EFFECTS ON J79 ENGINE COMBUSTION SYSTE.9. J79 Engine Combustor, Pictorial View. 25 10. J79 Engine...

AFAPL-TH-79-2015CEEDO-TR-79.-06 v L . -

EVALUATION OF FUEL CHARACTER EFFECTS

ON J79 ENGINE COMBUSTION SYSTE.7.-1

General Electric CompanyAdvanced Engineering & Technology Programs "

00 I Neumann WayCincinnati, Ohio 45215

JUNE 1979 -

FINAL REPORT June 1977 - August i978 ]

Approved for public. release; distribution unlimited

21D DC.

Air Force Aero-Propulsion Laboratory F-.. ljff77LA- Air Force Wright Aeronautical Laboratories 1

Air Force Systems Command UUT,... M -1i 1Wright-Patterson Air Force Base, Ohio 45433 U

U ,

-2 .A4b1 2 Awl. ib

REPRODUCTION QUALITY NOTICE

This document is the best quality available. The copy furnishedto DTIC contained pages that may have the following qualityproblems:

"* Pages smaller or larger than normal.

"* Pages with backgrouna color or light colored printing.

"* Pages with small type or ooor pnnting; and or

"* Pages with continuous tone material or colorphotographs.

Due to various output media available these conditions may ormay not cause poor legibility in the microfiche or hardcopy outputyou receive.

D7 If this block is checKea. the copy furnished to DTICcontained pages with color printing, that when reproduced inBlack and White, may change detail of the original copy.

NOTICE

When Government drawings, specifications, -r other data are used for any pur-pose other than in connection with a definitely reltted Government procurementopexation, the United States Government thereby incurv no responsibility nor anyobligation whatsoever; and the fact that the government may have formulated,furnished, or- in any way supplied the said drawings, specifications, or otherdata, is not to be regarded by implication or otherwise as in any manner licen-sing the holder or any other person or corporation, or conveying any rights orpermission to manufacture, use, or sell any patented invention that a&y in anyway be related thereto.

This report has been reviewed by the Information Office COT) and is releasableto the National Tc.hnical Information Service (NTIS). At NTIS, it will be avail-able to the general public, including foreign nations.

This technical report has been reviewed and is approved for publication.

THOMAS A. JACKSOQ1 ARTHUR V. CHURCHILLFucls Branch Chief, Fuels Bxcanuh A

Fuels and Lubrication Division Fuels and Lubrication Division

FOR THL COMMANDER

BLACKWELL C. DUNNAM, ChiefFuels and Lubrication Division7

"If your address has changed, if you wish to be removed from our mailing list,or if the addressee is no longer employed by your organization please notify

AF•APXS/SFF ,W-PAFB, OH 45433 to help us maintain a current mailing list".

Copies of this report should net be returned unless return is required by se-curity considerations, contractual obligations, or n3tice on a specific document.

AIR FORCE/i67SO/14 Nonmb., 297g-S3o

Unclassified

SI.CUqIrI," CLASSilCAItoN or THIS PAGE (tW-'• Veale Fnterud

V REPORT DoCI~iAENTAT;ON PAGE READ INSTRUCTIONS6- M POR DOC MEN A M N ACEBEIORE CUMPLE.TING FORM2. GOUT ACCESSION O•O 3. RECIPIENT'S CATALOG NUMBERL/ CE_,DO •R-79-0$6

.4. TITL.E (ond $ubt -•. TYPE OF RUPORT A PERIOD COVEREDEvaluation of Fuel Character Effects on J79 Final Technical J.eeit.Engine Combustion System p •J3un04J77-3l Aug'7fln78j

(C.C. ýGleason, T.L. Oller, M.W. Shayeson•pow"--i D.W. P1ahr / ._"F336,15- 77-C- 2P42 -

•79 PERFORM:4N, ORGANIZATION NAME- AND ADDRESS 10. PRO'GRAM EL FMENT, PROJECT. TASKAREA & WORI( UNIT NUMBERS

General Electric Company - 62203F

Ai f E Group 3Cincinnati, Ohio 45215 1 1•80-1••,----

11. CONTROLLING OFFICE NAME AND ADDRESS 11 REPOR(T OATE

Air Forcr, Aero Propulsion Laboratory (APAPL/SF JIurnc 1979Air Force Wright Aeronautical Laboratories 13, NUMBER OF PACES

Wri ht-Pattersn AMB. Ohio 45433 19714. MONITORING AGENC~Y NAM Q ADORESS(II diffaeren fItm Controlling v.jrcw' IS. SECURITY CLASS. (o. gle t¢ orQ.)

Unclassified154. OSCLASS FICATIONIOOWNGBAIN"SCHEDULE

1F. DISTRIBLTION STATEMENT (of thi. Report)

Approved for putc release; aistribution unlimiTed.

V7 DISTRIBUTION STATEMENT (of the ab•tract entered In block 20, LI diflerent frevtR, Ppt')

I8 SI.PPLEMENTARY NOTES

Partial funding and technical support in the area of the measurement andanalysis of gaseous emissions and smoke datz were provided by the Environ-mental Sciences Branch of the Enviropics Divisiio. in the Research andDevelopment Directorate of HQ Air Force Engineering and Services Center.

IS. KEYr WORDS (CotI.u. or. . -e.es .si•de .et....,y and Idernely by block .nbu.w. (I'-AFESC/RDVC)

Fuels J79 Engine CombustorAlternate FuelsGas Turbine CombustionExhaust Emissions

21 ABSTRACT (ContInnr on oeld.re e ie If n .e sry eid Identity by block number)Results of a program to determine the eLffects of broad variatio .s in fuelproperties on the performance, emissions, and durability of the J79-17Aturbojet engine combustion systemn are presented. Combustor tests conductedat engine id.e,takeoff, subsonic cruise, supersonic dash, cold day groundstart, and altitude relight operating conditions with 13 diffetent fuels aredescribed. The test fuels covered a range of hydrogen contents (12.0 to14.5 percent), aromatic type (monocyclic and bicyclic), initial boilingpoint (285 to 393 K)t final tuiling point (552 to 679 K) and vic-e..i tv r--

DD MJAN , 1473 EDITION OF I Nov BS OBSOLETE UnclassifiedSECURITY CLASSIFICATION OF THIS P-GC" (*hen Dfl. E, eE d)

¼t

-Uncl assified _'_-_SECURITY CLASSIFICATION OF THIS PAGE(Whn ,DA n.r.•• 0'd)

BLO0CK 20 (ConL'd)

-- - >0-83 to 3.25 MM2 /s at 300 K). ,

At high power operating coaditions, fuel hydrogen con Leilt was fouid tobe a very significant fuel property with respect to liner temperature,flame radiation, smoke, and Ox' emission levels. Carbon monoxide and HCemissions were very low at these conditions with all of the fucs.

SAt engine idle operating conditions, CO, RIC,"-ai d NO, emission levels

were found to be independent of fuel Lydrogen content,"l-ut a small effect

of fuel volatility and/or viscos-a-C #as found.

At cold day ground start conditions (to 329 K) lightoff was obtainedwith all fuels, but the required fuel-air ratio increased with the moreviscous fuels.

At altitude conditions, the current engine relight limits with JP-4/JP-5 fuel were essentially met or wxceeded with all of the JP-4 or JP-8 "based fuel blends. However, a very significant reduction in altituderelight capability was found when a No. 2 diesel fuel was tested.

Combustor liner life analyses, based on the test data, yielded rVlativelife predictions of 1.00, 0.78, 0.52, and 3D.35 for fuel hydrogen contentsof 14.5, 14.0, i3.o, and 12.0 percent, respectively. Turbine life waspredicted to be unaffected by any of the fuc2s tested.

Fuel nozzle fouling tests wtore also conducted with the 13 test fuels.No significant -ftects were noted. This was ,.ot surprising, in vi"- of theservice history of the J79 engine and the shý,rt (5-hour) test runs made.

UnclassifiedSECURITY C, ASSIPICATION OF THIS PAOE(Wh-n Dana Ente.edj

PREFACE

Thib final report is submitted by the General Electric Company, Aircraft

Engine Group, Evendale, Ohio. The work was conducted under Contract

No. F33615-

77-C-2042. Program sponsorship and guidance were provided b., the Air Force

Acro

Propulsion Laboratory (AFAPL), Air Force Wright Aero-autical Laboratories,

Air

Force Systems Command, Wrlght-Patterson Air Force Bane, Ohio under Project

3048.

Task 05, and Work Unit 81. Thomas A. Jackson was the government project

engineer.

Supplemental funding and technical guidance were provided in the

area of

gaseous emissions and smoke measurement and analysis by the Environmental

Scienceb

Branch of the Environics Division in the Research and DeveiopmentL

Directorate of

) q Air Force Engineering and Services Center, located at Tyndall Air Force Base,

Florida. This organization has been formerly referred to as CEEDO

or the Civil

Engineering Center.

Test fuel analysis was provided by AFAPL, the Monsanto Research

Laboratory

(under contract to AFAPL), and the Air Force Logistics Command

Aerospace ruels

La' oratory (SFQLA)- The cooperation of these organizations is appreciated.

AcOeejsicn F'or

.d

ior

This document containsblank pages that werenot filmed

TABLE OF CONTENTS

Section Page

INTRODUCTION 1

II 3

">< III TEST FUEL DESCRIPTION 5

A. General Description 5

B. Physical and Chemical Properties 5

C. Thermal Stability Characteristics 10

D. Computed Combustion Parameters 15

IV J79 ENGINE COMBUSTION SYSTEM DESCRIPTION 20

A. Overall Engine Description 20

B. Combustion System Description 20

C. Combustor Operating Conditions 26

V APPARATUS AND PROCEDURES 32

A. Performance/Emission/Durability Tests 32

1. High Pressure Test Rig Description 352. High Pressure Test Rig Instrumentation 353. High Pressure Test Procedure 52

B. Cold-Day Ground Start/Altitude Relight Tests 52

1. Low Pressure Test Rig Description 522. Cold-Day Ground Start Procedure 543. Altitude Relight Test Procedure 54

C. Fuel Nozzle Fouling Tests 57

D. Test Fuel Handling Procedures 59

E. Data Analysis Procedures 62

1. Fuel Property Correlation Procedures 622. Combustor Life Prediction Procedures 63"J. Turbine Life Prcdiction Proccdures

V

I-mi 4111 1 T--

TABLE OF CONTENTS (Continued)

"Section Page

Vi RESULTS AND DISCUSSION 70

A. Experimental Test Results 70

1. CO and HC Emissions 701 2. NOx Emissions 73

"3. Smoke Eaissions 804. Carbon Deposition and Emission 85. Liner Temperature and Flame Radiation 886. Combustor Exit Profile and Pattern Factor 997. Cold-Day Ground Starting and Idle Stability 998. Altitude Relight 1089. Fuel Nozzle Fouling 114

