of a modern counterrotating unducted fan engine … · 2013-08-30 · of a modern counterrotating...
TRANSCRIPT
:. 2J,
CR-180869
N/ SANational Aeronautics and
Space Adm_istratlon
FULL SCALE TECHNOLOGY DEMONSTRATIONOF A MODERN COUNTERROTATINGUNDUCTED FAN ENGINE CONCEPT
ENGINE TEST
December 1987
by
GE Aircraft Engines
GE36 Project Department
Cincinnati, Ohio 45215
Prepared for
National Aeronautics and Space Administration
UNjucrt q CA_,t _-r, lC;,lN,_ C'_NC/_I. rh, qJ-N_ l" .,T
(".i::) 3"*(.' p Cq_L _£.c oncl,Js,]__/07 07371S-,
I These limitations shall be considered void alter two (2) years after date or such data.This legend shall be marked on any reproduction of these data in whole or in part.
NASA-Lewis Research Center
Contract NAS3-24210
https://ntrs.nasa.gov/search.jsp?R=19900000733 2020-04-08T18:32:32+00:00Z
u
,w,i
i
.m/
i
!i
lut
___J
i
.._1.1
Zi
ill
CR-180869
N/ ANational Aeronautics and
S_)ace Administration
FULL SCALE TECHNOLOGY DEMONSTRATIONOF A MODERN COUNTERROTATINGUNDUCTED FAN ENGINE CONCEPT
ENGINE TEST
FOREWORD
This report presents the results of engine tests and
a discussion thereof, as conducted by GE Aircraft Engines,
Cincinnati, Ohio. These engine tests were performed on
behalf of the NASA Lewis Research Center, Cleveland, Ohio,
under Contract NAS3-24210. The program was carried out
under the technical cognizance of Mr. R.D. Hager of the
Advanced Turboprop Project Office. The contract effort
was conducted at the Evendale Plant of GE Aircraft Engines
by the GE36 Project Department.
PRECEDING PAGE BLANK NOT FILMED
!
iii
TABLE OF CONTENTS
Section
1.0
2.0
3.0
4.0
ENGINE TEST SUMMARY
I. I SIGNIFICANT HARDWARE CHANGES AND MODIFICATIONS
1.2 OVERALL HISTORY OF UDF TM ENGINE TESTING
INTRODUCTION
2.1 ENGINE DESCRIPTION
2.2 INSTRUMENTATION
REVIEW OF UDF TM TESTING
3.1 FORWARD STATIONARY CARBON SEAL FAILURE
3.2 FORWARD TELEMETRY ANTENNA REPAIR
3.3 TELEMETRY SYSTEM TEMPERATURE PROBLEMS
3.4 STAGE 1 TURBINE BLADE FAILURE AND STALL EVENT
3.5 TURBINE BLADE DAMPER EFFECTIVENESS
3.6 SUBIDLE OIL LEAK PROBLEM
3.7 STATIONARY EXHAUST NOZZLE (CENTERBODY) REPLACEMENT
3.8 FAN BYPASS BLEED VALVE CALIBRATION
3.9 FAN BYPASS BLEED VALVE DIFFUSER FAILURE/REPLACEMENT
3. I0 PROPULSOR FAN-BLADE-LOSS EVENT
3.11 PROPULSOR FAN BLADE HISTORY AND TEST DATA
3.11.I Fan Blade Test History3.11.2 Fan Blade Test Data
3 12 POWER TURBINE FRAME STRESS
3 13 EFFECT OF VORTEX DESTROYER ON STRESS
3 14 EFFECT OF TEST SITE CHANGE ON STRESS
3 15 ROTOR LOCKUP AFTER SHUTDOWN RESULTING FROM FUEL LEAK
3 16 LOW/HIGH CYCLE FATIGUE TESTING
3 17 ROTOR-TO-ROTOR LOCKUP/PROPULSOR DISASSEMBLY AND REBUILD
3 18 MISCELLANEOUS HARDWARE STRESS DATA
3.19 OIL LEAK/GULPING PROBLEM
HEAT TRANSFER AND SECONDARY FLOW SYSTEM
4.1 MIXER FRAME HEAT TRANSFER
4.2 POWER TURBINE SECONDARY FLOW SYSTEM
V
'! ..... , .k ,* r TM - _- ....... -r ...... _i! . [;'_
1
6
7
II
II
18
19
20
20
20
32
37
51
52
52
52
57
6O
6O
6O
73
73
8O
85
85
85
99
101
II0
II0
117
TABLE OF CONTENTS (Continued)
Section
7.0
8.0
9.0
4.3 POWER TURBINE HEAT TRANSFER
4.4 NACELLE VENTILATION
4.5 CENTER CAVITY VENTILATION
ENGINE SYSTEM DYNAMICS
BEARINGS AND SEALS
6.1 HARDWARE CONDITION
6.1.I Rotor Support Main Bearings
6.1.2 Inner Actuation Bearings
6.1.3 Actuation Tapered Roller Bearings
6.1.4 Fan Retention Bearings and Seals
6.2 MEASURED TEMPERATURES
6.2.1 Rotor Support Main Bearings
6.2.2 Fan Retention Thrust Bearings
PERFORMANCE
DATA SUMMARY
DATA REDUCTION METHODOLOGY
7.1
7.2
7.3 COMPARISON OF PRETEST PREDICTIONS AND REDUCED
TEST DATA
7.4 LCF CYCLIC TESTING AND RESULTANT DETERIORATION
7.5 DEFINITION OF PREFLIGHT TEST CYCLE
7.6 COMPARISON BETWEEN GROUND PRETEST CYCLES AND
REDUCED TEST DATA
7.7 PERFORMANCE SUMMARY
7.8 KEY RESULTS
ENGINE OPERABILITY
8.1 IPC STALL MARGIN
8.1.1 IPC Bleed Control System8.1.2 Control Threshold for Throttle Retard Rate
8.1.3 Maximum Allowable IPC Pressure Ratio Schedule
8.1.4 Final IPC Bleed Control Status
8.2 TRANSIENT TESTING EXPERIENCE
8.3 ENGINE TRANSIENT PERFORMANCE ANALYSIS AND PREDICTIONS
ENGINE CONTROL
Page
121
121
127
129
143
143
143
147
147
147
151
151
151
156
156
156
160
174
181
181
185
185
192
192
197
197199
199
2O2
202
208
vi
TABLE OF CONTENTS (Concluded)
Section
10.0
II.0
12.0
13.0
14.0
9.1
9.2
9.3
9.4
BUILD 1 - ENGINE CONTROL TESTING
9.1.1 Speed Sensing Anomaly9.1.2 Pitch Control
9.1.3 Gas Generator Control
9.1.4 Throttle System
9.1.5 Engine Starting
9.1.6 Off-Engine Harnesses
9.1.7 Lube Oil Bypass
BUILD 2 - ENGINE CONTROL TESTING
9.2.1 Speed Control9.2.2 Duct Bleed
9.2.3 Transient Testing9.2.4 Control Parameters
9.2.5 Vibration System
9.2.6 Overspeed System
9.2.7 Reverse Testing
9.2.8 Control System Modifications
BUILD 3 - ENGINE CONTROL TESTING
Endurance Testing
Core Response Modifications
Reverse Thrust Testing
UPS (Uninterruptable Power Supply) Systems
Transient Testing
SUMMARY
9.3.1
9.3.2
9.3.3
9.3.4
9.3.5
NACELLE STRUCTURES
10.1 FAN BLADE AIRFOIL LOSS
10.2 STRAIN EVALUATION OF THE STRUT
10.3 ACOUSTIC FATIGUE
RESULTS
CONCLUSIONS
SYMBOLS/ABBREVIATIONS
REFERENCES
APPENDIX A - STEADY-STATE TEST DATA - BUILD 3
APPENDIX B - STRAIN DATA FOR THE STRUT
208
208
208
2O8
211
211
211
211
211
212
212
212
219
219
219
219
219
224
224
224
224
224
226
226
231
231
235
235
239
241
243
246
247
279
vii
LIST OF ILLUSTRATIONS
I-I.
1-2.
I-3.
I-4.
2-I.
2-2.
2-3.
2-4.
2-5.
3-1.
3-2.
3-3.
3-4.
3-5.
3-6.
3-7.
3-8.
3-9.
3-10
3-11
3-12
3-13
3-14
3-15
3-16.
3-17.
3-18.
3-19.
Total Run Time at Various XN48, Times as of July 8, 1986.
Total Run Times at Various XN49, Times as of July 8, 1986.
Total Run Time at Various T46, Time as of July 8, 1986.
Run Time at Various Thrust Levels, Build 082-001/03 Only.
Cross Section of Unducted Fan Engine.
Cross Section of UDF TM Engine (Enlarged View).
UDF TM Engine at Peebles Test Site 4A.
UDF TM Engine at Peebles Test Site 4A.
UDF TM Engine at Peebles Test Site 3D.
Carbon Seal Assembly Fix.
Forward Telemetry System Support.
GE36 Telemetry System.
Telemetry System Scoop Designs.
Forward Telemetry Temperatures - Raw Data.
Forward Telemetry Temperatures - Normalized with TAM B.
Aft Telemetry Temperatures - Raw Data.
Aft Telemetry Temperatures - Normalized with TAM B.
Forward Telemetry Temperatures - Reverse Testing.
Aft Telemetry Temperatures - Reverse Testing.
Summary of Hardware Damage.
Turbine Blade IF Stress Trend.
History of Turbine Blade Response, Run to Thrust and Stall.
Turbine Blade Fix.
Power Turbine Blade Vibratory Response, Damper
Effectiveness - Stage I.
Power Turbine Blade Vibratory Response, Build I (Undamped)
Versus Build 2 (Damped).
Stage 1 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage 2 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage.3 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
2
3
4
5
12
13
14
15
16
21
22
24
25
26
27
28
29
30
31
33
34
35
38
39
40
41
42
43
viii
3-21.
3-22.
3-23.
3-24.
3-25.
3-26.
3-27.
3-28.3-29
3-30
3-313-32
3-333-34
3-353-36
3-373-38
3-39
3-40
3-41.
3-42.
3-43.
LIST OF ILLUSTRATIONS (Continued)
Stage 4 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage 6 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage 7 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage 8 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage 9 Turbine Blade Response, Accel to 25,000 ibf Thrust:
February 4, 1986.
Stage 10 Turbine Blade Response, Accel to 25,000 Ibf Thrust:
February 4, 1986.
Stage II Turbine Blade Response, Accel to 22,000 ibf Thrust:
January II, 1986.
UDF TM Engine Centerbody Separation Parameter, Cruise
Condition M = 0.72; 35,000; PT8/P 0 = 1.4.
UDF TM Engine - Redesigned Centerbody.
Bypass Bleed Valve Diffuser.
Redesigned Bypass Bleed Valve Diffuser.
Spar Schematic.
Total Run Until Fan Failure.
Original Fan Blade Airfoil Mechanical Design.
Improved Fan Blade Mechanical Design.
Stage 2 Fan Blades Engine Test History.
Fan Blade Structural Improvement.
Fan Blade Response Summary - ksi pp.
Stage I Fan Blade Response Summary.
Stage 2 Fan Blade Response Summary.
Fan Blade Response Summary - February 4, 1986 Maximum
Power Run (25,000 ibf Corrected Thrust).
Stage I Fan Blade Response, Accel to 24,000 ibf:
June 24, 1986.
Stage 2 Fan Blade Response, Accel to 24,000 ibf:
June 24, 1986.
Fan Blade Strain Gage Locations.
ix
44
45
46
47
48
49
50
53
54
55
56
58
61
62
63
64
65
66
67
68
69
70
71
72
LIST OF ILLUSTRATIONS (Continued)
Figure
3-44.
3-45.
3-46.
3-47.
3-48.
3-49.
3-50.
3-51.
3-54.
3-55.
3-56.
3-57.
3-58.
3-59.
3-60.
3-61.
3-62.
3-63.
3-64.
3-65.
3-66.
3-67.
3-68.
3-69.
Fan Blade Response with Facility Fans Turned On.
Power Turbine Frames Vibratory Response Summary.
Power Turbine Frames Vibratory Response, Updated Scope
Limits Versus Site 4A Response.
Power Turbine Frames Vibratory Response, Stage 12 Component
Strain Distribution Test Results.
Power Turbine Frames Vibratory Response, Material HCF
Test Results.
Power Frame Vibratory Response, Build 3.
Fan Blades Vibratory Response, Effect of Vortex Destroyer.
Power Turbine Frames Vibratory Response, Effect of Vortex
Destroyer.
Fan Blade Response Comparison, Site 3D Versus Site 4A.
Power Turbine Frames Vibratory Response, Effect of
Site Change.
Suspected IGV/Spool Rub.
Low Cycle Fatigue Cycle.
IGV and 1-4 Inner Spool.
IGV Crack/Tear at Braze Joint (ALF).
Suspected Stage I Rub.
Stage One Blade Cracks at Leading Edge Root.
Aft Outer Flowpath Seals.
Midsump Seals.
Suspected Stage II Rub.
IGV and 1-4 Inner Spool.
Power Turbine OGV - Engine Test Data.
Power Turbine IGV - Engine Test Data.
Forward Outer Rotating Seal Response, Decel from 25,000
ibf Thrust: February 4, 1986.
Aft Outer Rotating Seal Response, Accel to 25,000 ibf
Thrust: February 4, 1986.
Outer Turbine Spool Response (1-4), Accel to 25,000 ibf
Thrust: February 4, 1986.
Outer Turbine Spool Response (7-11), Acceleration to
35,000 ibf Thrust: February 4, 1986.
Page
74
75
76
77
78
79
81
82
83
84
86
87
90
91
92
93
95
97
98
I00
102
103
104
105
I06
107
LIST OF ILLUSTRATIONS (Continued)
Figure
3-70.
4-3.
4-4.
4-5.
4-6.
4-7.
4-8.
4-9.
4-10.
4-11.
4-12.
4-13.
4-14.
5-1.
5-2.
5-3.
5-4.
5-5.
5-6.
5-7.
5--8.
5-9.
5-10.
5-11.
5-12.
Inner Turbine Spool (6-11) Response, Accel to 25,000 Ibf
Thrust: February 4, 1986.
Mixer Frame Flows at Takeoff Power, Test Versus Predicted.
Mixer Frame Pressure Losses Comparison of Measured and
Predicted Pressures.
Mixer Frame Flowpath Static Pressure Distribution Takeoff
Power.
Mixer Frame Backflow Margin - Build 01 Engine.
Mixer Frame Backflow Margin - Build 02 Engine.
Mixer Frame Temperatures.
Power Turbine Secondary Flow Circuit.
Potential Forward Seal Problem.
Rotor Cavity Temperatures at Takeoff Power.
Power Turbine Temperatures Based on Analysis.
Spool Temperatures, Hot Day Takeoff.
Nacelle Temperatures Based on Test Data at Takeoff Power.
Forward Fan Blade Hub Temperatures; Hot Day Takeoff,
Mach= 0.0, Full Thrust.
Center Cavity Ventilation at Takeoff Power.
IPC One-Per-Rev Signature Comparison.
HPC One-Per-Rev Signature Comparison.
IPC One-Per-Rev Signature Comparison.
HPC One-Per-Rev Signature Comparison.
Subidle UDF TM One-Per-Rev Vibration Signature Comparison.
Subidle UDF TM One-Per-Rev Vibration Signature Comparison.
UDF TM One-Per-Rev Vibration Signature on 2-Minute Accel to
24,000 ib Thrust, Housing Vertical.
UDF TM One-Per-Rev Vibration Signature on 2-Minute Accel to
24,000 ib Thrust, Housing Horizontal.
Maximum Static Deflection.
Unbalance Forces Lineup in Vertical Plane.
Unbalance Forces Lineup in Horizontal Plane.
GE36 Demonstrator Engine 082-001/3 T/D 7-1-86 Time History
Plots at Ground Idle Power.
Page
108
111
112
113
114
115
116
118
120
122
123
124
125
126
128
130
130
132
132
133
133
135
135
137
139
140
141
xi
LIST OF ILLUSTRATIONS (Continued)
_re
6-I.
6-2.
6-3.
6-4.
6-5.
6-6.
6-7.
6-8.
6-9.
6-10.
6-11.
6-12.
7-I
7-2
7-3
7-4
7-5
7-6
7-7
7-8
7-9.
7-10.
7-11.
7-12.
7-13.
7-14.
7-15.
7-16.
7-17.
7-18.
7-19.
Seal Cavity Closure.
Possible Running Condition with IR Spanner Nut Loose.
Disassembly Problem and Roller Corner Damage.
Problem Solution - Positive Nut Locking.
Gearbox Stackup Problem Causing Incorrect Loading of
Ball Bearing and Cap Looseness.
Bearing Housing Deflection.
Irregular Wear Pattern of Roller "Path."
Main Propulsor Bearings.
Main Propulsor Bearings, 2-Minute Accel to 24,000 ib Thrust.
Main Propulsor Bearings, 4-Minute Accel/Decel to 1200 rpm.
Forward and Aft Fan Hub Temperatures.
Forward and Aft Fan Hub Temperatures, Accel to Full Power.
Instrumentation for UDFm/F404 Demo.
Core Engine Temperature Ratio Versus Pressure Ratio.
IPC Stall Margin Versus IPC Corrected Flow.
Power Turbine Flow Function Versus Energy Function.
Power Turbine Efficiency Versus Energy Function.
UDF TM Thrust Coefficient Versus Power Coefficient.
Front Rotor Blade Angle Versus Core Engine Pressure Ratio.
Rear Rotor Blade Angle Versus Core Engine Pressure Ratio.
Corrected SFC Versus Corrected Thrust.
UDF TM Engine - Redesigned Centerbody.
UDF TM Engine Centerbody Separation Parameter.
UDF TM Engine - Core Plug Static Pressures.
Nozzle Flow Coefficient Versus Core Engine Pressure Ratio.
Corrected SFC Versus Corrected EPR Power Hook Evaluation.
Corrected SFC Versus Corrected EPR Power Hook Evaluation.
Wind Velocity Versus Wind Angle.
Corrected SFC Versus Corrected Thrust.
Corrected Thrust Versus Corrected Pressure Ratio.
Core Temperature Ratio Versus Control Pressure Ratio.
Page
145
145
146
146
148
150
150
152
153
153
155
155
158
161
162
163
164
166
167
168
169
170
171
172
173
175
176
177
178
179
180
xii
Figure
7-20.
7-21.
7-22,
7-23.
7-24.
7-25.
7-26.
7-27.
8-I.
8-2.
8-3.
8-4.
8-5.
8-6.
8-7.
8-8.
8-9.
8-I0.
8-11.
9-1
9-2
9-3
9-4
9-5
9-6
9-7
9-8
9-9
9-10.
9-11.
LIST OF ILLUSTRATIONS (Continued)
IPC Efficiency Versus Control Pressure Ratio. 182
IPC Flow Versus Control Pressure Ratio. 183
IPC Flow Versus IPC Corrected Speed. 184
Compare Pretest Cycle Versus Posttest Cycle Versus Test Data. 186
Compare Pretest Cycle Versus Posttest Cycle Versus Test Data. 187
Compare Pretest Cycle Versus Posttest Cycle Versus Test Data. 188
Compare Pretest Cycle Versus Posttest Cycle Versus Test Data. 189
Compare Pretest Cycle Versus Posttest Cycle Versus Test Data. 190
IPC Operating and Stall Line Sea Level Static. 193
Steady-State IPC Operating Lines; 2,750 ft. 194
Steady-State IPC Operating Lines; 38,000 ft. 195
IPC Stator Schedule. 196
Predicted IPC APRS Usage During Throttle Decels; 2,750 ft. 198
Predicted IPC APRS Usage During Throttle Decels; 38,000 ft. 200
Maximum Allowable IPC Pressure Ratio. 201
Predicted IPC APRS Usage During Throttle Decels; 2,750 ft. 203
Predicted IPC APRS Usage During Throttle Decels; 38,000 ft. 204
Transient Testing Analytical Comparison with Updated
Transient Cycle Deck T/C from 97% Takeoff Thrust. 206
Transient Testing Analytical Comparison with Updated
Transient Cycle Deck Burst to 20,000 ib Thrust. 207
Fan Speed Sensor Segments. 209
Fan Speed Sensor Segments. 210
Duct Bleed System. 213
Throttle Chop Decels. 214
Throttle Chop. 215
Throttle Chop. 216
Throttle Chop. 217
Throttle Chop. 218
Reverse Thrust. 220
Reverse Thrust. 221
Reverse Thrust. 222
xiii
LIST OF ILLUSTRATIONS (Concluded)
Figure
9-12
9-13
9-14
9-15
9-16
9-17
10-1.
Reverse Thrust.
PS3 Accumulator Configuration.
Throttle Burst.
Throttle Burst.
Throttle Burst.
Throttle Burst.
UDF TM Engine 082-001Accelerometer Definition.
223
225
227
228
229
230
232
xiv
LIST OF TABLES
Table
1-1.
1-2.
1-3.
3-1
3-2
3-3
4-I
7-1
7-2
8-1
I0-I
10-2
10-3.
GE36 Test History - Engine 082-001, Build I/IA.
GE36 Test History - Engine 082-001, Build 2.
GE36 Test History - Engine 082-001, Build 3.
Power Turbine Vibratory Response Resolution.
GE36 HCF Cycle Count.
Turbine Blade Inspection Summary.
Power Turbine Secondary Flow (%W25), Takeoff Power.
Data Summary.
Chronology of the Acquired Data.
Significant Transients - GE36 Proof-of-Concept.
Vibration Response.
UDF TM Engine Mount Loads Due to Airfoil Loss.
UDF TM Engine Comparison of Mount Loads for
Airfoil-Out Event.
UDF TM Engine Airfoil-Out Event Comparison of Deflections.
Stress Data - Strut.
Page
8
9
I0
80
88
96
119
157
157
205
233
234
236
236
237
XV
1.0 ENGINE TEST SUMMARY
This Engine Test Report covers the UDF TM (unducted fan) Engine 082-001
ground testing at the GE Aircraft Engines Peebles Test Facility and includes
Builds I through 3 of Engine 082-001. Note: A new build number indicates a
significant change in engine hardware.
The UDF TM engine successfully completed a ground-test program exceeding
100-hours duration. The basic concepts of the engine have been successfully
demonstrated. Some of the accomplishments are as follows:
• Full thrust (25,000 pounds corrected) demonstrated
• Full propulsor rotor speed demonstrated (1393+ rpm)
• Specific fuel consumption (sfc) was better than predicted; sfc
of < 0.24 ib/hr/ib was demonstrated
• Flawless operation of the F404 gas generator
• Counterrotation of structures, turbines, and fan blades
• Actuation system operation successfully demonstrated
• Control system operation successfully demonstrated
• New fan blade design successfully demonstrated
• Reverse thrust successfully demonstrated.
The UDF TM was ground tested at Peebles for a total run time of 100:51
hours. This was split between Builds I through 3 as follows.
Run Time Total
Build per Build Run Time
1 5:24 5:24
2 29:20 43:44
3 66:07 100:51
In Figures I-I through I-4, the engine run time is a function of Stage 1
propulsor fan speed (XN48), Stage 2 propulsor fan speed (XN49), exhaust gas
temperature (T46), and thrust, respectively. Thrust data were available for
Build 3 only. Note that on the plots of run time as a function of propulsor
_N
0 0 0
a_ 0i li
NND
u')
NNNNNN
o _
¢,D
O
(spuesnoq.L) selnU!lN 'euJ!.L
E°_,.tp_
..1"
0
tO:>
_00
p_
_J0 _,_
E-- 0
!
¢;
_D
0
0 0 0
II II II
BNN
o I I I I I I I I i I I I I I
&
&
I
I
I0
0
(spuesnoq.L) selnU!lN 'eLuu.
_3
_JE
=-..-r
o
:>
¢o
0J oO
•pi r-i
_J
p_ o
I
_J
o_
3
I:
= ck
_ _i mini _ _i ,--,-- ,- ,- ,- ¢i ¢i ¢i ¢i
(spuesnoq.L) se_,nu!lN 'emu.
oOC_
0
u_
E_
ul
0
<11_>
QJ
P_
0
I
_J
_D
O_
cO_m
O_O
OO
!(_i¢OO
"OmmaiI
m
¢N
T"
,-: ,z ,_ ,- ,- (5 (5 (5
04,im
!aO
¢O!
O
OI
O
I
(spuesnoq£) salnuflN 'auu!l
O
O
OO
Ic-400O
u_
_J
QJ_J
¢n
J:
O
_>
0J
b.
I
speed, there is a large amount of time spent in the ranges of 600 to 700 rpm
and 700 to 800 rpm. Most of this time was spent in the immediate vicinity of
700 rpm, which is idle for this engine. Also note that the vast majority of
time spent in the 1400 to 1500 rpm range was actually spent near 1400 rpm.
i.i SIGNIFICANT HARDWARE CHANGES AND MODIFICATIONS
Significant hardware changes and modifications which occurred during the
UDF TM ground test are listed below. A more detailed discussion is provided in
the sections enumerated.
Build I to IA
Forward stationary carbon seal and 2R bearing replaced
Build IA
Forward telemetry system antenna modified
Added pipe-elbow air scoops to exhaust nozzle to aid aft
telemetry system cooling
Build IA to Build 2
All propulsor turbine blades replaced (Stages I-4 and 6-11)
Propulsor turbine blade tip clearances increased
Damper pins added between all propulsor turbine blades
Spool distress repaired
Pipe-elbow scoops were replaced with aerodynamic air scoops
Leading edge plugs were installed in propulsor fan blades
Installed redesigned stationary exhaust nozzle (centerbody)
Installed hardware to solve subidle oil leak problem
Build 2
Propulsor Stage 2, No. 7 fan blade replaced after blade-out
Added additional holes to telemetry system air scoops
Installed compressor discharge pressure (PS3) accumulator
system
Installed redesigned fan bypass bleed valve diffuser
Section
3.1
3.4
3.4
3.4,3.5
3.4
3.3
3.11
3.7, 7.3
3.6
3.10
3.3
Build 2 to Build 3
Installed redesigned propulsor Stages 1 and 2 fan blades
Installed improved PS3 accumulator system
Build 3
No significant changes
Postbuild 3
Replaced all propulsor Stage I turbine blades
Replaced three propulsor Stage II turbine blades
Added positive mechanical retention feature to IR bearing nut
Replaced IR and 2R bearings
Repaired IGV (inlet guide vane) - lip replaced with honeycomb
Drilled oil drain holes in propulsor
Borescope ports added to mixer frame to gain better
access to Stage 1 turbine blades
Replaced actuator control rods
Installed additional air scoops for cooling aft
telemetry system
1.2 OVERALL HISTORY OF UDF TM ENGINE TESTING
Tables I-I through 1-3 present an overall history of UDF TM testing.
Section
3.11
6.0
3.17
3.17
3.17
3.17
3.17
3.17
3.17
3.17
3.3
Table I-I.
Date (1985)i
Aug. 29
Aug. 30-Sept. II
Sept. 14
Sept. 15-17
Sept. 18
Sept. 19
Sept. 20-26
Sept. 27
Sept. 28-0ct. 1
Oct. 2
Oct. 3
GE36 Test History - Engine 082-001, Build I/IA.
