MIT 2 

(min:sec) 

PIL 3 

(mm) 

OCTAFLO 50/50 DILUTION—WATER SPRAY ENDURANCE TEST 


WS2081 

99-08-26 

PI 

4.98 ±0.10 

P2 

5:45 

mm 


Unsheared 


i , 

P3 

4.95 ±0.11 

P4 

5:45 


300 




P5 

4.96 ±0.12 

P6 

5:50 

6:30 

300 


WS2082 

99-08-26 

PI 

5.02 ±0.10 

P2 

5:15 

6:15 

300 

Sheared 



P3 

4.98 ±0.09 

P4 

5:20 

6:15 

300 




P5 

4.97 ±0.09 

P6 

5:20 

6:15 

300 

OCTAFLO 50/50 DILUTION—HIGH HUMIDITY ENDURANCE TEST 




PI 

0.31 ±0.01 

P2 

30:37 

32:20 


Unsheared 

HH1148 

99-09-10 

P3 

0.29 ±0.00 

P4 

33:38 

35:52 





P5 

0.30 ±0.01 

P6 

34:30 

35:20 

300 


HH1150 

99-09-13 

PI 

0.34 ±0.00 

P2 

29:05 

33:40 

300 

Sheared 



P3 

0.31 ±0.01 

P4 

29:45 

35:30 

300 




P5 

0.31 ±0.02 

P6 

32:20 

34:10 

300 

ADF CONCENTRATE 50/50 DILUTION—WATER SPRAY ENDURANCE TEST 




pi 

5.03 ±0.09 

P2 

3:10 

5:35 

n.m. 

Unsheared 

WS2079 

99-08-25 

P3 

5.06 ±0.09 

P4 

4:20 

5:35 

n.m. 




P5 

5.01 ±0.11 

P6 

4:05 

5:30 

n.m. 




PI 

5.03 ±0.09 

P2 

3:45 

5:45 

mssm 

Sheared 

WS2080 

99-08-25 

P3 

5.01 ±0.08 

P4 

4:20 

5:55 

mm 




P5 

4.99 ±0.09 

P6 

4:30 

5:45 

300 

ADF CONCENTRATE 50/50 DILUTION—HIGH HUMIDITY ENDURANCE TEST 




PI 

0.32 ±0.00 

P2 

31:05 

31:50 

KM 

Unsheared 

HH1201 

99-10-28 

P3 

0.27 ±0.01 

P4 

34:05 

35:40 

mM 




P5 

0.27 ±0.01 

P6 

33:55 

35:15 

300 




PI 

0.32 ±0.00 

P2 

27:55 

33:20 

KM 

Sheared 

HH1212 

99-11-30 

P3 

0.29 ±0.01 

P4 

30:45 

32:05 

mm 




P5 

0.30 ±0.01 

P6 

28:50 

32:50 

300 I 


1 FIE: First Ice Event: time for the first ice crystal to reach 25 mm in length. 

2 MIT: Mean Icing Time: time for the ice to reach a mean length of 25 mm. 

3 PIL: Plate Icing Length: Length of plate covered by ice at the end of the test. 


intensity varies only 4%, the main error comes from the nucleation time which can be 30 seconds 
to 1 min. In the HHET, in which holdover time is over 20 minutes, the experimental error is 
estimated to the 3 minutes corresponding to the variation in icing intensity. According to the 
sample selection procedures [8], the two candidate fluids selected for the AET testing program 
can be considered as approved SAE Type I fluids and consequentially are suitable for the present 
study. 


22 
















































































SHEARED 

. 

UNSHEARED 

» 

♦ADF 

A OCTAFLO 

OADF 

A OCTAFLO 

•adf 

■OCTAFLO 

oadf 

□OCTAFLO 



Icing Rate (g/dm 2 h) 


FIGURE 16. ANTI-ICING ENDURANCE TIME AS MEASURED IN WSET AND 

HHET TESTS 






































































4.2 ANTI-ICING ENDURANCE TIME TEST RESULTS. 


4.2.1 Methodology . 

The AET testing procedures used in this study are based on the procedures discussed at the 
Montreal G-12 Fluid Subcommittee meeting held in March 1999 [9] and revised at two 
subsequent meetings held between APS and AMIL representatives with the participation of the 
FAA and TDC, the first in Montreal on July 30, 1999, [3] and the second in Chicoutimi, on 

October 6, 1999 [4]. 

' » , 

These testing procedures detail general testing conditions, including fluid preparation and fluid 
failure criterion. General testing conditions and fluid application can be summarized as follows: 

• Test plates and panels: 300- by 100-mm mirror polished plates for frost tests and 
500- by 300-mm mirror polished panels in all other AET tests; 

• Volume of fluid applied: 120 mL applied on 300- by 100-mm plates and 500 mL on 
500- by 300-mm panels; 

• Fluid temperature at application and shearing condition: all fluids are applied at 
20° ±5°C, unsheared; 

• Exposure to freezing precipitation: there is a 5-minute delay between fluid application 
and exposure of fluid-coated plate to icing; 

• Water quality: all supercooled precipitation was generated using ASTM D1193 Type IV 
water. 

4.2.1.1 Sample Dilution Preparation . 

Fluid dilutions were prepared using hard water 1 , as per AMS 1424 B paragraph 3.3.3.1, diluted, 
as when applied on an aircraft, to a 10°C buffer from the OAT. The 10°C buffer was calculated 
using charts supplied by each manufacturer giving freezing point versus dilution. 

Table 6 depicts the identification number and water dilution ratios selected for the six different 
samples prepared for AET tests. The ADF fluid was diluted by increments of 1% whereas 
OCTAFLO samples were prepared using 5% dilution increments as specified by the 
manufacturer regarding fluid usage. As can be seen on the calculated buffer given in table 6, 
freezing points of ADF diluted samples are exactly 10°C below test temperatures whereas 
OCTAFLO dilution have freezing points of more than 10°C below the testing temperatures. For 
instance, the OCTAFLO 60/40 dilution used in AET tests at -25°C has a buffer of 5°C higher 
than the 10°C buffer, while the 45/55 and 35/65 dilutions used in tests at -10° and -3°C present a 
buffer of 2°C higher than the target value of 10°C. 


1 Composition of hard water: dissolve 400 mg ±5 calcium acetate dihydrate (Ca(C 2 H 3 ) 2 ) 2 . 2 H 2 0 ), and 280 mg ±5 
magnesium sulfate heptahydrate (MgSO 4 .7H 2 0), both of analytical reagent quality, in 1 liter of ASTM D 1193, 
Type IV, water. 


24 




TABLE 6. FLUID DILUTION SELECTION AND IDENTIFICATION 


No 

AMIL Code 

Manufacturer 

Fluid 

Lot No. 

Dilution 

Freezing Point 

Buffer 

1.0 

C418 



F-21104R 

NEAT 

— 

— 

m 


OCTAGON 

OCTAFLO 

F-21104R 


-40°C 

15°C 


KB 

OCTAGON 

OCTAFLO 

F-21104R 


-22°C 

12°C 

ID 

IB 

OCTAGON 

OCTAFLO 

F-21104R 

35/65 

-15°C 

12°C 

ID 


1S1SHBSB 


F-21104R 

30/70 

-11°C 

12°C 

IB 


OCTAGON 

OCTAFLO 

F-21104R 

50/50 

-28°C 

— 

2.0 

C397 

UCAR 

ADF 

67-CHC-12-B 

NEAT 

-28°C 

— 

Bl 


| 

ADF 

67-CHC-12-B 


-35°C 

10°C 

EH 



ADF 

67-CHC-12-B 


-20°C 

10°C 


C485 

UCAR 

ADF 

67-CHC-12-B 

28/72 

-13°C 

10°C 

m 

C511 

UCAR 


67-CHC-12-B 

21/79 

-9°C 

10°C 

pm 

C499 

UCAR 

ADF 

67-CHC-12-B 

50/50 

-34°C 

— 


4.2.1.2 Failure Criterion and Type . 

In all the AET tests with the exception of frost, failure is called when 30% of the plate is covered 
with frozen contamination. In frost tests, failure is called when 50% of the plate is covered by 
ice. Pen marks on the plate are used to estimate the area of failure. For instance, a line drawn 
across the 300- by 500-mm panel at 150 mm from the top edge will delineate an area 
corresponding to 30% of the plate. 

The frozen contamination at failure may appear under different forms. Examples of such 
appearances include, but are not limited to: 

• Ice front 

• Ice sheet 

• Slush, in clusters or as a front 

• Disseminated fine ice crystals 

• Frost on surface 

• Clear ice pieces partially or totally imbedded in fluid 

Usually, Type I fluid failure appears as an ice front, except in snow tests which involve slush in 
clusters. Normally, in the case of an ice front, the ice grows slowly, beginning at the top of the 
plate and moving down at a rate dependent on the failure time of the fluid. In some cases, 
however, a thin layer of fluid on the plate may freeze spontaneously, with no gradual growth. In 
such cases, nucleation may need to be initiated and the test repeated. In either case, should the 
ice cover more than 30% of the plate, the test is valid and must be repeated. No suspected 
delayed nucleation was observed during this study. 