B. Engine System Life Predictions 117

1. Combustor Life Predictions 1172. Turbine Life Predictions 117

C. Assessment of Results 117

VIl CONCLUSIONS AND RECOMMENDATIONS 120

A. Conclusions 120

B. Recolmendations 120

Appendix A High Pressure Test Data 121

Appendix B Carbon Deposition Data 136

Appendix C Low Pressure Test Data 163

Appendix D Fuel Nozzle Fouling Test Data 1i5

Appendix E Smoke Data Calculation 179

Appendix F No=enclaturc 181

References 183

vi

-_III__III__lii__II _ i --- . -. -.. . -...... .. -

LIST OF ILLUSTRATIONS

Figure_ Page

1. Comparison of Fuel Aromatic Content by Two TestMethods. 11

2. Variation of Fuel Aromatic Content with HydrogenContent. 12

3. Comparison of Gas Chromatographic Simulated Dis-

tillation Characteristics of the Test Fuel&.-.- 13

4. Comparison of Fuel Recovery Points by Distillationand Gas Chromatography. 14

5. General Electric J79 Turbojet Engine. 21

6. J79 Engine Combustion System, Flowpath. 22

7. J79 Engine Combustion System, Exploded PictorialView. 23

8. J79 Engine Combustion System, Partial AssemblyPhoto. 24

9. J79 Engine Combustor, Pictorial View. 25

10. J79 Engine Dual Orifice Fuel Nozzle. 27

11. J79 Engine Fuel Nozzle Flow Characteristics. 28

12. J79 Engine Altitude Windmilling/Relight RequirementMap. 31

13. Inlet Quarter View of High Pressure J79 CombustorTest Rig. 36

14. Combustor Installation in High Pressure J79 Test Rig. 37

15. Side View of High Pressure J79 Combustor Test Rig. 38

16. Combustor Exit Instrumentation Rake. 39

17. Tip Details, Combustor Exit Instrumentation Rake. 40

18. High Pressure Test Rig Exit Instrumentation. 42

19. General Electric Smoke Measurement Conisole. 43

vii

LIST OF ILLUSTRATIONS (Continued)

Fligure page

20. General Electric Emissions Measurement Console(CAROL IV). 44

21. Combustor Liner Temperature Measurement Locations. 46

22. Combustor Liner Instrumentation at 60 DegreesCWALF. 47

23. Combustor Liner instrumentacion at 164 .an 196Degrees CWALF. 48

24. Combustor Liner Instrumentation at 300 DegreesCWALF. 49

25. Combustor Dome Instrumentation. 50

26-.... Optical Pyrometer Setup in High Pressure Test Rig. 51

27. Low Pressure J79 Combustor Test Rig. 55

28. Fuel Nozzle Fouling Test Setup. 58

29. Effect of Temperature Gradient on Crack Propaga-tion Rate. 66

30. Effect of Fuel Hydrogen Content on Aft LinerTemperature. 67

31. Turbine Nozzle Diaphragm Flame View Factor. 69

32. Effect of Operating Conditions on CO EmissionLevels.

71

33. Effect of Fuel Hydrogen Content on CO EmisssionLevels. 72

34. Effect of Fuel Atomization and Volatility on IdleCO Emission Levels. 74

35. Variation of HC Emission Levels with CO EmissionLevels. 75

36. Effect of Fuel Hydrogen Content on HC EmissionLevels. 77

viii


Figure Page

37. Effect of Fuel Atomization and Volatility on IdleHC Emission Levels. 78

38. Effect of Operating Conditions on NOx EmissionLevels. 79

39. Effect of Fuel Hydiogen Content on NO, EmissionLevels. 82

40. Effect of Flame Temperature on NO, Emission Levels. 83

41. Effect of Operating Conditions on Rig Smoke EmissionLevels. 84

42. Effect of Fuel Hydrogen Content on Smoke EmissionLevels. 86

43. Effect of Fuel Hydrogen Content on Soot EmissionIndex. 87

44. Combustor Posttest Condition with Relatively LowCarbon Deposition (Fuel No. 2). 90

45. Combustor Posttest Condition with Relatively HighCarbon Deposition (Fuel No. 4). 91

46. Effect of Fuel Hydrogen Content on Combustor CarbonDeposition. 92

47. Effect of Fuel Hydrogen Content on Large CarbonParticle Emission. 93

48. Typical Inner Liner Temperature Distributions. 94

49. Effect of Operating Conditions on Inner Liner Tem-perature Rise. 95

50. Effect of Fuel Hydrogen Content on Inner Liner Tear-perature Rise. 97

51. Effect of Fuel Hydrogen Content on Liner TemperatureParameter at Cruise Operating Conditions. 98

52. Effect of Combustor Operating Conditions on FlameRadiation. 100

ix


Figure Page

53. Effect of Fuel Hydrogen Content on Flame Radiation. 102

54. Typical Combustor Exit Temperature Profiles. 103

55. Effect of Combustor Operating Conditions on PatternFactor. 104

56. Effect of Fuel Hydrogen Content on Pattern Factor. 106

57. Typical Ground Start Characteristics. 107

58. Effect of Fuel Atomization and Volatility on ColdDay Ground Start. 110

59. Effect of Fuel Thermal Stabi'lty Rating on FuelNozzle Fouling Tendency. 116

60. Predicted Effect of Fiel Hydrogen Content on Com-

buator Life. 118

B-1. Posttest Photograph of Liner After Test of Fuel 1. 137

B-2. Posttest Photograph of Liner After Test of Fuel 1R. 138


3-4. Posttest Photograph of Liner After Test of Fuel 3. 140

B-5. Poattest Photograph of Liner After Test of Fuel 4. 141









LIST OF ILLUSTRATIONS (Concluded)

Filture ?!Le

B-14. Posttest Photograph of Fuel Nozzle After Test ofFuel 1. 150

B-15. Posttest Photograph of Fuel Nozzle After Test ofFuel 1R. 151


B-17. Poattest Photograph of Fuel Nozzle After Test ofFuel 3. 153

3-18. Posttest Photograph of Fuel Nozzle After Test ofFuel 4. 154

9-19. Posttest Photograph of Fuel Nozzle After Test ofFuel 5. 155



B-22. Posttest Photograph or ruci Nozzle After test ofFuel 9. 158





E-l. Experimental Relationship Between Smoke Number andExhaust Gas Carbon Concentration. 180

xi

LIST OF TABLES

Table page

1. Test Fuel Chemical and Physical Properties. 6

2. Test Fuel Hydrocarbon Type Analyses (Mess Spectroscopy#ASMh Method D2789-71). 7

3. Test ?uel Gas Chromatographic Simulated Distillation(ASTh Method D2887). 8

4. Test Fuel Conventional Inspection Date. 9

5. Fuel Sample Thermal Stability Test Results(ASTM Method D3241). 16

6. Estimated Thermal Stability Ratings of Test Fuels(ASTM Method D3241). 17

7. Test Fuel Combustion Parsmeters. 18

8. J79-17A Engine Combustor Operating Conditions. 30

9. J79-17A Engine Combustor Test Parts List. 33

10. Test Combustr PFlow Calibration Results. 34

11. Summary of Measured and Calculated Combustor Parametersin High Pressure Combustor Tests. 41

12. High Pressure Test-Point Schedule. 53

13. Low Pressure Test-Point Schedule. 56

14. Fuel VerIfiction Analyses. 61

15. Summary of CO Emission Test Results. 64

16. Summary of HC Emission Test Results. 76

17. Summary of NO Emission Test Results. 81

18. Summary of Smoke Emission Test Results. 85

19. Summary of Carhon Deposition and Eml:ision Test Results. 89

20. Summary of Liner Temperature Test Results. 96

xil

LIST OF TABLES (Conctinud)

Table Page

21. Suamwry of Radiant Heat Flux Test Results. 101

22. Summary of Pattern Factor and Radial Profile Test Results. 105

23. umary of Ground Start Test Results. 109

24. Sumary of Idle Stability Test Results. 111

25. Summary of Altitude Relight Test Results. 112

26. Suaumary of Fuel Nozzle Fouling Test Results. 115

A-1. Basic High Pressure Test Data 122

A-2. Supplementary High Pressure Test Data. 124

A-3. CO Emission Test Data Correlation. 126

A-4. Hydrocarbon Emission Test Data Correlation. 127

A-5. NOx Emission Test Data Correlation. 128

A-6. Smoke Emission Test Data Correlation. 129

A-7. Detailed Liner Temperature Data. 130

A-$. Flame Radiation Data Correlation. 132

A-9. Pattern Factor Correlation. 133

A-10. Combustor Exit Temperature Profile Data. 134

C-1. Altitude Relight Test Restu!ts, Fuel Number 1. 164

C-2. Altitude Relight Test Results, Fuel Numbers 2 and 3. 165





C-7. Altitude Relight Test Results. Fuel Numbers 12 and 13. 170

xii

LIST OF TABLES (Concluded)

Table .,_"C-8. Ground Start Test Results, Fuels 1 Through 3. 171

(C-9. Ground Start Test Rtesults, Fuels 4 Through 7. 172

C-1O. Ground Start Test Results, Fuels 8 Through 11. 173I.

C-11. Ground Start Test Results, Fuels 12 and 13. 174

D-1. Fuel Nozzle Fouling Test Results with BlockedSecondary Orifice Valve. 176

D-2. Fuel Nozzle Fouling Test Aesults with Standard DualOrifice Fuel Nozzle. 177

D-3. Normlized Fuel Nozzle Fouling Test Results withStandard Dual Orifice Fuel Nozzle. 178

rl v

SECTION I

INTRODUCTION

For more than 25 years, the primary fuel for USArF gas-turbine-poweredaircraft has been JP-4, a wide-cut distillate with excellent combustioncharacteristics and low-temperature capability. Typically, its heating valuehas been over 43.5 MJ1kg (18,700 BTU/lb), its freezing point below 219 K (-65' F),and its aromatic content quite low, around 11 percent by volume. A primeconsideration in the definition of JP-4 was that during wartime, a large per-centage of domestic crude oil could be converted into this product with minimumdelay and minimum impact on other major users of petroleum products.

Convc:sion from high volatility JP-4 to lower volatility JP-8, which issimilar to commercial Jet A-1, as the primary USAF aircraft turbine fuel hasbeen under consideration since 1978. The strong motives for the change are NATOstandardization and reduced combat vulnerability.

Domestic crude oil production peaked in 1971 and has been steadily decliningsince that time, while demand has continued to increase. Thus, particularlyvisce 197', the cost and availability of high-grade aircraft turbine fuels havedrastically changed. These considerations have 6polrred efforts to determine theextent to which current USAF fuel specifications can be broadened to increase theyield from available petroleum crudes and, ultimately, to permit production fromother sources such as coal, oil shale, and tar sands.

As a result of the current and projected fuel situation, the USAF hasestablished an aviation turbine fuel technology program to identify JP-4 and/orJP-j fuel specifications which:

1) Allow usage of key world-wide resources to assure availability.

2) Minimize the total coLt o1 aircra.t system re-ration.

3) Avoid sacrifices of engine performance, flight safety, orenvironmental impact.

Engine, airframe, logistic and fuel processing data are being acquired to establishthese specifications. This report contributes to the needed data base by

describing zhe effects of fuel property variations on the General Electric J79-17Aengine main combustion system with respect to perfoimance, exhaust emissions,and durability. Similar programs, based on the General Electric F101 engine andthe Detroit Diesel Allison TF41 and High Mach engines, are also being conducted.Collectively, these programs will provide representative data for the engineclasses that are expected to be in substantial use by the USAF in the 1980's.