TRTi
0:02First Run to Idle; Broken Carbon Seal
Caused Internal Oil Leak Resulting in
a Rotor Unbalance; Caused a Turbine Rub
Which Cracked Some Stage II Turbine Blades
Engine Removed from Site; Carbon Seal and
2R Bearing Replaced
Build IA
Engine Returned to Test; Idle Achieved
Worked Instrumentation Faults
Mechanical Check-Out
Mechanical Check-Out
Reached Full Propulsor Speed
Modified Forward Telemetry Antenna
Reached 22,000 Ibf; Shutdown Due to
High Aft Telemetry Temperatures
Added Air Scoops to Exhaust Nozzle to
Cool Aft Telemetry System
Reached 24,000 ibf; Engine S/D Following
Stall (Stage I Turbine Blade Failure)
Engine Removed for Repair
0:II
0:14
1:10
3:43
5:24
Table I-2. GE36Test History - Engine 082-001, Build 2.
Date (1986)
Jan. 30-31
Feb. 2-3
Feb. 5-8
Feb. 9
Feb. 17-18
Feb. 28
March 5
March 6-30
April I-2
April 3-7
April 8
April 18
April 19-23
April 24
Resumed Testing; Mechanical Check-Out;
Reached 22,000 Thrust
Testing With Facility Fans On; Fan Bypass
Bleed Valve Calibration; Bleed Valve
Diffuser Can Failure; Engine Trim Balance
(20,000 Maximum Thrust)
Repaired Instrumentation and Aft
Telemetry System
Telemetry System Check-Out (to 1200 rpm
Fan Speed); Started Down Power Calibration
Ran Twice to 24,000 Thrust; on 2nd 24,000
Pt. Lost Stage 2, No. 7 Fan Blade Shell
Engine Health Check - With New Blade
Reached 1200 rpm Fan Speed; 16,000 Thrust;
Tested with Vortex Destroyer
Turbine Frame Stress Investigation;
I000 rpm Maximum Fan Speed
Stress Survey; 1029 rpm Maximum Fan Speed
Moved Engine from Site 4A to 3D
Stress Survey; 1150 rpm Maximum Fan Speed
Added Additional Holes to Aft Telemetry
System Cooling Scoops
Reverse Testing/Cooling Scoop Testing
Control Verification; Control Fault Caused
Stage 20verspeed; Engine Shutdown
Installed PS3 Accumulator to Slow
Propulsor Accel Rate
Engine Test With PS3 Accumulator;
1150 rpm Maximum Fan Speed
Build 2
Run Time
3:38
9:49
13:28
15:16
16:30
20:36
22:55
25:44
26:22
28:30
29:20
TRT
9:02
15:13
18:52
20:40
21:54
26:00
28:19
30:68
31:46
33:54
34:44
Table 1-3. GE36Test History - Engine 082-001, Build 3.
Date (1986)
June 20
June 23
June 24
June 25
June 26
June 29
June 30
July 1
July 3
July 4
July 4-7
July 8
Returned to Test; Achieved Idle
Reverse Testing;Mechanical Check-Out; 1270 rpmMaximumFan Speed
Mechanical Check-Out/Trim BalanceReached24,000 MaximumThrust
Trim Balance/Bleed Valve CalibrationReached18,000 Thrust
Bleed Valve Calibration/Trim BalanceReached23,000 Thrust
Bleed Valve Calibration; Reached19,000 Thrust
DownPowerCalibration; Shutdownfrom19,000 Thrust Due to Fuel Leak; AftPropulsor Rotor Locked up UntilEngine Cooled
DownPower Calibration; Reached24,000 Thrust;PerformanceOptimization Testing
PerformanceOptimization/VibrationSurvey/Trim Balance; Reached21,000Thrust Control Tests; Reached24,000 Ibf
Reverse Testing (to 850 rpm Fan Speed)
LCF Cycles (lO0)
Bodes; Reached 21,000 Thrust; Trim
Balance; Reached 22,000 Thrust;
Propulsor Rotors Locked Togetherafter a Normal Shutdown
End of Ground Testing at Peebles
Build 3
Run Time
0:16
0:45
2:07
4:33
8:18
I0:00
12:17
12:51
17:36
20:18
25:38
29:53
30:34
59:45
62:31
66:07
TRT
35:02
35:29
36:51
39:17
43:02
44:44
47:01
47:35
52:20
55:02
60:22
64:37
65:28
94:29
97:15
i00:51
IO
2.0 INTRODUCTION
2.1 ENGINE DESCRIPTION
The UDF TM engine is a new aircraft engine concept that is based on an
ungeared, counterrotating, unducted ultra-high-bypass turbofan configuration.
This engine is being developed by General Electric to provide a high thrust-
to-weight ratio power plant with exceptional fuel efficiency for subsonic
aircraft application.
The engine encompasses the operational flexibility and fuel efficiency of
a two-spool core gas generator with the propulsive efficiency of a propeller
(moderate diameter and tip speed). The engine is based on an aft-mounted,
counterrotating power turbine that aerodynamically couples with a basic gas
generator engine and provides for direct conversion of the gas generator
engine power into propulsive thrust without requiring a gearbox or additional
shafting. The concept of counterrotating fan blades is being utilized to
capitalize on the full propulsive efficiency of this configuration; that is,
the exit swirl from the first blade row is recovered by the second row and
converted into propulsive thrust. The turbine transmits its power through two
counterrotating power turbine frames which, in turn, transmit power to the
UDF TM blades through the polygonal support rings which act as the primary-load
carrying support structure for the fan blades. This isolates the turbine
flowpath from out-of-round distortions from the fan blade loads.
The counterrotating turbine rotors, power turbine frames, fan blades, and
static structures are components which comprise the "propulsor" for the UDF TM
engine. Mounted in front of the propulsor is a gas generator engine which
provides the required gas horsepower. The gas generator is a modified produc-
tion F404 turbofan engine.
Figure 2-I shows a cross-sectional view of the UDF TM engine. An enlarged
cross section of the propulsor is presented in Figure 2-2. Figures 2-3 and
2-4 are UDF TM photos at Peebles Test Site 4A, and Figure 2-5 depicts a UDF TM at
Test Site 3D.
ii
12
_0
"0Q;
"0
q_0
0
,&.J
0
,-4I
_J
_J
O
0
_e3
0
I
=
13
ORiGiNAL FA_ IS
Q_TY
¢J
.r-I
09
l/3
qJ
¢J¢J
¢J
z
r._
!
L,
14
ORIGINAL PAGE
AND Wl_ITE PHOTOGRAPH
i5
ORIGINAL PAGE
8LACK AND WHITE P_APH
Figure 2-5. UDF'M Engine at Peebles Test Site 3D.
160F _O_T'_)!"J'ALIT y
The major design characteristics of the UDF TM engine are as follows:
Gas Generator
Model F404-GE-400
Type Low bypass turbofan
Fan 3-stage axial flow
Compressor 7-stage axial flow
Turbines
Low pressure One stage
High pressure One stage
Rotor speeds
Fan 13,270 rpm (100%)
Compressor 16,810 rpm (100%)
Maximum airflow Approximately 140 ibm/s
Thrust 16,000 ibf class
Thrust-to-weight ratio 8:1 class
Overall compression ratio 26:1
(maximum climb)
UDF TM Engine Propulsor
Maximum nacelle diameter 76.4 inches
Fan blade tip diameter 11.67 feet
Fan design point tip 780 feet/second
speed, physical (maximum
cruise; 35,000 ft; 0.80 Mach;
ISA)
Fan rotor speed 1393 rpm (100%)
Fan disk loading, class SHP/A 87 HP/ft 2
(maximum cruise; 35,000 ft;
0.80 Mach; ISA)
Fan blade radius ratio 0.415
Power turbine inlet temper- 1310 ° F
ature (SL, T/O, ISA +27 ° F).
The gas generator/propulsor combination produces an engine with a net
thrust of 25,000 pounds. Other significant design features which have been
incorporated into this engine are:
• Advanced unducted fan aerodynamics that incorporate custom-
tailored composite fan blades over an inner titanium spar that
17
serves as the attachment mechanism to the engine for the fanblades.
Fully developed and available gas generator to provide the
necessary power for the engine.
DEC (digital electronic control) that provides overall engine
control by monitoring gas generator power and speed and propul-
sor speeds and pitch angles. The engine uses the existing gas
generator control and a separate propulsor control to minimize
development costs without sacrificing control flexibility.
Hydraulic/mechanical actuation system enabling setting the fan
blade pitch angle of the two fan blade rotors either together
or differentially; this system is driven by the control system.
Modular assembly of the gas generator and propulsor.
Individually replaceable propulsor fan blades with the engineinstalled on the aircraft or test stand.
2.2 INSTRUMENTATION
The UDF TM instrumentation consists both of static and rotating instrumen-
tation, with the rotating instrumentation being read out by telemetry on the
rotors. Static instrumentation includes temperature, pressure, kulite, strain
gage, and accelerometer instrumentation on the engine, pylon, and nacelle.
Rotating instrumentation includes temperature, pressure, and strain gage
instrumentation on the counterrotating rotors. Detailed information on UDF TM
instrumentation is contained in the Instrumentation Plan (Statement of Work
Paragraph 4.2 of Contract No. NAS-24210). Included in this is Drawing No.
4013341-034, which shows the engine cross section with the location of engine
instrumentation and the corresponding parameter names. Reference GE36 Second
Ground Test TPS No. MA-0004 for a detailed description of the instrumentation
system and for a list and description of all parameters.
18
3.0 REVIEW OF UDF TM TESTING
The following significant events, data points, and problem areas from
UDF TM testing will be expanded upon in the indicated sections:
Builds
Involved Section
Forward stationary carbon seal failure 1 3.1
Forward telemetry antenna repair I 3.2
Telemetry system temperature problem
(includes reverse thrust test data) 1,2,3 3.3
Stage I turbine blade failure and stall event I 3.4
Turbine blade damper effectiveness 1,2,3 3.5
Subidle oil leak problem 1,2 3.6
Stationary exhaust nozzle (centerbody) replacement 1,2 3.7,7.3
Fan bypass bleed valve calibration 2,3 3.8
Fan bypass bleed valve diffuser failure/replacement 2 3.9
Propulsor fan-blade-loss event 2 3.10
Propulsor fan blade history and test data 1,2,3 3.11
Power turbine frame stress investigation 2 3.12
Effect of "vortex destroyer" on stress 2,3 3.13
Effect of test site change on stress 2,3 3.14
Rotor lockup after shutdown resulting from fuel leak 3 3.15
LCF/HCF (low cycle fatigue/high cycle
fatigue) testing 1,2,3 3.16
Rotor-to-rotor lockup/propulsor disassembly and
rebuild 3 3.17
Miscellaneous hardware: stress data 1,2,3 3.18
Oil leak/oil gulping problem 2,3 3.19
Heat transfer and secondary flow system 1,2,3 4.0
Engine systems dynamics 1,2,3 5.0
Bearings and seals 1,2,3 6.0
Performance 1,2,3 7.0
Engine operability 1,2,3 8.0
Engine control 1,2,3 9.0
Nacelle structures 1,2,3 I0.0
19
2O
3.1 FORWARD STATIONARY CARBON SEAL FAILURE
After the first engine start, the engine was shut down due to a low oil
level warning. Carbon seal pieces were found in the propulsor scavenge
screens. The engine was removed from the test site, and the propulsor was
separated from the gas generator. After separation, it was found that the
forward stationary carbon seal was damaged. It was determined that the seal
was damaged during assembly. The 2R bearing was also found damaged, this due
to carbon seal debris. New hardware was added to the engine to help guide the
propulsor rotors together to avoid any damage to the seal (Figure 3-I), which
was completely replaced with new hardware, along with the 2R bearing. The
propulsor was remated with the gas generator with no problems, and the engine
was put back on test. There were no oil leaks when the engine tests resumed.
While running the engine with the damaged carbon seal, the resulting
internal oil leak caused an unbalance in the rotors. This unbalance caused
the Stages 7, 9, and II propulsor turbine blades to rub hard against the inner
6-11 spool. Twenty-two Stage II blades were found to have cracks in the root.
It was decided to not replace these blades but to closely monitor their stress
levels.
3.2 FORWARD TELEMETRY ANTENNA REPAIR
Operational problems with the forward telemetry system were caused by
insufficient clearance between the static and rotating forward telemetry
antenna components. These components were modified to increase the cold run-
ning clearance from 0.135 to 0.250 inch, as diagrammed in Figure 3-2.
3.3 TELEMETRY SYSTEM TEMPERATURE PROBLEMS
Problems with the telemetry system, due to high temperatures, resulted in
design changes to increase cooling. Cooling scoops were added to the rotating
exhaust nozzle to bring in additional ambient air to cool the telemetry system
(especially that of Stage 2). Prior to the addition of cooling scoops, there
were flush vent holes in the exhaust nozzle. The following summarizes the
design changes:
"0im
0I-
"0!.--
(I)C
CU.I
Z"0
"0<
VI
0
Cj
I
21
OE POOR _'_:d._-'_-
"== ..o =_
-.==m. =®
=u _,,'-
o _o _o =cmm_
o_=_8 ° =
0
0
I
0,0
Run Time With
ConfigurationConfiguration
No Scoops - flush holes (0.468-inch diameter) in 3:43 hr
exhaust nozzle, 60 total. Build I; August 29 to
September 27, 1985
Data Symbol: None
Pipe Elbow Scoops - with 0.375-inch diameter 1:41 hr
openings, 30 total. Build I; October 2, 1985
Data Symbol: Pipe
Aerodynamic Scoops - with 0.500-inch diameter 25:44 hr
openings, 30 total. Build 2; January 30 to
April 2, 1986
Data Symbol: Aero
Modified Aerodynamic Scoops four 0.188-inch diameter 69:43 hr
holes added to side of scoops, 30 total. Build 2;
April 8 to 24, 1986. Build 3; June 20 to July 8, 1986
Data Symbol: Mod Aero
The addition of four holes to the side of the scoops were an attempt to
increase flow to the telemetry system during static ground testing. Sixty of
the modified aerodynamic scoops will be used for flight test. Figures 3-3 and
3-4 illustrate the telemetry system and cooling scoop designs.
Figures 3-5 through 3-8 show both raw telemetry temperature data and data
normalized with the ambient temperature (TAMB). The data was normalized due
to the large variation in ambient temperature which has a direct effect on the
telemetry temperatures. Note that temperatures recorded during Configuration
3 were cooler than those of Configuration 4. This shows the effect of ambient
temperature on the telemetry temperatures. Also during Configuration 4, there
was more test time and more time at power which would tend to increase temper-
atures. Also note that there is no data for the first two configurations for
Stage i; the thermocouple was not reading at those times. The telemetry
temperature limit set during ground testing, after the first configuration,
was 300 ° F.
During reverse thrust testing, Stage 2 telemetry temperatures increased
throughout the time the engine was in reverse. Figures 3-9 and 3-10 show the
telemetry temperatures during a reverse thrust cycle. This was the fifth of a
series of consecutively run reverse cycles.
23
!
24
o<
+÷
+ ÷
o
oo o
o. cuo oo oo _4oo
(D _-d•H o
o :_
<
o
oo
_oo,r4D_
00
03
OJ.IJ
:>,
:>,
OJE
,j!
Q;
_D
25
0
00
01
Oo
0
<]
I I I I
0 0 0 0
c_ _ cN c'_
<I <_ <I0
<I
<3
0
0 0
0 <>
<>0
oo o4> o
<>
L_i
m
t-o
vx
O_-c;
-c_
LOI I I I I I I I d
o o o o o o o o oo a0 _ ,_- o_ o a0 _o
I
m.,
4,m
,-t
"o
o
I
26
3 E)3C] -- LOOAV].
ini
<3(D
(./1
oorl-
.<
w 0
I I I I
i 4
m
u_
A_
En_
o I
Zvx
0
0
u_
0
"O
N
0Z
!
m
e_N
.1.1
,-..t
0
IC'3
_J
O0
) I. O0_-IV.L
27
0
O_ 07n_
0+_
0<J
oo _r_
0
r-t 0
[]0
0 0
0
0
0
°0 0 _ 0
O ,_I0 0 o
0
_ .i_i-I- 0 Oi
0_ + _c_
I I I I I I I I I I I I I I0 0 0 0
LO
n-o
Zvx
oi
c_
0
:)
I
0_
_J
.,!-I
Q;
m
4-)
q-_,<
I
.,-I
"4 O3O -- t Ocr_J.
28
n 0
o
<
o_ 0Z _ .< "5
[3+0_
[] 0
[]
[]0
0
0
0
oo0
0
<>
[]
00 D
[] D
D
+ o .°o
o .ea*.0
[]
%
[]<1
I I I I I I
In
m
A_
t-
o I
zvx
0
I".0
0
;3
-o0JN
.ml
0Z
GO
_J
c_
oJ
,-4
<
o_I
b/)
(>4 )S_nV.L/(a ) I. Od_±
29
,,.n '0 20 40 _0 _0 100 LEO
T ItiE - 5EC
Figure 3-9. Forward Telemetry Temperatures - Reverse Testing.
30
Figure 3-i0. Aft Telemetry Temperatures - Reverse Testing.
31
32
3.4 STAGE i TURBINE BLADE FAILURE AND STALL EVENT
Sequence of Events
• Engine operating on point at takeoff power (24,000 ibf), 1,350
rpm propulsor fan speed for 4 minutes
• Stage 1 turbine blade failure: high cycle fatigue (first flex)
cracking at blade root resulting in aft lean of blades
• Contact of Stage 1 blades with Stage 2
• Reduction of Stages I and 2 flow area
• IPC stall
• Engine stopcocked about 1.5 seconds after stall initiation
• Rotor axial excursion
• Secondary damage to remaining stages resulting from failure of
Stage I
• Additional rubs experienced by Stages 7, 9, and II
• Fast coast-down
• Rotor-to-rotor lockup.
Figure 3-11 shows a diagram summarizing the hardware damage.
Results
During testing, prior to the stall, significant first flexural vibratory
response had been observed on all stages of power turbine blading (Figures
3-12 and 3-13). Teardown revealed significant damage to Stage I. Several
blades had large leading edge root cracks and were leaning aft toward Stage 2.
One local segment of blades was buckled aft into Stage 2. All Stage 2 leading
and trailing edges, as well as all Stage 3 leading edges, were bent circum-
ferentially and severely rubbed. Stages I, 7, 9, and 11 exhibited severe tip
rub. While Stages 7, 9, and II had rubbed previously during Build No. I due
to oil leakage past a broken carbon seal, the poststall rubs were more severe.
Stage 4 had very light tip rub only at the trailing edge. Stages 2, 3, 6, and
8 exhibited no tip rub. Stage I0 rubbed only one blade, which was bent at the
platform. Stages 7, 9, and II had several bent airfoils. The rotor inner
0
o
,,--t
,-,-I
I
(_
b0,_1
33
0
0o
+=D0
m,IB
C
U,, "_O'J
O'J
_xxx x)Y//////]]////J//////I
k\\\\\\\\\\\\\\\\\\\\_
It///]///]//
r-_
_Xxxxx_V/////////_
t0
_r
! !
(ePlS)l) ssaJIS :IL )lead
9"
001OS
(n
QJ
(n_J
_Q
_J
,..q
I
34
i
;S
I__- _- _._--
Isd_l lsd_
!
I
t_
.l,,J
0
0
.el
4
Ic_
C_
F,_
ORIGINAL PAGE _S
OF POOR QUAL!TY
35
spools exhibited local burn spots at Stages I, 7, 9, and 11 rub locations,
with cracking evident at Stages 7 and II. Domestic object debris damage was
confined to the power turbine flowpath. Nicks and dents were significant only
on the forward turbine blade stages. The power frame and OGV (outlet guide
vanes) airfoils exhibited only minor debris damage.
Metallurgical evaluation of the Stage 1 blades found root leading edge
cracks on 117 of 124 airfoils. However, none of the airfoils were completely
separated. Crack lengths varied from approximately 0.02 to 1.2 inch. SEM
(scanning electron microscopy) revealed evidence of high cycle fatigue along
the entire fracture surface. Evidence of root fatigue cracking was also found
on Stages 2, 4, 6, and 11. Stages 7 and 9, and the one bent Stage I0 blade,
were found to have tensile cracks at the root braze joints and tensile tears
in some of the platform braze joints. Local burn spots on the spools showed
evidence of severe overtemperature of the Inco 718 substrate. Four locations
at Stage II and one at Stage 7 were confirmed cracked.
Posttest review of recorded data indicated stable engine operation while
sitting on point until approximately 0.5 to 1 second prior to the stall. At
that time, propulsor speed started dropping slowly while the control increased
fuel flow to compensate, and vibratory response on the forward turbine blade
stages showed a sudden increase.
Conclusions
Based on the responsiveness of all stages of power turbine blading at
first flexural frequency, it is concluded that the blade and spool design con-
figuration did not provide effective dovetail Coulomb damping as anticipated.
Hardware condition indicates that the Stage I blades were cracked to varying
degrees prior to stall and that these cracks propagated in high cycle fatigue
while sitting on point on October 2, 1985. As the cracks grew, the blades
leaned further aft until one or more contacted Stage 2, causing the slow drop-
off in propulsor rotor speed prior to the stall. Contact with Stage 2 rolled
the leading edge, resulting in a sudden reduction in power turbine flow func-
tion. The reduction in flow function caused the F404 gas generator to stall.
The remainder of the damage was secondary during the rapid shutdown.
36
Corrective Action
All power turbine blades were replaced for Build 2. Spanning across
adjacent blades, coulomb dampers (in the form of Ren4 41 pins) were designed
for all stages of blading as illustrated in Figure 3-14. Effectiveness of the
damper design has been substantiated by component rig spin ("whirligig") and
bench wear tests. Although no evidence of fatigue was indicated for the power
frames, dampers were also added to Stage 12 as a precaution due to a predicted
first flexural response with the outlet guide vane passing frequency. Strain
gages have also been added to the Stages 5 and 12 power frame airfoils for
Build 2.
Low cycle fatigue and crack growth analyses of the spools indicate that
they have sufficient life capability for the remainder of planned testing.
The spools have been weld repaired and the rub coats stripped and reapplied to
provide additional margin.
Because of the severity of rub during Build I when oil leaked past the
forward intershaft carbon seal and during the Build IA stall, power turbine
blade tip clearances have been increased an additional 0.050-inch on the outer
attached blades and 0.030-inch on the inner attached blades, and all tips have
been tapered to 0.005 to 0.015-inch thickness as a precautionary measure.
3.5 TURBINE BLADE DAMPER EFFECTIVENESS
Coulomb damper pins were added to all turbine blade stages (I-4, 6-11)
and to Stage 12 power turbine frame airfoils (Section 3.4 and Figure 3-14).
Figure 3-15 demonstrates the Stage 1 turbine blade vibratory-stress reduction
resulting from the addition of the damper pins. Figure 3-16 shows the Stage 1
vibratory stress in relation to the scope limits before and after addition of
the dampers. Figures 3-17 through 3-26 illustrate the vibratory stress for
all turbine blade stages after installation of the dampers. Data are from an
accel to 25,000 ibf, except for Stage 11 blades data (bad telemetry signal)
which is from an accel to 22,000 ibf. Section 3.11 contains stress data for
the turbine frame blades.
37
A I1--"
._m .__- .c Q.o.m I- _a ca :"oI ,,,. _' C P"o _-- _T
c_
I
ClO.el
38
ORIGINALOF POOR PAGE IS
QUALITy
=m
"omii
m
(ZH) Aouanl_J=l
im
U
LU
m
J_
im
E
g
o
m
o
,r,I
= :>
o
o
I
_J
=
,r.I
P_
39
40
o+-_. -@
o +,mi
/
o ÷i. |[1
/121 ÷ ,_
I ! I 1
0 m
mE5-; =
° 4: JOI
_.7 ° +1®® __
° o / -?-,, , , , ,', ;1; o
xO_
.r-I=
Ul
S
0 _-__._
'0
W_J _J
E
O_
I
=
ORtG!NAL P,,_E"'" IS
OF POOR QUALITY
I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,,
ooo
0
_Jtj
m
Q
m
41
OF POOR QUALiT'f
I,,,,I,,,,I,,.,I,,,,I,,,.I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,.,,I,,,,I,,,,,8 S _ St a 2;
000
¢N
0
0
¢J
_J"0
m _
i
.M
42
ORIGINAL PAGE IS
OF POOR QUALITY
I .... I,,,,I,,,,I,,,,I .... I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I,,,,I, ....
I • III • e
!! X
!I
!
0
if3o4
0
e_0
!
43
I....I....I....I....I....I....I....I....I....I....I....I....I....I....I....I....I....I....I....I.....
.! _ # ! # _ # m
|
!i
u_
000
u_o4
0_J
_J
_J<
gm
o
m
m _
¢xli
.4
44
OF PGOR QU:,"_,L;TY
I .... I .... I .... I .... I .... 1 .... I .... I .... I,,,,I .....$ 11 II
it t i ! !t !! ! tf tl !
i °I,.i.Ii:l:I.L
ii
i1I:l:
O'i
ZX
II
ul.,-iI,,.i
I.i,,.i
o
(.,l(,,l
o
li)
<=,,dco
I,-i
I
45
OF POOR (_b_Li?V
I,,,,I.,,,I,.,,I,,,,I,,,, I,,,,I,,,,I,,,,I,,,,I,,,,,
ql_ 11 8 e
1| I I ! ! 1 t 1I 11 ! "
E
n
¢v.
O0'T
Z
X
2;
OO0
u_e_
0_J
¢;(J_J<
O
IlJ
m _
!o'1
4B
OF POOR Q,SAt..!TY
!
u_
0OO
Lc_
0
{.J
0
QJ
_J"0
m_
!
47
I,,,,I,,,,I,,,,I,,,,!,,,,I,,,,I,,,,I,,,,I,,,,I,,,,,
.=
r-4
000
0
0
vl
m
I¢¢3
.r,,I
,d
48
_ J _ _ _ _ _ '_ _ _" _ "'r,_ E _ _
OF POOR QUALITY
OF PO0_ Q_,aL_Y
2;
,.Q
000
u_
0
0¢J0
0
m
O _
m_
I
_d
49
I , ,,, i,, , , I , , , , i .... I , , , , ,8 o e
i;1
!i_! ! ! ! s o!
i wr_la.
00
¢N
0
aAUU
m
0
IlJ
1)
m _
I
=
5O
OF POOR QUALITY
3.6 SUBIDLE OIL LEAK PROBLEH
After the carbon seal failure from assembly damage was repaired, the
propulsor still leaked oil into the rotors, flowpath, and nacelles. 0il loss
was visible, both during operation and on shutdown, in the form of smoke and
oil drips. The oil was leaking through the carbon seals during low speed ope-
ration. Although the amount of oil lost was not large, it was decided that it
was desirable to reduce the amount of smoke after shutdown.
During early operation, much of the engine testing was at subidle speeds
and below, as control and starting problems were worked out. The design idle
speed was 650 rpm. Initial testing proved that idle speed had to be increased
to 700 rpm for several reasons: the 650-rpm idle speed was too close to the
subidle rotor critical; the vibration level was unacceptable; the telemetry
cooling system was not effective at the lower speed; and propulsor stresses
were lower at the higher idle speed.
Test data showed that the pressure drop on the aft carbon seal approached
zero using the 650-rpm idle speed. The carbon seal leaked without a positive
pressure drop. At 700 rpm, the seal pressure drop is increased enough to
limit the oil loss out of the aft carbon seal. At higher power settings, the
pressure drop across the seal is well above that which is required for oil
sealing.
At low speeds (subidle), for very short runs, and during starts until the
elevated idle speed is reached, the propulsor sump is not scavenged properly.
At low speeds, the oil is not pumped to the ends of the rotating sumps where
it can be scavenged. Due to the frame design, scavenge lines at each end of
the sump run uphill initially; this forms a trap that requires some small sump
pressure to overcome prior to flowing to the scavenge element. The scavenge
pump used (from a CF6-50 engine) has a reputation as one that is difficult to
get primed and establish full scavenge flow. For these reasons, a subidle
lube bypass valve was designed and installed on the propulsor lube system.