25 



























































4.2.1.3 Measurements and Failure Recordings . 

The following measurements were performed: recordings during the course of each AET test of 
air and plate temperatures, relative humidity, icing intensity, anti-icing endurance times, and 
photographs of ice fronts at failure. 

Over one hundred tests, including twenty-five calibration and fifty fluid tests, were conducted 
under various temperature and icing intensity conditions as prescribed in the AET procedures 
[5], Tested fluids and dilutions are identified in table 6. 

4.2.2 Frost Tests . 

Frost tests include three calibration and six fluid tests with the results being summarized in 
table 7. In frost tests, the failure is called when the observed ice front covers 50% of the plate. 
The dilution level of each sample is given in the fluid label column. Measured anti-icing 
endurance times are shown in bold on table 7, where they can be compared to the times taken 
to cover 30% and 100% of the plate. All measured icing intensities and temperatures correspond 
to target values, varying well within the prescribed variations of ±0.5°C for temperature and 
±0.02 g/dm 2 h (0.01 g/dm 2 h at -25°C) for the icing intensity. 

Anti-icing endurance times measured with both fluid samples at air temperatures of 0°C 
(triangles), -10°C (diamonds), and -25°C (circles) are plotted in figure 17 as a function of the 
icing intensities. The lowest value of icing intensity (0.06 g/dm 2 h) is obtained at the lowest 
temperature of -25°C. OCTAFLO samples are represented using solid shapes while ADF fluids 
are identified by open shapes. The maximum variation observed in the measured anti-icing 
endurance times between the three plates is during tests performed at 0°C. Here the variation is 
of ±6 min with OCTAFLO and ±5 min with ADF corresponding to 7%. The smallest variation 
of about ±3% observed with the two fluids is at -25°C, the lowest testing temperature. 

The holdover times of the current SAE Type I fluid HOT table (table 1) are represented in 
figure 17 by a dotted line for all temperatures. Most of the measured anti-icing endurance times 
fall above this line, with the exception of ADF at -10° and -25°C. 

As shown in figure 17, OCTAFLO samples present endurance times generally greater than ADF 
fluids, the difference increasing in average from 8 min at 0°C, to 17 and 48 min at -10° and 
-25°C respectively. For both fluids, endurance times reach a minimum value at -10°C to 
increase at -25° and 0°C. This behavior can be explained by two factors which contribute to 
increase anti-icing times when temperature decreases: the higher glycol concentration of fluid, 
which lowers the freezing point of the sample when the test temperature is lowered and the frost 
intensity that is lower for lower temperatures. The greater endurance times of OCTAFLO as 
compared to ADF samples could being explained, in part, by its freezing point which is 5°C 
lower than the ADF dilution used in tests at -25°C, due to the fact that the fluid is more 
concentrated. 


26 



TABLE 7. FROST TESTS RESULTS 


CALIBRATION TESTS 


Test Code 

Air Temp. 
Plate Temp. 
(°C) 

Date 

(y-m-d) 

Intensity 

(g/dm 2 h) 

CAFRSTA 

0.0 ±0.0 
-3.0 ±0.1 

99-10-05 

0.19 ±0.02 

CAFRSTC 

-9.9 ±0.1 

-12.9 ±0.2 

99-10-09 

0.15 ±0.01 

CAFRSTE 

-25.2 ±0.5 

-28.3 ±0.1 

99-10-14 

0.08 ±0.01 


OCTAFLO (12° AND 15°C BUFFER DILUTION) 


Sample 

Dilution 

(fluid/water) 

Test Code 
(date) 

Air Temp. 
Plate Temp. 
(°C) 

Ice Data 

Fluid Data 

Plate 

Intensity 

(g/dm 2 h) 

Plate 

30% 1 
(min) 

50% 2 
(min) 

100% 3 4 
(min) 

■EgM 

FRSTA001 

0.0 ±0.0 

PI 

0.21 ±0.01 

P2 

4 

n.m. 

74 

82 


(99-10-06) 

-2.9 ±0.1 

P3 

0.20 ±0.01 

P4 

n.m. 

86 

99 




P5 

0.20 ±0.01 

P6 

n.m. 

79 

95 


FRSTC003 

-9.9 ±0.1 

PI 

0.16 ±0.00 

mm 

mm 

46 

60 


(99-10-09) 

-12.9 ±0.2 

P3 

0.15 ±0.00 

mm 

K9 

51 

68 

m 



P5 

0.15 ±0.01 


45 

51 

66 


FRSTE005 

-25.0 ±0.7 

PI 

0.07 ±0.00 

mm 

62 

84 

n.m. 

60/40 

(99-10-14) 

-27.9 ±0.4 

P3 

0.06 ±0.01 

wm 

60 

87 

n.m. 




P5 

0.06 ±0.01 

wBm 

61 

86 

n.m. 


ADF (10°C BUFFER DILUTION) 


Sample 

Dilution 

(fluid/water) 

Test Code 
(date) 

Air Temp. 
Plate Temp. 
(°C) 

Ice Data 

Fluid Data 

Plate 

Intensity 

(g/dm 2 h) 

Plate 

30% 1 
(min) 

50% 2 
(min) 

100% 3 
(min) 

■Bl 

FRSTA002 

0.0 ±0.0 

PI 

0.19 ±0.01 

P2 

55 


79 


(99-10-06) 

-2.9 ±0.1 

P3 

0.18 ±0.01 

P4 

59 


91 




P5 

0.19 ±0.01 

P6 

60 


90 

C484 

FRSTC004 

-9.9 ±0.1 

PI 

0.14 ±0.01 

P2 

28 

31 

39 

36/64 

(99-10-09) 

-12.9 ±0.3 

P3 

0.13 ±0.00 

P4 

31 

33 

45 




P5 

0.14 ±0.00 

P6 

29 

32 

43 

C483 

FRSTE006 

-25.7 ±0.2 

PI 

0.07 ±0.00 

P2 

34 

37 

57 

51/49 

(99-10-15) 

-27.9 ±0.4 

P3 

0.06 ±0.01 

P4 

34 

36 

57 




P5 

0.07 ±0.01 

P6 

33 

35 

56 


1 30%: Time for the ice to cover 30% of the test plate. 

2 50%: Anti-icing endurance time (failure): Time for the ice to cover 50% of the test plate. 

3 100%: Time for the ice to cover 100% of the test plate. 

4 n.m. means not measured 


27 

























































































0°C AOCTAFLO AaDF -10°C ♦OCTAFLO <>ADF -25°C *OCTAFLO OADF 



FIGURE 17. FROST ANTI-ICING ENDURANCE TIMES 

Failure appears in each frost test as a very thin ice front because of the relatively low icing 
intensity involved. The failure is illustrated in figure 18 which shows the ice as observed in the 
frost test performed with ADF samples at -10°C. This picture is typical of the failure observed 
with frost, which consists of a thin ice deposit grown from surrounding water vapor. This 
process is entirely different from the others in which ice is accreted from the freezing of 
supercooled droplets as observed in freezing fog and freezing drizzle tests. 


28 









































































FIGURE 18. ICE FRONT AT -10°C, TYPICAL OF A FROST FAILURE 


4.2.3 Freezing Fog Tests . 

Freezing fog tests comprise 6 calibration and 12 fluid tests performed at air temperatures of -3°, 
-10°, and -25°C and icing intensities of 2.0 ±0.2 and 5.0 ±0.3 g/dm 2 h. Results of these tests 
involving two plates per test are summarized in table 8. In freezing fog tests, the failure is called 
when the observed ice front covers 30% of the plate, the area being estimated by means of a 
straight line drawn across a plate 150 mm down from the top edge (see figure 19). The time 
measured at failure was in minutes and seconds. The dilution level of each sample is given in 
the fluid label column of table 8. All measured icing intensities correspond to target values of 
2.0 and 5.0 g/dm 2 h with variations maintained well within the prescribed tolerances of ±0.2 for 
2.0 g/dm 2 h, and ±0.3 for 5.0 g/dm 2 h. Measured plate and air temperatures are also within the 
target values with variations within the allowable value of ±0.5°C. 

Anti-icing endurance times measured with both fluids at the two icing intensities and air 
temperatures of 0°C (triangles), -10°C (diamonds), and -25°C (circles) are plotted in figure 20. 
OCTAFLO samples are represented using solid shapes whereas ADF fluids are identified by 
open shapes. In the tests performed with OCTAFLO and ADF samples, differences of less than 
30 seconds are observed between the anti-icing endurance times measured on the two different 
panels. 

The holdover times of the current SAE Type I fluid HOT table (table 1) are represented in 
figure 20 by two dotted lines, for 0° to -10°C and below -10°C ranges. All anti-icing endurance 
times measured fall between or above these lines, with the exception of ADF at -25°C and 
5.0 g/dm 2 h. 


29 





TABLE 8. FREEZING FOG RESULTS 


CALIBRATION TESTS 


Test Code 

Temp. 