This report summarizes the results of a 13-month, three-task program whichwas conducted to clearly identify which fuel propertiea are important to J79-17Aengine combustor operation and quantitatively relate fuel property variationsto combustor performance, emission characteristics and durability characteristics.

-, - _ _ ._ _ .. . .. . .- -

Thirteen text fuels provided by the USAF were utilized. Descriptions andproperties of these fuels are presented in Section III. In Task I of the program,test planning and preparations were made, based on use of the J79 engine combustionsystem components and operating characteristics described in Section IV, and onthe three test rigs and procedures described in Section V. In Task I1 of theprogram, 46 tests (14 high pressure/temperature combustor performance/emissions/durability tests, 14 low pressure/temperature combustor cold-day ground start/altitude relight tests, and 18 high tmnperature fuel nozzle fouling tests) wereconducted. The" are suarized in Section VI-A. In Task III of the program

these test data wre analyzed to establish the fuel property correlations alsopresented in Section VI-A and to establish the engine system life predictions

S! presented in Section V7-3.

'I

!2

SECTION II

SUMMARY5A

The purpose of this program was to determine, by combustor rig tests and dataanalyses, the effects of fuel property variations on the performance, exhaustemission, and durability characteristics of the General Electric J79-17A turbojetengine main combustion system. Thirteen refined and blended fuels whichincorporated systematic variations in hydrogen content (12.0 to 14.5 wci'htpercent), aromatic type (monocyclic or bicyclic), initial boiling point (285 to393 K by gas chromatograph), final boiling point (552 to 679 K, also by gaschromatograph), and viscosity (0.83 to 3.25 mm2 /s at 300 K) were evaluated in:(a) 14 high pressure/temperature combustor performance/emissions/durabilitytests; (b) 14 low pressure/temperature combustor cold-day ground start altitudereltght tests; and, (c) 18 high temperature fuel nozzle fouling tests.

I

At high engine power operating conditions (takeoff, subsonic cruise, super-sonic dash) fuel hydrogen content was found to be a very significant fuel propertywith respect to liner temperature, flame radiation, smoke and NOx emission levels,and carbon deposition. Each of these parameters decreased with Increah.a.g fuet

hydr nontent, b no dfscerrnle effect of any of the other fuel propertieswas found. Carbon monoxide and HU emissions were very low at each of theseoperating conditions with all of the fuels. Combustor exit temperature profileand pattern factor were essentialli the same with all fuels, but a constantdifference between the two assemblies used was noted.

At engine idle operating conditions, the same strong effects of fuelhydrogen content on smoke level, liner temperature, and flame radiation wereevident. Carbon monoxide, HC, and NOx levels were found to be independent offuel hydrogen content, but a small effect of fuel volatility (as indicated by 10percent recovery temperature) on CO and hC levels was found.

At cold-day ground start conditions (to 329 K) lightoff was obtained with allfuels, but the required fuel-air ratio increased with the more viscous fuels,primarily as a result of the associated increases in relative fuel spray dropletsize.

At altitude relight conditions, the current engine relight limits with JP-4/JP--5 fuel were essentially met or exceeded with all of the JP-4- or JP-8-basedfuel blends. However, a very significant reduction in altitude relightcapability was found when a standard No. 2 diesel fuel was tested, indicatingagain a strong effect of fuel atomization characteristics. Improved fuelinjectors and/or higher ignition energies would be needed with a diesel-type fuel.

Combustor liner life analyses, based on the test data, were conducted. Theseanalyses resulted in relative life predictions of 1.00, 0.78, 0.52 and 0.35 forfuel hydrogen contents of 14.5 14.0, 13.0, and 12.0 percent, respectively. Turbinesyste, life is not predicted to change for any fuels with properties within thematri. tested.

3

.1

A series of short but severe fuel nozzle tests did not reveal any majorproblems with the fuels in the matrix. This was expected, based on servicehistory. However, considerable additional long-time tests are needed to fullyassess the effects of fuel thermal stability characteristics on fuel systemperformance and combustor/turbine life.

4

•4

SECTION III

TEST FUEL DESCRIPTION

A. General Description

Thirteen test fuels wcre supplied by the USAF for combustion systemevaluation in this program. These fuels included a current JP-4, a current JP-8(which was out of specification on freeze point), five blends of the JP-4, fiveblends of the JP-R, and a No. 2 diesel. The blends were made up by the USAF toachieve three different levels of hydrogen content: 12, 13, and about 14 percentby weight. Twc different types of aromatics were used to reduce the hydrogencontent of the base fuels: a monocyclic aromatic (xylene bottoms), and a bicyclicaromatic described by the supplier as "2040 solvent" (a naphthalene concentrate)A third blend component, used to increase the final boiling point and theviscosity of two blends, is described as a Mineral Seal Oil, a predominantly (90percent) paraffinic white oil.

The rationale for the selection of this test fuel matrix was to span systematic-ally the possible future variations in key properties that might be dictated byaveilatility, cost, the change from JP-4 to JP-8 as the prime USAF aviationturbine fuel, and the use of nonpetroleum sources for jet fuel production. TheNo. 2 diese was reiecced to appzuxiiate the Experimentas Referee BroaISpecification (ERB.) aviation turbine fuel that evolved in the NASA-Lewis workshopon Jet Aircraft Hydrocarbon Fuel Technology (Reference 1).

1. PhsQical and Chemical Prcpertles

Fuel properties shown in Tables 1, 2, and 3 were determined, for the mostpart, by Monsanto Research Corporation under contract to the USAF. Table 4presents conventional fuel inspectioi data determined by the Aerospace FuelsLaboratory, WPAFB. These data may be useful for assessing the accuracy of test-cthodc and com-arina theep fuPel to those used in other investigations.

In Table 1, density, viscosity, surface tension, and vapor pressure arepresented at a common temperature, together with temperature coefficients whichwere calculated by GE from Monsanto three-point data. Also shown in Table 1 are Ithe fuel cumponents, hydrogen content determined by the USAF using ASTM MethodD3701 (Nuclear Magnetic Resonance), and heating value determined by Monsantousing ASTM Method D240-64. Heating value of these fuels e MJ/kg) is very Inearly a unique function of hydrogen content (H, %) which can be closelyapproximated by:

Qnet w 35.08 + 0.5849 H (I)

Surface tension is virtually the same for all of the fuels. The other propertiesare, In general, quite dependent upon fuel components as well as on hydrogen co~tent.

Table 2 shows hydrocarbon type analyses by mass spectroscopy (ASTY Method

5

aa

oot 9 - m

41 3 5. O . . ~ .Q-).

.0 T. 9 1 no -ý- C ý

- 9

* do 0

- 0 -n

OD 0 'a (N S a ~1P

0Q

II-

-4toa e

0.4 M

-~~4 U

fa aý0U

*l a * 0 a ,* a-o o'C a ' a aC~

06 -(

-CIm CA t & S a *

0- 10 CL 96 *

(71co 47 0' ( ( -3 a r- or V 'i en 10' n kr C Ln sr) \r) tn- kn 0t kr' %ni \.o '.D

0\ 0' % o' 0 n 00 - 00' I I I I % V q.-~ 3 a 3 tnco in n %AS kp f '0 '

0' I' '.)u n -4 -43 U-% -1 Go- 'i (No f' r% in U-) in in Vin i) in in kn Sfn \C 1

00 - %.0 0\ -I'0 ' 0 C ' o ri-I \r. ' ' rfS C-4 r-I Cs-i a C 'C 0

w f'. in sj' U in n V.) %r) Ui% U. %ni Si \.0

Vj 4 r i % iN n in n an% sri cf 4 4 sf

(UE ( S5 .0 (N - C) i M ~ (N4 V) L' N M1.j- r- or-C 0 - 0' Un'C ''U f' n W\' Ln' -7 if'. Sf54 U4.4 .4 Sf

C3 MI -I - o L-51 0- 0 0i Mf' %C C> W .0 f17 -'

kO - n V-i kn -Z -4 U-1 -1 '4 . - 4

.54 k^ o\ (1- 0 -7 O - n e

& J( : ) 0 0 0 ,; 0 , ' oi C (N,3 r

IL Ir 0 .4 'Co 'C in% c-N '

a, ffN C'4 (-- O. ' C - N (N

en -a I" 4 - 4 -4 --t C^ in cn en C"'5.4

C) ~ fI \C fn C r ~ N ( 5'in~~~~~1 - -(4.r' N ( "

S1ý 0'n -. -.3 I- - r n (N cn M- -2

.01 1 1' 00 U) 0". 0'C\0 ' 0 ' )

I-'t\ a, CO4 G\ '5-

i ON- 0\r n ~ (

,mu - (N in - LI' c. r., 0 0' C) - (-4

cc c- 3 C 3 E ' '4 & - '

'3' C N

- c w a.. AL C r I r C. o

'C~(7 '1' 0 3 ~ - 'C 0,0 Ce I ' '.4' - -.3 t3

"r4 7 13-

%r, t14 k"

I'-. ~ ~ ~ ~ % C2 * f I " .3 nU

k-4 LM.

N C -N e4 -7 ' - 3 -4 - 3

Q. ~ C Lf - ' C-. -4vN N 'm-'~ IN ~ c- 'L3 'C3 -3 C'

K w *' .

-3 m

>1 N'C w C. = 'N OZN 0

- 0' - ,

> W CCw 1-4 EMU 0

Q6 U 3 0-

U C U c > '9

Best Available Copy

D2789) and Figure 1 shows a comparison of total aromatics determined by massspectroscopy (from Table 2) and by fluorescent indicator absorption (ASTM MethodD1319 from Table *.). It is apparent that there is a consisten. bias between theresults by the two methods, with the mass spectrometer yielding che more favorable(lower aromatic) results, particularly with the JP-8-basAu fuels. Aromatic type(monocyclic or bicyclic) does not appear to affect this bias.

Figure 2 shows the variation in fuel aromatic content (Vy mass spectroscopy"with hydrogen content for these fuels. There is, of cout-:e, strong negativecorrelation, but it is apoarent that both base fuel type and arowoapic comoonentstructures affect this relationship.

Table 3 lists the Gas Chromatographic Simulated Diitillation (ASTH MethodD2887) data for each of the test fuels. Among the blended fu ls, those containingthe Mineral Seal Oil (Fuels No. 3 and No. 12) had the |. hahst final boiling points.Figure 3 shove the complete simulated distillation cu. *,s for the three basic fuelsand the variation in initial and end points for all o" zhe blends. Points vorth7of note are:

1) All of the JP-A blends had initial boiling points (IBP's)essentially identical to that nf the base JP-4 fuel(about 300 K).

2) All of the JP-8 blends and the diesel fuel had IBP's essentiallyidentical to that of the base JP-8 fuel (ak'ut 385 K).

3) All of the JP-8 blends had final boiling points (FBP's) notgreatly different from that of the base JP-8 fuel (about 590 K).

4) The JP-4 blends had a broad range of FBP's, spanning those ofihe JP-8 blends (about 585 * 35 K).

5) The diesel fuel had a 3ignificantl' higher FBP (about 680 K).