The initial bypass system was a facility design - air actuated and con-
trolled by the engine stopcock. In order to preclude complete loss of oil to
the engine in a control or valve failure, the system was designed to divert
most, but not all, of the supply oil to the propulsor. At subidle, supply oil
51
52
was diverted to the scavenge circuit just at the exit of the air oil cooler.
Hoses and hardware used were madeon site and were not flight quality. The
system, combinedwith the new higher idle speed, reduced or eliminated the lowspeed leak/smoke problem.
For flight-test, the system had to be replaced with a digital electronic-controlled solenoid valve. The facility system as configured above did not
function as desired on engine shutdown. The scavengepressure decay lagged
the supply pressure decay, so that the propulsor supply was actually divertedto a higher pressure circuit, and briefly, oil flow was increased to the pro-
pulsor. This condition was somewhatalleviated by removing the air oil cooler
from the scavengecircuit (which wasnot required) and by lowering line lossesin the bypass circuit with flight quality hardware. These changes ended the
leak and smokeproblems.
3.7 STATIONARY EXHAUST NOZZLE (CENTERBODY) REPLACEMENT
Between Builds 1 and 2, a redesigned centerbody was installed on the
engine. This redesign was necessary, because analysis and wind tunnel testing
indicated that at some flight conditions there would be an undesirable flow
separation from the centerbody. The predicted flow separation data for both
centerbodies is shown in Figure 3-27. No flow separation was detected or pre-
dicted for ground testing. Figure 3-28 compares the two centerbody designs.
Further information is provided in Section 7.3, Performance.
3.8 FAN BYPASS BLEED VALVE CALIBRATION
During Builds 1 and 2, tests were performed to calibrate the fan bypass
air (from the gas generator) bleed valve. This was done to provide necessary
information about the amount of fan bleed air required to maintain stall
margin in the UDF TM. These data are extrapolated for use in finding the amount
of bleed air required in flight.
3.9 FAN BYPASS BLEED VALVE DIFFUSER FAILURE/REPLACEMENT
During Build 2, the fan bypass bleed valve diffuser failed in high cycle
fatigue (Figure 3-29). Until a redesigned diffuser (Figure 3-30) could be
! )c_
oc_
cn
2c_
c_
0I o
O0 CD 0 C_c_ c_
or)
dOS.l
J iiGJ
[-_m
0 0
m L'_
0 II
_ 0_.1 .,._
_ C._ 0
_ .,._
°r_
Ic_
=
53
54
©®
"00
_J
_J_J
_J
.,.._
_J"0
I
_J
.,-I
I
¢'1!
I
Ii!
0
c_ c_o_1
Q
i
c_
*,-t
,--4
¢NI
¢¢3
o_..t
55
Y \!
I
0
-ib- _ 4i-
_J
I
0
o_
4..
_J
_JOJ
I
L_
56
designed and manufactured, the damaged portion of the diffuser was cut off and
used "as is." When the redesigned diffuser was available, it was installed on
the engine.
3. i0 PROPULSOR FAN-BLADE-LOSS EVENT
Summary of Event
The engine reached test point:
fan speed.
24,000 ibf, 1371 rpm propulsor
Stage 2, No. 7 composite fan blade shell separated from tita-
nium spar and was released; however, the spar remained attached
to the trunnion.
No high fan stress or other anomalies prior to blade loss.
Shutdown initiated at blade loss - I second chop to idle and
stopcock within 9 seconds.
No gas generator stall.
Propulsor spool-down was normal.
Due to a large imbalance (approximately 260,000 gm/inch) caused
by the missing composite, a noticeable amount of engine vibra-
tion was experienced.
Control/actuation system functioned normally after blade loss.
Spar Damage
Inspection of the spar following shutdown indicated the titanium had
cracked at the EB (electron beam) weld line the entire width of the spar. The
EB weld crack was clearly visible over the midspan of the blade (Figure 3-31).
Secondary Damage
Blade No. 6, which was next to the released blade, had a slight nick in
the polyurethane coating where the composite material from Blade 7 hit Blade
6 before striking the ground. The nick was an indication that only light
contact occurred.
The isolators contain an absorption material which is intended to yield
under high unbalance conditions such as blade loss. Deformation of the aft
57
Figure 3-31. Spar Schematic.
58
isolator pads was observed, and the isolators were returned to the vendor for
inspection. The aft isolators had their original values of spring rate and
damping coefficient. Since there was deformation of the aft isolator pads,
they were refurbished as a precautionary measure and returned to engine test.
No damage to the forward isolator was observed.
There was also concern that the transfer gearbox attached to released
blade trunnion No. 7 may have suffered some damage. The gearbox was torn down
and both the MPI (magnetic particle inspection) and FPI (fluorescent penetrant
inspection) of the gears and pinnions indicated no abnormal wear.
Bolts attaching the Stage 12 power frame to the rotating exhaust nozzle
showed interference with the OGV assembly, with the heaviest wear at blade
Location 7. The OGV assembly is designed to clear the bolt circle by 0.I00
to 0.160 inch. The OGV assembly showed light wear all the way around, with
a l-inch section indicating a harder rub. No repair was required. The OOV
assembly had no other distress.
The Stage II turbine blades rubbed the inner spool at some time during
the event, leaving a 4-inch x 3/8-inch rub mark on the inner rotor. The rub
coincided with Stage 2 fan blade Location 4. The indication was a surface
discoloration with no measurable depth. The Stage II blade trailing edge tip
did not have any discoloration or tip curl that would suggest a heavy rub.
Both Stages I and 2 fan blades were returned to Evendale for inspection.
No debonding was found in the Stage I set, but one Stage 2 blade was found to
have a section of composite separating from the spar.
Fan Blade Corrective Action
Corrective action included the addition of a portable ultrasonic scan of
all fan blades following each hour of engine run time. Ultrasonic scanning at
Evendale proved to give accurate results when trying to determine if debonding
has occurred. The ultrasonic scan used when the blades were returned to Even-
dale indicated a Stage 2 blade had started to debond. Portable scan equipment
also verified debonding of the composite. Because the previous test method,
ping checking, indicated the debonded blade was acceptable, the ping test was
discarded as a debonding check.
59
6O
Test History of Failed Fan Blade
The blade that failed had been run 20:40 hours prior to the failure. A
breakdown of run time versus propulsor fan speed is provided in Figure 3-32.
3.11 PROPULSOR FAN BLADE HISTORY AND TEST DATA
3.11.1 Fan Blade Test History
None of the original design Stage 1 fan blades had to be replaced due to
failure or debonding. After Build 2, the blades were replaced with redesigned
blades with mechanical retention features. Figures 3-33 and 3-34 compare the
original and redesigned fan blades. All of the redesigned Stage l fan blades
ran through Build 3 without problems.
Some of the original design Stage 2 fan blades failed inspection during
Builds 1 and 2 due to debonding. These were replaced once any discrepancy was
found. Figure 3-35 depicts the history of the Stage 2 blades during Builds I
and 2. Investigation into the two debonded blades found after Build 1 deter-
mined that the probable cause of debonding was from propagation from cracks in
the foam inside the blades. This foam filled the cavity between the composite
shell and the titanium spar. Cracks in the foam during manufacturing propa-
gated into the shell/spar bond. A design change was made (Figure 3-36) to try
to prevent this from happening.
Section 3.10 discusses the Stage 2 fan blade failure in Build 2. After
this fan blade failure, the engine was limited to 1000-rpm fan speed until the
redesigned fan blades were available (Build 3 testing).
3.11.2 Fan Blade Test Data
Figure 3-37 summarizes fan blade vibratory stresses for Builds l, 2, and
3. Figures 3-38 through "3-42 present specific examples of stress data. Note
there are differences in fan blade stress levels with seemingly equivalent
test points. This difference is caused to a large degree by differing wind
conditions (that is, wind direction, velocity, gusting). Figure 3-43 depicts
the strain gage locations. Location No. 4, which gives the highest first flex
vibratory stress, is the strain gage used for the given data.
O)t_
00,q.
f
&,r-
Ln :
--_, _ ..o °
CO O0 _ x
"---, Zx
qr- O0
!
t_
I0
t_
I
m 0
I I I I I I
I,n 0 tS_ 0 It)t_ t_l _1 ,r.,. ,r-
e0
E 3-- C
I- E
o
i
QJ
_0,r-I
61
• Composite Shell(Carbon Fiber-S-Glass Epoxy Matrix)
• Titanium Spar• Foam Filled Cavities
!
, 0=.i =
iS 'If !
tBJ
, !
_| ;,
Figure 3-33. Original Fan Blade Airfoil
Mechanical Design.
r_
U
0
E
oq!o_
_J
_0
G3
mql D
e_L-am,@
_C
Dq_
!
II
I!
i
m
m
qlm
__-o
t o0o,- _,,-,-,-,- __
c
• v m"u
|
mm o
0
Q
Im
0
_J
E_
_O
c_
c_
c_
a_0_
!
C_
64
F Laminated Composite
Stiffener Bonded
CompositeDowel
Pins
• Plug Retrofit Creates Stiff'Closed-Box' To Preclude Crack
Propagation And Increase ToleranceTo Thermal Cycling
Figure 3-36. Fan Blade Structural Improvement.
65
m
m
(/)
rrOd
rr
rr
I I
0 0T- T- _- GO
! | | I
0 0qp-
o. o. o o
00 co! I
cO f_l I | n
0 0,i--
0 0 0 0
i._
3: co
D. n-
"0 m q.._ a. ¢_1
._ "(2o
•
I I I I
q-@d
I
0
U
m
Uu
U I_U
,m
-el
I
U'?
r/l
0
r_
"0
_._I
_-_
66
._ ._ _ • m _- Z
____.....-o u- _,,_ , =
,....,...._..._..L._._-,-_-,--_ _ "" '_ .... ' .... ' .... ' .... ' .... ' .... ' .....I l 1 /
I1 ., ',. • 2_.. ,._;_._.'._lJ_' Io I1
• 1" " `% " " I .._ ,.o.. _ o
•'._':.'.". "" ; _ _ .." o• ,_.. ., ._ "*._. _. . •
,' ..'..' .k _ ".l:._":;l': _ i•,..._-..:.:_-:!.:.:._,"_l,l_.ro It .,
-. !I'" "_':'l(i.'..F.I_ I: ", ,.,_ ..,_ , •
.. : .;_:, , o
H ' \.x._,l_ _ I]!t ........ .,.-_,h,. _,_o _1 , ....
c/_
0
Ic_
6?
41m,
+-+'3O) Incoo_ ,, :,, _
m _Q"c= o,,_c,,+ _:I' w w It• Q,, .'. q 0 a a.-
'_ • -w- .C "-'---'u
q= u__ =_ ' i" ++
• • • + +
o-- _ + +IC i_ CC mC'mJ('_II Im ,,a _'
I.. I .... I .... I .... I .... I .... I .... I ..... (_ C_l
AouenbeJd AouonbeJd
68
• n_
,= a.
, ® _ "" m •_,. I if" _ _
_o ., J
-= c0)®"'® ._o _}I,i,. m m ,_
"" !i• • ,,., .,.,
• _ _ _
,,...___._.,._,....,...,..._,...,...__,...,....,._.._....__,_.o.....u._,_.,....,...................,............,...,..............,......s _==_. ,, .... _. , _ _ | _i"
_"e" _"r
. u.
69
_> _>1.... I _>1.... I,,,,,I.... I.... I .... I .... I .... I.... I.... h,,,I,,,,I,,,,l_l,,,,I,,_,,,I .... I=
4r
000
cN
0
o
@
_N
m _
I
70
1::_! e",, __ i ..%,_
_,i,,,_ f....I_.,,.,,,,,,I.... ho,,l .... I.... I.... I.... I .... I.... I .... h,o,l ,_ ,,,I,, =
\
a
•, b.
I I I I I
,,q
0
0
-.1"
N
0
U
U
U
0
C_
II/
_4
I
c"l
l..l
71
(nr/)
0). ot_ tl
m
(nmI-
a.-o
o)t_ G)
"0r3(.
mm
(on_
01
...(_0i_q"
"0
,<r.o
mm
(ns-
O
I--
0
=" 0
U) ,_
4). o(/) ,-,0
0) 0)
!
®
®
®
11.m
"0r"<x0
m
1.1.
(_1
0
0im•_ 0
C _4) 4)(n rr, "0
(JOIL(O 0(3u.
!
®
®
"0
atom
m
x q)
o o
4) x
e_ mm
"" i_,e_
(/)r" '-4)
(/) ,-,0
010, J(_ ..m
I
®
V)U)4)I,..
(n
(n(n
_n _n
Im
II
-r->,Im
"" 00_"
mr" ,_O.Q
imm ii
m_J x
m
--ILL.I
4)
t_ r"O<
I
®
0
C
!
72
Fan Blade Response with Facility Fans
At the first test site (4A) a set of 12 facility fans were arranged in
front of the UDF TM inlet. Figure 3-44 shows a schematic of the facility fan
configuration. These fans attempt to simulate forward velocity of an engine
and try to smooth out airflow through and around the engine. It was desired
to see if some combination of these fans would lower fan blade stress. The
results can be seen in Figure 3-44; the first data point (no fans on) gave a
lower vibratory stress level than with any number of fans on. Stress levels
varied widely, depending on the number of fans that were on, but all levels
were higher than those without any fans. Since the engine centerline and
the facility fan centerline were not in line, it was believed that the fans
created additional disturbances in the flow field instead of smoothing the
airflow.
3.12 POWER TURBINE FRAME STRESS
During Build 2, power turbine frame vibratory stresses were higher than
predicted. The stresses were found to be predominantly 2/rev forced response.
This was caused from the fan Rotor 2 nodal, first flex mode (note that the
turbine frame and the fan blade rotor are mechanically linked). This causes
the frame stress to track the fan blade stress. Figure 3-45 provides a stress
comparison.
A detailed investigation into the frame stress allowed the vibratory
stress limits to be increased for the forced response mode (Figure 3-46). A
summary of this investigation and its results is in Table 3-I; supporting data
is shown in Figures 3-47 and 3-48. Build 3 data is presented in Figure 3-49.
3.13 EFFECT OF VORTEX DESTROYER ON STRESS
Vortices were seen between the ground and the fan blades at the bottom of
the engine. Visualization of these vortices was aided by having moisture on
the ground or by releasing smoke bombs. To see if the placement of a vortex
destroyer (a large metal grating) under the fan blades would reduce the fan
blade and power turbine frame vibratory stress levels, a vortex destroyer was
placed under the fan blades approximately 6 inches off the ground. Fan blade
73
40
200
///
I\i \
/ \
/
II
//
• Fans Turned
On Produce
Higher
ResponseThan Without
1 2 3 4 5 6 7 8 9 10 11 12 13
Fan Turn On Sequence
Figure 3-44. Fan Blade Response with Facility Fans Turned On.
74
"0
t1
4
I
q
0
.=U.
G)01(I1
U)
0 0C_l
(n
Eim
U.
C_I
4)01
U)
I_,_
_J
_,_._I
(ep!s)l) ssaJls/uoleJq!A IleJa^O
?5
Results:• Updated Scope Limits Versus Site 4A
Response• HCF Testing
Stage 5 Strut Gage #KD4H06
5O
40:°_.V
==
°110.7 0.9 1.1 1.3
Rotor Speed (rpm/thousand=)
2/17 2/18 2/18 2/28 2/28o 4- # • x
Stage 12 Strut Gage #KD4S07
i:io.L,_..0.7 0.9 1.1 1.3
Rotor Speed (rpm/thouunda)
2/17 2/18/ 2/18 2/28 2/28o .$- # • x
Figure 3-46. Power Turbine Frames Vibratory Response, Updated Scope
Limits Versus Site 4A Response.
?6
"0
°° {
_:E'r "_ Z:O 0 Z
(!s_l) sse_|s sse_|S pez!letUJON
8
g
5O
_J
@
OU
o4
0_
_n
0
_j •
0 u_
0_0 .,.4
_J ,,_
0
p_
I
_J
_0.M
??
• Frame Properties Better ThanMaterial Handbook Average
900°F Smooth Bar vs. 1100°F Smooth Bar vs.Material Handbook Material Handbook
50 50
_'_40 _ 40
I _ _ Tested Frame30 Tested Frame
_30 !- _'_ Minimum _ [_ __Minimum I
P .nd okAv..,.iI ..od_ook Minimum I I , Handbook Minlmum I
0_ 0_
0 20 40 60 80 100 0 20 40 60 80 100
Mean Stress (ksi) Mean Stress (ksi)
Figure 3-48. Power Turbine Frames Vibratory Response, Material HCF
Test Results.
?8
Peak Levels
- Consistent withBuild 2
- Below Scope Limits
STAGE 5
KD4H06O' "
STAGE 5
KD4H09
20
Figure 3-49. Power Frame Vibratory Response, Build 3.
79
stress data both with and without the vortex destroyer is contained in Figure
3-50. The same type of data are shown for the turbine frames in Figure 3-51.
Since this testing was performed after the Stage 2 fan blade loss, and before
installation of the redesigned fan blades, the engine was limited to 1000-rpm
fan speed, thus limiting available analytical data for comparison. The vortex
destroyer seemed to have a small positive or negligible effect on stresses;
however, it was decided to complete all remaining engine testing with the
destroyer in place. Wind conditions and the limited data (due to the 1000-rpm
fan speed limit) did make comparisons difficult.
Table 3-1. Power Turbine Vibratory Response Resolution.
Vortex Destroyer
Change Test Sites
Component Strain Distribution Test
Material HCF Testing
Updated Scope Limits
No Significant Effect (Section 3.13)
Slight Improvement (Section 3.14)
Mean Stress Analysis Verified
Stage 12 Engine Gage Poorly Located
for 2N Mode
New Stage 12 Gage Applied Which
Read Maximum Stress
Significant Property Improvement
Over Handbook Average
Stage 5 Frame Okay
Stage 12 Frame Marginally Okay
8O
3.14 EFFECT OF TEST SITE CHANGE ON STRESS
To try to reduce vibratory stress levels in the fan blades and the forced
vibration of the power turbine frames, the engine was moved from Site 4A to
3D. The site change increased the clearance between the ground and fan blades
from 28 inches to 64 inches as well as eliminated the frontal blockage area at
Site 4A caused by the permanent facility fan system that was in front of the
UDF TM inlet. Stress data are shown from both sites for the fan blades (Figure
3-52) and the turbine frames (Figure 3-53). There was no significant decrease
in either fan blade or turbine frame stresses. Fan blade stress data, after
the site change, was prior to the installation of the redesigned fan blades
>-O
Stage 1 Fan Blade Gage #KD9108Baseline
24
o 2-17-86
20 + 2-18-860 2-18-86
_ 16 A 2-28-86x 2-28-86 • " ° °¢o 12 • • l . .
• o ii a
=_ ." : ! 0 ° ° o ,, a8 "'. ; :, :: .
i! +I
RotorSpeed(rpm)
24
"o 20.
=_.,
j 18w 12
• 80
Stage 2 Fan Blade Gage #KD9212Baseline
o 2-17-86
+ 2-18-86
0 2-18-86
• 2-28-86
x 2-28-86
ou
o °o
o on
o u o. 1 f
i o ° ° °n +
iiii 'i4 :
07oo mo 90o I+ +too I+ I_ 1,oo
Rotor Speed (rpm)
Stage I Fan Blade Gage #KD9108With Vortex Destroyer
24
_" _ 2-28-862o
+ 3-5-86
= 18
!u_ 12
; 8o
4 z ; ;t 1 ; '
lO0 81)0 900 11)00 1100 1200 1300 1400
Rotor Speed (ri m )
Stage 2 Fan Blade Gage #KD9212With Vortex Destroyer
24-
20'
:1 16e
u) 12
O• 8O
4
O 2-28-86
+ 3-5-86
io
' ; _i ,_ 9 aQ
• II
0
700 800 900 1000 1100 1200 1300 1400
Rotor Speed (rpm)
Figure 3-50. Fan Blades Vibratory Response, Effect of Vortex Destroyer.
81
+
mC3_x
"2,C_
0
D.IC
III_O
01_1111
II l I I I I I I I I LIA I I I I l I I I I I I_L "_ _. o _ r_--E -- E
ep!s_ 'ssaJlS IleJaa 0 _ ep!s)l 'ssaalS IleJaa 0
3 33 3cc _ ¢
o_ _ .... o
o ooo _
o° _,_co oo
÷_ o _,-_-_ oo _,_:= oo _
o
P +
0
It')
O_t_
U)
0
k_0
_._
_Jk_
0
_J
L_0
_J
if'3
I
82
==
--.o. _"o_
• _.= Q. .=J CO(/) I I I_ I I I
,<D
(n
e.
Ul
Vl
4)Im
U)
m
n
Qco
<1:
E_- _ _ _ _ _o oi=,
0 Aauenb_Jd0o
m_1_
01 _. ?
-- 2
" -_- 2
M -
o _
m
0
0
0
AU
E_p
o
o
_3uenbeJ_
"'t [
:1 !'1 !
E "I !
II, o
i
0 AouenbaJ:l
04
u_
I--
o
0
1.°E
.° i:
o
I
o o 0 oo
Aouenbe._.,l
.-.T
q)
U3
u3
_J
::>
c_
_J
O
c3
Orj
_J
O
r.n
c_
_J
I
c_
_J
83
Stage 5 Strut Stage 12 Strut
100, . _ _ 100 , , v _. ..
120 I-_"l u. 20
0 -- I 0 I I I I I I I I
600 800 1000 1200 600 800 1000 1200
XN48 (rpm) XN48 (rpm)
_,_ _: -_t _ _ t._.., ":
o_ 20 _ 20_ .____.._-_0 0 I I I I I I
600 800 1000 1200 600 800 I000 1200
XN49 (rpm) XN49 (rpm)
Figure 3-53. Power Turbine Frames Vibratory Response, Effect of
Site Change.
84
so that a direct comparison of fan blade stress could be made. Fan speed was
limited to I000 rpm until the redesigned fan blades were installed. Turbine
frame stress data, after the site change, is from Build 3 (after installation
of the redesigned fan blades).
3.15 ROTOR LOCKUP AFTER SHUTDOWN RESULTING FROM FUEL LEAK
While running a power-down calibration at 19,000 ibf on June 30, 1986, a
fuel leak was observed, and the engine was quickly shut down. After propulsor
spoo_ down, the Stage 2 rotor would not rotate until the engine had cooled for
several hours. At that time, the exact cause was not known; however, based on
a later teardown, it is believed that the IGV lip seal had deflected thermally
and bound to the 1-4 inner spool. When the engine had cooled sufficiently,
the IGV and the spool separated enough to allow rotation of the Stage 2 rotor.
From looking at strain gage data for Stages i and 2 turbine blades, it appears
(Figure 3-54) that the IGV and spool had started to rub as early as June 24.
Heat generated from this interference eventually closed the Stage I turbine
blade clearance causing severe turbine blade tip rubs; Section 3.17 discusses
this further.
3.16 LOW/HIGH CYCLE FATIGUE TESTING
During Build 3, 100 low cycle fatigue cycles were run for endurance test-
ing. The cycle that was run is shown in Figure 3-55.
No dedicated high cycle fatigue testing was necessary because cycles in
excess of I x 106 were accumulated through the course of normal testing. The
number of cycles accumulated on each component was calculated using the compo-
nent first flex natural frequency and the engine run time versus propulsor rpm
relationship.
Table 3-2 shows the number of high- and low-cycle fatigue cycles that the
turbine blades, fan blades, and power frame airfoils experienced.
3.17 ROTOR-TO-ROTOR LOCKUP/PROPULSOR DISASSEMBLY AND REBUILD
During an attempt to make a normal start, it was noticed that the forward
and aft rotors although locked together could be turned with force. During
85
6-24-86, 1260 rpm
9:53:09
9:53:10
9:53:12
9:53:12.5
9:53:12.8
9:53:13.5
-_"_1 !'
_. r-: i L. I.
,. " 2
_'!i.l _":
_i _ I_i T
!P. _ ....... :p. _, .....
Stage 2Blade
GageKD4E02
i_:_t- !"!-_-t'_ ,__:, i _1_! I ! ;: '_i i,i ec
i_ _,-.",:;
I
_I -1̧ -I ._- _-._ •_-"_
Figure 3-54. Suspected ICV/Spoo! Rub.
86
3.0
2.6
2.2
1.8
o. 1.4
ul
1.0
0.6 _
m
0.2-
0
i
-- Flight Idle
I I I I L I I I I I I I = =2 4 6 8 10 12 14
Time, minutes
16
Figure 3-55. Low Cycle Fatigue Cycle.
87
trim balancing, previously, the engine had run up to 22,000 ibf. No problems
occurred before the rotors locked. The force required to turn the rotors
increased when the blade actuation system was commanded to full reverse. It
was found that the forward actuator would not move to full reverse (-20°), but
only went to -12 ° . The rotors did not free up as the engine cooled, and the
decision was made to remove it from the test site to investigate the problem.
Turbine Blades
Stage 1
Stage 2
Stage 3
Stage 4
Stage 6
Stage 7
Stage 8
Stage 9
Stage I0
Stage l1
Fan Blades
Stages I/2
Power Frames
Stage 5/12
Table 3-2.
HCF Cycles
6.23 x 106
4.09 × 106
5.00 x 106
18.18 x 106
14. I0 × 106
7.5 × 106
15.15 x 10 6
1.2 × 106
2.5 × 106
1.3 x 106
3.69 × 106/
3.78 x 106
4.04 x 106/
4.2 x 106
GE36 HCF Cycle Count.
LCF Cycles
100
100
100
100
100
100
100
100
100
100
100
100
100
100
Remarks
53 Cracked Blades
No Discrepancies
No Discrepancies
No Discrepancies
No Discrepancies
No Discrepancies
No Discrepancies
No Discrepancies
No Discrepancies
3 Cracked Blade
No Discrepancies
No Discrepancies
When the engine was returned to the vertical build stand, it was noted
that the rotors rotated freely. The actuation system was exercised and found
that the forward system still would not travel to the full reverse extension.
With the actuation system in the stopped reverse position, the rotors became
bound in the same manner as seen on the test stand. A scale was utilized to
measure the force required to rotate the forward rotor holding the aft rotor
stationary. With the engine vertical and the actuator fully forward (feather
or 90°), the force to turn the Stage i rotor was 4 lb. When the actuator was
88
driven fully aft (full reverse or -20°), the rotors bound, and the force which
was required to turn the forward rotor increased to 40 lb.
Full Forward Full Reverse
Forward Rotor 4 pounds 40 pounds
Aft Rotor 5 pounds 10 pounds
The above tabulation provides a comparison of the forces required to turn
the rotors independently, with the other held stationary. It was also found
that the lack of full travel in the actuator could easily be seen when watch-
ing the actuator. The forward system lacked 0.5 inch of its full travel. As
the pressure in the hydraulic system was increased (actuator in reverse), the
force required to turn the forward rotor increased to 40 Ib at 600 psi. The
decision was made to pull the propulsor from the gas generator and begin the
teardown to investigate the rotor binding.
Removal of the propulsor revealed problems unrelated to the rotor bind-
ing. Inlet guide vanes to the propulsor had cracks at the trailing edge ID
(inner diameter) braze. The ID was very irregular, and evidence of contact
with the Stage I-4 inner spool was noted. The area between the IGV and the
inner seal was black from oil coking, and the Stage I-4 inner spool was dis-
colored from varnishing. Figure 3-56 diagrams the IGV seal area as-designed,
and after test. Figure 3-57 shows the ALF (aft looking forward) view of the
tear/crack.