(°C) 

Date 

(y-m-d) 

Ice Data 

Plate 

Intensity 

(g/dm 2 h) 

CAFOGA 

-3.0 ±0.0 

99-10-11 

A 

2.01 ±0.05 




B 

2.05 ±0.04 

CAFOGB 

-3.0 ±0.1 

99-10-06 

A 

4.9 ±0.2 




B 

4.7 ±0.2 

CAFOGC 

-9.9 ±0.0 

99-10-08 

A 

1.82 ±0.04 




B 

1.89 ±0.06 

CAFOGD 

-9.9 ±0.0 

99-10-07 

A 

5.2 ±0.3 




B 

4.9 ±0.2 

CAFOGE 

-26.0 ±0.1 

99-10-12 

A 

2.06 ±0.08 




B 

2.04 ±0.07 

CAFOGF 

-25.5 ±0.3 

99-10-13 

A 

4.9 ±0.1 




B 

4.8 ±0.1 


Sample 

Dilution 

(fluid/water) 


C482 

35/65 


C482 

35/65 


C481 

45/55 


C481 

45/55 


C480 

60/40 


C480 

60/40 


OCTAFLO, 12° AND 15°C BUFFER DILUTION 


Test Code 
(date) 


FOGAOOl 

(99-10-11) 


FOGB003 

(99-10-06) 


FOGC005 

(99-10-09) 


FOGD007 

(99-10-07) 


FOGE009 

(99-10-13) 


FODF011 

(99-10-14) 


Temp. 

(°C) 


-3.1 ±0.1 


-3.0 ±0.1 


-9.9 ±0.0 


-9.9 ±0.0 


-25.5 ±0.6 


-25.0 ±0.5 


Ice Data 


Plate 


A 

B 


A 

B 


A 

B 


A 

B 


A 

B 


A 

B 


Intensity 

(g/dm 2 h) 


1.99 

2.06 


4.7 

4.8 


1.90 

2.03 


5.2 

5.0 


1.93 

1.87 


4.8 

4.8 


Fluid Data 


Plate 


A 

B 


A 

B 


A 

B 


A 

B 


A 

B 


A 

B 


31 

(mi 


19 

19 


10 

10 


12:3( 

12:31 


6:4 

6:4 

10 

10 


6:3< 

6:3( 


ADF CONCENTRATE, 10 °C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 


Test Code 
(date) 


Temp. 

(°C) 


Ice Data 


Plate 


Intensity 

(g/dm 2 h) 


Fluid Data 


Plate 


3( 

(mi 


C485 

28/72 


FOGA002 

(99-10-11) 


-3.0 ±0.0 


A 

B 


1.97 

2.02 


A 

B 


18:30 

18:30 


C485 

28/72 


FOGA004 

(99-10-06) 


-3.0 ±0.0 


A 

B 


4.7 

4.6 


A 

B 


10:30 

10:15 


C484 

36/64 


FOGC006 

(99-10-09) 


-9.9 ±0.0 


A 

B 


1.88 

1.98 


A 

B 


12 

12 


C484 

36/64 


C483 

51/49 


C483 

51/49 


FOGD008 

(99-10-08) 


-9.9 ±0.0 


A 

B 


FOGEOIO 

(99-10-13) 


-25.2 ±0.4 


A 

B 


FOGF012 

(99-10-14) 


-23.6 ±0.9 


A 

B 


5.3 

4.8 


1.90 

1.95 


4.9 

4.8 


A 

B 


A 

B 


A 

B 


35 

45 


Anti-icing endurance time (failure): Time for the ice to cover 30% of the test plate. 












































































































FOGD008 

FIGURE 19. ICE FRONT AT -10°C, TYPICAL OF A FAILURE IN A FREEZING FOG TEST 

According to figure 20, endurance times of OCTAFLO and ADF samples are very comparable at 
temperatures of -3° and -10°C. Indeed, the less than 1 min variations observed are within 
experimental error, even if the difference could be the results of its 12°C buffer for OCTAFLO 
as compared to that of 10°C of the ADF samples. However, at -25°C, endurance times of 
OCTAFLO are between 2 to 3 minutes longer than those obtained with ADF fluids. The greater 
endurance times of OCTAFLO as compared to ADF samples could partially be explained by its 
5°C higher buffer than the prescribed value of 10°C at -25°C. As expected for both fluids, 
measured endurance times are shorter at the lower temperatures and higher icing intensity. 

Failure appears in freezing fog tests as an ice front with separated pieces of ice. Indeed, ice 
deposit in the freezing fog test forms and grows from supercooled droplets freezing upon the 
plate, whereas under frost conditions, ice forms from water vapor condensing on the plate 
maintained a few degrees below the air temperature. 

The ice front formed in the freezing fog tests for ADF at -10°C is illustrated in figure 19. The 
photo was taken when the ice front and separated pieces of ice covered the 30% failure area. 
This picture is typical of failures observed in freezing fog tests performed that consist of an ice 
front at the top and side edge with an ice sheet on the fluid surface. 


31 


-3°C AOCTAFLO aadf -10°c ♦ OCTAFLO OADF -25°c • OCTAFLO OADF 



FIGURE 20. FREEZING FOG ENDURANCE TIMES 

4.2.4 Snow . 

Snow tests consist of 5 calibration and 10 fluid tests performed at air temperatures of -4°, -10° 
and -25°C and icing intensities of 5.0 ±0.3, 10.0 ±0.5 and 25.0 ±1.0 g/dm 2 h. Although the 
current requirement of the proposed procedures specify -3°C as a test temperature, this test was 
conducted at -4°C due to difficulty in maintaining proper calibration. Reducing the temperature 
by 1°C eliminated this difficulty and the proper snow parameters were achieved. Each test 
involves two fluid-coated panels subjected to artificial snow one at a time. A fluid snow test is 
performed immediately following a validated calibration test under the specified icing condition. 
Results of the snow calibration and fluid tests are summarized in table 9. For snow tests, the 
failure is called when the fluid fails to absorb snow covering 30% of the test panel, which is 
estimated by an area delineated by a straight line drawn across plate 150 mm down from the top 
edge. The failure time is measured in minutes and seconds. The dilution level of each sample is 
given in the fluid label column. 


32 




























TABLE 9. SNOW TEST RESULTS 


CALIBRATION TESTS 


Test Code 

mmSMm 

Date 

(y-m-d) 

Intensity 

(g/dm 2 h) 

CASNWA 

-4.2 ±0.0 

99-10-15 


CASNWB 

-4 ±? 

99-10-15 

hseeeeh 

CASNWC 

-10.0+0.0 

99-10-13 

9.9 ±0.8 

CASNWD 

-10 ±? 

99-10-11 

25.4 ±1.6 

CASNWE 

-25 ±? 

99-10-05 

4.9 ±0.3 


OCTAFLO, 12° and 15°C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 

Test Code 

Temp. 

(°C) 

Ice Data 

Fluid Data | 

Plate 

Intensity 

(g/dm 2 h) 

Plate 

iP^M 1 

■ 

SNWA001 

-4.1 ±0.0 

A 

9.9 

A 

4:35 


SNWA002 

-4.1 ±0.0 

B 

9.9 

B 

4:30 


SNWB005 

-4.1 ±0.0 

A 


A 



SNWB006 

-4.1 ±0.0 

B 

BUI ■ 

B 

WBSM 

■ 

SNWC009 

-10.0 ±0.0 

A 

9.8 

A 

2:30 


SNWC010 

-10.0 ±0.0 

B 

9.8 

B 

2:35 

■ 

SNWD013 

-10 r 0 ±0.1 

A 

■ B 

A 

1:25 

l 

SNWD014 

-10.0 ±0.0 

B 


B 

1:30 

C480 

SNWE017 

-25 ±? 

A 

4.9 

A 

■ ... ■ ■ 

60/40 

SNWE018 

-24.9 ±0.1 

B 

4.9 

B 



ADF CONCENTRATE, 10°C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 

Test Code 

Temp. 

(°C) 

Ice Data 

Fluid Data | 

Plate 

B 


mu 

C485 

SNWA003 

-4.1 ±0.0 

mm 

9.9 

A 


28/72 

SNWA004 

-4.1 ±0.0 

mm 

9.9 

B 

WEm 

C485 

SNWB007 

-4.1 ±0.0 

mm 


A 

mm 

28/72 

SNWB008 

. -4.1 ±0.0 

mm 

bub 

B 

m 

C484 

SNWC011 

-10.0 ±0.0 

A 

9.8 

A 


36/64 

SNWC012 

-10.0 ±0.0 

B 

9.8 

B 

n 

C484 

SNWD015 

-10.0 ±0.0 

A 

25.4 

A 

B 

36/64 

SNWD016 

-10.0 ±0.0 

B 

25.4 

B 



SNWE019 

-24.9 ±0.1 

A 

4.9 

A 

4:30 


SNWE020 

-24.9 ±0.1 

B 

4.9 

B 

4:25 


1 Anti-icing endurance time (failure): Time for the fluid to fail to absorb snow covering 30% of the plate. 


All measured icing intensities correspond to target values of 5.0, 10.0, and 25.0 g/dm 2 h with 
variations kept well within the prescribed tolerances of ±0.3 at 5.0 g/dm 2 h, ±0.5 at 10.0 g/dm 2 h, 
and ±1.0 at 25 g/dm 2 h. Measured air temperatures are also within the target values with 
variations smaller than the allowable value of ±0.5°C, with the exception of the snow test at 
-3.0°C. Under that particular temperature condition, it was very difficult to obtain an even snow 
distribution, so tests were performed at -4°C where the prescribed even icing intensity could be 
achieved. 