Figure 4 compares fuel volatility characteristics as measured by gaschromatography and conventional distillation (ASTM Method D8E). It is apparentthat gas chromatography signiticantly extends the apparent boiling range in bothdirections; the IRP and 10 percent recovery temperatures are lowered while the90 percent recovery and Fr? temperatures are raised. Temperature differencesof up to about 70 K are obtzined by the two procedures.

C. Thermal Stabilitv Characteristics

The thermal stability of test fuels was determined by the Jet Fuel ThermalOxidation Tester (JFTOT) described in ASTM Method D3241. The actual thermalstability is given in terms of the breakpoint, which is defined as the highest(metal) temperature at which the fuel "passes" by both filter pressure drop andtube rating. A fail on pressure drop is 25 mm Hg pressure drop or more in lessthan 150 minutes. A "fail" oh. the tube is a color code of 3 or darker as describedin the ASTM procedure. In practice', the fuel is tested first at the estimatedbreakpoint, and then, depending upon whether it fails or passes, is rerun at atemperature 10K lowey or higher until the breakpoint is established.

lO

Best Available Copy

80O JP-4 Based Fuels

0o QJP-8 Based Fuels

ANo. 2 Diesel Fuel0

o 60

ideal,

4 JP-4 Based Fuels,

/ y 0 5. ,x + 1.89

U All Fuels,*y l.llx 4. 1.25

0 !_.P-8 a'Ised Fuels,

, - 1.26x - 0.06

0 2" 40 60 80

.r'M , tlime 'ercent by mss spectroscopv

r .1 ,l Aro1mitc Content-wo '=, T ' -t lethds.

11

Bezt Available Copy!

60_ __ _

ho

ie

40

A-0.941

U W

*1"

20

OJP-4 Based Fuelsn iP-S Bane#. fuel*

G0

6XI , M4es•el Fuel

Tailed Subols:

Mmocyclr Aroaat.c blendsj

12 13 14

Fuel H4ydroge-n Cinmtent, ¶aeig•ht percent

Figtire 2. V.•irint'm of F's'I Aromateic Content 41ith Hydrogen fomtent.

12

UI

II

800. 1

700

JF4Blends

JP-8 BlendsI

390~~Bb JJ--JP4 led

209 200 4 (1 S

Percent Eccove red

ure 3. Cparicsan hof Gas Itromatographie sl-ilateeDistilla~tion Caracteristtca of the it-at fuels,

13

A

700 -Final Boiling Point,

600 ~ ~~y - 0.772x + 91.71-• '---- - ••

90% Rec. v rC,

50011

-c. I__ 'Q 4L lde~a!, _ _Idal

400 102 Recovery.

E

C. y = 0.930x + 50.6

300 - "---IInitial Boiling Po-nt,j I/ I =l 1.032z P3•-.5/ /,

200 390 400 500 600 700

Temperature by Gas Chromatograph, YK

i

Figure 4. Cowarison of -ut: kecovery Points by Distillation

and Gas Chromatography.

Table 5 shows the JFTOT data that were provided by the USAF. Occasionally,anomalous or indeterminate results were obtained, and sometimes the fuel sample(one gallon) was expended before the breakpoint was determined. For these reasons,breakpoints are not shown for all of the test fuels. Anamolous results are thosein which two or more tests of the same fuel at the same temperature showed botha "pass" and a "fail." Indeterminate results are those in which a fuel passes orfails by both tube color and pressure drop, but in which no additional tests wererun at higher or lower temperature to determine by which criterion it would failfirst. Generally, it appears that the repeatability and reproducibility of thebreakpoint is greater than the difference in thermal stability of the base fuelsand their blends.

Table 6 is an attempt to assign ratings to the fuels, despite some apparentlack of precision in the test results. The JP-8 base fuel appears to besignificantly more stable than the JP-4 base fuel. The addition of Mineral SealOil has no apparent effect on the thermal szability. This would be expected,since it is a high-purity white oil, suitable for medicinal and food applications.The addition of both types of aromatics appears to have little or no adverseeffect on thermal stability.

D. CoEputed Combustion Parameters

Table / show's several fusel pa-rameters which w~r ccrputed fr= thc Ph-,;sical1and chemical properties for use in conducting the combustion tests and analysesof the results.

Fuel hydrogen-carbo'i atom ratio (n) was used in the exhaust gas samplecalculation. It was calculated directly from the hydrogen weight percent (H)by the relationship:

11.915 Hn 100 - H (2)

and ranged from 1.625 to 2.021 as hydrogen content increased.

Stcichionetric fuel-air ratio (fst) was used to calculate comparativeadiabatic flame temperatures. It wzs calculated fro= the fuel hydrogen-to-carbonratio (n) by the relationship;

f - 0.0072324 (1.008 n + 12.01) (3)st ( + 0.25n)

which assumes that the fuel is Cfl%. that the 3ir is 20.9"95 vohume-pýrcent oxygen, 4and that the air has a molecular weight of 28.9666. For the test fuels thestofchiometric fuel-air ratio ranged from 67.50 to 70.19 g-fuel/kg-air as hydrogencontent decreased.

Stoichiometric flame temperature was used in analyses of NOx emissions. Itwas calculated at takeoff operating conditions (13 = 66. K, P 3 = 1.359 !0a) usinga standard equilibrium-thermodynamics computer program (Reference 2) and rangedfrom 2494 to 2515 V as hydrogen content decreased.

15

Tabh( 5. Fuel Sample Thermal Stability Test Results (ASTMMethod D3241).

Mode ofFuel No. Breakpoint, K Failure

1

Table 6. Estimated Thermal Stability Ratings ofTest Fuels (ASTM Method D3241).

Breakpoint EstimatedFuel No. Range, K Rating, K

1 533-573 573

8 513-533 523 + 10

10 553 553

11 543-553 548 + 5

17

12 543 5I

1,7

S.4 N

m-40

0 - i), 'c 0f 'C m 04 0 i) .

.14 -.1 -.7 In 1N7f

p" -- N C- q

'.a.

-11 4 N ( ' 0 ' 0' C4 0) C' 00 (

* ~ O -f - - -) - -) - ) - C C a

~ -. ) 0) C 0 70 0, C) 0' 0) n

Relative required fuel flow rate was used in all combustion tests to adjustthe JP-4 fueled engine cycle operating fuel flow rates for the reduccd heatingvalues of the other fuels. The factor is meruly the ratio (QJp_4/Q) aid rangedfrom 1.000 to 1.0395.

RelatLve fuel spray droplet size was used In analyses of the low-poweremissions and relight performance. The 379-17A combustion system employs pressure-atomizing fuel nozzles. so Jusaja's correlatio'n parameter for this type atomizer(Reference 3) was used to estimate the relative fuel spray droplet Sauter MeanDiameter (SMIJ) from the test fuel density (W). surface tension (o) andviscosity (v) by the relationship:

v )0,16 ( ).6 )O.43 (4)(SMD) JP_-4 Vjpý4 aJP-4 PJP--4

Ab shown in 'sable 7, none of the blending agents appreciably changed the predictedrelative droplet size of the babe fuel. However, the JP-8-based fuels arepredicted to produce mean droplet sizes about 23 percent larger than those of theJP-4 fuel, Further, the diesel fuel is expected to produce mean droplet sizesabouL 41 percenL larger Lhan tOwe@ of the JH-4 fuel.

I!

Il

SECTION IV

J79 ENGINE AND COMBUSTION SYSTE1M DESCRIPTION

A. Overall Engine Description

The J79 engine is a lightweight, high-thrust, axial-flow turbojet enginewith variable afterburner thrust. This engine was originally qualified in 1956.Since that time various models with improved life and thrust have been developed.The model currently in use by the USAF, the 379-17A, was the reference enginefor this programii. An overall view of the engine Js presented in Figure 5. The379 has a 17 -stage compressor in which the inlet guide vanes and the first sixstator stages are variable- The compressor pressure ratio is approximatel,13.4:1. "lite combustion system is cannular with ten louvered combustcrs. Theturbine has an air-cooled first-stage stator and a three-stage uncooled rotorthat is coupled directly to the compressor. The engine rotor is supported bythree main bearings. The afterburner is fully modulatiigg with a three-ring V"

gutter flameholder. Afterburner thrust variation is accomplished by means offucl flow scheduling and a variable area exhaust nozzle.

B. Com~bustion Systcm. Dcscript ion

The J79 engine employs a cannular combustion system with ten combustion cans.The ten cans are located between an inner and outer combustion casing forming anannular passage. The combustion system flowpath is illustrated in Figure 6. Anexploded view of the systen with the various components, including the compressorrear frame and the turbine first-stage nozzle, is shown in Figure 7. Figure 8shows the combustor cans (one omitted) mounted in an engine transition duct. Apictorial drawing of a combustor can assembly' is presented in Figure 9. EachScombustor consists of three parts riveted together to form an assembly. Theforward outer liner is an airflow guide to assure proper flow distribution to theinner liner. The leadinig edgo of the out-er liner has n sno-t -ch extends into

the diffuser flowpath. The snout incorporates internal vanes which distributeair to the dome in the desired flow pattern. A slot in the snout permits accessof the fuel nozzle to the Inner liner dome. The rear liners are oval shaped andoblique at the rear to facilitate removal from the engine during overhaul orinspection. Cooling air for the inner surfaces of the inner and rear liners isadmitted through punched and formed louvers. Combustion and dilution air isadmitted through a series of thimbles in the inner and aft liners. These thimblesare arranged to provide flow patterns for flame stabilization in the primary zoneand mixing and turbine inlet temperature profile control at the aft end. In anengine assembly, two of the cans are provided with spark igniters for starting.Adjacent cans are joined near the forward ends by cross ignition tubes to allowpropagation of the flame from the cans with a spark igniter to the other cans.The flanges of adjacent cross-ignition tube bosses on the liners are held by V-band clamps which can be removed for engine disassembly or inspection. The linersare each positioned and held in place by mounting bolts at the forward ends. Axialstack-up and thermal growth are accommodated by a sliding seal between the combustor

can and the transition duct. The major liner material is Hastelloy X.

20

V C|

__ --- Ji

-.