The IGV had worn a O.O10-inch groove in the Stage 1-4 inner spool around
the entire circumference. The heat generated by the continual interference
between the IGV and spool resulted in a spool growth that closed the Stage 1
blade tip clearance. The spool had heavy rub indications where the Stage 1
blade tips rubbed. The Stage 1 blade rub appears to have begun as early as
June 24, after 3 hours of Build 3 testing (Figure 3-58). Examination of the
strain gage data for Stages 1 and 2 of the power turbine indicates the IGV-to-
inner spool rub occurred prior to the Stage 1 blade rub (Section 3.15).
The Stage 1 blades were found to have HCF cracks that initiated at the
leading edge root. The blades with the most visible cracks are illustrated in
Figure 3-59. There is a total of 124 Stage 1 blades; of these, 19 blades had
89
360 ° Braze Tear/(:
360 ° Hard Rub
360 ° Rub-
(0.01 Deep, No Discoloration)
Figure 3-56. IGV and 1-4 Inner Spool.
9O
OF Peefr £_'rY
<
I::=
0
b_
:>
u_
I
-_-I
ORIGINAL PAG£
B/..4,C_ AhlD '_"/JTE PHOTOGRAP, H
9].
.IJ
,.C:
0
0
0
Oq
0o,
0
0
p_
if'300
E
0
C-I
u'3
0
L_
Oq
1(.)
if3
I
92
OF POOR QUALITY
OF_ITY
4-100
_J
r._
ODC
r,.)
,--Iccl
_JC
0
qJO.C
4.1r../3
Lr_I
_J
O0
ORIGINAL PAGE
BL&CK AND WI}-BTE P1-KTT'IgGRAP_I
93
94
cracks visible to the naked eye, and an additional 34 were found cracked when
examined at 40x. The entire set was removed, and new blades were installed.
Tip clearance was increased an additional 0.037 inch at the blade leading edge
and 0.016 inch at the trailing edge for the new set of blades.
The Stage 2 polygonal ring was removed, followed by the outer seal. The
seal teeth had no damage, and coating wear was normal for the engine run time.
A cross-sectional view of the seal is shown in Figure 3-60.
Following removal of the Stage 1 polygonal ring, the aft stationary hard-
ware was removed. The eight actuator control rods which penetrate the No. 2
bearing static housing were all observed to have varying degrees of wear. The
heaviest wear was on the rods at 12 and 6 o'clock. The aft support had heavy
scoring on the forward OD (outer diameter) where the aft actuator travels over
it. Wear marks were in the same position as the sting tube support brackets.
Blocks providing support were reduced to lightweight springs during rebuild.
The scored area was cleaned and covered with a Teflon coating (Emralon 333) to
reduce friction between the actuator and support. The coating when scratched
will not flake but will lubricate to protect the material below.
Tile aft actuation subassembly was lifted out followed by a borescope of
the forward sumps. Entering the sump through the mid-driveshaft actuator rod
cutouts revealed the source of the binding rotors. The No. I roller bearing
(IR) outer race nut had come completely off (Reference Figures 6-2 and 6-3,
Section I.I). The actuator could not travel the full distance because the nut
was interfering with the last 0.5 inch of actuator travel.
The 6-12 assembly, less the midshaft, was removed for further inspection
and set in the teardown tooling.
Upon removal of the forward actuator, the IR bearing housing was removed
so that access to the IR bearing could be made. The outer race nut was laying
inside the actuator assembly, and the outer race was found almost completely
off. The rollers were pushed inward and the two shafts were easily separated.
The rollers sustained disassembly damage, making the bearing not serviceable
but repairable. The outer race had no visible grooves. The outer race nut
backed off as a result of Stage i-5 rotor aft shaft growth caused by thermal
expansion.
A
_J
@
0
c_I
_J
95
Because the carbon seal aft of the IR bearing, sealing the cavity, was in
remarkably good condition and repairable, the seal was removed for inspection,
the carbon segments were replaced, and the seal assembly installed. The mid-
sump seal aft of the carbon seal had rubbed grooves about O.14-inch long ×
0.04-inch deep in the honeycomb. As shown in Figure 3-61, the honeycomb and
seal teeth were all serviceable and reused.
All Stage II blades were heavily rubbed, and Stage I0 was lightly rubbed
(Figure 3-62); 3 Stage II blades were discovered with cracks in the leading
edge root, and the inner spool had evidence of a hard rub. The outer spool
(odd rotor) aft stages are cantilevered from the Stage 5 power frame. If the
IR bearing were not providing adequate radial support, it would account for a
rub at the aft stages of the outer spool. The 3 cracked Stage II blades were
replaced prior to rebuild. Table 3-3 summarizes the results of the turbine
blade inspection.
96
Table 3-3.
Stage
No. of Blades
No. Cracked
(IX Visual)
Additional No.
Cracked
(Borescope)
Additional
No. Cracked
(fOx Visual)
Additional
No. Cracked
(40× Visual)
Total Cracked
Turbine Blade Inspection Summary.
1 2 3 4 6 7 8 9 i0 11
124 118 120 94 90 72 84 54
19 0 0 0 0 - -
82 56
- 0
0 0 0 0 0* 0* 0* 0 3
0 0
34
53 0 0 0 0 0 0 0 0 3
(42.7%) (5.4%)
* Approximately 25% Sample; Others Inspected 100%
o'-"
I.-
I
0 m
coo
oodo
u_
=
I
._
97
V_
00c_
o
EooIcOIf....
0Z-.T
<0
0.0
c']
I
98
OF '_..........
The 2R bearing was removed for a detailed examination at Evendale. One
roller had a rounded micro dent which would not affect the design intent of
the part. The 2R bearing was serviceable and was stored for later use, if
required.
Corrective action for the binding rotors included an improved locking
feature for the 1R outer race nut and other spanner nuts in the propulsor
assembly. Upon installation of a new No. 1 roller bearing, the spanner nut
was installed with a small amount of Loctite applied axially in four places
across the threads of the nut. Both threaded surfaces were cleaned and dried
prior to nut installation. An additional locking feature was added with two
pins installed between the shaft and nut, which were also tack welded in place
(Reference Figure 6-4). The inner race nut was installed only with Loctite.
A new carbon seal was also installed to replace the one damaged from the 1R
bearing problem. In addition, the 1B inner race nut had Loctite applied; all
others were untouched and remained as they were.
The IGV was repaired and modified to include a honeycomb seal to prevent
spool damage if a reoccurrence of the rub persists in later engine operation.
Figure 3-63 is a view of the IGV/spool seal area. During assembly of the pro-
pulsor to the gas generator, the gap between the IGV and spool was checked and
measured at 0.125 to 0.188 inch, with a minimum of 0.125 inch per the drawing.
Oil drain holes were drilled into the 1-4 inner, 1-2 outer, and 3-4 outer
turbine spools to drain any accumulated oil.
Borescope ports were added through the mixer frame flange for better
visibility to the IGV and Stage 1 blades. The inspection interval was set at
10 hours to keep better records of Stage 1 blade activity and IGV spool gap.
3.18 MISCELLANEOUS HARDWARE STRESS DATA
This section presents data on hardware that had no indication of high
stress (vibratory stress was always under limits) during testing and has not
been previously presented in other sections.
99
L i 1-4 Inner Spool
_Honeycomb
Figure 3-63. IGV and 1-4 Inner Spool.
I00
Nomenclature
Outer Guide Vanes (OGV)
Inner Guide Vanes (IGV)
Forward Outer Rotating Seal
Aft Outer Rotating Seal
Turbine Spools
I-4 Outer
7-11 Outer
6-11 Inner
* I0.0 for Turbine Blade Rubs
Limits
(ksida)
20.0
3.4
74.0 (2/rev)
II.0 (3,4/rev)
8.0 (5/rev)
15.0
30.0*
Fi__ure
3-64
3-65
3-66
3-67
3-68
3-69
3-70
3.19 OIL LEAK/GULPING PROBLEM
During the latter portion of Build 2 and, more markedly, during Build 3,
the propulsor oil consumption limited the UDF TM test time (oil consumption
increased from 0.35 to 1.58 quarts/hour from Build 2 to Build 3). The engine
was limited to about 2.5 hours of testing to allow for reservicing of the pro-
pulsor oil tank. By this stage of testing, the test site facility remote oil
fill system had been removed since it would not be available for flight-test.
Two major leaks contributed to the high oil consumption. The carbon seal
in the starter gearbox was leaking into the core nacelle and then back into
the fan, and oil leaked through worn actuation rods and seals in the aft sump
wall into the sting tube area in the center of the propulsor (aft stationary
support). Oil lost into the sting tube could be pumped out after engine shut-
down. A total of 39 quarts was recovered during Build 3 testing (66 hours).
The fan nacelle and propulsor rotors remained dry, indicating the oil loss was
not through the main propulsor carbon seals. Oil was also found leaking from
an instrumentation fitting in the aft sump wall. The propulsor lube oil leak
limited the engine test time in two ways.
First, as previously stated, the low lube level inside the sting tube
required careful monitoring to ensure that enough oil did not collect to flow
out of the mixer frame. The cavity was pumped approximately every 2.5 hours,
and about 2.5 quarts were removed each time (toward the end of Build 3).
101
2000
1600
Frequency
1200
8OO
400
1st
Torsion
1stFlex
0
700 800 900 1000 1100 12001300 1400
XN49 rpm
Maximum Stress
1 ksipp
• 1F And 1T Responses
• Low Amplitude
• Maximum Allowable 20 ksipp
• Predicted Values
- 1F 600 Hz
- 1T 970 Hz
• Data Typical For Engine
Testing To Date
Figure 3-64. Power Turbine OGV - Engine Test Data.
i0_
Frequency
2000
1600
1200
80O
400
IGV Frequency Prior To Stall Event
1 ksida
XN48-- 1350 rpm
T46 -- 1360 ° F
i "-----T----I ----I .... I - I ---I I I I I I-- I | I
0 20 40 60 80 I00 120 140
Time (Sec)
• Engine Test Dynamic Strain
Gage Frequency -- 1475 Hz(1 Flex)
• Predicted 1st Flex Frequency Is1425 Hz
• Allowable Stress Level Is3.4 ksida
• Trace Is Typical Of VaneData Seen During Engine Test
Figure 3-65. Power Turbine IGV - Engine Test Data.
103
450
400
350
300
= 250
@
200 -
150
500 600 700 800 900 i000 ii00 1200 1300 1400 1500f
I w _I _ ° I XN48, rpm
YIIIEINM.J. i• IN TOP[ tlS8 lidlqPU[ Mlrl[. 1600 OENSITIUITY- I SO
TKI6 I(DAS0! lue _s Buy,, cvcu. o ee Kect o;m. It|HCZL ,_L_ CUt- O _,rrxe,, cem- t
£3G e82-eo ,2 2-4-86 now
Figure 3-66. Forward Outer Rotating Seal Response, Decel from
25,000 ibf Thrust: February 4, 1986.
104
ORIGINAL PAGE IS
OF POOR QUALITY
450
400
350
150
i00
50
#20--
o
15
10"
i
5"
500 600
INOaqlI m " [
Om
TNKI? KD4$04
700 800 900 i000 ii00 1200 1300 1400 1500
XN49, rpmT*I'( Iml] _ k_. t_ II[nllTlUI_. |U/ICCEL 15'13 J4tS IIUTVeke" t _ K0_ IIU* |1| NI
FILTI[J _" 0 WUNCTIm GON* I
GE36 eBa-_l/_ 2-4-86
Figure 3-67. Aft Outer Rotating Seal Response, Accel to
25,000 ibf Thrust: February 4, 1986.
105
i000
900
800
700
600
500
400
300
200
i00
0
500 600
I Ml_qll I_ * J
'l_gllllm,Jh 0 |lION
TKIII?, KD4EOS
700
TAP(I|I|
GE36
800 900 i000 ii00 1200 1300
XN49, rpm _m_EM_. B_
FILYI_I cI- ?
eB2-eel/e 2-4-86
1400 1500
I[IltlTIUllrV* I SOlllll* IM4
JrkeCTlm COU- I
mlJ
Figure 3-68. Outer Turbine Spool Response (I-4), Accel to
25,000 ibf Thrust: February 4, 1986.
106
ORIGINAL ..... _
OF POOR Qu_
CR:C;;4_- _,, .....
OF POOR QUALITy
450
400
350
300
L3
250
200
150
.
50
.._+
JOO 600l"1 00 ipel l
_8IIIOLI- I 1O81O
TKI4 KD4L$3
+.++..
..++
+.
700 800 900 i000 ii00 1200 1300 1400 15OO
T_ U_I XN48, rpm _ b_- IlMII IINIIIIIVIW. t N
till tells ItttV ¢_tJr. t IS IILO_ Jill- ltlACCIL Ir|LTIN COE- lJ IrMI_TIONCON- t
GE36 082-e_)1/2 2-4-86 m
Figure 3-69. Outer Turbine Spool Response (7-11), Acceleration
to 35,000 ibf Thrust: February 4, 1986.
107
ORIGINAL P_.C_ iSOE POOR QUALITY
I,,,,I,,,,I,,,,I .... I,,,,I,,,,I,,,,I,,,,I .... I,,,,,
- t i il i
w
tlli
-II
-tl
b
-ll
-|
-t
O
,l.,i
0
O
.l-I
<
0
!
0
<_1
i.--4
e_
108
Second, combined with the excessive leakage, the gulp limited testing
duration. The propulsor gulps about 2.5 quarts of oil with increasing power
from idle to maximum. This is consistent over all engine testing. As speed
is reduced, the oil level will return. That is, it will return if the propul-
sor is not losing oil from leaks in the sump wall or gearbox carbon seal.
With a normal oil consumption rate, oil gulping is not a problem with the II
quart tank.
Following Build 3, the leaks were fixed; new actuation rods and seals
were installed in the actuation system, a new starter adapter gearbox was
installed, and the instrumentation fittings were replaced. A larger scavenge
port was installed in the new starter adapter gearbox to improve scavenging
and to keep oil away from the aft carbon seal that leaked on Build 3. A new
flight-type static vent air demister was installed, and oil drain holes were
added to the turbine spools.
109
4.0 HEAT TRANSFER AND SECONDARY FLOW SYSTEM
4.1 MIXER FRAME HEAT TRANSFER
The mixer frame flow circuit is shown in Figure 4-i. At takeoff power,
film-hole flow is 15% less than predicted, outer flowpath secondary flow is
10% less than predicted, and total sump flow is 6% higher than predicted. The
outer flowpath flow and sump flow are determined through use of pressure tap
readings, and film-hole flow is determined by subtracting these flows from the
total fan bypass flow.
Figure 4-2 provides a comparison of mixer frame predicted and measured
internal pressures at takeoff power; whereas, a comparison of predicted and
measured flowpath static pressures at takeoff power is depicted in Figure 4-3.
Although very few pressure readings are available, test data indicates that
the pressure drop across the film holes is close to that predicted. The low
film-hole flow appears, therefore, to be due to undersized film flow area.
The total pressure drop available to the frame film holes is very low,
making BFM (backflow margin) a concern. The regions of minimum BFM are the
inner flowpath fairing upstream of the strut leading edge and strut leading
edge cavity at the root. A comparison of predicted and measured pressures for
engine Build 1 is shown in Figure 4-4. While the BFM across the inner flow-
path fairing film holes is slightly less than predicted, the BFM across the
strut leading edge cavity film holes is much higher than predicted.
Installation of power turbine blade dampers for engine Build 2 resulted
in a lower level BFM. The inner flowpath fairing pressures for engine Build 2
are shown in Figure 4-5 for takeoff and maximum cruise, the minimum backflow
condition. Takeoff pressures are taken from test data, and the maximum cruise
pressures are predictions based on test results. The 0.79% BFM at maximum
cruise condition is low, but it is in a very localized region. The BFM in all
other regions is significantly higher.
Figure 4-6 illustrates mixer frame metal temperatures, both predicted and
actual, for hot day takeoff power. All thermocouple data have been scaled up
to hot day conditions by multiplying the raw data by the ratio of the hot day
iiO
..... ,"L :o/:l,i_F i_
I Test iPredictedJ
All Flows in %W25
1.41
.56
4.27
5.63
0.32
0.31
9.77
2.13
2.45
2.62
2.47
0.32
0.32
Figure 4-I. Mixer Frame Flows at Takeoff Power, Test
Versus Predicted.
III
47.4
47.2
47.0
C_ 46.8
46.6
46.4
46.2
46.0 "15
? Predicted Total Pressure
I I Ii 16 2
Station
e2
112
Figure 4-2. Mixer Frame Pressure Losses Comparison of Measured
and Predicted Pressures.
45
44.el
43
c_
42-el
i ath "_
i_ _ Outer Fl°wpath Predlcte d
Q Test Data
40 _ _ i i I-4 -2 0 2 4 6 8 10 12
Axial Distance from Strut Leading Edge, inches
14
Figure 4-3. Mixer Frame Flowpath Static Pressure Distribution Takeoff Power.
113
.I
B
C,d
1.4_
Z
ZZ
N
Q_
r_
N
z
_.JId_
_w
I.LJ ,--,
_ I..LJi.._ _"
o-
zlp__-._
_'_
.._
I--
I--
z
._.1
..-.1 I---
_I.LJ
Id-- I.I--
----I
I'--"
._I I.LJ
Z Id.
r,.- _ _..j
L_ _
I
,r-I
o
!
114
L.I-L.L-0
I"--
L.J
X
I..L0L._
h--
N,---I
II
L.L.
0
h'-"
0 _ _..I
X (-_ 0
t--- ¢_11.LI _'_:=
_ _ LJ- -
I B D
\
o
I
o
u_
I
i18
u_ oo
u_
QJ
E
I
116
ambient temperature over the test day ambient temperature. No thermocouplesare available along flowpath portions of the frame, but the existing thermo-
couple readings indicate that the frame is adequately cooled. Also, there are
no visible signs of overtemperature.
4.2 POWER TURBINE SECONDARY FLOW SYSTEM
Figure 4-7 illustrates the power turbine secondary flow system. The flow
rates, predicted and actual, are tabulated in Table 4-I. Although the total
secondary flow for engine Build I neared prediction, the Stage I inner flow-
path purge flow was significantly lower than predicted. This redistribution
of flow is a result of a mixer frame supply hole flow coefficient that is
lower than predicted and shafting hole flow coefficients that are greater than
predicted. The increased pressure drop across the mixer frame supply holes,
combined with reduced pressure drop through the shafting annular holes, serves
to lower the supply pressure to the forward seal. Also, the sump vent flow
was higher than predicted due to large carbon seal leakage areas. This addi-
tional vent flow results in a further decrease in Stage I purge flow.
During Build 1 testing, Stage I cavity temperature was 830 ° F at takeoff
power, as compared to 795 ° F predicted. Although there was no evidence of gas
ingestion, it was observed that the forward seal AP decreases with increase in
power setting. Figure 4-8 plots AP versus P46Q2.
Extrapolation of the test data indicates that for Build 1 testing, back-
flow could occur at maximum power. To maintain positive flow in all cases,
the mixer frame supply hole area was increased from 5.4 in 2 to 5.84 in 2. The
resulting increase in forward seal AP from Build 1 testing to Build 2 testing
is significant (Figure 4-8); however, the impact of this area increase on the
remaining purge flows is negligible. The aft labyrinth seal flow increased
significantly from Build 1 to Build 2 testing, presumably due to an increased
seal clearance, indicated by a reduction in pressure drop across the seal.
During Build 3 engine testing, rubs occurred on two middle inner flowpath
seals (G and H). There is an increase in the flows of these two seals, but
because the system is metered for the most part by the supply hole and the
shafting holes, the total secondary flow increases only slightly (Figure 4-7).
i17
|
0
0U
C¢1
,r.,t
0
I
=
I18 ORIGINAL PAGE l_
OF POOR QJAL,, Y
0
0_J
0
0
_Jo_
_D
0n,
T
_D
E-_ oo oo oo
119
PROBLEM. EXTRAPOLATIONOF TEST DATA
INDICATESTHAT FORWARDSEALCOULDBACKFLOWAT MAX POWER,
| CAUSE: UNDERSIZEDSUPPLYHOLE AND
OVERSIZEDANNULUSHOLESREDUCETHE SEAL SUPPLY
PRESSURE,
| SOLUTION:INCREASESUPPLYHOLEAREA,
1.0
.8--
°6 _
,Q .5_
,=l
.3_0b.,
0 --
-1
1
EST DATA
BUILD Ol
I I2 3 4
P46/P2
Figure 4-8. Potential Forward Seal Problem.
120
It should be noted that the very limited amountof instrumentation makes
direct measurementof most of the individual flows impossible. The available
pressure and temperature readings were used to rebalance the analytical flownetwork model which yields the best estimate of the flow distribution. The
two flows leaving the mixer frame (S and T) are directly measured.
According to thermocouple data, all cavities are adequately purged. Acomparison of predicted and actual temperatures is presented in Figure 4-9.
4.3 POWER TURBINE HEAT TRANSFER
The power turbine spool and frame temperatures, predicted and actual, are
shown on Figures 4-10 and 4-11 for hot day takeoff power. The original pre-
dictions (Figure 4-10) were determined using a finite difference heat transfer
program. The actual temperatures (Figure 4-]I) have been scaled up to hot day
conditions, as described in the mixer frame section. Because the fan bypass
air temperature is lower than predicted, most of the cavity temperatures are
also lower. The cavity temperatures have a direct impact on the inner spool
temperatures. Unfortunately, there was too little instrumentation to be able
to draw any firm conclusions about inner spool temperatures, as compared to
predictions.
4.4 NACELLE VENTILATION
The nacelle ventilation system is depicted on Figure 4-12, with component
temperatures at takeoff condition. All thermocouple data have been scaled up
to hot day conditions, as described in the mixer frame section. Although some
components are higher in temperature than predicted, all of the hardware is
adequately cooled.
During initial engine Build 1 testing, ambient air was to be brought in
to the nacelle cavity aft of the Stage 2 telemetry through radial holes in the
cowling. However, poor ventilation in that region led to an overheating of
the aft telemetry, probably due to a low level of static pressure at the inner
flowpath. A total of 30 air scoops were then mounted to the holes. The addi-
tional ventilation air resulted in the temperatures presented in Figure 4-13.
121
122
|
r
/
i
rJ
.._
bo
<
o,,-.i
o c_
0 _ "_ _-_
0 _ cd 03
_._ ©
._
0
©
m_
[-_
:>
0
©
I
I.-i
DRIG_NAL ',_,-c'_':
OF POOR _UAL_TY
ORIG!N,_L P/_GE fS
OF POOR QUALITY
h : 60
h =
h 5
h = 97
1023° F
6O
285 ° Fh=17 --E J3
miii
"_76 'F_• h
S11OF_. -h = 15Jr
FI
Hot Day (+27 ° F) Rated Take-Off
Steady-State Conditions
h - [BTU/hr. ft.2 ° F]
Figure 4-10. Power Turbine Temperatures Based on Analysis.
123
oO
• p
//
i
du_
0
,._
_3
:>,
e3
0
0
0
!
4_
124
i"_t_l_ _ _ _ ___ _, ._. _
OF. POOR QUAL_TI
o
l
_=_!",.I"
_D
125
ITest DataPredicted
]'Air = i00° F 174 ° F
152 ° F
281 ° F
226 ° F
l404 ° F
Figure 4-13. Forward Fan Blade Hub Temperatures; Hot Day Takeoff,
Mach= 0.0, Full Thrust.
126
It was determined during initial Build 1 testing that, although the aft
telemetry modules were adequately cooled, the solder which fixes the thermo-
couple leads to the aft telemetry circuit board, was overtemperatured. Eventhough the solder temperature is not measureddirectly, it was found that thesolder melts when the nearby air thermocouple reaches a temperature of just
over 300° F.
The solder, although adequately cooled during high power, is marginallycool at idle conditions. Set to match the prop flow at takeoff, the scoop
direction is misaligned with the flow direction at idle. For +27 DTAMBcondi-
tions, the solder temperature is 219° F at takeoff, 235° F at flight idle, and
283° F at ground idle.
Figure 4-13 gives a breakdownof the forward fan blade hub temperaturesfor hot day takeoff. The 404° F trunnion temperature is taken from thermo-
couple data. All other temperatures are determined by rebalancing the heattransfer model on the basis of the trunnion temperature. All temperatures are
higher than predicted, but are still within allowable limits.
4.5 CENTER CAVITY VENTILATION
The center cavity is portrayed (Figure 4-14) with temperatures at takeoff
power; normal operating temperatures are as listed. The actuator axial posi-
tion sensor (LVDT) temperature of 312 ° F is well below the normal operating
temperature of 360 ° F (the maximum temperature limit is 400 ° F). The speed
sensor temperature was not measured directly, but it is probably less than the
LVDT temperature since it sees less radiation from the cavity wall than does
the LVDT. In addition, the sump oil may provide a sink to the speed sensor.
There is no indication of speed sensor or LVDT overtemperature problems.
127
c_
_J
0.0
o
Z
!pl
AI
II
o
00...1-
I
0
o_
ny
I
,j
0
0
,._
0
,,_
_J
I
oO
128
5.0 ENGINE SYSTEM DYNANICS
Valuable engine dynamics experience was obtained during ground testing of
the UDF TM Demonstrator Engine (GE36 S/N 082-001). This ground testing, along
with a series of mechanical impedance tests conducted on the support system
and Peehles test facility, was used to modify and verify the analytical model
of the engine. Topics that are discussed in a later section regarding this
subject include such gas generator vibration signatures as: synchronous I/rev
IPC and HPC responses, linear subidle UDF TM I/rev signature and its dependency
on support structure/test facility stiffness, linear UDF TM I/rev response
recorded in the operational speed range, and nonlinear UDF TM I/rev response
observed during the Stage 2 propulsor airfoil separation event.
The UDF TM, with its counterrotating propulsion system, also demonstrated
that more sophisticated methods and hardware are required beyond turbofan
engine experience with regards to propulsor trim balancing and vibration
measuring techniques.
Gas Generator Vibration Signature
The gas generator used to power the GE36 Demonstrator is an F404 engine.
Its rear frame is replaced by the mixer frame, and the gas generator is mated
to the UDF TM. Gas generator synchronous IPC (intermediate pressure compressor)
and HPC (high pressure compressor) vibration levels were well within the pre-
scribed F404 limits throughout ground testing. Both the maximum IPC and HPC
1/rev levels observed in the operational speed range during testing occurred
at the F404 fan case vertical location.
0.24-inch/second (average velocity) at
Build 3 testing.
The maximum IPC 1/rev response was
11,700 rpm and was observed during
Figure 5-I presents IPC signatures for Builds IA, 2, and 3, demonstrating
that the IPC I/rev levels remained similar and low throughout ground testing.
The maximum HPC 1/rev response was 0.28-inch/second at 13,700 rpm as observed
on Build 2. Figure 5-2 compares HPC I/rev vibration signatures for Builds IA,
2, and 3 (F404 fan case vertical location). Like the IPC vibration signature,
the HPC synchronous response remained similar and low throughout testing.
129
spuoaas/saqau$(_AV) _aTaOlaA
d0m
e_
0
_J
4-Jc_
Iu_
>
i
I
G
_._
;°°
i !
0oO
0
oO
o8
>q_
rz:I
I%
0
_4!
130
The F404 predicted engine dynamics show little changebetween the turbo-
fan engine and the GE36gas generator application. These predictions wereverified by comparing the vibration data from the check-out (at GE-Lynn) of
the turbofan configuration with the results observed during the GE36groundtest. Demonstrating little change between both configurations, the results
are shown in Figures 5-3 and 5-4; readings were taken at the F404 midframe
horizontal location (only commonaccelerometer location between the two con-
figurations) for the IPC and HPC1/rev vibration signatures, respectively.