33 


































































































The holdover times of the current SAE Type I fluid HOT table (table 1) are represented in 
figure 21 by two dotted lines for all temperatures. All anti-icing endurance times measured are 
shorter than the minimum SAE Type I fluid HOT table values. 



Anti-icing endurance times measured with the two fluids at the three values of icing intensities 
and air temperatures are plotted in figure 21. Triangles, diamond, and circles correspond to tests 
performed at 0°, -10° and -25°C respectively. OCTAFLO samples are represented using solid 
shapes whereas ADF fluids are identified by open shapes. In the ten tests performed with 
OCTAFLO and ADF samples, time variations of less than 1 minute are observed between anti¬ 
icing endurance times measured on two different panels coated with the same fluid. These 
variations do not appear to be dependent on fluid nor the testing temperature, but are within the 
experimental error of measurement. 

OCTAFLO endurance times are 30 sec to 2 min longer than ADF values (figure 21) at the three 
test temperatures of -3°, -10°, and -25°C. As expected with both fluids, measured endurance 


34 








































times are shorter at lower temperatures and at higher icing intensities, with the exception of the 
tests performed at -25°C for 5 g/dm 2 h. 

As opposed to the frost and freezing fog tests, snow tests involve solid ice particles impacting the 
fluid surface. In the first minutes that the fluid is exposed to the artificial snow, ice particles are 
easily dissolved in the fluid. Depending on the air temperature and the intensity of the snowfall, 
after a few minutes, ice particles take more and more time to be absorbed and a slush begins to 
form on top of and within the fluid. Therefore, ice contamination at failure consists mainly of 
slush, which is a mixture of partially diluted fluid and artificial snow particles. The artificial 
snow particles are both embedded in the fluid and floating on its surface. Even if the amount of 
observed slush usually is greater at the top of the plate, where the film of fluid is thinner, failure 
may appear anywhere on the test plate and is called when the fluid fails to absorb snow covering 
30% of the test panel surface. 

The slush formed in snow tests is shown in figure 22, which shows the ice contamination formed 
during the snow test performed at -10°C with ADF. The photo was taken when slush covered 
30% of the plate, i.e., at failure. This picture is typical of slush in clusters—failure observed in 
all ten snow tests performed. The percentage of contamination is evaluated in the course of the 
test by an observer inside the test chamber. 



SNWC010 


FIGURE 22. SLUSH AT -10°C, TYPICAL SNOW TEST FAILURE 
4.2.5 Freezing Drizzle . 

Freezing drizzle tests include four calibration and eight fluid tests performed at air temperatures 
of-3°, and -10°C and icing intensities of 5.0 ±0.3 and 13.0 ±0.5 g/dm 2 h. Each test involves two 
fluid-coated panels, labeled A and B, placed side by side in the test icing area. Results of these 
tests are summarized in table 10. In freezing drizzle tests, the failure is called when the observed 
ice front covers 30% of the plate, as in the freezing fog and snow tests. The dilution level of 


35 







each sample is given in the fluid label column of table 10. All measured icing intensities 
correspond to target values of 5.0 and 13.0 g/dm 2 h with variations kept well within or much less 
than the prescribed tolerances of ±0.3 at 5.0 g/dm 2 h and ±0.5 at 13.0 g/dm 2 h. Measured plate 
and air temperatures are also within the target values with variations smaller than the allowable 
value of ±0.5°C. 


TABLE 10. FREEZING DRIZZLE TEST RESULTS 


CALIB 

RATION TESTS 

Test Code 

Temp. 

(°C) 

Date 

(y-m-d) 

Ice Data 

Plate 

Intensity 

(g/dm 2 h) 

CAZLA 

-3.0 ±0.1 

99-09-22 

A 

B 

4.9 ±0.2 

4.9 ±0.1 

CAZLB 

-3.0 ±0.1 

99-09-10 

A 

B 

13.0 ±0.6 

13.4 ±0.5 

CAZLC 

-10.0 ±0.0 

99-09-20 

A 

B 

5.0 ±0.3 

4.9 ±0.3 

CAZLD 

-10.0 ±0.1 

99-09-07 

A 

B 

13.0 ±0.7 
13.2 ±0.4 


OCTAFLO, 12° AND 15°C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 

Test Code 
(date) 

Temp. 

(°C) 

Ice Data 

Fluid Data | 

Plate 

Intensity 

(g/dm^h) 

Plate 

K» 

C482 

ZLA001 

-3.0 ±0.0 j 

A 

4.9 

A 

11:50 

35/65 

(99-09-22) 


B 

4.9 

B 

12:40 

C482 

ZLA001A 

-3.0 ±0.0 

A 

5.1 

A 

14:00 

35/65 

(99-09-23) 


B 

4.9 

B 

15:10 

C482 

ZLB003 

-3.0 ±0.1 

A 

12.5 

A 

KB 

35/65 

(99-09-10) 


B 

12.6 

B 

HI 


ZLC005 

-10.0 ±0.0 

A 

5.1 

A 


I 

(99-09-21) 


B 


B 


C481 

ZLD007 

-10.0 ±0.2 

A 

HUH 

A 


45/55 

(99-09-09) 


B 


B 



ADF CONCENTRATE, 10°C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 

Test Code 
(date) 

Temp. 

(°C) 

Ice Data 

Fluid Data | 

Plate 

Intensity 

(g/dm 2 h) 

Plate 

■ 

C485 

ZLA002 

-3.0 ±0.0 

A 

4.9 

A 

10:50 

28/72 

(99-09-22) 


B 

4.8 

B 

10:10 

C485 

ZLB004 

-3.0 ±0.1 

A 

12.8 

A 

5:10 

28/72 

(99-09-11) 


B 

13.5 

B 

6:00 

C484 

ZLC006 

-10.0 ±0.1 

A 

5.1 

A 

6:40 

36/64 

(99-09-20) 


B 

4.9 

B 

6:35 

C484 

ZLD008 

-10.0 ±0.1 

A 

12.8 

A 

ns 

36/64 

(99-09-09) 


B 

13.5 

B 

11 


1 Anti-icing endurance time (failure): Time for the ice to cover 30% of the test plate. 


36 











































































Anti-icing endurance times measured with the two candidate fluids at the two icing intensities 
and air temperatures of -3° and -10°C are plotted in figure 23 where air temperatures of -3° and 
-10°C are represented by triangles and diamonds respectively. OCTAFLO samples correspond 
to solid shapes whereas ADF fluids are identified by open shapes. In the eight tests performed 
with OCTAFLO and ADF samples, endurance time variations of 1 minute or less are observed 
between the anti-icing endurance times measured on the two different panels. These variations 
do not appear to be dependent on the fluid nor the testing temperature and are within 
experimental error of measurement. 


-3°C A OCTAFLO A ADF -10°C ♦OCTAFLO <>ADF 



FIGURE 23. FREEZING DRIZZLE ENDURANCE TIMES 

The holdover times of the current SAE Type I fluid HOT table (table 1) are represented in figure 
23 by two dotted lines for all temperatures. All anti-icing endurance times measured fall 
between or above these lines, with the exception of both fluids at -10°C and 13.0 g/dm 2 h. 


37 






















According to results shown in figure 23, OCTAFLO endurance times observed under these 
conditions of temperature and icing intensity are found to be between 1 and 2 minutes longer 
than those obtained with ADF samples. As expected for both fluids, measured endurance times 
are shorter at lower temperatures and higher icing intensities. 

In freezing drizzle tests, failure appears as an ice front, similar to the freezing fog tests. 
Moreover, as in the freezing fog tests, ice deposits in freezing drizzle tests, form and grow from 
supercooled droplets freezing upon plate, the difference being the droplet sizes which are 10 to 
20 times larger than those of freezing fog. 

The ice front as formed in freezing drizzle for OCTAFLO at -10°C is illustrated in figure 24. 
This photo was taken when the ice front covered 30% of the area. This photo is typical of the ice 
contamination observed at failure in other freezing drizzle tests. 


ZLC005 

FIGURE 24. ICE FRONT AT - 10°C. TYPICAL OF A FAILURE IN 
A FREEZING DRIZZLE TEST 



4.2.6 Light Freezing Rain . 

Light freezing rain tests consist of four calibration and eight fluid tests performed at air 
temperatures of -3° and -10°C and icing intensities of 13.0 ±0.5 and 25.0 ±1.0 g/dm 2 h. Each 
fluid test involves two panels placed side by side exposed to the freezing rain. Results of 
calibration and fluid tests are summarized in table 11. In light freezing rain tests, the failure is 
called when the ice is covering 30% of the plate, as in the freezing drizzle tests. All measured 
icing intensities correspond to targeted values of 13.0 and 25.0 g/dm 2 h with variations well 
within the prescribed tolerances of ±0.5 at 13.0 g/dm 2 h and ±1.0 at 25.0 g/dm 2 h. Measured plate 
and air temperatures are also within the target values with variations smaller than the allowable 
value of ±0.5°C. 


38 





TABLE 11. LIGHT FREEZING RAIN TEST RESULTS 


CALIBRATION TESTS 


Test Code 

Temp. 