0 - - I•

21

iA

IF

IL

II

I7

4

C7.

tx

7r;-

kk&

4A0

Best Available Copy

NOII ow

IIFFOR%% AýLLINE

iNAt' PII f

411

""AS-: tr;,w"-Altlor41, CL $,'JXta rJTh? 0 ten ova] nj e p,(Jrtl t £4 ;'SAt-t 0-1AL V j u4tulI Vj( crt -s 1Wax- i-i / St1 b-tr -I 'A e t;r tl tiIier Ctj it I ,ariiuI a i to wu:tt' i i up se I 01

i o tuijrb1,It: I ',vwpat. '1sv L.r;jJYAi tIJisb -i -ttlC), /~ t area ý ;3 a/ppr v/ ir-a tt j u-iw ;fc(f O t't,tJ exit :i ,v f ti':t4-; ttprcvvdlnr' ara.ccervi4 1iw tt0re-as IntoCoLust l-e. Vhi! tr~Tanpitivnf1dwjtt a uWs"iftelly UYiv'- a,'ý)touvprt pf:1:; 1:) t(h

alJi'rdlrsr fvr Cpftrrýt:s ttt-ita yr--sh btj m tbfr MSI tai 0*axa otti

a I f I r P4 t Fig, f -%)' f, ~ cf tkit;, 479w~j bli2 1t, kitb "Inrvve-3) OLe ?YW,--

e~ha. , 1rA,;sd-vt~pco-~ at r 04t Thv'r 'n,'t, dv ota-ns stl;;uu:. it) d&JO'ns'rA~It1eeta y.~htttn a ~'a Cle~tuos-k sus,,!-',itt4 rear IIr.rs -t

%g:if thlrybutIi1tta the s. fo lrd etow.-r3nr r-t alt t-enuay L-4-.z)icralt at0P4, 13_,(* atri: Om t. I't-r'}dl lrn-pev, t0;et I;, beisr1]

f £41t~tt 41v ~$~yJr iI'i lty, a f'srl-II-nw dfvidirnz VAvlv vxttrnab tQ ow i3 ~vwt-t ~iuyi- Th w~i:joiaij tip, vtl'.1, ddij-/Ce,-t- fs-3 to, OCe CMwbuttor a; -a

t~i~gu~t~iie-4l's-] It pvro4vide by a2() u-if t,' 2-A pt'-.-burt' Mtijir 6'ptr-a. 4IW n'i; *pw-flp, a 49) %ri, l miiiy;, prAL-nt'n'-11 vZr ( 4vemrtt r -. oQf punxp), a nd a

0, 0)enI t valve4- i'sl-6t Ulj a;V eryi: fUtr f1'm rate- Qf np;,,-vrlmjat ey

L, i tk- Cd

i/rjj~t- Wij~j)J!- vft, ;;rlm-iry or)f 1"t-e, iis 4 O .9 . da is dic-tsr.ii64- 1 jjt- fi, w t' fe'J lo ;, I -n l fI -.. 7?' It 111w it Ier j1 It ti

I W 1 Y j tr o ' IJ / 9h e , e P U t ( V r d i : O t e ; ' s ' o a b

'1ji JO 4/9 ý fip-.nzzJ- triiuw(.: r' curd 1:, ket-irslly sAsuit), no crti,sq~ ~ ~ ~ ~ ~ ~ ~~~l .~Jymun-;ri~'w~,hoz e~iia o ur ret1,ty p-orwdat 129(A

tp'-tgnted) Isir- ic'verbaijit-&. bh- ynInny rcrzlera are con: !ru-wed Ine te-rvic-ias. 's410torapsI 2i'Y) Yi,)umn at S440,:1 (Sit- Usey are- 41uehold-Til]t LUleve-dC

to ,"a ioras-i~tlvr it:(L(.4V the:ý ny-zzltvý jJ4V t J4,ldt-rtC Ito have; prac~tj(a) y

The osebu,tv~r tbuut uJ"-ruAc- uv4r a wldt. Tanj&- uf fue flow, Inle-A airfl]vid,

Iu~~~~1lo W~ j oUIECA

now~4 IVJ iww'

AtI

figure Ii.yw i~m Duail orilz~e lue-I N,'~]

27

250 -I - - - ___Test Fluid:

KIL-F-7024.

_ ///200 yp~e 11 at 3003 K

150-

Takeoff

Dash

I.-

Cruise

Reli h Ileh

0 2 3 4

Fuel N~ozzle Pressure Drop, !Pa

Figure 11. J79 Engine Fuel Nozzle Flow Characteristics.

28

temperature, ard pressure. Table 8 presents the combustor operating paran-etersat four inportant steady-state engine conditions which are typically evcountered.At these conditions, fuel effects on combustor perforuance emissions and cobustorand turbine life are of particular interest.

In addition to steady-state operation. thz zzc sticz syste- t.u!st -r(-videfor starting over a wi&i range of conditions ranging froM cold day ground startto relight at high altit-ade and high aircraft Nach pumber. Figure 12 presentsthe required J79 relight envelope.

29I

It-.-S-

I

29

Table 8. J79-17A Engine Combustor Operating Conditions.

Cround Subsoni. S.upersonic Groun 3)

Idle Takeoff Cruise Dash Start

Flight Altitude, km 0 0 10.7 10.7 0

Flight Mach Number 0 0 0.9 2.0 0'1 Tctal Airflow. kg/s 18-3 75.2 29.5 V7.5 3.2(1). Combustor Airflow, kg/s i5li v)2.7 2-4.6 3.3 \3.2

Wf( Fuel Flow, kg/s 0.144 1.259 0.335 1,171 0.042

T3 Inliet Total Temperature, K 421 664 559 781 -239

P3% Inlet Total Pressure, HPa 0.248 1.359 0.471 1.589 0.101

f4 F'uel-Air Ratio, g/kg 9.42 20.07 13.60 1b.0i 148.0

T4 ,Exit Total Temperature

(Ideal), K 792 1064 1362 1335 -

Vr( )Reference Velocity, x/s 24.2 28.6 27.2 33.5 >7.1

(1) Engine flows indicated (ten coabustor cans).2(2) 9xsed on W 3 and Ar 3684 cm "

(3) 1000 rpm, typical starting speed.

(4) Minimum engine ftuel flow schedule (normal).

40

30(

S-• , • 4 a _, *0• b 0•m • - - -

I ~20.0

Estimat ed Current

379-17A Starting LimitsWithout Starter Assist,

• .1P-4 Fuel

,,.; .- //i- -

5.0

--/1

- - -~~~o -.- - - ..

'I

SECTION V

APPARATUS AND PROCEDURES

All combustor and fuel nozzle testing in this program was conducted inspecialized facilities at General Electric's Evendale, Ohio plant, using apparatusand procedures which are described in the following sections. Combuscor performance/emissions/durability tests were conducted in a high pressure/temperature single-can combustor rig which is described in Section V-A. In parallel, combustorcold-day ground start/altitude relight tests were conducted in a low pressure/temperature single can rig which is described in Section V-B. Also in parallel,high temperature fuel nozzle fouling tests were conducted in a small specializedtest rig described in Section V-C. Special fuel handling procedures used in allof these tests are described in Section V-D. Finally, procedures employed inanalyses of these data are described in Section V-E.

Actual engine-quality, current-model J79-17A engine combustion systemcomponents, listed in Table 9, were provided by the USAF for use in these tests.The combustor assemblies were newly repaired units obtained from USAF stores.Four udis were obtained, which visually were identical. Before testing, theywere airflow calibrated; the results arc listed in Table 10. Assemblies 1 and2 were -gssined to the high pressure tests because they wcrc the most nearly

alike with respect to flow calibration. Assembly 3 was used in low pressuretesting.

A. Performance/Emission/Durabilit%, Tests

High pressure/temperature single-can-combuator rig tests were conducted atsimulated J79 engine idle, cruise, takeoff, and dash operating conditions witbeach of the fuels to determine the following characteristics:

I) Gaseous emissions (CO, HC, and NO '.

2) Smoke emissions.

3) Carbon deposition.

4) Carbon particle emissions.

5) Liner temperatures.

6) Flame radiation.

S7) Combustor exit temperature profile and pattern factor.8) Idle stability (lean blowout and ignition) limits.

Thus, a large part of the total data was obtained in these tests using apparatusand procedures described in the following sections.

32

Table 9. J79-17A Enginv Combustor TelL Parts Lint,

rPart Name I.: i'nrt Number National Sutock Number

Ignition Combustor 106C332OG73 2840-01-004-1728-NN

Li;wr Aussembly

Fuel Nozzle 577C796P10 7915-00-110-55141'1.

(Parker-flannfifin 1345-633327)

Main Tgnition Unit 106C5281P3 2 9 25 -00-992-7904PI,

(B~endix 10-358765-1"I liOVAC, 400 cps)

I Main Spiark Plug 696D256P02 2925-00-925-05445PI.(Chaniploti FHE J87-2.)

Main Spark Plug Leand 517D377P2 29 25-00-956-0293PL

(Mfr M1482 59393.)

33

Table 10. Test Combustor Flow Calibration Results.

Combustor Effective Airflow Area, cm

Combustor 1 2 3 4

Front Liner Assembly 23.7 24.5 23.2 24.4

Aft Thimbles 58.5 59.0 59-1 58.3

Aft Cooling Louvers and Seal 21.0 19.9 23.9 24.5

Total 103.2 103.3 106.3 107.2

34

344

1. Htgh Prssaure Test Rig De-scription

These tests were conducted in the Small Combustor Test Facility. Cell AS.located in building 304 of the Evendale Plant. This test cell is equippMd withall of the ducting, fuel and air supplies, controls, and inscrmsentation requiredfor conducting small combustox high pressuxe/temperature tests. High presurtair is obtained from a central air supply system, and a gas-fired indirect airheater is located adjacent to the test cell. For the singlc-can-ccabustor rig,J79 engine idle, cruise, and takeoff operating conditions can be exactl'.duplicated. Dash operatii, .- corditions can be exactly duplicated with respectto temperature, velocity, and fuel-air ratio, but pressure and flat, rates Mustbe reduced about 25 percent in order to be within the facility airflowcapability.

The High Pressure Crmbustor Test Rig, shot•n in Figures 13, Ii and 15.exactly duplicates a one-tenth sector of the engine flou-path from th, cceresscroutlet guide vane (OGV) to the first-stage turbine nozzle diaphragn (.). Asshotn in Figure 13, the test rig inlet flange which bolts u; to a plenum chamberin the test cell, incorporates a Beilmouth transition te the situlated k%"IV plane.The combustor housing is a ribbed, thick--alled vessel which forms the innErflotpath contour and side walls, covered by a thick lid that forms the outerflw-parh contour. Figure l. shows a combustor -nstalle& in the pressure Vesseiwith the lid removed. The combustor exit engages an actual segment of at. engiýe

sector section which contains an array of %ater cooled ibistrmn-:ation rakes inthe TD plane, indicated by an arrow in Figure 15. Additic'nal derails of the exitinstrraw-tat ion rakes are shc-our. in Figures 16 and 17. Dotnsrrea of the rTaes.the combustion gases are water-quenched and the sector flcw-atk transiticns tocircular, which is bolted up to a remote-operated backpressure valve ir. thc testcell ducting. Two other features of the test rig ca3 be_ seen in Figure 15.: aflme radiation pyrometer, -noted to view the combustor primary zone rErr-hacrossfire duct wino-d; and bleed air pipes to withdraw, collect, ar.d metersimulated turbine cooling airflow.

2. High Pressure Test lnýsranratton

A sumary of the important ccnatstor operating, perforimance. and emissio.parameters which were measured or calcrdlated in these test- is shon in Tabie 11.Airflow rates were measured with stainard )S'4Z orifices- Fuel flow races ueremeasured with calibrated turbine flotaMcers corrected for the density aanviscosity of each test fuel at the measured supgly temperature. Combastor inalettemperature and pressure were measured with plenum chmaber probes.