UDF TM Vibration Signature
The test facility and aircraft structure play a major role in the overall
support structure stiffness of the engine and resulting rigid body mode defi-
nition of the entire system. Mechanical impedance tests were conducted during
January 1986 (between Builds IA and 2) to obtain the support structure stiff-
ness. The pylon/isolators system was tested with the pylon attached to ground
to obtain pylon and isolator stiffnesses and also mounted at Peebles Site 4A
facility to obtain the entire support system stiffness. These results were
incorporated in both updated linear and nonlinear (propulsor blade-out) ana-
lytical dynamic models and will be referenced in subsequent discussion of UDF TM
vibration results.
The rigid-body modes occur primarily in the subidle speed range. Since
the demo engine was tested at both Sites 4A and 3D, subidle resonances were
subject to change due to the differences in facility stiffness properties. To
demonstrate these differences, a comparison of Build 2 UDF TM I/rev vibration
signature obtained during engine starts at both sites are shown in Figures 5-5
and 5-6. Figure 5-5 compares the subidle vibration signatures at the F404 fan
case vertical location (vertical direction for Peebles test is in line with
the strut). The signatures indicate that the overall support system stiffness
was softer at Site 3D in this direction. Figure 5-6 compares the two subidle
signatures at the same location in the horizontal (normal-to-strut) direction.
Overall system stiffness effects at Site 3D acted to decouple the predominant
rigid-body modes observed at Site 4A.
The UDF TM I/rev operational speed range vibration signature at the No. 2
ball bearing housing location is illustrated in Figure 5-7 for the vertical
131
0
0.!-I
1-4q.I
&l"L3
o_qJ
0,z_ 0
0 _q
q3
I'II!
I
000¢q
00
00
C
0 r._
0
C _
p_
0
0
00
--0
000
spuoa_s/s_q_uT (_AV) K]!aOl_A
d0
°,q
0___
q,l
4-1
O,O
o_
&l
I
&l
I&l
L.)
I
q,l
0
04-1
&lio
•_ L3O.O
0,-_ 0
[--, ,_
q3
IIII
II
!
I
00
00
00
.q,-
_2(11
0
00
000
0000
spuoaes/s_qauT (gAy) Xa_aOI_A
0
0cJ
&l
4-1
0,0
>e_
r_I
q_
I
I
132
\ i
0
iiiLx'_p..
0
0
0
0
I
I
0
._
I
.,-I
0
c_0
(V_) SlIm ']uama_Ids!c
0
0
0
0
>
>
I
I_D
0
o_
I
133
direction and in Figure 5-8 for the horizontal direction; each indicates the
deflection in mils (DA) as a function of propulsor speed for accels conducted
to 24,000 ibf at Sites 4A (Build 2) and 3D (Build 3). Although the signatures
are similar, differences in levels were noted; the differences in levels are
accounted for by two explanations. The first is that the propulsor blades
were modified, and a nominal correction weight was added to the Stage 2 rotor
between runs. The second explanation deals with the repeatability of vibra-
tion on a closely synchronized counterrotating propulsion system; this subject
will be discussed in a later section.
Stage 2 Airfoil Separation Event
The Stage 2 propulsor airfoil separation event that occurred on Build 2
(Site 4A) led to an opportunity to modify and verify the nonlinear dynamic
model predictions (by mechanical impedance test results) for this event and,
subsequently, to apply this knowledge to the blade-out design criteria of the
aircraft application. The dynamics, both actual and predicted, were presented
during the May 13, 1986 Quarterly Review at the Nasa-Lewis Research Center in
Cleveland, Ohio. A major benefit resulting from this event was proof that the
pylon isolators function as designed. This was demonstrated by a significant
reduction of motion through the isolators, and the fact that the isolators
soften with the increased loading and, thus, lower the system resonant speed.
Trim Balance Experience
The initial efforts to trim balance the propulsors during Build 2 were
hampered by the hardware used. The basic trim balance test sequence was to
separate propulsor speeds by 100 rpm (1.67 Hz) and take amplitude and phase
readings from the existing SDll9C Trim Balance Analyzer unit at predetermined
steady-state speed conditions. Making the initial trim balance unsuccessful,
the built-in 3 Hz (41.5 Hz) bandwidth tracking filter of the SDll9C unit did
not adequately separate the vibration response of the two rotors and, thus,
did not give correct amplitude and phase information.
A modified SDll9C unit was purchased with a 1.0 Hz (40.5 Hz) bandwidth
tracking filter and was evaluated during Build 3 testing. Improvements were
achieved in the balance of each rotor utilizing this new unit and applying
134
20.0
_o:2_aLL_eaLr_n__°u_ngVertical
<_vv15.0J .................. , .............................................................................................................
| I sS_ u_,o.d...................,_.........._.-/..................................................:..-........-t/ ' I' l_'^ /-- Build 3
| ...................
_.d.................,,....../'.,.':._.........,,,.,,..,,...._.\ .....,,,.,......../._................/ t, ,, \ _--'" .... ,X,' V _/ L;
/500 700 1100 1 00
Forward and Aft Propulsor Speed, rpm
Figure 5-7. UDF TM One-Per-Rev Vibration Signature on 2 Minute Accel to
24,000 ib Thrust, Housing Vertical.
15.0
<c_
10.(._.-4
Q.I5.
C_
I
No. 2 Ball Bearing Housing Horizontal I
T
.............:i:!-:/..................... ..........'--_- .:,,d,,--,f/_--_1-....................',-'II_.....',, ...................................... " ............
_Ioo " ",'oo...... _,'oo"" " ,Joo......... ' I " ! I ' ' ' !• ' ' i.'O0
500
Forward and Aft Propulsor Speed, rpm
Figure 5-8. UDF TM One-Per-Rev Vibration Signature on 2 Minute Accel to
24,000 ib Thrust, Housing Horizontal•
135
force correction weights. The sensitivity to the force unbalance was deter-
mined to be 325 gram-inches/mil which is 2 to 3 times less sensitive than fan
unbalance in a large bypass turbofan engine. But even with the improvements
noted above, the vibration response observed during power hooks (constant rpm
thrust excursions) was still considered undesirable. The test on this build
ended before a good set of high thrust data could be obtained during the power
hook excursions.
The amount of data accumulated in these balance exercises did, however,
indicate that one or more excitation sources, other than a pure force unbal-
ance, was contributing significantly to the UDF TM vibration signature of the
engine. Efforts are currently underway to evaluate each of the following
possible sources:
• Changes in either force or mechanical moment unbalance due to
the blade actuation system (from low- to high-power condition)
• Aerodynamic moment unbalance being introduced by any propulsor-
blade tracking problem.
Measuring Techniques
For a simpie one degree of freedom system, the maximum sensed deflection
would lag the forcing function (rotor unbalance) by 90 ° at its first natural
frequency. The deflection level is a function of the unbalance force magni-
tude, the damping, and the mode shape at this given frequency. For a single
rotor application, such as a high bypass turbofan engine, the maximum static
deflection at any circumferential location, even though not necessarily equal,
would occur in one revolution of the rotor and wouid be repeatable from one
revolution to another, provided the unbalance or rpm were unchanged. Using
this simpie one degree of freedom system and assuming that the mode shapes in
the vertical and horizontal directions are identical, Figure 5-9 demonstrates
this point.
The UDF TM engine, on the other hand, may not sense the maximum deflection
at a given circumferential location during a revolution of its counterrotating
propulsor rotors. For example, assume that both the forward and aft rotors
are exactly synchronized, that the forcing functions are equal at each rotor,
and the mode shapes are identical in both the vertical and horizontal planes.
136
0
4.;
u3.
t,,
4-I
-f
,7
0>
4.1
4.1
0°_
t_
4-J
_q
E
I
c_
137
Now, using the single rotor exampleabove, we have two cases (Figures 5-10 and
5-11) which showeither the vertical or the horizontal deflection at a maximum
level, and the other plane at a minimum.
This example indicates that for a given revolution, if the vertical loca-tion is at the maximumlevel, the horizontal response is zero, and vice versa.
This example also shows that the maximumdeflections in each plane are equal,
and that the minimumdeflections are not only equal but are, in fact, zero.Expanding this example into real-world application, we find that the maximum
response in a given plane will always result in the minimumresponse occurring
in its orthogonal plane.
First, let the force unbalance of each rotor be different. In this case,
maximumdeflection for both the vertical and horizontal planes will still beequal, as will the minimumlevels, but these minimumvalues will no longer bezero. Second, let the modeshape definition and force unbalance be different.
At this point, neither maximumnor minimumdeflections between the two planeswould be expected to be equal. The final step in approaching the real world
is to leave the one degree of freedom system and enter the actual multidegree
freedom system situation. This step does nothing more than to vary the phase-angle lag between the lined up forces, and the resultant maximumdeflection
circumferential location, as a function of the engine dynamics.
Test data is provided in Figure 5-12 demonstrating the above discussion
and illustrating the UDFrMI/rev response at ground idle power for the approxi-mately orthogonally mountedaccelerometers at the No. 2 ball bearing housing
location. Also shownis the phase-angle lag between the forward and aft pro-pulsor i/rev indicators. The phase data indicates that the rotors are close
but not totally synchronized. The vibration data shows the 1/rev levels are
modulating at the samefrequency as the propulsor rotors difference frequency.These data demonstrate, as earlier stated in the discussion, that the maximum
response in one plane occurs at the minimumresponse of the other plane, andvice versa. The two rotors unbalance line up at a phase-angle lag of 240° to
yield the maximumvertical response and at a phase-angle lag of 60° to yield
the maximumhorizontal response.
138
Cx,I
0
_ g
T "_
¢n
©.r-t4_
0>¢)
cI
rJ
,.-!.r-I
r_
0
U
I
¢1
139
f_
0
¢nLI
J
-t4.J
2
:>X
/
0
0::>
q
B
0:>
._]_
c_4-J
0
0
_-]
r_
0_._
c_
,--t
I
0C
140
NmZ
CD"
_4
<J
O-
L_
50--
No. 2 Ball Bearing Housing Vertical
40-
30- _UDF One-Per-Rev Velocity Response
20- . ' i _IU_,,,,,,.
I0 ,J ,_l l i. I
0
5O
200.
_J
"o I00.
g_
No. 2 Ball Bearing Housing Horizontal
40-
30- /--UDF One-Per-Rev Velocity Response
04 ..... [ ..... I''"''1 ..... I'"'''1 ..... } ..... t ..... I ..... I .....
-I00-
iiiiiiiiiiiiiiiii":: ..- iiiiiiiii!iiii
....
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 I0.0
Time, minutes
Figure 5-12. GE36 Demonstrator Engine 082-001/3
T/D 7-1-86 Time History Plots at
Ground Idle Power.
141
This discussion explains why variation in vibration would be expected in
the UDFTM I/rev levels shown in Figures 5-7 and 5-8. Although the rotors are
not precisely synchronized during the throttle advances, they are close, and
therefore, the response at any given speed is dependent upon the orientation
of the unbalances to each other and to the sensing device.
Realizing that this characteristic existed prior to initiation of ground
testing, GE Aircraft Engines developed a system to optimize the chance of cap-
turing the maximum vibration level each revolution. The automatic vibration
engine shutdown/aircraft monitoring system has two sets of orthogonal acceler-
ometers mounted on the engine. Each signal of the orthogonal accelerometer
goes through a software package that takes the square root of the sums of the
responses squared. This method vastly increases the probability of capturing
the maximum vibration response each revolution and is, by far, a more accurate
measuring tool for synchronized counterrotating rotors than the conventional
method utilized to measure high bypass turbofan engine vibration.
142
6.0 BEARINGS AND SEALS
This section covers the main propulsor bearings and all actuation system
support bearings. All of the bearings performed well during ground testing,
from a design standpoint. Some problems arose as a result of debris produced
during manufacture of other engine parts; however, engine operability was not
compromised. Seals in the fan blade retention system performed well during
testing except for a few torn fan blade thrust bearing seals resulting from
insufficient lubrication.
6.1 HARDWARE CONDITION
6.1.i Rotor Support Main Bearings
Debris Ingestion
The main bearings operated well, which was to be expected, as they had
ample capacity.
The two roller bearings were replaced twice during ground testing when
they were exposed at disassembly. They were replaced because debris damage
had scored some of the rollers. The debris was comprised almost exclusively
of O.007-inch-diameter steel shot particles, with some other manufacturing and
wear debris present in very small quantities. Prior to engine teardown, oil
sample analysis had identified the possibility of a debris problem.
A blind cavity in the forward intershaft carbon seal land, designed to
prevent possible thermal coning, had trapped a quantity of steel shot during
the peening process. The shot did not wash out during manufacture but was
sluiced out by the hot lube oil during engine running.
Causing extensive scoring of the rollers, the shot became imbedded in the
soft silver plate of the rolling element retainer of the bearing. This would
cause both a breakdown of the hydrodynamic lube film because of the high loads
and low rotational speeds and a diminished fatigue life. However, no surface
distress due to rubbing was observed.
Bearings were returned to the manufacturer for refurbishment, although
only inner rings and cages were usable. The rings were rehoned and the cages
143
stripped and replated. The design of the roller retention features permitted
roller removal without damage to the cage.
To prevent further debris damage, the blind cavity was ultrasonically
cleaned and sealed closed with a nichrome strip (Figure 6-I). This was fairly
successful, although one or two pieces of shot were still found in subsequent
oil samples.
Retent_n Nut Loosening
The outer ring retention nut of the IR intershaft bearing came loose and
completely disengaged during the final stage of ground testing; this permitted
axial movement of the outer ring.
The nut is believed to have jammed the actuation system preventing blade-
angle adjustment. On teardown, forward and aft rotors could not be separated
in the normal disassembly sequence. This was caused by the rollers no longer
being in contact with the outer raceway. Borescope inspection prior to dis-
assembly had shown the rollers still engaged, so running in this disengaged
condition had not occurred. When the rotors were separated, the bearing was
found to be in remarkably good condition, except for the damage to the roller
corners caused when the rollers dropped into the region of the outer shaft,
from which the ring had moved, and hung up on the shaft shoulder (Figures 6-2
and 6-3).
Examination of the spanner nut indicated no obvious thread damage, and
the Vespel insert retention feature appeared undamaged. The appearance of the
raceway did not indicate any running off the normal roller path; however, some
coning was indicated. The shaft coning was in the direction to move the shaft
radially away from the nut, and the wedge effect would exert an axial force on
the nut in the direction to promote untorquing. There was no damage from ring
spinning on either the outer ring OD or the shaft bore.
For subsequent engine testing, a hole was line-drilled axially in the
thread, and a roll-pin was mechanically locked in the thread to prevent the
nut becoming untorqued (Figure 6-4).
144
_ircumferentia!Nichrome Strip
Figure 6-1. Seal Cavity Closure.
SpannerNutOuter Race
Figure 6-2. Possible Running Condition withIR SpannerNut Loose.
145
Rollers Having MovedRadiallyOutwardContact Shaft Sho_
Figure 6-3. Disassembly Problem and RollerCorner Damage.
Axial Roll PinWeldedin Place
Figure 6-4. Problem Solution - PositiveNut Locking.
146
6.1.2 Inner Actuation Bearings
Actuation Ball Bearings
No problems were encountered with the large mainshaft actuation ball
bearings. These bearings were examined at the last teardown for any evidence
of irregular wear or cage impact that might occur due to the calculated axial
distortion of the outer ring from the four gear rack loads, but no indication
of abnormalities was seen.
Radial Quill Shaft Gearbox Bearings
The inner bearings supporting the radial shafts could not be seen without
teardown of the gearboxes; however, since no problem existed and they operated
flawlessly, no teardown was performed.
The outer gearboxes were examined because of the loosening of the outer
housing nut. The nut butts against and clamps the outer ball bearing, which
locates the pinion gear. The cause of the loosening was found to be a stackup
problem incurred by a change in a washer thickness (Figure 6-5), and the prob-
lem was corrected. Although the bearings are grease lubricated, most of the
grease had been expelled, and some bearings showed wear.
6.1.3 Actuation Tapered Roller Bearings
The tapered roller bearings supporting the counterbalance torque tubes
and bevel gears were examined and showed no deleterious effect except for some
slight corrosion. This problem was manifest because the grease lubricating
these bearings is centrifuged outwards, and the bearings are not adequately
protected from the high humidity experienced during testing, some of which
occurred during moderate rain.
6.1.4 Fan Retention Bearings and Seals
Setup Bearing
The setup bearing is incorporated to react the blade overturning moment
and carry most of that load. However, at speed, the centrifugal load is suf-
ficiently high that a very large moment would be required to unseat the main
147
ii
m n
Oversize Vespel Washer
Cap
Figure 6-5. Gearbox Stackup Problem Causing Incorrect Loading
of Ball Bearing and Cap Looseness.
148
thrust bearing. Therefore, the setup bearing has a relatively easy life. The
only problem associated with these bearings was, again, corrosion. Although
protected by a silicone seal and lubricated by grease, because the grease ispermeable, somecorrosion was seen, but not severe and none on the active sur-
faces of the bearing.
Fan Thrust Bearing
Bearing Condition and Seal Condition - The main fan thrust bearing reacts
the considerable centrifugal load of the blade and is subjected to dither from
blade vibrations and actuation system load fluctuations. Therefore, false
brinelling and fretting were the major concerns.
Most noticeable about the condition of the bearings was the disparity in
appearance between forward and aft rotor parts. The front rotor was in a much
hotter environment, sometimes as much as i00 ° F hotter. Bearings from this
rotor were slightly blued, but the aft rotor bearings were not discolored.
Some fretting corrosion was present on both rotor bearings; however, this had
not progressed to the point were the bearings were unserviceable.
The lubricant in the forward rotor was dry and discolored; whereas, the
aft rotor grease looked like new. Grease seals in some of the forward rotor
parts had been torn due to lack of lubrication.
The roller wear paths on the raceway of the cup and cone showed evidence
of bearing distortion due to the unsymmetrical loads caused by the housings
deflecting, as was predicted by finite element analysis. The bearing cone had
been made more flexible to compensate; however, even the cone bending was not
enough to provide an even loading. Deflection analysis was performed using a
lighter blade weight than we currently have (Figures 6-6 and 6-7).
Thread Clamp Condition - During ground check-out for flight, play in the
trunnion support bearings was discovered which caused blade tip movement. The
clamp load was measured using a unique eddy-current technique for determining
the stretch in the trunnion threads. The technique had been calibrated during
ground testing at Peebles to substantiate the torques used at assembly. The
results revealed a loss of clamp load, and the rings were returned to Evendale
for retorquing the trunnions. This gave us the opportunity to examine all the
149
Actuation GearStub-Shaft
90°
Figure 6-6. Bearing Housing Deflection.
0.08 Min
0.170
0+
Min
(a)
0,12
0.095
Cup
0.08 Min
0.170
Max.
Min
i(b) Cone
Figure 6-7. Irregular Wear Pattern of Roller "Path".
150
trunnion bearings. Thus, the condition of all of the retention bearings was
known. Due to minor surface distress, coupled with increased trunnion/bearing
thread torque, two bearings were replaced; however, all other bearings were
acceptable for flight testing as is. The consumable parts (seals, shims, and
grease) were also replaced where necessary.
6.2 MEASURED TEMPERATURES
6.2.1 Rotor Support Main Bearings
The main bearings were instrumented initially with three thermocouples at
each bearing location. The thermocouples were at 120 ° circumferential loca-
tions; however, loss of signal and lack of available recording channels left
only one gage at each position providing good data.
Steady-State Temperatures
The bearing temperatures were influenced by environment (oil, metal, and
cavity temperatures) and by engine speed. The intershaft bearings ran hotter
than the Stage 2 rotor supports, as would be expected due to the higher rela-
tive speed (Figure 6-8).
The Stage I rotor bearing temperatures increased from 240 ° F at 700 rpm
(idle) to 340 ° F at 1393 rpm (maximum thrust). The Stage 2 rotor bearing
temperatures were lower, with that of the 2B bearing rising 50 ° F, to 250 ° F
at maximum thrust. Oil supply temperature rose 100 ° F over the same ranges.
Transient Temperatures
The bearings did not pick up temperature very rapidly during fast accels.
A 40 ° F rise in intershaft bearing temperatures is attributable to increased
centrifugal loads. The Stage 2 rotor support bearings changed little during a
2-minute accel to full power. However, a 4-minute accel/decel demonstrated a
similar 40 ° to 50 ° F rise in all bearings (Figures 6-9 and 6-10).
6.2.2 Fan Retention Thrust Bearings
The fan retention tapered roller bearings were not instrumented; however,
thermoeouples were placed inside the trunnion adjacent to the bearing. Since
151
i,
3
hl
2OO
I_0
100
_0
0
3OO
I00
_0
(a)
u
i
(b)
0,_
1R
b
Steady State RDGS
I I I l I I
4 8 12 16 20
¢n_u,or_.)24
1R
IB
2R
2B
-- • .___%_----- - _;- _
[] 2R + 1R 0 1R _ oR
IIJILD 00.3 - R00t B1-102
Steady State RDGS
I I I I I I I I
0.7 0.9 1,1 1.3
O_umnd=)
P
Figure 6-8. Main Propulsor Bearings.
152
LL
W
F-
LU
LL
Ld
Ld
1GO
1OO
O
2R
Transient RDGS
| I I I I I I I i I I I I I i i I |
20 40 60 80 100 120 140 180 180 :2CX)
Figure 6-9.
•_ (_=,)
Main Propulsor Bearings, 2 Minute Accel to
24,000 ibs Thrust.
Q 21R + 1B @ IlR A ,._
3:)O
f_
0
o
Transient RDGS
40 BO 1"_ I00 2QQ 240 :2;BO
_ME (..=,)
Figure 6-10. Main Propulsor Bearings, 4 Minute Accel/Decel
to 1200 rpm.
153
this is a closed cavity, the temperature would not be too far from that of the
bearing.
Steady-State Temperatures
The forward hub temperatures rose from approximately 340 ° F at 700 rpm to
about 390 ° F at 1393 rpm. The aft hub went from 190 ° to 250 ° F in the same
range (Figure 6-11).
Transient Temperatures
There was little difference between a 2-minute accel and a 4-minute
decel, in that the temperatures of both forward and aft hubs rose about 20 ° to
40 ° F. This is due to the fact that the thermocouples were in a closed cavity
anti, although influenced by turbine air gas temperatures, they are shielded
from the immediate effect by the fan ring bulkhead plate (Figure 6-12).
154
4OO
.3OO
2OO
100-
o
0.5
o _ HI.IR 4" 6Fr I-IjR
Steady State Readings
| I I I I I I I I
0.7 0.9' 1.1 1.:q(T_u,,',_s)
1.5
Figure 6-ii. Forward and Aft Fan Hub Temperatures.
I,
W
a.3
4OO
3OO
2OO
10000
o _o HUB A _ HUBTransient Readings
I I I I I I I I I I I I I I I I I I I
0 20 40 60 £0 1(:]0 120 140 1_ laO 200
TIME (_s)
Figure 6-12. Forward and Aft Fan Hub Temperatures, Accel to
Full Power.
155
7.0 PERFORMANCE
This section describes the steady-state data acquired during Build 3
testing at Peebles Site 3D and the various analyses employed to support the
test effort. The major items covered are as follows:
• Data summary
• Data reduction methodology
• Comparison between pretest predictions and reduced test data
• LCF cyclic testing and resultant deterioration
• Comparison between ground and flight pretest cycles and reducedtest data
• Overall summary
• Key results
• Conclusions.
7.1 DATA SUMMARY
During Build 1 testing at Site 4A, a total of 46 steady-state DMS (data
management system readings were taken, none of which were usable for perform-
ance analysis since all the points were recorded mainly for mechanical check-
out and were not stable.
During Build 2 testing, 104 steady-state DMS readings were taken at Site
4A, and an additional 31 at Site 3D. Some of the data points were usable for
analysis.
Table 7-i shows the breakdown of steady-state DMS readings for Build 3.
Note that of the total of 358 data points recorded, 27 readings were expressly
taken to define the baseline performance, with an additional 48 data points
recorded to map the UDF 'M performance at off-schedule conditions. Table 7-2
shows the chronology of the acquired data.
Appendix A presents a listing of the Build 3 steady-state data points.
7.2 DATA REDUCTION METHODOLOGY
Figure 7-1 shows the schematic of the engine and the positioning of the
performance-related instrumentation.
156
Table 7-1. Data Summary.
Breakdown of Steady-State DMS Readings
for Build -03.
Zero Readings
Ground Idle Readings
Flight Idle Readings
Trim Balance Data Readings
Bleed Evaluation Readings
EPR Power Hooks
Down Power Calibration
LCF CyclesMiscellaneous Data Points
Points
86
2
16
36
28
48
27
102
15
Total 358
Table 7-2. Chronology of the Acquired Data.
Breakdown of Steady-State DMS Readings
for Build -03
Readings Date Points
1 to 5
6 to 15
16 to 31
32 to 47
48 to 56
57 to 70
71 to 77
78 to 149
150 to 197
198 to 209
210 to 251
252 to 304
305 to 330
331 to 358
6/20/86
6/23/86
6/24/86
6/25/86
6/26/86
6/29/86
6/30/86
7/I/867/3/86
7/4/86
7/5/86
7/6/867/7/86
7/8/86
5
I0
16
16
9
14
7
72
48
12
42
53
26
28
Total 358
157
idr
|
• "_ • C
_ ,
>
+
4(1;
0
0
0
_J
m
158
IPC
Inlet flow - From inlet total rakes, circumferential pressure
statics, physical area, flow coefficient
Pressure ratio - From inlet and exit total pressure rakes
Efficiency - From inlet and exit total pressure and temperature
rakes.
HPC
Flow - From HPT flow function (iteration)
Pressure ratio - From IPC exit conditions and stratification logic to
define inlet pressure, PS3 correlation to define P3
Efficiency - From matching core overall performance.
Bypass Duct
Flow - From IPC exit conditions and stratification logic
Pressure drops - Measurements at IPC exit and duct exit
Temperatures - Measurements at IPC exit and duct exit.
Combustor
Efficiency - Assumed (map value)
Pressure drops - Assumed (map value)
Fuel flow - Measured (including all parameters for corrections).
HPT
All parameters assumed (map).
IPT
Efficiency
Exit from IPT
- Energy balance with IPC
- (T46) - From fuel flow and inlet airflow, together
with assumed secondary flows
- (P46) - Measured.
Mixer Frame
Assumed losses, mixing characteristics.
Secondary Flows
Assumed level and distribution (based on model test and some
measurements).
Power Turbine
Inlet - Mixer frame exit conditions, secondary flows
Exit - From total pressure and temperature rakes
159
Power
Efficiency
Core Nozzle
Inlet
Exit
Thrust
Installed
- From airflow, delta temperature
- From delta temperature and delta pressure.
- PT (power turbine) exit conditions
- Ambient conditions.
- Measured
Core component - From PT exit conditions, nozzle coefficients
Uninstalled From installed thrust, assumed drags
UDF TM component From uninstalled thrust and core thrust.
7.3 COMPARISON OF PRETEST PREDICTIONS AND REDUCED TEST DATA
The steady-state data points used for the performance evaluation were
restricted to the down power calibration and maximum power points recorded on
July I and July 3, 1986. Data points taken for UDF TM mappingEPR (engine pres-
sure ratio) power hooks were recorded on July I, 1986. The pretest prediction
used to compare the test data is the Status D5C cycle. The Status D5C cycle
was expressly defined for Build 3, and it includes the following major items:
• PT derates on efficiency due to open clearances and the blade
damper pins
• PT flow function adjustments for open clearances and the blade
damper pins
• Revised exhaust nozzle characteristics as a result of new
nozzle hardware.