(°C) 

Date 

(y-m-d) 

Ice Data 

Plate 


CALZRA 

-3.0 ±0.0 

99-10-04 

A 

B 

13.5 ±0.5 
13.0 ±0.4 

CALZRP 

-3.0 ±0.1 

99-10-01 

A 

B 

25.3 ±1.1 

24.1 ±0.6 

CALZRC 

-10.0 ±0.1 

99-10-03 

A 

B 

13.4 ±0.6 

13.0 ±0.5 

CALZRD 

-10.0 ±0.1 

99-09-30 

A 

B 

24.8 ±1.1 
24.3 ±0.6 


OCTAFLO, 12° AND 15°C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 

Test Code 
(date) 

Temp. 

(°C) 

Ice Data 

Fluid Data | 




m 


LZRA001 

-3.0 ±0.0 

A 

13.2 

A 

7:00 


(99-10-05) 


B 

12.6 

B 

7:00 


LZRA001A 

-3.0 ±0.0 

A 

12.7 

A 


1 

(99-10-05) 


B 


B 

KB! 


LZRB003 

-3.0 ±0.1 

A 

wmam 

A 



(99-10-01) 


B 

WM55M 

B 

5:00 

KM 

LZRC005 

-10.0 ±0.2 

A 

13.1 

A 

3:30 


(99-10-03) 


B 

13.1 

B 

3:25 


LZRD007 

-10.0 ±0.1 

A 

25.9 

A 

WMk 


(99-09-30) 


B 

25.1 

B 

— 


ADF CONCENTRATE, 10°C BUFFER DILUTION 


Sample 

Dilution 

(fluid/water) 

Test Code 
(date) 

Temp. 

(°C) 

Ice Data 

Fluid Data | 

Plate 

Intensity 

(g/dm 2 h) 

Plate 

mm 


LZRA002 

-3.0 ±0.0 

A 

12.9 

A 



(99-10-04) 


B 

13.1 

B 


■ ■ 

LZRB004 

-3.0 ±0.0 

A 

25.9 

A 

4:20 

1 

(99-10-01) 


B 

25.4 

B 

4:30 


LZRC006 

-10.0±0.1 

A 

13.4 

A 



(99-10-03) 


B 

13.5 

B 


C484 

LZRD008 

-10.0 ±0.1 

A 

24.7 

A 

■ 

36/64 

(99-09-30) 


B 

25.3 

B 

mmSM 


1 Anti-icing endurance time (failure): Time for the ice to cover 30% of the test plate. 


Anti-icing endurance times measured with the two candidate fluids at the two icing intensities 
and air temperatures of -3° and -10°C, are plotted in figure 25 where air temperatures of -3° and 
-10°C correspond to triangles and diamonds respectively. The OCTAFLO samples are identified 
by solid shapes, and the ADF samples are identified by open shapes. In the eight tests performed 
with OCTAFLO and ADF samples, time variations of 1 minute or less are observed between the 
anti-icing endurance times measured on the two different panels. These variations do not appear 


39 
































































































to be dependent of the fluid nor the 
error of measurement. 


testing temperature and can be attributed to experimental 


The holdover times of the current SAE Type I fluid HOT table (table 1) are represented in 

figure 25 by two dotted lines for all temperatures. All anti-icing endurance times measured fall 
between or above these lines. 



FIGURE 25. LIGHT FREEZING RAIN ENDURANCE TIMES 


40 






























































In light freezing rain tests, failure appears as a speckled ice front. Ice deposits, in light freezing 
rain tests, form and grow from supercooled droplets freezing on the plate. 

The speckled ice front as formed in light freezing rain with OCTAFLO at -10°C is illustrated in 
figure 26. This photo was taken when the ice front covered 30% of the area. This picture is 
typical of the speckled ice front observed at failure in other light freezing rain tests. 



LZRB003 


FIGURE 26. SPECKLED ICE FRONT AT -10°C, TYPICAL OF A 
FAILURE IN A LIGHT FREEZING RAIN TEST 

4.2.7 Rain on a Cold-Soaked Wing . 

Rain on a cold-soaked wing tests consist of two calibration and four fluid tests performed at an 
air temperature of +1°C and icing intensities of 5.0 ±0.3 and 75 ±3.0 g/dm 2 h. Two water droplet 
sizes are used to produce the freezing rain: 237 pm MVD at 5 g/dm 2 h and 1400 pm MVD at 75 
g/dm 2 h. For this AET testing procedure, the cold soak box is first chilled to 
-17° ±0.5°C then the test panel is placed on it. After the panels top surface is adequately covered 
to prevent frost formation due to condensation, the cold box is allowed to warm up. When the 
temperature sensor that is located between the panel and the cold box surface reads 
-10°C, the test plate is coated with the fluid sample and exposed to freezing rain after the 
standard 5-minute delay. 

Each fluid test involves two panels placed one at a time on the cold-soaked stand. The fluids 
used for these tests are ADF 21/79 (fluid/water) diluted samples with a freezing point of -9°C 
and OCTAFLO 30/70 dilution with a freezing point of -11°C. For comparison purpose, the two 
fluid tests were also performed at 5.0 g/dm 2 h icing intensity with a 10°C buffer with respect to 
the cold soak box temperature of -10°C; one with ADF 36/64 diluted samples (F.R* = -20°C) and 


41 



the other with OCTAFLO 45/55 dilution (F.P. = -22°C). A valid ice catch calibration test at the 
two prescribed icing intensities is performed prior to each test condition. 

The holdover times of the current SAE Type I fluid HOT table (table 1) are represented in 
figure 27 by two dotted lines for all temperatures. Anti-icing endurance times measured at 5 
g/dm 2 h fall between or above these lines and those measured at 75 g/dm 2 h are shorter than the 
SAE Type I fluid HOT table values. 

Results of calibration and fluid'tests are summarized in table 12. The failure is called when the 
observed ice front covers 30% of the test panel. All measured icing intensities correspond to 
target values of 5.0 and 75 g/dm 2 h with variations kept within the prescribed ±0.3 at 5.0 g/dm 2 h 
and ±3.0 at 75 g/dm 2 h. Measured plate and air temperatures are also at the target value of +1°C 
with variations smaller than the allowable value of ±0.5°C. 

TABLE 12. RAIN ON A COLD SOAK BOX TEST RESULTS 


CALIBRATION TESTS 


Test 

Code 

Temp. 

(°C) 

Date 

(y-m-d) 

Ice Data | 

Plate 

Intensity 

(g/dm 2 h) 

CACSWA 

1.0 ±0.0 

99-10-07 

A 

4.9 ±0.2 

CACSWB 

0.9 ±0.1 

99-10-11 

A 

76 ±3 


OCTAFLO, 12° AND 23°C BUFFER DILUTION 


Sample 

Test 


' Ice Data 

Fluid Data | 

Dilution 

(fluid/water) 

Code 

(date) 

Temp. 

(°C) 


RBI 

Plate 

HjinimSSK 

C512 

CSWA001 

1.0 ±0.0 

A 

4.9 

A 


30/70 

(99-10-08) 


B 

5.2 

B 

■H 

C481 

45/55 

CSWA003 

(99-10-09) 

1.0 ±0.0 

A 


A 


C512 

CSW B005 

1.0 ±0.0 

A 

75.4 

A 


30/70 

(99-10-12) 


B 

72.1 

B 

1 


ADF CONCENTRATE, 10° AND 21°C BUFFER DILUTION 


Sample 

Test Code 
(date) 

nu 

Ice Data 

Fluid Data i 

Dilution 

(fluid/water) 

■9 

Plate 

Intensity 

(g/dm 2 h) 

Plate 


C511 

CSWA002 

1.0 ±0.0 

A 

4.8 

A 


21/79 

(99-10-09) 


B 

5.1 

B 

i 

C484 

36/64 

CSWA004 

(99-10-09) 

1.0 ±0.0 

A 

5.2 

A 

7:00 

C511 

CSWB006 

1.0 ±0.0 

A 

74.5 

A 

00:35 

21/79 

(99-10-12) 


B 

73.6 

B 

00:30 


1 Anti-icing endurance time (failure): Time for the ice to cover 30% of the test plate. 


42 





















































STANDARD TESTS 

FP -11°C 

■ OCTAFLO 

FP -9°C 

□ ADF 


NON-STANDARD TESTS 

FP -22°C 

® OCTAFLO 

FP -20°C 

OADF 




FIGURE 27. RAIN ON A COLD-SOAKED WING ENDURANCE TIMES 

Anti-icing endurance times measured with the two diluted samples are shown in figure 27 for the 
icing intensities of 5 and 75 g/dm 2 h. The OCTAFLO are identified by using solid shapes, and 
the ADF samples are identified by open shapes. Diluted samples with freezing point of 
-10° ±1°C are identified by using squares, whereas those with freezing points near -20°C are 
identified by circles. In the tests performed with samples of the same fluid, time variations of 
less than 1 minute are observed between the anti-icing endurance times measured on the two 
different panels. These variations do not appear to be dependent on the fluid nor the testing 
temperature and are within the experimental error of measurement. 


43 



























































As shown in figure 27, OCTAFLO and ADF endurance times in tests performed at the 75 g/dm 2 h 
icing intensity are comparable within the experimental error, which can be estimated to less than 
30 sec. However, at the low 5 g/dm 2 h icing intensity, OCTAFLO endurance times are found to 
be 3 to 4 minutes longer than those obtained with ADF samples. Nevertheless, the number of 
tests, limited to only two is not sufficient to analyze this discrepancy, these tests need to be 
repeated. As expected for both fluids, endurance times are shorter with higher icing intensity. 