Combustor outlet temp)erature. pressure, and gas samples _ere reasurt_' witzha fixed array of seven water-cooled rakes, arranged and hooked up as shoW-nin Figure 18. Each rake contained five capped chromel-aI~ei thermccoupleprobes, located on radial centers ef area, and fcur impact pressureigas sampleprobes, located idwa-; between therwcouple elements- As sh iri Figure IS,eight of the impact probe elements were hooked up for total-pressure measurement,and the other 20 elements w~ere manifelded to three heated gas sample transferlines leading out of the test cell to th. gas compo-siticn mfasurement inscrtrzntats-Transfer Lines I and I1 were connected through selector VL1!ves to a smokemeasurement console (Figure 19) and a gas analysis conscit (Figure :0)_ Transfer

35

ala

I3

m ww

Ia UM

3Z

r --4L

- e - - _ _z

tA

F -- - -t�?tintF --------------.-- -�----a a - - -�

-V.

7 *

( II j& I

III

I'

I

N

t

N

c2 I, �teuura"

"a * * * "' • P "•

S.

'ell

Ilk-

A-

404

7~a Vr :rt

9LC 91W

go Id UN Cd

s- O4L.I

vt ,a vi o m c

r4 E > 4c -LS- Q Q

41 S, C0 2£t £ C, C x a- c c

: Co -'D aF c i .

-1 CC -u I . -. -OE .4 -. i 1 C-cL) .C I 4 aJ S C i s

To -c fo4 =6 0. 7 V 7s 41 C 0 8 0 '

4-wC- r L. v n w 7 + 4.-

41 -

3 0

t~Thermocouple 5 Element/Rake, 35 Total0 Total Pressure, 8 Elements Read Individually

IiSmoke/Gas Sample, 6 Elements Manifold to Line I -0) smoke/Gas sample, 6 Elements Manifold to Line 11

RakEEie~ii L~Carbnan EmI.S51iu, 8 Elements Uanilold to Line III

AlYl

04 -Gas Sample Each at

G1l Sample Points 2-10Console

B2 (CAROL)_________I1

F3

Flow

El -Valve CletOtA2- CI uinuiat ive3i.2 scf gas

C _ III Sample (0.8 sef gas_3i- at Points 3, 5, 7E3 Flowand 9).

D4 Carbon Meter TestF4 Collector Ts{Crucible Meter

Figure 18. High Pressure Test Rig Exit Instrumentation.

42L

AIL--

40 A

I I lwX 40W1117

43

JAM-;

*Ale

44L

Line I11 was connected to a -_arbon collection crucible located near the testrig, and then continues on to flow control/metering apparatus in the gasmeasurement area.

The General Electric smoke measurement console (Figure 19) containsstandard test equipment which fully conforms to SAF. ARP 1179 (Reference 4).Smoke spot samples are collected on standard filter paper at the prescribed gasflow rate and at four soiling rates spanning the specified quoted soiling rate.The spot samples are later delivered to the data processing area, where thereflectances are measured and the SAE Smoke Number is calculated.

The gaseous emission measurement console, shown in Figure 20, is one ofseveral that were assembled to General Electric specifications for CAROL systems(Contaminants Analyzed and Recorded On-Line) and that conform, generally, toSAE ARP 1256 (Reference 5). This system consists of four basic instruments: aflame ionization detector (Beckman Model 402) to measure total hydrocarbon (Wc)concentrations, two nondispersive infrared analyzers (Be'kman Models 865 and 864)for measuring carbon monoxide (CO) and carbon dioxide (CO 2 ) concentrations, and aheating chemiluminescent analyzer (Beckman Model 951) for measuring oxides ornitrogen (NO or NOx = NO + N02) concentrations. Each of the instruments arefully calibrated with certified gases before and after each test run; andperiodically during testing, zero and span checks are made. Readings from theinstruments are continuously recorded on strip charts and hand-logged on testand calibration points for latec Calculation of concentration, fuel-air ratio,and emission indices, using a computer program which incorporates the equationscontained in ARP 1256. Gas sample validity was checked by comparison ofsample to metered fuel-air ratios.

Carbon deposition tendencies wei2 assessed by installing a clean combustorfor each test, and inspecting it after the test, assigning it a visual rating.

A cumulative carbon collection technique was also employed (Line III in Figure18) in an attempt to quantify the amount ef large carbon particles emitteddirectly and/or that were shed periodically from the depositions. A fixed volumeof combustor exhaust gas at idle cruise, takeoff, and dash was filtered through

Sporous c~ruible. At the completiono the •t-., the .•cibl• . . rcmovcd f,-laboratory analysis to detennine the total quantity of carbon collected.

Combustor liner temperature was measured by an array of 24 thermocoupleslocated on the inner liner as shown in Figure 21. This instrumentation patternwas selected to provide detailed data in the vicinity of the known hottestregions of the combustor. There are no thermocouples downstream of the cross-fire ducts or on the aft liner, since those regions are usually at least 150 Kcooler than the forward section of the inner liner (Reference 6). Figures 22through 25 show the actual liner thermocouple installations on the inner liner.

Flame radiation in the primary burning zone of the combustor was measuredby a total-radiation pyrometer (Browr. Radiamatic, Type R-12), which can be seenin Figure 15. A diagram of the optical view path is shown in Figure 26. Thepyrometer sensing element is a thermopile which provides a direct currentvoltage outpuL. The flame radiation is focused on the thermopile with acalcium fluoride lens vh-ich !s transparent to radiation of wavelengths lessthan three microns. The pressure seal at the test rig wall is formed by a

45

t

V

_ I! +, y, > I> . , . , .

Tt0Jir--•'• Y. ,•,,, , ,.•.._: • .• .-. / .0

".,'; . .. __- .• • ,-• •,•.. . -- ' »-. -- )" v . L -A• , •.TV .•- -• .• -•,.•,t .. •• .• •. .,,. o - . .. , . ) 1 _ .- '

" ..." ." . ' ,' I" - " ) " •"N C " "

,: . ,ti . . , , , \ I . .. .J.. . • ', N ..' , ,, ," . • •t, ,• • :, • / I • l , • "- - 4 . { I" \i ,J%'. . " I

• " I " ([ '. • , • • • /l-I e .e. - .. ..* I.

-).. .: ; . . . .. •,,,. ,, • . ,. -> - % V, c , ,). .. ,- C,9- ! r ': r .• -• - \'.SI' I " •t, .- •,- - z] 1 N : . .

" Ct•_.---- -• _; g \N, " ') . ½ • " ',.- -,, , -+,,

4' -'W \¾'-,-""7_ "% r " 4 .- "j¾ --.-# " \'N - "2•, .> - ,- ) .* .

di ) . ' "e),. - - - . -• , , ' . -, , .

-I.- ,, + .1o

J •J r I '?i N-- " . ' '/V ._I_> • ) ¾ -N -, _ /b.:t 0 _

46-

F i -1t 2. (or. Itst ur 1.3 i t i I i timciit -it tion -1 60( Dk'grcl * ( WAI 1'

47

tilrC2.Cormhui-,t, r 1. itic r I nst riinun t .kt Uif on t I 6. ind I ')t Pb9 [kgr'S

48

F i k2ý. CL'Mbu '-ý L I iz ncr I t~ t rnC 11 t 1 F' OlA t 300 1k r,' Cz CW'A IF.

49

IIC6I

lea.

AO.

sapphire wintdow which is transparent to radiation of even loNrgt' waveýlengths.The sapphirt windo~a was swept cl-ean by a small flow of filtered air. Thepyrk-meter was calibrated by viewing atrssac-vt~-abnbcbd furnacethrough the same ocptical system usted in the combuistor test. The furnxacetemperature was measured with a dlsappearing-filameut opticol pyroatetr.

3. High Prt'ssurt' Test Procedures

A total of 14 high pressure rig tests were run., oike for each test fue.plus a repeat test with fuel No. 1 to establish test variability. Twol differentifuel nozzles and combustors were alternated in the tests. st. that while otxe setwas undergoing test, the other set was bein& cleaned, calibrated, and re-inistrumented as needicd.

Each test was conductetq to the ten-point test scheýdule showix tit Table 12.Ona Point No. 1, minimum lightot t antd lean blow-out 1limits at idle inlet condit ionst

weedetermined. onl Point-s No. 2 thro.ugh 10, steady-state operatitug. perforwUnce.and emissions- measurvemets ut-ýre obtainecd at simul0.ated enl'inek idle, crilise -takeoff, and dash operating ctliin.At each of these simulated engineoperating conditions, data were reco.,rded at two noutinal fuel-air ratios: 30and 100 per:enit of the enginxe cycle valuev correckted for the test fuel heat tugvalue. Ho~wever * if the 80 ucrecent fuel-air rat to point Indicated that th.e 100pet cent fuel-air ta it on.wol oaly exceed thle gas temperature lintitsof the exit thermocouple rakes, a lc4 er fuel-air ratito point was substituted.This limit was usually exc:,&ee~e at simulated takeoff ckenditdons, s-o a t'0 percentfuel-air ra't it' point wasl usull,1y substitulted for thle 100 per~enit point,

Test Voints, 2 through 10 were changed every thirty zutatdies (aiominallv) iWeatch test in order to s,,ubject thle ct.ombustor to- approximat ely thle same 4.S hourcarboni deposit ion clevi' with eaceh f telI. At t he colmpletionl of eahtest the%combustor was removed for visual inspect ion and phtgahcdocutlL'nt at ioNl Otthe car-bonl accumulationll Pre- 'and post -t est ailftlow Cal thrat ions" of tit' COMn-busto-rs were also periodically condulctedl to dtermine- it Othe carbon accutmo1ia-

cauý:k:"I t- 1: f 'I R.- 1: tN U.% 0; 11' I & -1 ii,'W ( it'

but no discernable changes Were ever foun~d.

It. Cod1,vAlýtitude Relighit Tests

Low' -pe *ure It u; erature s~ingle cant combustot: rig tests were con~iducted at"sialulatc J'I~79 eng ine groun11d cr-ankin~g anld allt itidt' wtidmillit ig opert'inigCondit ions1- to determline the cld-day grounld st art Anld al t i tu rol iglat

cha-acer~t ~sof each ot thle test fuels, The apparatus and procedures whichwere uttiltevd art' desc-ribed in thle fol klowing c ins

ct';W.I,-konldu~cted inl feil d tklllug' 30t Colnhusi~ ionl Laboratory attit( Yvll.Utl 11111. Th. tk~lty last~aabiitivs olo test lug small :ombliust tnrig oerýtW~erugk o -Aultu ýrotxil ýtttand atod klg cllit ionls

Liquid lirgl e'l xch1angers are. used to obtainl low Ituel anld air tempelkratureS.ati srikticjctos itthe, exhaust dc t aiv usýed to ~a in low cox11Ubstox tinlet

pressures

-

-� - . -

'4

A' a

A

jK*1

A 4, V �

'b tAk:�4 (4.' -t

� -,

- �. �,

-

-3

4, .. '

-

-' N" a .4, � '.-' A '3

N I '�

4 - I �N�'*" � � -' A

A r t 2 A3> .-. -

A

4) Mi- -- -- -.

k. It'-"''

4'.

2

'I N. '.4 " � A.

U) i '4 4,'. A

�

-- 4

'1

.4

-I

-4 ________

4)4-

*. -.

teA A A' Ž1,

F' a- a

� "� N '3 -, . N

.- 4. 4'

t �

-'44. -4'

A ' ii

'4 A '4<

'I ' .