Core Performance - Figure 7-2 shows the overall core (F404) temperature
ratio versus pressure ratio. Note that the core performance is approximately
as predicted at takeoff power conditions and is better than predicted at lower
powers (70% takeoff and below). Figure 7-3 presents IPC stall margin versus
corrected flow, illustrating that the IPC operating line was approximately as
predicted.
Power Turbine Performance - Figure 7-4 shows the PT flow function versus
the PT energy function. The PT flow function was within 0.5% of the predicted
value at higher powers (80% takeoff and above). Figure 7-5 illustrates the PT
160
; I
\ E1\
\
'\
\\,
\
e4 _ e4 e4 e4 e4
o!lel:l aJn|e_adtual au!Su-_ aJoo
Q,)I.l=
=
>
0
E
=
0
!r_
=
161
A
0
,G.em
0. I-
_j
/
/_j
J
C,O _ 0_1 0 CO QD
% 'u!6JeW IIRIS Odl
0
ql""
o
0"0
I.I.
tJ
0
0
0
0
QJ
_J
_J
0
_Q
Q;
,,_
_J
CQ
(0
I
r-.
_J
=
A
U
0
(/)
.0 0
'10 =
hi,-
00
o-e.
\
\
\\
in o in o In
uo!loun-I MOI.-I J.d
0
0
Lt)Ln0
0Lno.
o.
LI.
Ul
p-n
0
0
#0
O
e.=
e-,
o
0
_A
0
I
163
F 7
\o \
O G0 _D _1" (_1 O
_0ua!0!ll3 ld
A
oina(/)-1
(/)
._o o" "O
O
O
In
O
O
o.
o,.=
• U,I
F-0.
O
O
Ul
O
O
O
Q
U=
=
=
>
U=_J
U
_J=
=
Q
!r_
_J
164
efficiency versus PT energy function. The efficiency was within +1.5% of the
predicted value and within the scatter range observed in LP turbines of large
turbofan engines.
UDF TM Performance Figure 7-6 depicts the UDF TM thrust coefficient versus
power coefficient. Note that the test data describe a shallower slope than
predicted. Thrust coefficient is approximately 4% worse than predicted at a
power coefficient of 1.95 (takeoff condition), nearly as predicted at power
coefficient of 1.70 (92% takeoff), and approximately 4% better than predicted
at power coefficient of around 1.32, which covers a range of power from 30% to
80% takeoff thrust. Figures 7-7 and 7-8 compare predicted versus actual UDF TM
rotor blade pitch angles for front and rear rotors, respectively.
Overall Performance - Figure 7-9 shows corrected installed specific fuel
consumption (SFCIIR) versus corrected installed thrust (FNIIQA). The SFCIIR
can be seen to be approximately 4% poorer than predicted at takeoff power. At
92% takeoff thrust the SFCIIR is about as predicted; whereas, at 80% takeoff
thrust and below, SFCIIR is 4% to 6% better than predicted. Most of the dis-
crepancy between actual and predicted performance is due to the characteristic
exhibited by the UDF TM, as discussed above under "UDF TM Performance." Magnitude
of overall improvement in excess of that expected from the UDF TM performance
at 70% takeoff thrust and below is due to the core engine performance being
better than predicted, as discussed in "Core Performance."
Nozzle Performance - Between Builds 1 and 2, a redesigned stationary
exhaust nozzle (centerbody) was installed on the engine. The original plug
was analytically predicted to have flow separation at cruise conditions. The
plug was redesigned so that no flow separation would occur at any flight con-
dition. Figure 7-10 compares the original and redesigned plug lines. Scale
models were made of both the original and redesigned plugs, and then tested.
Figure 7-11 makes a comparison of the separation parameter (Fse p) for the two
plugs at cruise conditions, illustrating that the new plug lines would keep
F below potential separation conditions. Figure 7-12 shows a comparisonsep
between predicted and actual engine test data for both plugs to demonstrate
good match between prediction and data. Figure 7-13 presents a comparison
between predicted and engine test nozzle flow coefficient versus core engine
pressure ratio (P46Q2). Note that at P46Q2 of 2.9 and below (approximately
165
\
_L
.0 0
m_a
G) V)
lua!o!lleoo lsnJq.L .-10n
0
o!.'!'"
=.
8
n
,.: ::)
,l,m
(_1
O,p..,iJ,
u
g0
g
0
cn
In
g;>
g
0
_n
,dI
_A
166
II
A
°° IoOIT,...,• _
m
'¢1" ¢O t'_
1.6NVI8 'qol!d Jo;ol:l luoJ-I
_.o
¢_a.
o_
,p,,
O'P",p-
C
=
_U
0U
=
_J>
_J
C<
2;
o_
0 _
_ m0 IIJ
p_I
167
A(,.)
J_ .u
n_ I-
t °
11
I
o u') c=
_6NV.L8
I ,
'qOl!d Jolou JeaU
I_ •c_
c_jc,-;
c_
O_
c_o
.=
a.
i._
I-. Oc,,i O
O')T"
f,,.
Ix)O -
c;
(g
u]r.n0J
._
_J
0CJ
_J
_J
_J
0
0n_
_J
I
_J
168
0
/./
a_ co _ co Ln
ql/Jq/sql 'HLIO-IS
¢N
oo¢N
ooo
oo
oo¢N
oo
ooo
0¢1'
000
_1_
0000
000cO
000
_J(J_J
0(0
_J:>
(0
_J
U_J
0(0
!r---
C;
_D
169
©
0
®0
,._
_J
_J
OD
_J.-__J
I
I
170
II
I-
II
0llm
:5
llm
0
._ C
_.,..
°I
\
I
U)I,I.
0
0
0
_J
0..._
©
I
171
wO.
1.02
1.00
Modified
PlugCrown
(
0
Predicted
.92 U Original Plug
O O Original Plugs90
Crown
o o A
Predicted
Modified Plug
Peebles Test Data
O Original Plug Test 1616,
Rdg. 45, Ps0/Pam b = 1.092
Z_ Modified Plug, Test 1685,
Rdg. 192, Ps0/Pamb
.88300 310 320 330 340 350 360 370 380
Station, in
Figure 7-12. UDF m Engine - Core Plug Static Pressures.
172
0 0 0 09" _==
8..10 ';ua!o!llaoo
O
L_ .._
_0
J
!
f
L
I
0 01
MOI:I alZZON
14'1
CO
.!,,..
O'J
0')
'Ill
mm
f/I
00
u_
_J
C_
_J
_J
0_J
>
_J
U
iJ0_J
0
_J
NN0Z
I
_J
173
80_ takeoff thrust and below> the calculated flow coefficient is within +I_ of
the prediction. However, at takeoff power there is a discrepancy of about 3_.
This is probably due to the fact that the flow coefficient is sensitive to the
[IDFTM exit pressure which was not a measured parameter, but rather, the value
being ass_ed from the performance maps. This discrepancy is not a concern,
since 3_ change in flow coefficient changes sfc at thrust by only 0.005_.
EPR Hooks - These test points were run to map UDF TM performance at off-
schedule speeds. The test was conducted at six different propulsor rotational
speeds where the propulsor speed would be held constant, and core power level
would be varied by demanding various EPR levels; hence, EPR hooks.
Figure 7-14 shows installed corrected specific fuel consumption (SFCIIR)
versus installed corrected thrust (FNIIQA) for on-speed schedule and constant
propulsor speed pretest prediction, with Figure 7-15 showing the equivalent
reduced test data. Note that the on-speed schedule lines on both figures are
the same as those found in Figure 7-9. The test further proved that the UDF TM
characteristics were different than predicted, with significantly higher speed
sensitivity than prediction.
7.4 LCF CYCLIC TESTING AND RESULTANT DETERIORATION
The LCF testing involved running the engine through a power cycle as is
illustrated in Figure 3-54; a total of 100 complete cycles were run. In the
following comparisons, a sample of early data is compared with a sample of
late data in order to quantify the magnitude of scatter as well as the magni-
tude and source of deterioration. Figure 7-16 shows that the wind conditions
were relatively consistent, and all data points, except one, were within the
prescribed performance testing wind envelope.
Figure 7-17 shows that at corrected installed thrust, corrected installed
specific fuel consumption deteriorated by approximately 0.040_.
Figure 7-18 demonstrates that at EPR, corrected installed thrust followed
the predicted trends and showed no signs of degradation. This indicates that
the power extraction at EPR by the LP turbine and the thrust produced by the
UDF TM for LP turbine power remained unchanged during the tests. Figure 7-19
174
o o o _ _-r-3::3::::11: ,. ,. ,.
• t,. _- r- I:: _: _:
"* Q. a. Q. '- "- '-
__ II II II II II II
"_ _ _ OD _ _ _
f =
'l J, t
_I_ .................................................... i_ "a'_ ..............................._ "
qll_qlSql '1:1LIO.-IS _ ::):IS peloaJ._oo
o
4.J
>
oo=:
om,
_D
co
orj
_J
o
I
175
176
r/)
_ _ _ 0 0 00 0 0 0 0 0
It II II
ZZZZZZ¢OXXXX X X
A
U
Co
el
Uim
"I0
II
Q.
Clh
I-I
i_
0.I
\
I\
/!1
UU--nLc/)
000
000
0
i
Zi.1..
I
0.00
00
_0
0
>
0
0
0
r,_
_J
0
0
!
0.0
o
..I
T
JJ
/
_.fJ/
//
JJ
Jf
I
J
u_
qdm ',_l!OOleA Pu!M
0GO
0
00
0r.O
_t
'I0
!
cO!
00
!
0_r
!
0GO
!
>.
©
:>
I
177
O=lS PellelSUl peloaJJoo
0
0
TP_
178
,\
0
0
sql 'lsnJqJ. pel:)eJ, oo
_3
_J
rj
t_0
L_
u_
GL_
!
O0
179
3.6
3.5
0im
(" 3.4ft.4)Im
(g"- 3.34)Q.E4)I--
3.2II
00
3.1
3.(_1.8
F Data
Late LCF Data
Predicted Data (Status D5C)
2.0 2.2 2.4 2.6 2.8 3.0 3.2 3.4
'1%J
Control EPR
Figure 7-19. Core Temperature Ratio Versus Control Pressure Ratio.
180
reveals that at EPR, core engine temperature ratio was essentially unchanged,
and the core showed little or no signs of deterioration.
Figure 7-20 illustrates that at EPR, intermediate power compressor (IPC)
efficiency was down 0.60 points during the latter part of the test compared to
the early part of the test. However, Figure 7-21 shows that at EPR, IPC flow
follows the predicted trends, confirming that inlet flow and core engine power
output remained consistent. This implies a dirty IPC, which is further sub-
stantiated by Figure 7-22 indicating that the flow versus speed characteristic
of the IPC deteriorated, with approximately 0.50% drop in corrected IPC flow
at corrected IPC speed.
In general, there was no evidence of mechanical deterioration, with
performance degradation due mainly to a dirty intermediate power compressor.
7.5 DEFINITION OF PREFLIGHT TEST CYCLE
The base cycle used for Build 3 pretest predictions was the Status D5C
cycle. Listed beiow are the changes made to the Status D5C cycle to define
the preflight test cycle; this being the Status D8B cycle:
• Modify IPC flow-speed characteristics
• Modify IPC efficiency characteristics
• Modify HPC flow-speed characteristics
• Modify HPC efficiency characteristics
• Revise bypass duct losses and effective area
• Scale IPT efficiency
• Modify LPT efficiency characteristics
• Modify LPT flow function characteristics
• Modify UDF m thrust coefficient characteristics at Mach No. = 0.00,
no change at Mach No. _ 0.20
• Use NASA MPS test-derived UDF _ maps for Mach number _ 0.67
• Extra cooling air scoops for telemetry system modeled.
7.6 COMPARISON BETWEEN GROUND PRETEST CYCLES AND REDUCED TEST
DATA
The following comparisons between the ground pretest cycle (Status D5C),
flight pretest cycle (Status D8B), and reduced test data are made to compare
181
I r
I_I .=IQ.
0
er
¢M .=.o
Co
q
ct_
O
I
_ouo!o!ll=lOdl
182
96
94
92th
.orail
0m
It.
¢,.)a. 88m
86
841.8
_3
_K
Early LCF DataLate LCF Data
Predicted Data (Status D5C)
r , T
2.0 2.2 2.4 2.6 2.8 3.0
Control EPR
3.2 3.4
Figure 7-21. IPC Flow Versus Control Pressure Ratio.
183
A
O
-J .J
o
U
L_
O
Oa.
<2
O
Cn
U_
(D
O
<.2
!
<Ut_
0_
184
_._o_ .Q
the starting status, ending status, and data used to define the ending status
against each other.
Figure 7-23 shows the core temperature ratio versus core pressure ratio,
Figure 7-24 illustrates LPT flow function versus LPT energy function. Figure
7-25 demonstrates IPC stall margin versus IPC corrected flow. Figure 7-26 is
a comparison of UDF TM thrust coefficient versus UDF TM power coefficient. Figure
7-27 depicts the overall performance, corrected installed sfc versus corrected
installed thrust. Note that in all of the above comparisons, the Status D8B
cycle more closely matches the test data than did the Status D5C cycle.
7.7 PERFORMANCE SUMMARY
In the testing of the UDF TM engine at Peebles, the following performance
and capabilities were demonstrated:
Performance
• 25,000 ibf installed corrected thrust
• 0.232 ib/hr/ibf installed corrected specific fuel consumption
• 15,000+ physical total shaft horsepower.
Capabilities
• Running at full power statically
• Data was repeatable statically
• Prediction of low speed, low pressure, counterrotating turbine
performance with an accuracy comparable to that of high speed,
conventional, low pressure turbines.
Comparing data recorded early and late in the LCF testing, performance
deterioration was confined to a dirty IPC; there was no evidence of mechanical
deterioration.
7.8 KEY RESULTS
The F404 gas generator provided repeatable and predictable performance
and was better than predicted at lower powers (below 70?0 takeoff power).
UDF TM blade performance sensitivity to rotational speed was much greater
than predicted.
185
186
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\
\\
\
, \\
, \\
\ , \, \
\
\
\
\
A
"m(gin
w
aa pmiga
\
, \, \
\
, \"\\ _
, \, \
2
.. _.__8
I-
r,o
!,--
0
B
0
:>
.
4i
r_
A
A_U_
_1 o.
o
!
o.
o_
.J
o.
o
_L
0
O9
E0 0
Cq
I
187
i tl
_//I I/
- /
jJ/
o_
o!
P_
0
0
r_
o
4_1
:/1
i
188
\
\
\
\
\
\
\
\
A
Aea
,c4
(.)_ _
\
\
\
\
\
\
\
\
i i \\
O
ii
OO
3OQ,
_. u.
OA •
09
_3O
L4
_J
O
q.3
_J
_J
_j c_
c_
mO
I
189
A
AW
m_w
w W
a. o.
!
i j
ii -
i1
Jt
CeJ _1
\
I
I
1
I
I
I
I
/!
!
/
0
00L_C_L
0
"0
00
0
0
>
0
>
0
cJ [--_
I
_0
190
At power coefficient = 1.32, thrust coefficient was 4% better than pre-
dicted (approximately 60% takeoff power). At power coefficient = 1.95, thrust
coefficient was 4% worse than predicted (approximately 100% takeoff power).
Power turbine efficiency was approximately as predicted, being within
+1.5 points over the major portion of the operating range. The flow function
was within +0.5% of prediction at high powers (above 80% takeoff thrust).
Overall performance was better than predicted up to 92% takeoff power
(23,000 ibf corrected installed thrust), but was poorer than predicted beyond
92% takeoff power (23,000 ibf corrected installed thrust) due to UDF TM speed
sensitivity as noted above.
At 60% takeoff thrust (15,000 ibf), sfc was approximately 5.50% better
than predicted. At 92% takeoff thrust (23,000 ibf), sfc was approximately as
predicted. At 100% takeoff thrust (25,000 ibf), sfc was approximately 4.00%
worse than predicted.
191
8.0 ENGINE OPERABILITY
The most important operability concern with the GE36 proof-of-concept
engine is the stability of the IPC (F404 compressor). Replacing the F404
variable area exhaust nozzle with the GE36 UDF TM propulsor assembly reduced the
Station 48 (F404 nozzle/UDF TM power turbine inlet) flow function by about 15_.
This raises the IPC operating line (Figure 8-1) and reduces the stall margin.
To help increase IPC stall margin and ensure stall-free operation of the IPC,
two design modifications were incorporated:
• Variable Stator 1 and more closed IGV schedule (raises the IPC
stall line)
• IPC bleed system (lowers the IPC operating line).
The GE36 proof-of-concept HPC has an adequate stability margin since the
stall and operating lines are similar to those for the F404. Due to the more
limited operating range and flight envelope encountered during GE36 ground and
flight test, the stall margin requirements are lower.
8.1 [PC STALL MARGIN
Steady-state IPC operating lines at 2,750 ft/OMn/+31 ° F and at 38,000 ft/
0.80Mn/ISA, as predicted by the Status D6C cycle model, are shown in Figures
8-2 and 8-3. Also shown is the nominal IPC stall line and the, statistically,
worst-case IPC stall line which includes analytical estimate of effects due to
deterioration, inlet pressure distortion, and IPC tracking error. The nominal
stall line shown in these figures is from the F404 green run results with the
GE36 variable geometry schedule (Figures 8-i and 8-4).
Transient cycle model predictions of IPC operating line migration during
decel transients from maximum power with no IPC bleed and with maximum IPC
bleed are included in Figures 8-2 and 8-3. These predictions demonstrate the
need for, and potential of, the IPC bleed system to prevent IPC stalls during
rapid decel transients.
192
0
i
4J
0
rj
0
Cb u'h
\
\\
Lr_
0C'q
C_0
0
C_
0
0 C'q
,'W
0
L_
0
o_eN a_nssa_ Ddl
(D>
L_
t_
0
Ioo
_J
._
193
\\\
0Z
\
\\\
oT_e_ a_nssa_ D_I
194
0u_p..o4
c_O
0._0
q.;
k_
I
4-1
IoO
m
0
o_e_ a_nssa_E DdI
0
0
0
0
qxl
0
0
0
c_
<
_J
0
0
L3
k_
0
0
0
gm
,r.l
,--1
4-1
I
4-1
I
O0
195
60
Closed
_0
50
oO
J 40,-.-t
<
O
*_ 30
0
m 20
0
_ 10
Open 0
-" "_--_- -- Ground
, _ _ Test
F404 Nominal_ Ik I
40 50 60 70 80 90 i00
Corrected IP Rotor Speed - % _ere 100% = 13,270 rpm
Figure 8-4. IPC Stator Schedule.
196
8.1.i IPC Bleed Control System
The IPC bleed valve position is controlled by the DEC. The control logic
opens the bleed valve in response to any one of the following inputs:
a. d(PLA)/dt throttle retard rate > threshold
b. PLA - throttle step > threshold
c. d(P)/dt - IPC exit pressure decay rate > threshold
d. IPC P/P - scheduled maximum allowable P/P (f[XN2R]).
Thresholds "a" and "b" indicate a decel transient condition to the con-
trol, in which case additional IPC stall margin could be required. Threshold
"c" would be exceeded in the event of engine surge. Schedule "d" is designed
to maintain a minimum level of IPC stall margin under all normal operating
conditions.
If thresholds a, b, or c are exceeded, the control logic is designed to
"kick" the valve full open, hold for 5 seconds after the last demand for full
open, and then ramp close in 5 seconds. The slow closing will avoid transient
pressure pulses. In addition, for Input c, the control downtrims fuel flow to
maximum authority, further decreasing engine system pressures. If scheduled
Value d is exceeded, the control modulates the bleed valve to maintain the
scheduled maximum allowable IPC pressure ratio. In the event multiple inputs
are received, the valve kicking logic takes precedent.
8.1.2 Control Threshold for Throttle Retard Rate
Figures 8-2 and 8-3 show the PRS usage (stall pressure ratio normalized
to stall line), as predicted by the transient cycle model, during decel tran-
sients from maximum power without IPC bleed. Shown are decels at various PLA
rates at 2,750 ft/OMn/+31 ° F and at 38,000 ft/0.8Mn/ISA. Also indicated in
these figures is the current status available PRS and the minimum available
PRS as a worst-case estimate. Figure 8-5 shows that with the current level of
available margin at SLS (sea level static) conditions, an IPC stall would be
predicted during a throttle chop from maximum power without IPC bleed, yet a
decel transient at -lO°/second d(PLA)/dt would not consume all of the current
available margin. A decel transient from high power at -10°/second d(PLA)/dt,
without IPC bleed, was successfully accomplished without stall during ground
197
14
12
_Du I0
g 8
_ 6
°,-Ir/lm
_ 4
< 2
0
2O
| l I |
2750ft/M.O/+31 ° F ATam b
_ I N° IPC B_eed I
Clean ---I
[ [ X I Available _I
I I X[ APES
< -9olsec--/// ] _"",,_,. Worst Case
30 40 50 60 70 80 90 i00 II0 120
W2R , ibm/sec
Figure 8-5. Predicted IPC APRS Usage During Throttle Decels; 2750 ft.
198
testing; thus, confirming the analytical prediction. At altitude, however,
the operating line migration during a decel exceeding -9°/second d(PLA)/dt is
expected to consume all of the available margin, even at the current quality
of the engine (Figure 8-6).
Considering the statistically worst-case IPC stability condition, Figures
8-5 and 8-6 show that even an IPC bleed kick threshold of -5°/second d(PLA)/dt
is not sufficient to assure stall-free operation. This illustrates the neces-
sity of the maximum allowable P/P schedule (bleed valve modulation).
8.1.3 Maximum AHowable IPC Pressure Ratio Schedule
The maximum allowable IPC pressure ratio (P15/P2 versus XN2R) was set at
approximately 1_ above the engine ground test operating level (Figure 8-7).
This was accomplished at the end of ground testing, after analysis of control
measurement data. The effect of the expected control measurement variation on
this maximum allowable P/P schedule was determined to be equivalent to approx-
imately +0.015 PRS (pressure ratio schedule).
This bleed modulation function will be effective during steady-state
operation if the IPC operating line migrates upward due to the altitude/Mach
effects (not predicted) or deterioration effects. It will also be effective
during decel transients which do not exceed IPC bleed kick threshold levels.
8.1.4 Final IPC Bleed Control Status
During the engine ground testing, the IPC bleed system was successfully
demonstrated; proper mechanical function and pressure relief capability were
verified. Each of the control inputs (a through d) were individually checked
and verified. By the end of the ground test, threshold levels were adjusted
to appropriate levels, as follows:
a. d(PLA)/dt (throttle rate threshold = -5°/second)
b. PLA (throttle step threshold = -2°/0.25 second)
c. d(Pl5)/dt (IPC exit pressure decay rate threshold = -20%/second
d. PIS/P2 versus XN2R (scheduled maximum allowable IPC pressure
ratio set at approximately I% above engine ground test operat-
ing level).
199
14
12
i0
8m
= 6
co
_ 4
a::
"_ 2
0
40 140
I l l38,000 ft/0.8Mn/ISA
o _ No IPC Bleed-15 /sec
--- _i3O/sec f _
-_/sec __ ]
_ o__--I I _[--- _--Clean
_-5°/sec _X _ I Avaiilable APRS
5{) 60 70 80 90 I00 Ii0 120 130
W2R, ibm/sec
Figure 8-6. Predicted IPC APRS Usage During Throttle Decels; 38,000 ft.
200
Cq
oQD
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
30
Control Schedule for PC Bleed Modulation
(Maximum P/P Schedule)
* Flight Test Schedule
40 50 60 70 80 90 i00
PCN2R, percent
Figure 8-7. Maximum Allowable IPC Pressure Ratio.
201
Figures 8-8 and 8-9 show the predicted PRS usage during decel transients
at 2,750 ft/0Mn/+31°F and at 38,000 ft/0.SMn/ISA, with bleed system control
thresholds set as defined previously. For decel transients which exceed -5°/
second d(PLA)/dt, the kick function is effective, and the PRS transient usage
is zero. This was demonstrated during a throttle chop from high power in
which the bleed valve kicked open, and the transient data indicated that the
IPC operating line during decel was lower than the steady-state level. For
decel transients which do not exceed -5°/second d(PLA)/dt, the bleed modula-
tion function is effective. If the operating line migrates above the steady-
state ground test level, the control will modulate the bleed to maintain the
maximum allowable P/P during the decel.
Figures 8-8 and 8-9 show that the IPC bleed control system should prevent
the IPC operating line from migrating to even the worst-case stall line, thus,
assuring stall-free operation.
8.2 TRANSIENT TESTING EXPERIENCE
Table 8-I summarizes the significant transient tests conducted. Small
PLA acce[s and decels were conducted to verify control functions and to adjust
control gains to appropriate levels to ensure control stability; control fault
trips were checked, resulting in throttle chops and stopcocks.
Several unintentional decel transients were encountered, both operator
and control initiated, due to instrumentation faults and operating limits.
For cycle operability evaluation, large PLA accels, decels, and bodes were
conducted. All transient testing of the demonstrator engine during ground
testing was accomplished without adverse results.
8.3 ENGINE TRANSIENT PERFORMANCE ANALYSIS AND PREDICTIONS
The transient cycle model is used to predict the operating line migration
of the compression components during engine transients. In order to foresee
any stall or operational problems prior to transient testing, good agreement
between the cycle model predictions and test data is necessary. Two engine
transients conducted during the ground test were simulated by the transient
cycle model with good results; a throttle chop from 97% thrust and a throttle
202
14
12
i0
_ 8
_ 6',-"1
m
1-4
_ 4
<3
2
I i I
2750 ft/M.O/+31 ° F &Tam b
_I IPC Bieed SY itiieaEifectiv i
I Available APRS - -
P --I ---- I -- 'A_aa&ilRble
-->-5 /_FRS : O) _-__Iwors_ica '' ""_.X_/---'_:--t Case --
<-5°/see IPC Sieed Modulation-_ _ Il i .i I i
20 30 40 50 60 70 80 90 i00 ll0 120
W2R , ibm/sec
Figure 8-8. Predicted IPC _PRS Usage During Throttle
Decels; 2750 ft.
203
14 I ; i i
38,000 ft/0.8 Mn/ISA
12_ii_i_ IPC Bleed System Effective_ lO
I Available APiS
_ .
_ ,I l "_"---_,1 I_-_:::::_:_y_
o .,_:o> ,\--_1 ,' I2--<-5 Isec IPC Bleed Modulation --_
40 50 60 70 80 90 i00 ii0 120 130 140
W2R , ibm/sec
Figure 8-9. Predicted IPC APRS Usage During Throttle
Decels; 38,000 ft.
204
burst to 20,000 thrust. Figures 8-10 and 8-11 show the transient cycle model
simulations, as compared to the test data, for selected parameters.
The results of these comparisons provide confidence in transient cycle
model predictions of IPC operating line migration during decel transients
which were used in designing and optimizing the IPC bleed control system.
Date
2-09-86
4-18-86
7-8-86
7-8-86
7-8-86
7-8-86
Table 8-1. Significant Transients - GE36 Proof-of-Concept.