The endurance times observed with the fluids diluted to a freezing point around -20°C (circles in 
figure 27) are 2 and 5 minutes longer with OCTAFLO and ADF respectively as compared to 
dilutions with a freezing point about 10°C higher. As mentioned before, the number of tests is 
not sufficient to analyze this difference. 

In rain, on a cold-soaked wing test, failure appears as an ice front, as observed in other AET 
tests, with the exception of snow. Although freezing fog and freezing drizzle consists of 
supercooled droplets which freeze on impact, ice contamination in rain oil a cold-soaked wing . 
test was formed in a similar manner. For the latter, rain droplets freeze because of contact with a 
surface whose temperature is below freezing. 

The ice front as observed in rain on a cold-soaked wing test for OCTAFLO at +PC and 
5 g/dm 2 h, is illustrated in figure 28. This photo was taken when the ice front covered 30% of the 
test plate area. This picture is typical of the ice fronts observed at failure in other rain on a cold- 
soaked wing tests in which an appreciable ice contamination were initiated on the edges and at 
the top of the plate. This phenomenon is called the edge effect. 



CASWA001B 

FIGURE 28. FAILURE IN RAIN ON A COLD-SOAKED.WING AT +1°C 


44 




5. COMPARISON BETWEEN AMIL AND APS RESULTS. 


5.1 SCOPE . 

One of the objectives of this study is to establish a comprehensive basis to compare and analyze 
results obtained by APS in the NRC laboratory with those of this study in order to determine the 
variation between the two methods and facilities, and to reconcile and establish a single set of 
standardized procedures. The procedures are, ultimately, to be published after concurrence with 
the SAE G-12 Fluids and Holdover Time Subcommittees. This way, any testing facility with the 
appropriate equipment will be able to perform AET tests by adhering to an approved published 
set of procedures. 

Another expected outcome of this study is the substantiation of the current generic Type I fluid 
HOT table. Indeed, once substantiated, new fluids will have to be tested according to the AET 
set of standardized tests, and the generic table will be adjusted accordingly prior to the use of the 
fluid. As with the establishment and the publication of approved standardized prpcedures, the 
substantiation of generic and specific HOT tables is also under the responsibility of the SAE 
G-12 Fluids and Holdover Time Subcommittees. This is why the results obtained in the present 
work are to be further discussed in subsequent subcommittee meetings. 

5.2 COMPARISON OF AMIL AND APS DATA . 

During the month of July 1999, APS conducted tests in NRC laboratory with samples of the 
same two Type I fluids. In fact, the two manufacturers sent a sample of the same lot to AMIL 
and APS to be sure that both facilities will perform tests on identical fluids. In the APS work 
program. Type I diluted samples were tested according to procedures which are similar or 
comparable to five of the six AET procedures used in the present work. APS, however did not 
test for frost. Since APS tests were performed prior to the meeting held in Montreal on July 30, 

1999 [3], the APS/NRC procedures were not subjected to any discussion with AMIL 
representatives. 


Results of APS/NRC tests performed with the two Type I fluids were forwarded to AMIL in 
mid-November 1999. The information released by APS did not contain the variations in icing 
intensities and air temperatures, which should be expressed by the standard deviation. 

The analysis of APS data obtained in the five different types of icing precipitation tested reveals 
that the aimed comparison is limited to tests performed at -10°C under freezing fog, snow, 
freezing drizzle, and light freezing rain conditions and at +1°C with rain on a cold-soaked wing 
test. The anti-icing endurance times, as measured by AMIL and APS in these five testing 
conditions, judged comparable, are presented in table 13. The difference between AMIL and 
APS failure times, expressed in absolute and percentage values, are shown in the two last 
columns of the table. 


45 




TABLE 13. COMPARISON OF AM EL AND APS RESULTS 


__ AMIL _ 

Failure Icing 
Fluid time Intensity 

Identification (min) (g/dm J h) 


OCTAFLO 45/55 


OCTAFLO 45/55 


OCTAFLO 45/55 


OCTAFLO 45/55 


ADF 36/64 


ADF 36/64 


ADF 36/64 


ADF 36/64 


OCTAFLO 45/55 


OCTAFLO 45/55 


ADF 36/64 I 1.5 


ADF 36/64 1.5 


OCTAFLO 45/55 


OCTAFLO 45/55 


OCTAFLO 45/55 


|lsisUGS2SSC£K{il 


ADF 36/64 


ADF 36/64 


ADF 36/64 


ADF 36/64 


OCTAFLO 45/55 I 3.5 


OCTAFLO 45/55 
OCTAFLO 45/55 I 2.3 



Test ] 

Temp.. Fluid 

(°C) Identification 

FREEZING FOG 


OCTAFLO 42/58 


OCTAFLO 42/58 


OCTAFLO 42/58 


OCTAFLO 42/58 


ADF 36/64 


ADF 36/64 


ADF 36/64 


ADF 36/64 


SNOW 


OCTAFLO 42/58 


OCTAFLO 42/58 


ADF 36/64 


ADF 36/64 


_FREEZING DRIZZLE 


-10.0 OCTAFLO 42/58 


-10.0 I OCTAFLO 42/58 


-10.0 OCTAFLO 42/58 


-10.0 | OCTAFLO 42/58 


ADF 36/64 


ADF 36/64 


ADF 36/64 


ADF 36/64 


LIGHT FREEZING RAIN 
-10.0 I OCTAFLO 42/58 1 


-10.0 | OCTAFLO 42/58 


APS 

Failure 

time 

(min) 


Icing 

Intensity 

(g/dm 2 h) 


Comparison 

A Fail. ““ 
Time 

(min) A % 



8.2 

5.6 

-10.0 


5. 

4 

-10.5 





2.5 

42% 

2.3 

41% 


-10.0 

1.7 

34% 

-10.1 

1.7 

'Kw 

-10.1 

1.7 






























































































































































































































In all cases, with the exception of freezing fog, APS endurance times, when expressed in 
percentage, are longer than AMIL measurements. Indeed, with freezing fog tests, APS-measured 
endurance times are very comparable to AMIL values, the time differences ranging from -16% to 
+9%, for a ±10% variation. With the four other types of tests, HOT values measured by APS are 
found to be, on average, 30% longer than those obtained by AMIL. That increases seems to 
depend on the nature and the intensity of the iced precipitation. In the snow tests, the average 
increase is 45%; in freezing drizzle and light freezing rain tests, increases range from low values 
of 4% to 33%, respectively, to high values of 42% to 50% for an average increase of 23% in 
freezing drizzle and 41% ih light freezing rain. In rain on a cold-soaked wing tests, APS 
endurance times are found to be longer by 1% to 71%, corresponding to an increase averaging 
of 36%. 

These endurance time differences, when expressed in percentage, correspond to very large 
values. However, in absolute values, i.e., in minutes, the time differences are 0.1 to 4 minutes. 
When the 4 min. differences, observed in the two rain on a cold-soaked wing tests with ADF 
samples, are excluded (see section 4.2.7), APS endurance times are on average 1.3 min. longer 
than AMIL values; more precisely, +0.5 min. in the case of freezing fog tests, +1.3 min. in the 
case of snow tests, +1.6 min. in the case of freezing drizzle tests, +2.0 min. in the case of light 
freezing rain tests, and +1.8 min. in the case of rain on a cold-soaked wing tests. Since the 
failure times of Type I fluids are rather short as compared to those of Type II and IV fluids, the 
relatively short times has the effect to overvalue the increase, considering a 1 minute time 
variation could be within the acceptable experimental error of measurement. 

Among the five AET tests conducted, freezing fog tests are those with the longest endurance 
times and for which the variations are the smallest as expressed in percentage. It is not, 
therefore, a surprise that APS and AMIL freezing fog tests are those for which the agreement is 
closest of the five types of icing tests. On the basis of this agreement, the experimental error, 
which includes observer evaluation of percent of ice coverage at failure can be approximated at 
about 15%. 

Nevertheless, even if the measured time differences, 1 to 2 minutes, are shorter, it appears to be 
systematic. At this point, we can look attentively at the testing procedures used in the two 
facilities to identify and isolate the main factors which could be at the source of that 1 to 2 min 
differences. 

5.3 COMPARISON OF AMIL AND APS TESTING PROCEDURE 


Thirteen test parameters of AMIL and APS/NRC are compared in table 14. These factors vary 
from the environmental conditions, plate and panel alloy, surface roughness to fluid dilution, 
sample application, and plate cleaning. Among these factors, only four parameters, i.e., testing 
of unsheared samples, use of hard water for dilution, fluid temperature at application, and 
freezing points of ADF samples dilution can be considered identical. 