I

- 'N 'N *' 4''''' V A

�7JY:T 71 '� 3a ¶ :4 4'"4'

The low-pressure, single-can J79 combustor rig used in these tests isshown in Figure 27. The combustor housing is made from actual engine partsand th3 rig exactly duplicates a one-tenth segment of the engine combustionsystem flow path. Combustor inlet temperatures and pressure are measured withprobes in the plenum chamber. The combustor assembly is installed from the rearof the combustor housing which bolts up to a segment of an engine combustortransition duct. An array of thermocouples is located in the transition duct

to sense ignition and blowout. This rig has no provY ions for turbine coolingair extraction.

Air obtained from the central supply system was dried at the facility to

a dew point of about 240 K and metered with a standard ASHE orifice. Fuel flowrates were measured with calibrated turbine meters corrected for the densityand viscosity of each test fuel at the measured supply temperature. All Lem-perature, pressure, and flow data were read on direct indicating instruments(manometers, poteutiome~ers, etc.) and hand logged by the test operator.

2. Cold-Day Ground Start Procedure

The firsL part of the test with each fuel was structured to evaluatecold-day ground starting characteristics. The test poi:at schedule is shown

in Table 13. The airflow rate (0.318 kg/s) and combustor Inlet pressure(ambient) were set to simulate typical engine ground starting conditions(1000 rpm). Fuel and air temperature were lowered from ambient to 239 K1HL~iIJIum (JF-B freeze point) in steps to simulatc progressively colder days.At each temperature step, minimum ignition and lean blowout fuel flow ratesware determined. Maximum fuel flow rate was taken as 7.56 g/s/can, which iswell above the current engine minimum fuel flow rate (4.22 g/s/can). Thetest sequecce was as follows:

1) With inlet condition. set, energize the igniter and slowly openthe fuel control valve until lightoff is obtained. Record light-off fuel flow rat,. Deenergize igniter.

2) Slowly decrease fuel flow rate to blowout. Record lean blowout] fuel flow rate.

3) Decrease fuel and air inlet temperatures in 5 to 8 K incrementsand repeat Steps I and 2.

When the minimum temperature limit was established, the second portion of thetest was begun: altitude relight.

3, Altitude Relight Test Procedure

The decond portion of the test with each fuel was structured to evoluat:altitude relight and stability characteristics. The test schedule is alsoshown in Table 13. Investigations were carried out at four airflow rates(0.23. 0.4., 0.50, and 0.91 kg/s) selected to span the J79 engine altituderelight requirement map (Figure 12). Air temperature was selected from thewindm'illing data and ranged from 244 K to ambient. Fuel temperature wasmatched to the air temperature. The test sequence was structured to determiue:

54l

r �---�--�-. - -

I

) I.

I I- �- I

I I

.JJ

.11

100

d) t. Q 10

C4- C4

0V,

-J-1-C

CL. 0 .- C

kn

~000000 0 00 00

00 0

Cs CC C4) .0

I 0 - - --0 - r4

'-5 Li U CO iS)56

I) The waximum relight and bloaout pressure altitudes with currentengitte minimum fuel flow rates (4.22 g/s/can).

2) The mininmum relight and lean blowout fuel flow rates at the relightaltitudes determined in (I).

The test sequence was as follows:

1) With 1.5.2 km (50,000 ft) altitude conditions- set, energize theigniter, set fuel flow rate at 4.22 g/s, then increase combustorinlet pressure (reduce altitudeand flight Mach number) until ignitionoccurs. Deenergize the igniter and record maximum relight altitude

conditions.

2) With fuel flow rate at 4.22 g/s, slowly reduce combustor inlet pres-

sure until blowout occurs. Record max'mum pressure altitude blowout

conditions.

g\ r. ize- ig.nicrn,.and 4nc-'rncn funl f1,yun, ,,i-4 liohi-nff-

Deenergize igniter and record minimum lightoff fuel flow rate at

maximum relight altitude.

4) Slowly reduce fuel flow rate until blo.wout. Record lean blowout

fuel flow rate at maximum relight altitude conditions.

5) Repeat Steps 1 rr.rough 4 at each airflow setting.

The 179-17A combustion system has excellent relight characteristics, so the

facility minimum pressure capability (about 40 kPa, corresponding to an al-

titude of over 18 km) was often encountered before a pressure blowout limitwas reached.

C. Fuel Nozzle Fouling Tests

Tests with each of the fuels were conducted to determine the relative

tendency to cause fuel nozzle fouling, which might be in the form of valve

sticking, metering slot plugging, or carbon buildt . in spin chambers and orifices.

The J79 fuel nozzle was known to have a long, troule-free service life and

to be quite tolerant of fuel property variation. It was anticipated; therefore,

that the test conditions would need to be extra severe to produce any signifi--

cant fouling in a short test.

The tests were conducted in the Building 304-1/2 Cumbustion Laboratoryusing the simple test rig shown in Figure 28. In this setup, hot fuel isflowed through the fuel nozzle which is inmnersed in a high velocity hot gasstream. Initially selected test conditions were:

57

ErCC

A-W-

F) -58

Gas Temperature 811 K

Gas Velocity at Fuel Nozzle Stem 304 m/s

Gas Pressure Drop Across Fuel Nozzle Air Shroud 13.8 kPa

Fuel Flow Rate 5.04 g/s

Fuel Temperature 436 K

Run Time 300 minutes (total)

Both the gas temperature (and velocity) and the fuel temperature were signifi-cantly higher than are normally encountered in .179 engine use, and this eleva-tion was expected to accelerate the fouling tendency. Moreover, the fuel flowrate was much lower than ever encountered at high temperature engine conditionsand this depression was expected to further aggravate the fouling tendency. Atthis low flow rate only the primary nozzle was flowing.

The fuel nozzles were cleaned before each test, then run for 300 minutes,with a shutdown at 100 minutes for an intermediate flow calibration. Fueltemperature was -nraedo7c-fe tescond test beccause no f-culling had becen

detected. Also, some changes in nozzle flow divider valve configuration weremade in later tests; these changes are described in Section VI-A-9.

D. Test Fuel Handling Procedures

Special procedures were followed in all of the tests to ensure that thetest fuels were not contaminated or mistakenly identified. Hand valves wereinstalled in the fuel lines near each of the test rigs for obtaining fuelsamples while a tcst was in progress.I ii

The fuels were delivered in tank trailers, as needed, and transferredinto three isolated, underground storage tanks of 40-0 3 capacity each.These tanks had previously contained only clean, light distillates. Never-theless, to assure their suitability for this program, they were first emp-tied as far as possible, using the permanently installed unloading pumps.The manhole covers were then removed, and the few inches of remaining liquidwere pumped out using a portable pump. The tanks were then inspected andfound to be in good condition with only a light, adherent coating of rust onall interior surfaces. These surfaces were washed down with a small quantityof the next test fuel, and this was then removed with the portable pump.After replacing each manhole cover, the test fuel was transferred into thetank, and a sign identifying the test fuel was placed on the switch contrl-ling the tank unloading pump. This procedure was repeated for each of thefirst 12 test fuels. The thirteenth test fuel, the diesel, was handled in asimilar manner except that it was stored in a tank trailer parked near theunderground storage tanks.

59

IF , ....- -

As fuel was needed for the high pressure tests, it was transferred fromthe appropriate tank, using a 4 m3 stainless steel tank trailer. This tankallso was drained and flushed with the next test fuel before being loaded.After filling, it was marked with the proper fuel number, hauled to the test.site, and parked adjacent to the test cell. The tank drain valve was con-nected directly to the cell system by a flexible hose, after flushing thehose with a suitable volume of test fuel. In the teat cell immediatelybefore the line entered the test rig, a valved line was teed off for drawingsamples. This location was selected because it insured that the fuelactu;?lly used in the test was being sampled.

For the tests planned, it was not intended to change fuels during atest, so the fuel sampling procedure consisted of drawing a sample justbefore the first data point was taken and again after the lest data pointwas taken. In each case, the sample container was rinsed twice with asmall portion of the fuel being sampled, before actually taking the sample.

Similar sampling procedures were followed in both the altitude relightand the fuel nozzle fouling tests. For both of these tests, fuel was trans-

ferred trom the trailer to clean drums whict were cle.arly marked and movedto the test sites. These drums were in good condition and had previouslycontained only clean materials, such as calibrating fluid. Before filling,they were drained, inspected, and rinsed with the next test fuel.

Pretest samples taken at the test sites were returned to Wright-Patterson Air Force Base for verification of significant characteristics,to determine whether fuel quality had been compromised during storage orhandling at the several test sites. Analyses included density, viscosity,surface tension, and vapor pressure. These analyses were performed byMonsanto Research Corporation. I

A compilation of these data is shown on Table 14. From a con.parisonof the properties of the original samples with those of samples returnedfrom the several test sites, it is apparent that no significant change infuel properties occurred. Therefore, it was concluded that fuel handlingprocedures were satisfactory, and analysis of samples of the remainingtest fuels was considered unwarranted.

60

60

-~~~~~ p tu .- .-

'.0 '. LA 'C' I *q

'-34

su 0

4.1U) 'n (n4' LA 0 0C

04 -14 oIc on

V4 -4.1

4-4

J.'A 0 'd 00 ( 00.4~C (N C-4 (N4 F - (A IA

Em ;' .4 -: -: .44 4-

'I

E. Data Analysis Procedures

Generally standard data reduction and presentation techniques wereemployed Key parameters and calculation procedures are Indicated inTable 11 and Appendix A. Some additional special procedures are describedin the following sections.

1. Fuel Property Correlation Procedures

Analyses of the experimental test results were conducted to: (1) cor-relate the performance and emission parameters with combustor operatingconditions; (2) as appropriate, correct the measured rig data to true stand-ard day engine operating conditions; and, (3) correlate the corrected datawith the appropriate fuel properties from SectLon I1i. To illustrate theprocedure, the CO emission data is outlined below.

Inspection of the CO emission data for the first two high pressuretests (JP-4 fuel, Table A-1) showed that as in .dference 7 the data were

Elc % ( [exp k - ke SCO (5)CO~~ Co4.

where the combustor operating parameters (Vr, P 3 , and T3 ) have been normal-ized to engine operating conditions at idle. Multiple regression techniqueswere employed to detenrine the constants k0 , kl, and k2 . A very good corre-lation was found, as shown in Figure 32. Additional analyses showed thatin general k0 was slightly fuel dependent, but kl and k2 essentially not.The CO emission operating nr severity parameter was then taken to be:

0. 2SCO - 1 (6)

which is tabulated in Table A-2, and was used as shown in Table A-3 tocorrect measured test data to true engine operating conditions by therelationship:

El El ( COi engine(7CO engine -ECO test S t

62

The correction was usually very small except at dash operating conditions

(75 percent density in rig tests) as shown in Table A-3. Corrections werethen tabulated (Table 15, for example) and plotted against appropriatefuel properties (Figures 33 and 34, for example). Equations for the effectof fuel hydrogen content on CO emissions shown in Figure 33 are the result

of regression analyses, and show, for example, that

-1.38

E cruise 14.8 H g/kg

with a correlation coefficient (r) of (-0.753).