Maneuver Reason
Throttle Chop from 24,000 Fn
Stopcock at II,000 Fn
I0 Throttle Bursts to 18,000 Fn
2 Throttle Bursts to 20,000 Fn
Throttle Burst to 21,000 Fn,
Chop to Idle
2 Bodes (Chop from 20,000 Fn
to Idle, Burst to 20,000 Fn)
No. 7 Aft Blade Debonded
XN49 O/S Trip
Planned
Planned
Planned/Overspeed
Planned
2O5
• I I _ i | | | ! I I 0 I S
l|ll|!|!°II. m4u4_p '_Ik_r _ wlow I_mAuo d ----_
_ t l l I I t l II,, I I
:-|z I fe._ (b.G M / '4
'°'"I ¢O L _ "r i_" KO41K4_
8oOJ64q_ _l_d
I I I _ I I I I I I !
Z-OL • _ ':)dl
I I t I ' I I I I I !
_-OL • IcLI ':_dH
"SOld Nn d
! o_
;i
..,i•
o
m
,I
I--
i
i
q
I I' II I I 0
_'_'_---''.L.OL • md_ 'JOslnd_ d irf
_J
_J
_J.,-t
C)
@r_
_ m
@
m
0
I
c_
206C i: i':,:L)G_': Q_AL_TY
OF POOP, _:_,...I, Y
el,
e,I
- ii!, _,, :
j'!l
, !E
! .m_ 6
i[!
i i .... ooS II Ii ,... :I., _ _
[- mui_ ".¢-id
,,.. S 8 II ,,. ,,- .:,. -,_,__
_ot x icb"_dHl , I i t ! t i i l • J
S :1 _- o 8 :_ g p: _ .° oz.OL x Jq_acl 'J_LO lend
4I4
k
:!E
Q_m
qr
t,i
_e
o o o o
,,,,,_,,,-_
!
L
oo
I14 T I I I
I l ' I I ; I I I I I
_-OL • u._ 'Joqnck_d Inr ------
I
E
o
vl)
_J
C_
.l..j
0(./)
b_
G
•_ (.r)
_>_
..__._,-_
I
,_,_:_
207
9.0 ENGINE CONTROL
This section provides a summary of the GE36 control system performance
during the ground test program performed at GE's engine test facility near
Peebles, Ohio. This summary is divided into three parts; each portion covers
the milestones associated with each of the three engine builds.
9.1 BUILD i - ENGINE CONTROL TESTING
The control system performance for Build I was successful. A limited
amount of testing was performed due to the propulsor turbine failure.
9.1.1 Speed Sensing Anomaly
During Build 1 testing, it was discovered that the speed being sensed by
the control for the forward propulsor rotor (XN48) was incorrect. A further
study as to the cause showed that the gaps in the target wheel, which was
divided into eight segments, were inducing a superfluous signal onto the mag-
netic speed-sensing pickups (Figure 9-1). This caused a higher sensed speed,
up to two times actual. The situation was initially corrected by a change
in speed signal processing. A more elegant solution was identified for imple-
mentation on Build 2. This solution consisted of locating the teeth on the
target wheel directly at the segment gaps (Figure 9-2).
9.1.2 Pitch Control
For Build I, the engine was controlled to pitch angle, rather than to
propulsor speed.
9.1.3 Gas Generator Control
The gas generator control performed as expected. The EPR (engine pres-
sure ratio), HP (high pressure) and IP (intermediate pressure) stator control,
and the control of the duct bleed system all exhibited stable operation during
this engine build testing. A new control strategy for stall avoidance and
recovery for the duct bleed system was identified during this build and was
incorporated into the control system during Build 2.
2O8
• Original Design
• 8 Segments Total
• 1 Tooth Centered
In Each Segment
,oth Locallon
Toolh - Top View 4-- 0.016 - 0.034 Gap
Figure 9-1. Fan Speed Sensor Segments.
209
• Redesign
• 8 Interlocking Segments
• 1 Tooth Per Segment
• Toolh Positioned On End
• Gap Width Reduced
0.012 - 0.028
Toolh - Top View
Figure 9-2. Fan Speed Sensor Segments.
210
9.1.4 Throttle System
The throttle system was redesigned to minimize hysteresis. The resolver
was relocated to the HMU (hydromechanical unit) fuel lever to provide a more
accurate indication of fuel lever angle. A rotary potentiometer was added to
the system at the throttle converter to provide an input for PLA which would
provide for a reverse indication and redundancy for the resolver.
9.1.5 Engine Starting
Starting tests during this build achieved expected results. The engine
was first dry motored to check instrumentation and control signals. After
successful dry motoring, the engine was wet motored (engine motoring with fuel
on), also without incident. After wet motoring, the engine was fired to idle
power. The control performed as expected, successfully controlling HP rotor
speed to the HP rotor starting speed schedule. Due to the aforementioned
speed sensing anomaly, limited propulsor speed control testing was performed.
9.1.6 Off-Engine Harnesses
Crosstalk was observed on mulLiconductor off-engine cables during testing
as well as inaccuracies on alternating current type sensors. This situation
was corrected by placing twisted pairs for each circuit inside the same shield
and removing unused pairs.
9.1.7 Lube Off Bypass
Scavenge capability of the propulsor lube oil system was marginal during
Build I testing. A solution was identified for use on Build 2. This solution
involved installing in the propulsor lube system a bypass valve which allowed
lube oil into the propulsor only after a light-off had been detected by the
control. This prevented excess lube oil from entering the propulsor during
start operations, thus minimizing the effects.
9.2 BUILD 2 - ENGINE CONTROL TESTING
Due to lessons learned during Build 1 testing, Build 2 testing involved
further testing of the engine in the realms of transient testing, verification
211
of control schedules, and modifications to the control. This testing further
demonstrated the stability of the control system. Full control was maintained
under a blade-out condition.
9.2.1 Speed Control
Closed-loop speed control (modulation of fan pitch to maintain scheduled
fan speed) was used by the control successfully for all power settings, with
the exception of reverse and windmill testing. The propulsor speed was demon-
strated to be stable from 550 rpm to 1400 rpm and from idle to 25,000 lbs,
corrected thrust. This modulation of pitch to control fan speed also was
demonstrated for Mach numbers up to 0.1 using facility fans.
9.2.2 Duct Bleed
Modifications to the duct (fan bypass) bleed control logic were verified
for PI5Q2, PLA chop, and simulated stall. A throttle chop from 97_ takeoff
thrust was performed without IPC stall; also, unrestricted throttle chops from
all power settings were performed. Figure 9-3 is a schematic of the engine
bleed system.
9.2.3 Transient Testing
A limited number of small, part, and full power throttle chops and small
part power accelerations were performed. During a throttle burst from idle
to 1150 rpm, an overspeed incident occurred. It was determined to have been
caused by the fan pitch actuator becoming force-limited due to a too rapid
response time of the gas generator. A fix was identified to slow down the
accel rate of the core by putting an accumulator on the CDP (compressor dis-
charge pressure) sensing line to the HMU. A preliminary fix was installed
on the engine which adequately slowed the core accel rate. A more polished
design was used on Build 3.
As a result of the blade-out incident, the engine was chopped from 24,000
corrected thrust to idle, then stopcocked approximately 10 seconds later. As
illustrated in Figures 9-4 through 9-8, the control system maintained complete
control during the blade-out, chop to idle, and subsequent stopcock.
212
STA 225
ManifoldJunction
Servo Fuel
Digital/o- ._p Electronic
.... :1 co_._ol
AftNacelleBulkhead
STA 255
ButterflyValve
' "91_- 2 ports2.85" I.D.* 'q"'" 1 porl3.60" I.D.
m
Mount Beam
Fuel
\ /
leed Manifold
Electrical
Air
Figure 9-3. Duct Bleed System.
213
, i
o
• • _!g_goooo°_" s_aJ6ap 'O00V'ldI ! I ! I ! ! I I !
0 _ _ _ 0 0 0 _ 0 0
z:OL x sqi 'lsnJqJ. o
,I0
I I I I I I I I l I,_ _, _ _ _ _ _ _ _ ,_,_
e!sd 'Vg LZ.Ld ,,
I I I I I I I I ! I !0 ¢0 ¢D _ ¢N 0 ¢0 ¢D _1" C_ ¢_
INdO 'MOl-.Ilan..-I
0
r.-,t4.J
0
[.-,
.<.TI
214
¢MoI"*
x
EO.¢._
-o"
Q.Or)
oO.m
o
200
180
160
140
120
100
80
60
40
20
0
• 24,000 FN/8 to Idle
• P46Q2 Control Prior to Chop
• Blade Out Incident
-- 180
0194 202 210 218 226 234 242 250
Time, seconds
Figure 9-5. Throttle Chop.
OF POOR QUALITY
2].5
• 24,000 FN/8 to Idle
• P46Q2 Control Prior to Chop
• Blade Out IncidentmFT ROTOW |LAO| LO_S - TIIBQT_LE [ilOP TO lULl
'_. ,-- [", ,i; -, " ';; I....
,.,i, i
i,!,!, I. :.i! :,',_,. ":" iI I i' ;'i P'
i :, I}..._
0 X" "" _'""--"-' _ _'"
190 198 206 214 222 230 238 246
Time, seconds
Figure 9-6. Throttle Chop.
216 OF POOR QUALITY
OR]O;,_L;,L " __':..,,.:,= ISOF POOR QUALITY
100
90
80
70
60m i
so_ 4o
30
2O
10
_
• 24,000 FN/8 to Idle
• P46Q2 Control Prior to Chop
• Blade Out Incldent,o"'''' _.o,,oAFT IOTOR 9LRO( LOS_ - TNRO
_ 100 ,_,,__,.,: ,_.....v ,. ,,,
I: IIIII,h ,If _-,_I!_!i!_I:::I_: ":,:; " _: :I
, l,,j,,,i,it,.';,,.:,-,-,-'-,-.i....:.._ .......i..i, i'_ T; ::' _ ;; .
'-. _n LLI-I}U_!!! 'I_!I!!!!,!!'!!_'!':'!'J ! ;!"!" I' F_, v.. !lll'!lfi!li;il;!!_!llill;:_iu_ :;__', i: _li:_' I:, /-o ,:. _-l_':'._!li!_.I I:1,1,1.1 I I i / f.= "" _!lli_!l!l;l!!_l!__i_1;' 1.,_.,, I:.. ::.
An 'i_]',J:!IIIilii!i,'_lliil::ili,q_;:li.!l:I_I;l;..l';,....
,,)n _-:,-_J; m, '"d__r_lrA_ ' :1: I ! / 1
t;l il, tl :: .' , ' '
•I:I.I.I ! '\ _o k,l:.;.:.... ..: , : ,194 202 210 218 226
Time, seconds
234 242 250
Figure 9-7. Throttle Chop.
217
0
o
u_
• • •
0
®IEm
0
! Io o.e_ ...
0-- 0
00')
m
0
0
m_<
0
4-1
0
E_
I<_
218
9.2.4 Control Parameters
All control parameters were stable.
9.2.5 Vibration System
The vibration monitoring system proved to be functional; however, it was
discovered that there was not enough gain in the amplifier for small vibration
signals. As a result of the blade-out incident, readings were gathered that
allowed for the scaling of the vibration signal. To increase its gain by a
factor of four, the amplifier was reworked; this change was implemented during
Build 3 testing.
9.2.6 Overspeed System
The capability of the overspeed system was demonstrated during an over-
speed incident. During a throttle burst from 550 to 1150 rpm, the fan pitch
actuator became force-limited and was unable to modulate fan pitch to control
rear rotor speed. An overspeed condition resulted, which was quickly sensed
by the overspeed system which immediately shut down the engine in a controlled
manner.
9.2.7 Reverse Testing
Limited by the telemetry system, this due to high temperatures, reverse
testing for Build 2 consisted of a throttle push from idle to 1000 rpm (3000
lb reverse thrust at static conditions). Figures 9-9 through 9-12 illustrate
a transient from forward idle, to reverse idle; then throttle is pushed until
a maximum propulsor speed of 1000 rpm is obtained, followed by a decel and
return to forward idle. The fan speed is controlied by modulating pitch in
forward thrust, and PLA is used to scheduie pitch directly in reverse thrust,
hence, the difference in propulsor speeds in reverse thrust.
9.2.8 Control System Modifications
Two areas of the control system hardware were modified during this build
to improve performance. The hardware modified was the pitch actuation system
transfer valves and the watchdog monitor circuit of the control computer.
219
20
18
16
14
8U,,,
<4 6
4
2
0
• Forward Idle to Reverse Idle
• Throttle Push to 1000 rpm Maximum Fan Speed• Return to Forward Idle
o_. -100
--_ -200
- _-300
0 -400
- -500
- -600
- -700 - -1000 40 80 120 160
Time, seconds200 240
Figure 9-9. Reverse Thrust.
220
• Forward Idle to Reverse Idle
• Throttle Push to 1000 rpm Maximum Fan Speed• Return to Forward Idle
TR_NSItNT PLiIT
140 - 140
130
_:_ 120I"'
x 110
_100
.._90
a,,m 8o(..)a. 70
O 60
5O
4O
•" 110 i_ _ _X ';;i............. _,,-;i_ -_........
E • "1' "'1 I ,I ,pt ; , ' I I"- 100 ..,,, ...:,; I!_r" ';i'::,' :.... 'z._
O¢Z,
-- (/)
L_
-- OO
- i-1
0 .i + =1 ,
• n.. | , i .| .. g .... | ,_,=....= ., .t_
70
60
so-t40
0
• ii ..... /;.:":I':'
j/4O
%.
80 120Time, seconds
iI
i'-- |
i
I I
160 200
Figure 9-10. Reverse Thrust.
221
• Forward Idle to Reverse Idle
• Throttle Push to 1000 rpm Maximum Fan Speed.• Return to Forward Idle
TmANS I( NT PLOT
110 -- 110
100
_o 9OX
E 80O.t_
-d 70
_ so
_ 5O
_. 40o
300 2O
10
--_o100 -_ , _ ............ ,.K__/-
-- 90 _'_
_ so
, i_ 70_ - .-_ L......
60 - _ f_¢/) i
-- _502I_. 40
_ 3o--020 ..............
I,I.
-- n 100 20 60 100 140 180 220
Time, seconds
Figure 9-11. Reverse Thrust.
222 ORIGINAL PPGE _SOF POOR QUALITY
ORIOLN_.Lr_RQ_ ISOF F_ro::_' Q_-,v",LITy
• Forward Idle to Reverse Idle
• Throttle Push to 1000 rpmMaximum Fan Speed
• Return to Forward Idle
TRONSI|N_ PLOT
40 80 120 160Time, seconds
200
Figure 9-12. Reverse Thrust.
223
The transfer valves exhibited erratic performance during the beginning
of this build. It was determined that the filtering of the hydraulic oil was
inadequate. To correct this problem, the filters for the hydraulic system
were changed from 5 micron to 3 micron.
The watchdog monitor circuit on the control initiated several inadvertent
automatic shutdowns, later discovered to be caused by noise on the backplane
of the control propagating into the watchdog circuit and inducing erroneous
pulses. This condition was corrected by adding filters to all high speed
lines on the watchdog circuit card.
9.3 BUILD 3 - ENGINE CONTROL TESTING
Build 3 testing was the most successful portion of the ground test, both
in terms of time on test and control system testing.
9.3.1 Endurance Testing
Pitch and EPR controller gains were optimized during the 100 LCF cycles
which were performed (Reference Figure 3-54).
9.3.2 Core Response Modifications
The gas generator response time was reduced with an orifice/accumulator
system (Figure 9-13) in the CDP (compressor discharge pressure) line to the
HMU (hydromechanical unit) and a control rate control. This modification was
due to an overspeed incident caused by a force-limited actuator. The orifice/
accumulator was sized to give an 8-second idle-to-rated-thrust response.
9.3.3 Reverse Thrust Testing
Reverse testing during Build 3 achieved maximum speeds of 850 rpm limited
by high telemetry temperatures.
9.3.4 UPS (Uninterruptable Power Supply) Systems
The UPS systems for the control were successfully demonstrated during
this build. The UPS powered the control computer, the overspeed unit, and the
peripheral computer. The UPS systems were tested, by removing/reconnecting
224
TO MEC
Ps3TAP-_--____ It_-_FAN DUCT CASE
Ps3 LAG ORIFICE (_ .03 e)
CHECK VALVE
L
.02
WEEP );(HOLE
TO Ps3 LAG VOLUME
(= 30 IN 3)
Figure 9-13. PS3 Accumulator Configuration.
225
input power, both statically and with the engine running without affecting
control performance in any way.
9.3.5 Transient Testing
Transient testing consisted of accelerations, decelerations, and throttle
bodes. The controller was found to exhibit an underdamped response above 2.85
engine pressure ratio.
Figures 9-14 through 9-17 depict a throttle burst from idle to approxi-
mately 18,500 lbf. Satisfactory control response was demonstrated.
9.4 SUMMARY
Testing of the UDF rM provided valuable data and demonstrated the viability
of the unducted fan concept and its control system.Q
Further, test results verified that the control system concepts of using
EPR as the thrust parameter and utilizing EPR to schedule propulsor speed are
sound. The modulation of pitch to maintain scheduled propulsor speed has been
demonstrated statically and for Mach numbers up to 0.I.
226
100 -
90
80
70
60
_ 5o
5 4on
Z_ 30
2O
10
2
Time, Seconds
Figure 9-14. Throttle Burst.
227
ORIGhNAL P_,GE iS
OF POOR QUALITY
100 - 160 - 160
9O
8O
7O
6O(/)
o_ 50
2O
10
- 150
- 140
- _ 130
- _ 120
110
- = 100
- 8O
130-- 120
_ A".+It
- _ 1010
u.
- [] 90
- 8O
- 7O
- 600 4 6 8 10 12 14 16
Time, Seconds
18 20
Figure 9-15. Throttle Burst.
228
OF PO0;-_ QUALITY
100
90
80
7O
CR
60014)
a 50
.JQ.
40ZX
30
20
10
200_
180
- 160
- 140Ea.
- .1=- 120
- _ 100
Un
- 800
- 60
- 40
- 20
- 0
- 2OO
- 180
- 160
I I 1 I I i
/ :t
!
i
._ _ 7.
i
I
I....... L --
i
: I
16 18 20
Figure 9-16. Throttle Burst.
229
OF POOR QUALITY
20-
18-
16-
=E 14 --O.
3" 120
Ix.
® 10 -=1u.
ZX 8
6
4 -
2--
O-
200 --
180
160
140
_120
gP 100
0 8O
6O
40
2O
0
100 , , , , ,_ , , _ , ,
90
eo _! , I ....
7O ,
60 - -
o,o /_: ,, I,o ]2°1 ._ . .... I i =
10 ._0_, L " - " I' ; -
0 2 4 6 8 10 12 14 16 18 20Time, Seconds
Figure 9-17. Throttle Burst.
230
10.0 NACELLE STRUCTURES
No problems occurred for any of the nacelle structures components during
ground testing conducted at Peebles, Ohio.
The fan blade airfoil loss provided valuable data pertaining to the
nacelle structural and dynamic integrity. No structural damage occurred, and
the isolators absorbed the unbalance as expected. Based on data taken during
the airfoil loss, the isolator/mount structure provided a minimum margin of
safety of 0.47 and an average margin of safety of 2.54 for all components.
Strain measurements taken on the strut during ground testing revealed
extremely low stress levels. The maximum stress level recorded, 1.4 kpsi peak
to peak, corresponds to a margin of safety of 80.4 based on shear strengths of
the strut material. The minimum margin of safety was found in the composite
mid-fairing. Based on the interlaminar shear strength, this margin was 2.7,
which is an acceptable figure. Based on the strain measurements, the acoustic
fatigue of the strut and fairings was deemed to be of little concern.
10.1 FAN BLADE AIRFOIL LOSS
The fan blade airfoil loss incurred on February 9, 1986 caused a 262,000
gm-inch (577 ib-inch) unbalance load. This failure occurred at a fan speed of
1371 rpm and a thrust level of 24,000 pounds. Engine, isolator, mount beam,
and strut acceierometers were located as shown in Figure 10-I.
At blade-out, the rear-mount horizontal vibration ievel reached 200 mils
double amplitude, while its normal level is 10 mils. The front mount along
the horizontai axis reached 95 miis, while its normal levei is I5 mils. The
engine was then chopped to idle (680 rpm) in 1.23 seconds, where it remained
for 9.0 seconds before being shut down. Vibration levels of the vertical axes
of the front and rear mounts reached 200 mils DA (double amplitude) from their
normal ievels of 2 miis each. As attested in Table 10-1, the maximum vibra-
tion leveIs occurred in the rigid body modes during coast down.
Mount loads were estimated based on relative motion across the isolators
and on the isolator dynamic spring rates (from the Barry Qualification Test
231
7-
_ < ---_
_ wo0_J
<
_ b
if)
m
0 °
0if)
0
o,r,,I
s__r,4e-'
4-4
4_
o
o
ooi
¢xl
o0o
x
7o
a0
232
Report). Table 10-2 shows that these loads were well below the design loads.
The minimum margin of safety turned out to be 0.47, which was for the rear
mount along the axial axis at 460 rpm; the load it carried (40.9 kips) was the
maximum load for the system. The average margin of safety for the isolators
was 2.54.
Sensor
Front Mount Vertical
Front Mount Horizontal
Right Rear Vertical
Right Rear Horizontal
Right Rear Axial
Left Rear Vertical*
Left Rear Horizontal
Left Rear Axial*
Table I0-I. Vibration Response.
Displacement, mils DA
1370 rpm
8O
95
170
200
95
120
Idle (680 rpm)
200
50
200
105
330
130
Coast Down (rpm)
560 (460)
385 (280)
450 (200)
165 (200)
980 (460)
430 (200)
* Sensor Inoperable Before Event
The overall response of the isolators indicated that they functioned as
designed, significantly reducing the motion at the pylon/mount interface and
softening with increased loading. The resonant speeds of the system decreased
for the airfoil loss event. Further, the isolators reduced mount loads on the
engine and pylon. These loads were well within design limits.
Physically, the isolators performed as designed during the airfoil loss.
No wear or deformation of the isolator structure was incurred (or in any other
structure). The Met-L-Flex wire mesh showed some minor deformation in the aft
isolators. The mesh in the front isolator was undamaged. All isolators were
sent to the manufacturer, Barry Controls, for inspection and testing. Despite
the fact that all the load-deflection data for the isolators indicated little
233
0
0u_
,.-6<
0
,._
0
0:E:
I
-r,4 _
0
0
;--..i
234
change and were still within limits, the wire mesh in the aft isolators was
replaced. The front isolators were used "as is."
Predictions that maximumloads would occur in subidle modeswere veri-
fied. However, modes were not matched for the front mount (Tables 10-3 and
10-4). Also, as predicted, the highest load occurs on the aft mount in theaxial direction. Both magnitude and speed were in good agreementwith predic-
tion; however, the front mount loads measured were higher than predicted.
i0.2 STRAIN EVALUATION OF THE STRUT
Extensive strain tests were performed on various strut components during
ground testing. Corresponding stress levels were calculated and recorded. As
illustrated in Table I0-5, the greatest stress level recorded was in the upper
middle portion of the mount beam at a high pressure compressor speed of 7800
rpm. This stress level, 1.4 kpsi peak-to-peak, is well within the acceptable
levels and corresponds to a margin of safety of 80.4 (compared with the ulti-
mate shear strength of the material of which it is made - AMS 5528 stainless
steel). The minimum margin of safety occurs in the mid-fairing in the axial
direction at an intermediate pressure compressor. The recorded stress level
is 0.96 kpsi, peak-to-peak. This corresponds to a margin of safety of 2.7
with respect to the interlaminar shear strength.
Appendix B shows the range of data tested in terms of calculated stress
versus frequencies and speeds; figures contained therein (Figures B-I through
and B-40) illustrate the positioning of the strain gages used. Note that some
are arranged in rosettes; while others are uniaxial.
10.3 ACOUSTIC FATIGUE
A preliminary acoustic fatigue analysis of the strut fairing calculated a
response frequency of 174 Hz (the fundamental blade passing frequency range
for the engine operating range, 600 to 1400 rpm, is 80 to 187 Hz); however,
this figure is questionable due to the numerous assumptions and uncertainties
involved in the calculation. The operating conditions assumed were those of
cruise. Further, the panel was assumed flat; whereas, a slight panel curva-
ture or irregularity would increase the response frequency; finally, the damp-
ing ratio was estimated rather than known.
235
Table 10-3. UDF TM Engine Comparison of Mount Loadsfor Airfoil-Out Event.
Maximum Load
Modified Nastran Model Test Results*
Location Load, ib rpm
FMV
FMB
RRMR
RRMA
LRMR
LRMA
8,800
7,500
10,500
44,000
12,000
31,000
228
528
228
444
228
420
Load, ib rpm
24,000 460
13,300 280
16,400 260
40,900 460
* Using Barry Mount Data and Relative Deflection,
Unbalance = 262,000 gm-inch
Table 10-4. UDF TM Engine Airfoil-Out Event Comparison
of Deflections (Absolute).
Maximum Displacement, mils DA
Location
FMV
FMH
RRI%R
RRMA
LRMR
LRMA
Note:
Modified Nastran Model
Deflection, mils
440
1120
6O0
II00
750
II00
rpm
228
540
450
420
228
420
Unbalance = 262,000 gm-inch
Measured Test Results
Deflection, mils
560
385
350
988
rpm
460
280
200
460
236
Table 10-5. Stress Data - Strut.
• Y
: 6A4JDEi..F ...... •
:FMRING, VERT
',FMRINE, 0IA6
:FAIRIN6, AXIAL
:FHD [SOi .......
:RROF SPAR,VERTi .....................
:RR OF SPAR, OIA6i .....................
:RR Of SPAR, HORIZ
:SIDE OF SPAR, VERT
ISlOE OF SPAR, HORZ
:SIDE OF SPAR, 0IA6
:LT AFT ISO CLEVISi: ..................
',HT BIt,TOPAFT,VERT.....................
:HT M,TOP AFTvOIA6
',RT OIl,TOPAFT,HORZII .....................
'.HT DH,BTHAFT,HORZ
:fiT M,TOP HIO,HORZ
:lit PIt,TOP IllO, VERT
:HT OII,OTHFIID,VIE]IT
;:J:AFT HT Dll SUPTFTE,HOII
:RT AFT lSO CLEVIS
:RT RRHT, TAN
'.RROF OPAR,LIIRAFTtVER'.
:SME, HORIZ
:RR OF SPAR,UPRAFT,VER
:SAHE, DIA6
:SklqE,HORIZ
XH2w
2 tO._7)
3 (O.O_)
2 (0.96)
3 (0.72)
0
0
0
0
0
0
2 (0.5_)
3 (0.12)
0
3 (0.20)
(0.22)
(O.Ol)
3 (0.44)
3 (0.1.4)
2 (1.08)
2 tO.011)
0
(0.36)
3 (0.1.8)
0
0
0
XN2_
3 (0.1_)
3 (0.091)
2 (0.20)
(0.74)
0
0
0
0
0
0
2 (0.5)*
3 (0. lO)
0
3 10.L8)
2 (0.17)
2 (1.4)
;3 (0.42)
3 (0.02)
2 (0.21)
3 (0.94)
0
(0.41)
;3 (0.15)
0
0
0
ZN48 :I
I (O. LS) :
1 (0.54) :
2 (0.19)
1 (0.91) :
4 i0,057)*:
4 (O.Ol_)*:
4 (0)* :
4 (0.271* '
4 (0.22)t
4 (O.Z2)*
4 (0.:9) :
4 (0.12)
"l
3 (0.20)
3 (0.2;3)
2 {0.20) :
2 IO.2l) :
2 or 4 (0):1
2 I0.[0) :
3 (0.92) :
4 (0.66)t
3 (0.;36)..... I
2 (0.17)
4 (0)*
4 (0.50)t
4 (O. L2)t
CODES:
0: NODATAAVAILABLE
l: START/ACCELERATE
2: START/ACCELERATETO 700
_: ACCELERATETO 24,400 LOS.4: ACCELERATETO 24K F6
,: ONLYONEDATAPOINTAVAILABLE
FORflATIS: CODE| (IMiIHUH STRESS[N (PSI PP)
237
For instance, if the curvature of the panel was as slight as 99 inches,
this frequency would then become383 Hz. What is known is that strain gage
data discussed in Section 10.2 and displayed in Appendix B, show that stressesare relatively low. Thus, there appears to be no resonant vibration within
the operating range of the engine and that acoustic fatigue is not a problem.