47 



TABLE 14. ANTI-ICING MATERIALS INTERNATIONAL LABORATORY VS APS/NRC 

TEST CONDITIONS 


Parameters 

AMIL 

APS/NRC 

Different 

Same 

1. Test temperatures 

+1°, -3°, -10° and -25°C 

+1°, -10° and -30°C 

y 


2. Relative humidity 

— 

>70% 

y 


3. Fluid shearing 

Not sheared 

Not sheared 


y 

4. Water used for dilution 

.Hard water 

AMS 1424B § 3.3.3.1 

Hard water 

AMS 1424B §3.3.3.1 


y 

5. Fluid application 
temperature 

20° ±5°C 

20° ±5°C 


y 

6. Volume of fluid applied 

500 mL 

1000 mL 



7. Delay between application 
and exposure to water spray 

5 minutes 

No delay 



8. Water for precipitation 

ASTMD 1193 

Type IV water 

Well water from NRC 
test site 

v' 


9. Test plate alloy 
and finish surface 

Aluminum alloy 

AMS 4037 (2024-T3) 

Ra = 0.1 to 0.2 pm 

Alclad Aluminum 
2024-T6 or 5052-H32 
polished standard roll 
mill finish 

y 


10. Plate size and 
working area 

300 x 500 x 3.2 mm 
with working area of 

300 x 500 mm 

300 x 500 x 3.2 mm 
with working area of 

250 x 450 mm 

y 


1 1. Test plate cleaning 

Plate cleaned with 
ethanol and hot water 

Scrape up and squeegee. 
Rinse with fluid and 
squeegee again 

y 


12. Icing rate measurement 

Calibration before test 
and calibration plates 
during test 

Weigh pans exposed to 
precipitation at different 
times during test 

y 


12. Dilution ratio (buffer) 

Tests-10°C OCTAFLO 
Tests-10°CADF 

Tests+1°C OCTAFLO 

Tests +1°C ADF 

45/55 (F.P. = -22°C) 
36/64 (F.P. = -20°C) 
30/70 (F.P. = -11°C) 

21/79 (F.P. = -9°C) 

42/58 (F.P. =-20°C) 
36/64 (F.P. = -20°C) 
27.5/72.5 (F.P. = -9°C) 
21.5/78.5 (F.P. =-9°C) 

y 

y 

y 

y 


The factors which are different can be divided into two groups. In the first group, there are 
second^' I f t SeC ° nd , a, J w * ich j ud § ed t0 ha ve nonsignificant or minor effects. These 

rZhne« 1 T h f °* idng generation ’ the P^te cleaning and 

Zt 't. m S 6 T gr °^’ th6re ^ thC factors J ud S ed relatively important and susceptible 
mn rlT 6 Cf CtS ' 656 important factors consist the plate working area, the 
flulfT an!I pH ° n ' * he pr f ipitatl0a rate measurement method, the sample dilution, the amount of 
fluid applied, and the 5-minute delay prior to the start of precipitation. As shown in table 14 
here are many parameters which can influence the anti-icing endurance time, probably with a 

flO ’ll and^r^TJ f faCt ,° rS Were studied and reported in previous investigations 

[ 0, 11 and 12], Unfortunately, most of these studies were done on Type E and Type IV fluids, 


48 




and therefore, there is limited test information regarding Type I fluids. The following discussion 

is devoted to the three' factors judged more important to explain and understand the observed 
differences. 

5.3.1 Sample Dilution . 

A sample selection procedure for SAE Type I fluids was approved as part of the AET testing 
procedure [8] and was recommended at the May 1999 Toronto Fluid Subcommittee meeting [6], 
In it, it states that the sample selected has to be diluted to a 10°C freezing point buffer. At 
AMIL, the buffer recommended by the manufacturer was used for each temperature. For the 
ADF fluid. Union Carbide recommends to dilute the fluid in 1% increments to always have a 
10°C buffer; this is how they recommend customers to use the fluid. For the OCTAFLO fluid. 
Octagon Process Inc. recommends to dilute the fluid in 5 % increments; again, this is how they 
recommend customers to use the fluid. At APS, they diluted both fluids (ADF and OCTAFLO) 
to a buffer of exactly 10°C; therefore, in the case of ADF, both laboratories tested the same 
dilutions. Whereas, in the case of OCTAFLO, the dilutions prepared by AMIL are somewhat 
more concentrated. For example, at -10°C the OCTAFLO dilution had a freezing point 2°C 
lower at AMIL than APS. This extra 2°C buffer may be, in part, responsible for the longer 
endurance times of OCTAFLO as compared to ADF. It is also important to do not forget they 

are two different fluids, made with different kinds of glycol. 

Table 15 shows that the differences between AMIL and APS endurance times are usually larger 
for ADF than OCTAFLO, with the exception of freezing fog and snow tests. This fact can 
possibly be attributed to the 2°C higher buffer used by AMIL for their dilutions. However, it 
does not explain the longer times observed by APS with respect to AMIL; on the contrary it 
should lead to longer endurance times for AMIL. 

TABLE 15. COMPARISON OF DIFFERENCES BETWEEN OCTAFLO AND ADF VALUES 


Condition 

Fluid 

Average Difference 
Between Laboratories 
(min) 

Difference OCTAFLO— 
Difference ADF 
(min) 

Freezing Fog 

OCTAFLO 

0.5 

0.9 

ADF 

-0.4 


Snow 

OCTAFLO 

1.4 

0.2 

ADF 

1.2 


Freezing Drizzle 

OCTAFLO 

1.1 

-1.0 

ADF 

2.1 


Light Freezing Rain 

OCTAFLO 

1.7 

-0.5 

ADF 

2.2 


Rain on a Cold- 
Soaked Wing 

OCTAFLO 

0.9 

-1.7 

ADF 

2.6 



49 




5.3.2 Amount of Fluid Applied and 5-Minute Delay . 

In the APS/NRC testing procedures, a volume in the order of 1L of fluid is applied on the test 

aw ^ 6 U 1S eX ?~ Sed , t0 P reci P ltat,on durin g application. In the procedures used by 
AMIL, the amount of flmd applied is limited to 500 mL, and the fluid-coated plate is exposed to 
precipitation only 5 minutes after fluid application. Five hundred mL of fluid is equivalent to a 
fluid thickness of 3 mm. The 5-minute wait allows the fluid to settle, coat the plate well and 
further reduce in thickness for a representative surface. One L of fluid is equivalent to a fluid 
thickness of 7 mm. Because tjiere was no wait time at APS, implies that full thickness of fluid 

vZ wSt /^p°i the T teSt ‘ The 5 ~ minUte Wait iS a practice commonI y used in standard tests 
like WSET and HHET. It contributes to obtain at the start of the test a settled fluid coating 

which is not disturbed by the falling precipitation. Moreover, previous AMIL experiments have 
shown that the thickness of the fluid film left on the plate after a 5-minute settling time was 
independent of the volume of fluid applied. 

hi order to investigate whether these two specific factors had an effect, two AET tests were 
performed at -3°C with OCTAFLO 35/65 (FP of -15°C) samples: a freezing drizzle test at 5 

g h™t, h 1 UUr hl ? h ^ o ^ 1S app,ied under lcin g’ and a light freezing rain test at 13 g/dm 2 h in 
which 1000 mL of fluid is applied on the panels and allowed to settle for 5 minutes Freezing 
drizzle temperature recordings (ZLA001 and ZLA001A), illustrating the effect of the 5-minute 

T 7 p y lnn?A?T n m f ! gUrC 29; whereas those of hg ht freezing rain tests (LZRA001 and 
figme 30 ’ mS ° f a 500 ^ additi0nal vo,ume of a PP lied fl uid, are depicted in 


Air and plate temperatures recorded in the four tests present all the same pattern. Air 
temperature is maintained exactly at the targeted value of -3.0° ±0.1 °C during the course of the 
test; whereas plate temperature, as measured by a RTD sensor located on the underside of the 
est plate, varies considerably once the fluid is applied at room temperature. Indeed plate 
temperature, which is at -3.0° ±0.5°C just before fluid application, rises sharply during fluid 
application to reach a maximum value and decrease thereafter. During this period, the fluid film 
is cooled by air convection and impacting supercooled water droplets. At the time of the failure 
call, the plate temperature reaches a minimum value which increases thereafter to the end of the 
test. This temperature rise is due to the latent heat released during the solidification of water. 

The effects of both factors are summarized in table 16. It can be noted that the time at the 
minimum temperature of the four recordings corresponds to the measured endurance time values. 


50 


TEST: ZLA001 


DATE: 99/09/22 



(a) Test with 5-minute delay 


















































TABLE 16. FIVE-MINUTE DELAY AND VOLUME OF APPLIED FLUID EFFECTS 


Waiting Time 
Volume of Fluid 

Test Label 

Plate 

Tair 

Tmax 

Tstart 

Tfailure 

AETfailure time 

Average 

Difference 

% 

Difference 

5 minutes 

500 mL 

ZLA001 

A 

B 

-3.0 

-3.0 

8.3 

10.8 

2.5 

2.5 

-2.5 

-2.5 

1 lm50s 
12m40s 

2m45s 

+18 

0 minutes 

500 mL 

ZLA001A 

A 

B 

-3.0 

-3.0 

7.3 

10.2 

7.3 

10.2 

-2.0 

-2.0 

14m50s 

15ml0s 



5 minutes 

500jnL 

LZRA001 

A 

B 

-3.0 

-3.0 

8.8 

10.0 

2.0 

2.5 

-0.9 

-1.8 

7m00s 

7m00s 

0ml5s 

+3 

5 minutes 

1000 mL 

LZRA001A 

A 

B 

-3.0 

-3.0 

13.8 

14.5 

5.0 

6.0 

-1.3 

-0.5 

7m00s 

7m30s 




5.3.2.1 Five-Minute Delay Effect . 

In the test conditions for freezing drizzle at -3.0°C, using a 5-minute settling time shortens the 
measured endurance times by 2 to 3 minutes. This decrease can be understood by the differences 
in the cooling and draining of the fluid coating which differs whether or not it is exposed to 
icing. According to table 16, the test without the 5-minute delay (test ZL001A) began when the 
thickness of the fluid film and the temperature of fluid-coated plate are both at their maximum. 
The fluid film at the start of the test is then about 6°C higher and thicker without the 5-minute 
delay than with it. The fluid is cooled by air convection and by impacting supercooled droplets. 
Because of the higher fluid temperature and the thicker fluid film, more water is required, i.e., a 
longer time, for the fluid to be cooled and diluted to a freezing point at -2.0°C. For the test with 
the 5-minute delay, the fluid thickness and temperature are smaller (by 6°C for the temperature) 
than for the test without the 5-mmute delay. As a result, it takes less time, i.e., a smaller amount 
of supercooled water, for the fluid film to be diluted and cooled by air convection to a freezing 
temperature of -2.5°C. On the basis of these test results, the delay of 5 minutes used by A MTT. in 
AET procedures appears to have contributed to shortening the measured failure times, in an 
order of magnitude of 2 to 3 minutes with respect to the APS/NRC results. 