2. Combustor Life Prediction Procedures

Field experience shows that thk, J79 crnmhbutor is life-limited by lowcycle fatigue crack propagation from the ends of the puniched louvers on rheinner liner. From the measured metal temperature rises, the stresses that

are calculated, including stress concentration factors at these cracks, arewell above the yield strength of the material, resulting in plastic yield-lng on each cycle. Since the metal temperatures are in the range where

creep can occur below the yield stress, additional plastic deformationoccurs beyond that which corresponds to relaxation to the yield stress.This total plastic deformation correlates reasonably well with low cycleJ fatigue life in many applications.

"The j79 COllubustLr, d _v..,,d before 1960, has a relatively short lift,and is less amenable to detailed analysis than are more recent combu.'hutrdesigns. The punched louvers of the J79 combustor liner have cracks alreadyinitiated at the ends of the punch when the part is new. This, crack resultsin a severe stress concentration region at which crack propagation proceeds.t..

Also, the thermal gradients in the vicinity of the ends of the louver punchare very high because of the presence of tdie fresh introduction of1 film airat this point creating a sudden transitilon from hot metal to highly coledmetal. Total life is influenced by changing stress levels as the cracks-propagate and changing metal temperatures due to metal distortion.

63

63

........................................................................................-, i .1

Table 15. Summary of CO Emission Test Results.

CO Etmis;ion Index, g/kg

Number Idle Cruise Takeoff Dash

1 68.2 16.5 5.7 2.4

IR 63.8 14.2 4.8 2.6

2 73.2 17.3 5.8 2.9

3 71.2 15.8 4.2 2.2

4 72.4 18.9 4.8 2.5

5 73.4 16.1 4.7 2.6

41 Z. v .11.- Q 1) U7 75.5 18.5 3.6 1.5

8 75.9 20.5 3.4 1.7

9 69.7 18.0 3.7 1.9

10 66.2 17.7 4.5 3.7

11 63.4 15.1 3.1 1.6

12 57.0 13.7 4.3

13_ 68.0 17.0 3.0 1.5

64

cv cli i tert ktsimple lovrseimos n h presence ) hragradients. have been econducted to Provide basic data for estimatiit;lu tiltliife of puneht'd louver coutbustors . Iin best' teats. tht v1rack propagq-atedlin approximately direct prtoport ton to the numbe~r of thermiztl exposure'Cyt' Its. Basted onl t hese I otuVt1r spec iml ttn id mU tugs and Ilow e ye It' fa t i gkiMaterial p'ropertit's, the detsign ('tlVVt' shown1 inl Figut'" 21 is inl use atGenera1 EIec t ric fot-r t he 17) ct'mibusvtor 11 iner1, f or e i thler tait LIgne fit' toinitial cracking or for crack plropa~gat ion rate'. Fromt this curve it Callbe test imat ed that a 4S K chanige iii litter temperx'ture, re~sults in aboutafact or of two cliangt' in lifIe.

Dci initivi' t'~xp-vrienct't Irom 179 engint's o' 'hstor ij li cits Witliidifferent fuels or with diffterent metal tt'mpeture does nt t'xist . 1~iI -Iferencve inl I ft' of about 2 , percen-It haS h)OC'n t1)`~'hcVCd be~tween Navy% (31P-S)and Air Force (31-4) oper~it jolalI exper jllon It is be hI ikxcd t' I litd dif ter-tuces inl operat ionalI tvs~lgk' are as 1 argi' all i of lut'ncek. tin liltt as is thli Ikedi4fference anid vmdv the1Lr~t'frt', mask any' at tt'mpt to attribuite this obst'r\'o'dlift' difterence to the itil. type, alone., llowvvcr, a 2') percentt lift' rcdhii't ionl is p~redicted b"' the, trend(s sh1own1 in1 Figure-L 29 folr a1 mI ll I 'tpetatnrchangeiL of 20K, R. ifllti bsed on1 th Ilk;& t a it1'. Re I o renecct'' , t~l 1 8 Iý ap l', ox iti it' INvItltt met al t'mpt'ra ni' change expected fot 'it cangc in ivti h\krog'.t't'l Clt 'nt kiot14 .9' to 13. 8 ptercent.

The aft combus tor' liner hias much I llenor I i ft' t han dot's hei I inn'er Iit i' 1 2,and it is not affected by fueil type' It wats nottet itt Rt-:t'renc, t, that tilt'metal temperaturtes inl the al t po~rt ion oft tilt' outer OV rear I iner, wt'rt', notat'tected by the lumi nos it t'chal nges- o f fulelIs thailt d id i f fet'(t tilt ups t retaminlner I lne r . 'Thtese data wet'Et not It' t aIL>'1 p~rt'tt oell t'din1 Rt' I'ttc' 6,el so t 0It'>'are presentutd hiero in Fi"Igurt' 30. 'l't'irlpt'rcl nix'Tasurt'nlcittt wo'rt, nott I aý'ttn illt tils a ft I ifilt'E re g ion ini thIi s p)r 4,, t's t program ht'-aus, kit !his;xerltttandt no dii terencet in atl t Iintr lift' withintl t'elt'g' iS p1 t'tlictetj.

3, Turhbit' L~if Pt lrediet ion P~rocedut tx.

If aF 1t ternatt' fut'l s ct' ~ated subst ant i a tilauiges in I t'flpt'rat urt' pat tetrnfact or or tempjerat ure' Prof i It' inl tilt contbs t or e'xit gast's , t'ilanges in ti,1-bine comtpontent 1i ft' would ht' predicted. how1ever , as discusst'd in Stect ionVL-A-6 nto cliange:s were founid in theste patramelters. As iitl icat ed above inFigure 30, ht'll aft 'liner1 t'mpt'Iratures whichd are wellI away from thlit hiighitflamte luminosity we're uinaf fected b' tilt' same fuel CIZchngt'S t hat a ft tV4'' IcsignifCicant ly the forward l inetr; theret on', tit' st ator \'&tflt' ill thei ttrit'hien

65

-A,

10' -

444

Sgo ýIý,

300 80

29. Ef ec of T m tr t*1Gr d v E tl

6I

O'V

Qo L0 +

I.

,r4

U(5 " !'.4'n 0)"

0 -n 0

0.''-

I I

O~I.?.

diaphragm which are even further away from these flame luminosity Offectsare also expected to be negligibly affected. Figure 31 illustrates the%t ry small view the stator leading edge has of the l'minous flame region.This small view factor results in regliglblc effects in the J79 engine.However, in other engines with annular combustors and shorter combustors,large view factors exist anc ifxame .luminosity may in some cases aecomesignificant to the vane leading edge temperalure.

I,.'

co~

144

00o5

SECTION VI

RESULTS AND DISCUSSION

All planned test series (44 total) were completed and no major problemswere encountered. In general., results were well ordered and consistent withpublished data insofar as comparisons could be made. Detailed test results,which are listed in Appendices A through E. are summarized and discussed in thefollowing section. Engine system lite prediction analyses based on these resultsare then presented in Section VI-B. ?urther, an overall assessment of thesetests and analyses is presented in Section VI-C.

A. Experimental Test Results

Fourteen high-pressure rig tests were conducted to obtain the performance/emissions/durability data. These data are listed in Appendices A and B andsummarized in Sections VI-A-I through VI-A-7. Fourteen low pressure rig testswere conducted in parallel to obtain the ground start and altitude relight data. IThese data are listed in Appendix C and summarized in Sections VI-A-7 and 8.Also in paralle], 18 fuel nozzle fouling tests were conducted to obtain the datalisted in Appendix D and summarized in Section VI-A-9.

1. CO and HC Emissions

Carbon monoxide (CO) and unburned hydrocarbons (KI) are both products ofincomplete combustion, and are, therefore, generally highest at low poweroperating conditions (idle and cruise). Figure 32 shows the strong effect ofcombustor operating conditions on CO emission levels with JP-4 fuel. At idleoperating conditions, the CO emission index is about 64 g/kg which corresponds toa combustion inefficiency of about 1.5 percent and is in good agrteement withengine measurements. At cruise, takeoff, and dash operating conditions, the COemission levels are approximately 29, 7, and 3 percent, respectively, of the idleCO emission level, which indicates the strong effect of combustor inlettemperature and pressure on combustion reaction rates, and hence, on combustionefficiency and CO emission levels. These pressure and temperature effects,determined by multiple regression curve-fit techniques of these data, are in goodagreement with previous results (Reference 7). Within these test limits, noeffect of fuel-air ratio is evident. Moreover, very good repeatability between

the two combustor assemblies and test runs is indicated.

Carbon monoxide emission results very similar to those shown in Figure 32 were•obtained with each of the other fuels, and all of the results are listed inTable 15. The effect of fuel hydrogen content on CO emirsion levels at

each of the power levels is shown in figure 33. At takeoff and dash operatingconditions CO levels are very low and virtually independent of fuel hydrogencontent, and any other fuel property. At cruise operating conditions a

significant fuel hydrogen effect (negative 1.38 exponent) is indicated, but noeffect of other fuel properties (aromatic type or base fuel) is evident. Atidle operating zonditions, a relatively weaker fuel hydrogen effect (negative0.47 exponent) is indicated, but the correlation is poor and other fuel property

70

200 T1I~SCombustor Assemnbly No. 1

100 ~ Cumbustor Assemibly No. 210*- JP-4 FitelI

* metered 7 to 16 g/kg ,

Oc 50

IT I eA

(n 10-

wlnt Ta e fk r u is e!c-- -5--

\_ -y 4x H.

4Fi-'ur-i 32. Ft i ect of Clpa dt jg Condit itoNi; onDid, Lei-l

71

Symbol 0 0

Base Fuel JP-4 JP-8 2-D

Tailed Symbols: Monocyclic AromaticBlends

80- _ - ---0

0'00'01

60 In

to Idle, y 65 ,9 - 0 '

"a Cruise, y = 14.8 / x

.o Takeoff, y 4.5 +

• • o o • , . • • . ( • ) 0 .0 7 -40 Dshy 2.2 x

~~~T 40- Dah5(j 0

20

S0-

11 12 13 14 15

Fuel Hydrogen Content, weight percent

Figure 33. Effect of Fuel Hydrogen Content or, CO Emission Levels.

72

effects are evident. In particular, the lowest idle CO levels are obtained with]the JP-4/monocyclic aromatic blends, suggesting a volatility or atomization effect.Figure 34 shows these same idle CO data plotted (without -egaid to hydrogencontent) against relative spray droplet size (from Table 7) or fuel 10 percentrecovery temperature (from Table 3), the latter being one of the more commonlyused indicators of fuel volatility. It appears that either of these parameterscorrelates the idle CO data better than does hydrogen content. Both of theseproperties can be expected 1,o affect idle emissions, but for thes

EFFECTS ON J79 ENGINE COMBUSTION SYSTE.9. J79 Engine Combustor, Pictorial View. 25 10. J79 Engine...

Documents

Transcript of EFFECTS ON J79 ENGINE COMBUSTION SYSTE.9. J79 Engine Combustor, Pictorial View. 25 10. J79 Engine...