5-3.2.2 One Thousand mL vs Five Hundred mL Applied Fluid Effect . 

In the test conditions of light freezing rain at -3.0° (table 16), the fact that 1000 mL instead of 
500 mL was applied, with a 5-minute delay, appears not to have any significant effect on the 
measured failure times. The fact that no significant effect is observed, even if the fluid 
temperature at the start of the test obtained with 1000 mL of applied fluid is by about 3°C higher 
than that observed with an applied volume of 500 mL, seems to indicate that the factor which is 
important will be the amount of fluid on the plate at the start of the test rather than the air cooling 
by convection. According to results of the preceding test, it can be assumed that, if testing was 
done without the 5-minute delay, the increase volume would likely have the effect of increasing 
the time to fail, mainly because of the greater amount of fluid at the start of the test, giving then a 
much thicker coating. However, no tests were performed to investigate this factor in AET tests, 
and therefore, the determination of the full importance of the volume of the fluid applied would 
required additional tests conducted in conjunction with the 5-minute delay. 


53 




5 3.3 Comparison of En d urance Time Under Similar Conditions 

It is interesting to compare values of the endurance times measured under similar conditions 
for instance, those obtained at 13 g/dm*h icing intensity in light freezing rain and freezing drizzle 
tests. Endurance time values measured in the two tests at -3° and -10°C (printed in bold in 
tables 10 and 11 ) are listed in table 17. b ° ld in 

Table 17 shows that endurance times as measured in freezing drizzle and light freezing rain tests 
are similar, within the experimental error of measurement. 

TABLE 17. LIGHT FREEZING RAIN VS FREEZING DRIZZLE AT 13 «/dnFh 


-3°C LZR -3°C ZL -10°CLZR 


- 1 0°C ZL 


OCTAFLO 

ADF 


In light freezing rain tests, failure appears as an ice front, as with freezing fog tests. Moreover 
as freezing fog and freezing drizzle tests, ice deposits formed in light freezing rain tests grows 
from supercooled droplets freezing the plate. The difference between freezing drizzle and light 
freezing ram is the water droplet size, the latter involves drops of 1000 4 m, 5 to 10 times larger 
than those of freezing drizzle (MVD of 250 |im). Table 17 shows that the size of droplets in The 

endurance^imes 65 “ thiS Study ’ d ° n0t ^gnificantly affect the measured anti-icing 

6. CONCLUSIONS 

On the basis of the results of AET tests performed on two certified SAE Type I aircraft deicing 
fluids, the following conclusions can be drawn: ® 

• The results obtained demonstrate the feasibility and the practicability of performing all 
six testing procedures within the prescribed accuracy and repeatability. Indeed 
environmental parameters in AET calibration and fluid tests were within the target values 
with variations within the prescribed allowable drifts. 

Moreover, AET results obtained under the six environmental conditions showed an 
expected inverse relationship between endurance times and precipitation rate, the shortest 
and longest failure times being obtained respectively under the highest and lowest icing 


Time variations of 1 minute or less were generally observed between the endurance times 

flniH n^th 1h t h f C Same Samp e ‘ These variations do not a PP ear to be dependent on the 
fluid nor the testing temperature; the 1 -minute variation is considered to be within the 
experimental error of measurement. 


54 



• When the AET results were compared with HOT test data obtained by APS in the NRC 
facility using a somewhat different testing method, AMIL failure times were 0.1 to 4 
minutes shorter than APS’s measured values, with the exception of freezing fog tests. 
Indeed, with freezing fog tests, APS’s measured endurance times are very comparable to 
AMIL’s values; the time difference range being from -16% to +9%, with a ±10% 
variation. With the four other types of tests, HOT values measured by APS are found to 
be on average 30% (about 1 to 2 minutes in most cases) longer than those obtained by 
AMIL. 

• Examination of APS and AMIL testing procedures allows for the identification of 13 
differences, among which the following six can be judged more significant: 

the plate working area 
the sample dilution 
the failure call 

the precipitation rate measurement method 

the amount of fluid applied 

the 5-minute delay prior to the start of precipitation 

The last two factors may partially explain the lower failure times observed in the AMIL facility. 
This interpretation is supported by the results obtained by two tests in which effects of these two 
parameters were compared. 

In the process of reducing the number of parameters which are not the same in the APS/NRC and 
AMIL procedures, it is recommended that each parameter for which a difference is identified in 
this report shall be analyzed and discussed. In order to realize this, real conditions and actual 
practices of using fluids in airports during deicing and anti-icing operations should be taken into 
consideration, as well as the feasibility of performing reproducible tests in a laboratory. 

7. REFERENCES . 

1. SAE Aerospace, Unconfirmed Minutes, G-12 Holdover Time Subcommittee Meeting, 
May 17-18, 1999, Toronto, Ontario. 

2. SAE Aerospace, Workgroup on Laboratory Methods to Derive Holdover Time 
Guidelines, November 20-21, 1997, Dorval, Quebec, 21 pages. 

3. SAE Aerospace, Workgroup on Laboratory Methods for Experimental Endurance Time 
Testing Meeting, July 30, 1999, Montreal, Quebec. 

4. SAE Aerospace, Workgroup on Laboratory Methods for Experimental Endurance Time 
Testing Meeting, October 6, 1999, Chicoutimi, Quebec. 

5. Leroux, Jacques, Aerospace Standard 5485, ''Draft Endurance Time Tests for Aircraft 
Deicing/Anti-Icing Fluid SAE Type I, n, m, and IV,” October 17, 1999, 50 pages. 


55 



6. SAE Aerospace, Unconfirmed Minutes, G-12 Holdover Time Subcommittee Meetin- 
May 21, 1996, Zurich, Switzerland. 

7. SAE Aerospace, Unconfirmed Minutes, G-12 Fluids Subcommittee Meeting May 18-19 

1999, Toronto, Ontario. & y ’ 

8. SAE Aerospace, Sample Selection Procedure for Endurance Time Testing for Fluids 
Meeting the Requirements of SAE AMS 1424, May 27, 1999, 3 pages. 

9. SAE Aerospace, Workgroup on Laboratory Methods for Experimental Endurance Time 
Testing Meeting, March 18-19, 1999, Montreal, Quebec. 

10. Bemardin, S„ Dubuisson, C., and Laforte, J.L., “Aircraft Ground De/Anti-Icing Fluid 
Holdover Time Laboratory Test Program: Freezing Drizzle and Freezing Rain.” report 
prepared for Transport Canada, TP13036E, May 1997, 60 pages. 

11. Bemardin, S., Dubuisson, C„ and Laforte, J.L., “Development of Laboratory Test 
Procedures to Replace Field Anti-Icing Fluid Tests,” report prepared for Transport 
Canada, TP3141E, November 1997, 110 pages. 

12. Bemardin, S., Beisswenger, A., and Laforte, J.L., “Holdover Time Field Tests,” report 
prepared for Transport Canada, TP 13590E, May 1999, 57 pages. 

8. ADDITIONAL INFORMATION 

1 Dawson, P., D' Avirro, J., and Potter, R.V., “Validation of a Methodology for Simulating 

a Cold-Soaked Wing,” report prepared for Transport Canada, TP12899E. October 1996 
92 pages. 

2: SAE Aerospace, Unconfirmed Minutes, G-12 Fluids Subcommittee Meeting, May 12-13, 

1998, Vienna, Austria. 

3. SAE Aerospace, Unconfirmed Minutes, G-12 Holdover Time Subcommittee Meeting, 
May 11-12, 1998, Vienna, Austria. 

4. SAE Aerospace, Unconfirmed Minutes, G-12 Fluids Subcommittee Meeting, October 22- 
23, 1997, Atlanta, Georgia. 

5. SAE Aerospace Unconfirmed Minutes, G-12 Holdover Time Subcommittee Meeting, 
June 10, 1997, Pittsburgh, Pennsylvania. 

6. SAE Standard AMS 1424B - Deicing/Anti-Icing Fluid, Aircraft, SAE Type I. 


56