238
11.0 RESULTS
The I00+ hour ground test program of the UDF TM engine demonstrated the
following :
• 25,000 ibf installed corrected thrust
High fuel efficiency: 0.232 ib/hr/Ibf installed corrected sfc
(specific fuel consumption)
• 15,000+ physical total shaft horsepower
• Full propulsor speed (1393+ rpm)
Advanced UDF TM aerodynamics that incorporates custom-tailored
composite fan blades over an inner titanium spar that serves as
the attachment mechanism to the engine for the fan blades
Individually replaceable propulsor fan blades with the engine
installed on the aircraft or test stand
DEC (digital electronic control) provides overall engine con-
trol by monitoring gas generator power and speed and propulsor
speeds and pitch angles; the engine utilizes the existing gas
generator control and a separate propulsor control to minimize
development costs without sacrificing control flexibility; this
control system drives a hydraulic/mechanical actuation system
that permits setting the fan blade pitch angle of the two fan
blade rotors either together or differentially
• Flawless operation of the F404 gas generator
• Counterrotation of structures, turbines, and fan blades
• Reverse thrust capability
Capability to withstand a fan blade airfoil loss with no struc-
tural or secondary damage
Failure of Stage I turbine blades and the subsequent damage to
following turbine blade rows; turbine structures withstood the
failures with little or no damage.
Also demonstrated were the following capabilities:
• Running at full power statically
• Data was repeatable statically
239
Prediction of low speed, low pressure, counterrotating turbine
performance with accuracy comparable to that of the high speed,
conventional, low pressure turbines
Comparing data recorded early and late in the LCF testing, per-
formance deterioration was confined to a dirty IPC; there was
no evidence of mechanical deterioration
F404 gas generator provided repeatable and predictable perform-
ance and was better than predicted at lower powers (below 70%
takeoff power)
UDF TM blade performance sensitivity to rotational speed was much
greater than predicted
At power coefficient = 1.32, thrust coefficient was 4% better
than predicted (approximately 60% takeoff power)
At power coefficient = 1.95, thrust coefficient was 4% worse
than predicted (approximately 100% takeoff power)
Power turbine efficiency was approximately as predicted, being
within +1.5 points over the major portion of the operating
range; the flow function was within +0.5% of prediction at high
powers (above 80% takeoff thrust)
Overall performance was better than predicted up to 92% takeoff
power (23,000 ibf corrected installed thrust), and was poorer
than predicted beyond 92% takeoff power (23,000 ibf corrected
installed thrust) due to UDF TM speed sensitivity as noted above
At 60% takeoff thrust (15,000 ibf), sfc was approximately 5.50%
better than predicted
At 92% takeoff thrust (23,000 ibf), sfc was approximately as
predicted
At 100% takeoff thrust (25,000 Ibf), sfc was nearly 4.00% worse
than predicted.
240
12.0 CONCLUSIONS
As a result of the I00+ hour UDF TM test program, it can be concluded that
all of the major objectives of this engine can be and have been met. Some of
these objectives are as follows:
The demonstrated feasibility of an unducted, ungeared, counter-
rotating ultra-high-bypass turbofan
Capability to produce at least 25,000 ibf and at least 15,000
shaft horsepower with an engine of this configuration
Exceptional fuel efficiency as compared to other turbofan or
turbojet engines
The capability to produce thrust with a new fan blade design
of composite materials over a titanium spar
The capability to control the engine and actuate the fan blades
with a digital electronic control
Capability to produce reverse thrust with the fan blades with-
out the use of a thrust reverser
• UDF TM propulsor capable of producing thrust as predicted
Current computer model cycle deck techniques can adequately
model a counterrotating turbofan
Propulsor deterioration (large seals, turbine, etc.) was not
encountered over the duration of testing, which exceeded I00
hours
Operation of the engine, and its performance, is stable at
takeoff power statically.
As expected in an engine program utilizing such new technologies and con-
cepts as the UDF TM, numerous problems have been discovered. However, it is
believed that none of these problems will present a major stumbling block to
future flight test of this engine or to the development of this concept into
an important new entry into the arena of subsonic commercial and military
transport aircraft. Every problem that has occurred during ground testing has
been addressed and adequately solved. Fine tuning may be necessary, and more
problems may be discovered of course, as the testing of this engine continues.
No significant fundamental aerodynamic or control problems were uncovered, and
241
only two mechanical problems created significant setbacks to the test program.
Both of these problems were solved by rather simple, but successful means:
1 • Fan Blade Airfoil Loss - Although static component tests veri-
fied the integrity of the airfoil bonding to the titanium spar,
actual engine testing brought into effect additional factors
leading to the loss of one airfoil. At that time, only the
adhesive qualities of the composite to titanium bonding agent
held the airfoil to the spar. By adding positive retention
features to the design (by adding fasteners), airfoil retention
proved to be of no problem for the duration of testing and is
not expected to present a problem in the future.
° Turbine Blade Failure Stage 1 turbine blade dynamic response
was excessive due to insufficient damping, which eventually led
to their failure. Damping, in the form of friction, was intro-
duced to all blade rows by placing simple damper pins between
each blade; this satisfactorily reduced the dynamic response.
242
13.0 SYMBOLS/ABBREVIATIONS
ALF
Beta Angle
BFM
BTAN91
BTAN92
BTWT
CV
DA
DEC
dia.
DMS
DOD
DTAMB
EB
EDM
EPR
F404 Fan
F404-GE-400
FHV
FMH
FMV
FNI IQA
FPI
Fsep
h
HCF
Hz
ID
IF
IGV
IPC
IT
LCF
Aft, Looking Forward
Fan Blade Pitch Angle, degrees
Backflow Margin, %
Front Rotor Pitch, degrees
Rear Rotor Pitch, degrees
Boeing Transonic Wind Tunnel
Convex
Double Amplitude
Digital Electronic Control
Diameter
Data Management System
Domestic Object Damage
AT from Standard Day/ISA Conditions
Electron Beam
Electrical Discharge Machining
Engine Pressure Ratio
Bypass Pressure Ratio
Low Bypass Turbofan Gas Generator
Fuel Heating Valve
Front Mount, Horizontal
Front Mount, Vertical
Installed Thrust, lb
Fluorescent Particle Inspection
Flow Separation Parameter
Coefficient of Heat Transfer (Btu/hr ft2, o F)
High Cycle Fatigue
Hertz, cycles/second
Inner Diameter
First-Flexural Frequency, cycles/second
Inlet Guide Vanes
Intermediate Pressure Compressor
First-Torsional Frequency, cycles/second
Low Cycle Fatigue
243
244
LE
LPT
LRMA
LRHR
LRHV
LVDT
Max
Min
Mn
MPI
OGV
ON
P15/Pz
P46Q2
P48Q2
PLA
PO
PRS
PS
PS3
PT
PT8
PTO
PZ
RDGS
RRMA
RRMR
S/D
sfc
SFCIIR
SO
S/N
T46
T46Q2
TAmb
Leading Edge
Low Pressure Turbine
Left Rear Mount, Axial
Left Rear Mount, Radial
Left Rear Mount, Vertical
Linear Variable Differential Transducer
Maximum
Minimum
Mach Number
Magnetic Particle Inspection
Outlet Guide-Vanes
O-Nodal
F404 Fan Bypass Pressure Ratio
Gas Generator Pressure Ratio
Engine Pressure Ratio
Power Lever Angle, degrees
Ambient Pressure, psi
Stall Pressure Ratio
Static Pressure, psi
Gas Generator Compressor Discharge Pressure, psi
Total Pressure, psi ... (also Power Turbine)
Power Turbine Exhaust Pressure, psi
Peebles Test Operation
Compressor Inlet Pressure, psi
Readings
Right Rear Mount, Axial
Right Rear Mount, Radial
Shutdown
Specific Fuel Consumption, ib/hr/ib
Installed Specific Fuel Consumption, ib/ib
Strain Gage
Serial Number
Propulsor Exhaust Gas Temperature, o F
Gas Generator Temperature Ratio
Ambient Temperature, o F
TAAP01
TAAP02
TAFC01TE
TT
2RUDFTM
W25
W48R
Xh
XV
XN48
XN49
ZN
Stage 1 Fan Telemetry Ring Cavity Thermocouple, o F
Stage 2 Fan Telemetry Ring Cavity Thermocouple, o F
Stage 1 Fan Telemetry Ring Cavity Thermocouple, o F
Trailing Edge
Total Temperature, o F
No. 2 Roller
GE36 Unducted Fan Engine
HPC Inlet Flow, ib/second
Power Turbine Flow Function, Corrected
Horizontal Displacement, inch
Vertical Displacement, inch
Stage 1Propulsor Fan Speed, rpm
Stage 2 Propulsor Fan Speed, rpm
Z-Nodal
245
14.0 Ref_
i. Carlson, S.L., "Engine Damage Report," No. GE36-003, March 18, 1986.
2. Carlson, S.L., "Engine Damage Report," No. GE36-004, August 14, 1986.
3. Cote, M.J., "GE36 Nacelle Fairing Sonic Fatigue Analysis," Memo of July
24, 1986.
.
o
o
.
8_
Gaffney, E., Review of Airfoil Loss, "In Tne Unducted Fan Engine; GE
Quarterly Review No. 8," May 13, 1986.
General Electric, "Full-Scale Technology Demonstration of a Modern
Counterrotating Unducted Fan Engine Concept - Ccmponent Test," NASA
Contract No. NAS3-24210, CR-180868, _ 1987.
General Electric, "Full-Scale Technology Demonstration of a Modern
Counterrotating Unducted Fan Engine Concept - Design," NASA Contract
No. NAS3-24210, (_-180867, December 1987.
General Electric, "The U_ Fan Engine - Quarterly Review No. 6,
"NASA Contract No. NAS3-24210, August 16, 1985.
General Electric, '_[he Unducted Fan Engine - Quarterly Review No. 7,"
NASA Contract No. NAS3-24210, November 18, 1986.
9. General Electric, '_Fne Unducted Fan Engine - Quarterly Review No. 8,"
NASA Contract No. NAS3-24210, May 13, 1987.
i0. General Electric, "UDF TM Airworthiness Executive Review," NASA
Contract No. NAS3-24210, July 1986.
Ii.
12.
13.
14.
Kirkpatrick, R.A., "Failure Report: Engine S/N 082-001/A: High Power
Stall of October 2, 1985," December 13, 1985.
Kroger, J.P., "GE36 Seoond Ground Test," Test Project Sheet No. MA-004,
January 13, 1986.
Nelson, E. (Barry Controls), "Qualification Test Report for 94624
Engine Isolators: GE-36 Demonstrator Engine Program," Document No.
WD04624-I-005, Revision A, May 14, 1986.
Smith, R.E., "Engine Damage Report," No. GE36-002, Revision A, January
6, 1986.
246
APPENDIX A
STEADY-STATE TEST DATA - BUILD 3
247
DATA LEGEND
XN2
XN2R
PCN2R
XN25
XN25R
PCN25R
XN48
XN49
UT91R2
UT92R2
BTAN91
BTAN92
WI3
WI5
WF36
WF36R2
FNI IQA
SFC184
TAHB
TI0
PAMB
PIOM
HUHSER
RELHUM
WINVAV
WINAAV
XM0
P46Q2
CT46
T46X
IPC Physical Speed (rpm)
IPC Corrected Speed (rpm)
Percent IPC Corrected Speed (%)
HPC Physical Speed (rpm)
HPC Corrected Speed (rpm)
Percent HPC Corrected Speed (%)
Stage 1 Physical Speed (rpm)
Stage 2 Physical Speed (rpm)
Corrected Stage 1 Tip Speed (feet/second)
Corrected Stage 2 Tip Speed (feet/second)
Stage 1 Pitch Angle (degree)
Stage 2 Pitch Angle (degree)
Bypass Duct Inlet Flow (ib/second)
Bypass Duct Exit Flow (ib/second)
Fuel Flow (ib/hr)
Corrected Fuel Flow (ib/hr)
Corrected Installed Net Thrust (ib)
Corrected Installed Net Specific Fuel Consumption,
FHV = 18,400 (ibfuel/ibthrust hr)
Ambient Temperature (o R)
Inlet Total Temperature (o R)
Ambient Pressure (psia)
Inlet Total Pressure (psia)
Specific Humidity (grains/ib dry air)
Relative Humidity (%)
Average Wind Velocity (knots)
Average Wind Angle (degree)
Mach Number
Engine Pressure Ratio
Calculated T46 (o R)
Measured T46 (o R)
PRECEDING PAGE BLAi_j:-_,i'_OT glL,"AED
J
249
(_JNI.,J • •
_00
ZZ
OQ
_="_ r'.-
I I aO
_ I"l N
<mmL,d _ C._ Z
UJ _
0 ° •
0 ° °
L.J _Z
Z
I !
250
II !
II
__*_ *_
i
__.. .._0
I
_ _.__1
I
• ._ ....... _. _ ----mo
|1
_0_** .0-_ __-_-_0 _*_ .*
II
_*_ _ _*._
_*_ _ _ -.
II
_,_
_m_.
._._
0 mO
_.
r_-oo -
_Nr_.__,-r-
_0_0
•.1" 00 -
P'-=O
_ro0 *
_0 r,.-_- i.-
0_.00 -
...o....
00000000
• *_ _.
00_0 .0• _ I
_*..*0-
__ ,,_
I
I
__*
.*__0_.._*_
_ _1
__-
_.__.
I
I
,,__.
0.,5" •
0.5"0
_3C3
t.- t..- ¢0, ,.r') ¢.g
*'_ r'- •
¢_ P.- ¢g
O_U*_
4- ÷ *
," u'_ 0° •
0 _0 _r_°--5"P3
C) • *
C) O0_,-¢0_
e.* r*- e-
* °00¢_
* r'-- _0
_- _o _r'_
__ ._
*_,*_* +÷*
__*_ _
_ .._
N
N
*-e
EN
C_
N
N
01..a;r*q
_J
0L._J
0S,
N
0S..(M
N
Ut
0_
_o=J
I
r't | O0
"-1-.,..I
,00 ._ lU
I")(_ tl I_I._ [ZI I1.¢
r,90 i-- m
--OO
ZZOO
u'_P-I ,- ..7
• _._1 _,__
f4 t'_
_,_0
O • •
I O-_
PJOO •
Z
7
WI !
,_mm
I
_ _...0o_
_66_66 "'®_ _. _.__ I .o
• -_.o_.° O_ _
II
.._• °
,._
_ I
NN
_N
NN
NN
OO
oo_-._
o .r _- .
t'- t-- i'g
.._0
-o_
.__._
_.__.
_-_.._
OPeN
_.
I
_o
I
OO_
_N__-
.__
_N_._
er_ . °
f-- e-- ,_=.-
,r-- i,,-'
i,,.- • •
r,,- N _'1
• f,,"_ i._I,,- e,- e-.
÷ ÷ •
_ ,.--Q_
.._ t'- eO• r" P"
° °
odor _e.- _ o+ ÷,._
QD ('_ID'-
-_ f'4 0
O --'.'.'._ •_0NO
0_/ ,--
I
x
I
°.
_II (.,_ I./'l. _-.-
N
NQ;
• Q;
W._3
N
N
m Orx __
I.
O
_2
_O
I
_. O 0_O
_ I 00
t'5 I'_ ." I
L_ m II r"L00 _ U_
251
...." ".
( )
_N
_00
ZZO0
I_
Q • °
_QO
OZZO
Z
II
252
0 00_cO
t_.cO
_0
00_
• °
I,,1 r_
_ P
t1'-- P
00
00
_,0 'WO
N_
m:"
oo
edN
• 0d_0oc:
_N
oo
• °
_JN
NNO0
• .
I !
-0
!
OJO, I
0_0_
r-P,.! I
O0
P,. r'-.
I I
ChO,,0_0_
• °
(_1 0,1
I
0 •
t.
_ __._
I
00_
_,_
._
_.
C_O
"N
," 0'_ 0 °• °
r,'} U",
_1 .0 -
r'.- _%
.,__,_
_.
il
_°_
N __ ,,_ _ _0 _
_n_rN
o_r_.
_.- eel (M
,-' (%1 r,-
°p._-
• °u'%
C_ Ch (_
.ON
_ .... _ ,___- N_
__. _
_..._ _
L ]=m o
Q.Q.
L.
1,.
m o _ t_tc;-c_ "_ 0
> ,,
I',.d
_pr_
o1,.
N
_J
0{J
I
.,._0
II
!
U. 0 0'0
C= I O0
i
Ou'%
.3"
W.. "_
OO
<mm _
O , •
= d
Z
Z
II
N
0
_.._. _._
• .__0 _0_
°°°o,...
00000000 00_0_
d_
000_0..
-._ o_o
O_00_-_O
O_NOO
-.__-
_°_
_..NO_
_NN
O0
OOu'%_%
Id_ "
"O!
r0_0
r" r"• °
! !
._r0
0dO• •
• o
• °
O0° o
_N
.0
o'} o'1
• °
0,1_1
2_"e
_._ __. _._• "_ _-_O _O.
_. _ _ _I
_. __ _ _I
I
• __
I
oo._m- _ r--CU O f'-
e.- _'- e-
O_O...
__.___1
OOP')
+ + •
0%C_N
• .&_,
%0c00 •
__.
_..._.
!
c0 ,-, %0
or_.o• c_ _,-
0_ 0_
,%0t4
_0 c,'),'-
_N_.
._O_ .....
0% •
¢W (D 0_._u'%
C%i _'- r-
_ON
_OOOO_OO• __.
_...._
O_OOr,_
(%1 _'-
i,'- •
Ou%,-c_OO
.7 O% o4_ O e4 .:T-- r-- r,3 .3"WD r'0 L,'%_c,_ ,--
_o_o°_.
I
¢'4 u'% C_r,% C%_ ,-
, " .kO_04 r'- _--
I
° e.-- _.-
P- "_ ,'- -=.'T
• ..'_,-" %0O --'_ •
.0%0
r-e-
O_O_
co_r
e0_o
° r_ W'%
CD
r--
eL_L _ _O
W_W
m(
C_N
oWN
O
S.
,o O O
CS. O
_m
o_
M
3 _r_
INc_. ,iT
b.O_O
g °%0 .,_
LU_ I r_,.] 0 I-- m
253
.)
U
m
W C__. ,.q
_0
Q" 0!
"00_ j
_3 0 i- _0
0JN1_2 . .
_QO
._,_OO
k¢'_ r,,-
I P-_"
! ! clo
{,3 ¢'¢)N
= ,de'_ . .O
W_C_Z
I,-
I--
; )©= ,
NO0= dQO_
tA! I
254
r_ 00000 O00 00_,_ .0 OC_O oo.._.o_* _÷
II I
I
i
_O_O_ZZO _
_UN
N
_-- 0
2- ii
_U
.°
b
)
_, . o °
W • ° •
--QOO
Z_Z000
_OO
-'..I
O • . °
_OOO°,,
_0OO
OZ_Z
OOO
Z
III
._0
.-(
_o_o_
_N,_-_
inO_Ot.N_.
O_.,
_N_
_N_O
OO_in°
O__N.N_
,°_
. ._ • ,_ _
• ._.._. _
__t.%l_ .__N°_•_in _•
•.in_g66_66_6 oo==
_4
OF POOR Q_-;_LiTY
_ _ .,_•_,_
"" inP_ _'_•_O_N _, __.*
I
Or")
0_0J _ __.__. __
_ ,,_ ,._._bin in_ _,._,_
• " _Nin inin_,,
0,, U'_U"',{_
r,,,u'_
u"_ u'_
N©O, .O_O_t.%lO• . _-__
!
I
_ in_ _,_,._
_ _N _ _,
-O!
m_
I I
_ _ _-_I
I
_ .._ ,__
0 0 • •
_0 in ,-
inn •
N_--
Ch0_p- • •
0d i,- _--
°(MN
inON(MPN
(%1 * •ChNIO
r'- P'- _• r-- P-
-inn
i-- l-- in
C_0_
t.NinP
P- .=" r--,i_.p
(Mr-
r-- P. ro
• i-- r,-N ,'-
I
.=r .:r
, •inr,I_.- f-
• ,.p.._. ++,
_ _ _ _ .._
.OI
i,.=e--
3o
N
= oo__0
ID
N
= g10 -=I
IU
O
O
m
"=ra.. o,,=4
L. ,--I
oua
,Z
o
L
0
a,,
L,
0
,-,¢
0 UJ
ii
k.
=O
.. '_
ele
I
,wO
n,
!
L_O _0_ I O0
_0 _I.O t_l III I_Ldm 1 r"
L.q 0 I-- m
255
_o°0
000
|||
_00
_°°°
0
f__._._
_-_N.-_._-.
_._.._
0-..
_00
_000
QZZ_QQO
IIi
_oo
Nr,-
F-
C_
r'-
Gr_
256
• oo. ....
_._
_NN
-_O_ • •
_.°°
0_0_
_._
_..
._
000_
_ _._
_ .._
II
_ _._
O0 .._P
• . __ _.
• 0 Od ;_" 0 •
| • •_00_r-- f,.-
II
E NNPN N_E E_ EE_ E_
I
I
I
!
-PP.NOm
_N_OO.P
°_°_
I
_PPP_P
I
I
I
_oo.o_
I
.-r N ,,_
o -1" • •;',-00_
_ r'- IN. /... p.
I_l i.- i--
n% P.. u'_p.. ° o
P- I."_ OJr... Q: .:r
• P.. I'-.
o_ r-- ......:rO_e,I
00,-
t'J _ CO
,'- r'- D-
NO0_i'g .'- r,-.
0% •
t_J I".- ----C'.I .'- CO
• e,_ OJ
0 ° -
0J_JO
• _'1 ¢MP P _
+
0_ Ch ._
• ot_t
',Of"- °
p- (_ p-
° i_. e-C_,"
O t'q u'_° °
O c0 ,-"°co_.
rd P- r_r..- . ••.':" .:" r..-
r-- .=r _. r.. r_
i_a e- p
{u •
:1o
w,
o
o
m
e-
.=
r-
o
o.
mot<
tn
4J
_0
I
_0 _0_ I O0
:3-.,_1
.0 0 .i.i II
I")IN II IIII.dm II.E
I
rO
O_
N
_00
_ZZ000
_000
0.°°
_OQ
NOO0
OZZZ
OQO
I1|
en
__°°
00_00
._ ._.
O0_N_O_
O0_00
._.,_
O0_NN
._.*_*
• _ ° °_
_o. _ _
• ,_ _
• -0_ 00
_oo
_0..
• ._. O0
_p.. ..
_N_._
_.o .°
,._
-*_
_O_N
N_O_
00_
• "-¢0 _'• ° cO _--
-3" U"%,_, •0") 0,') .-"
*._
• °¢_ •
O_O,-.Q'.T
._I._ - 0,1 i'_
_r 00 o., i'll
p o
_._.._P
I
I
°°__°
I
o__°
_PP_PPI
_-_N_O
°_o_
.o_o
I
__I
_O00
_,-e-p
0
000
_..f..p
_oo_r_o
¢o_,o,-.
o _ u._
or,-_
o¢or..
o •_'--_r_;
N .N
ur%_o_O
Od_-
-.5"_0
.f_.p..
I
=> O_X
r_
¢..
.o
II
/
*-_11
,,,,4
_ _ _ °
o.
U_
= 7:61 9"
,,_ Ill {Xli n, ,,_0
e. 0
I
b, 0 _.0_ I O0
,00_ II
tO 0 _ CQ
257
0
-_
_000
Z_Z
OO0
0III
Z _N
_oo,
_ZZ_
O
if!
258
_.._
_--_N
-__
E
NNZNNZ_
E XX_XX_XX
N_
,-_N
,_
_NN
_o_
._N._
_0_
NP_
_PP
-__,o
_NPN
_0,-
NNPN
(M(M
• 0p-p..
N(M
p,-p-
p.p,.
p.p,.W_W_
i._ p.-
(_I(M
OOp.p-
OOe,- r-
• oOOe.- I,-
• •
e--p
• •
PN
• 0
_a(M• •
W_W_
,=r_-
i,,. _I
• 0
• 0
:=:=
_0
N_N
O,i ;',-0 _w_
sisI_ ,
t'_ _ r',- ,--
_NN
.{Mfl,;_, r'-.,_
._" o,_ _o _
i.- i,.-p..
_o_
o_._
* eO ',_ 0
_0
°'oN
O0_N
o.o_-NN_P
PP_.
PP_.
0,,_0_0_
_o
I
__._ -_
I
._o_o
I
I
_oo,o_
I
I
NN_P_.O
!
!
W_m
p-
W_OC_0._-_
o fe_ ,=..
COal,--"
,-ow_
,NNP P _
NW_(M
,NNP p p
_=, I/"i •
_ _,_ ,,=r_") P" N
• ('_Ie.-
r_ _'_ .=r.P-N
P _1 P
e--
i._ • o
OC_N
p- e- _,
0
e--v--e-
_O_O_JZZO _
{11
N
=
o
[.,iw
r".,-I
o
0 ,_'
0
e- _ e-
o,
I
in
{- .=_
i 0,-
I
!
_i I O0
_0 ,¢J-i_ 0 .w IiK; (_i m il
WOO iil e"
m 0 _-- U)
o._
(M
O_J
e-
• o_ o_ _
r_._
co
I,-'
f..
i-
_ P.- (M
OOP-
_=_0
000
I_ _
III P _"
_... __ _ .._
._ _ oo_
I_ _ _-
000 _ ._
|||
o_7_=© _ _._
_oo. _
0
I
I_0D
O
O_PO
-_N_
{_1,-.0
0 -
_I,-IM
PP _
o_,o
r_u'_
.0
0 •t%l,-
_.
PP _
.o_ I_.(M
,--u_
0_.Toa
_ur_
,-'0o_-
_.. o
rr_u%
ON
P_QPP_
E _NPN
N_
m_
00
p_p.
_0
NN
_r_"
Mr'_
• o
_O
cQ00
OO
I'_P-(M(M
NN
_P.
"_ • .._" _ 0_,_0 _ • "
• 0'_ O_ f'--
P- e--t,-
-_0
o._
._ _N
NO0
p-
_0_
o. _.PP_ON
¢0
C_O
!
,"0p-
N_0i
-0 N
_PPN
P P _PP_N
_'_0
p-(M
|
P" _" P NP p-
I
t
00_o_
I
o.
I
N_PN_PP_NP
I
u% -_'l
0 IZ_ _r
OJ _- I,-
•_" 0 Ch
e-e-I-
p... p _
I"- f- I,-
C_ f_ Cl_
or_
r_
_- .h-
C_J _" _-"
• .=r -.'r
w'% • •
(M_N
•=" -.T ,-
EZ>>
P=
0(:,
L.
0
=
r.
2
g.
e,
I.. r-la
0 0 _
au .,._
00•< _, :Z:
oo
259
t_
M
.,_OIll
I
"_ ! O0
'_ 0 /-- U)
0
N
{N
N
N
d
2 60
oo.ooooo o._ _o _._
00000000 000_ -0 _00 •0_ ! .-
_ - ._ _ _ _• ._P • • -_0 _ • • ._
_.._ _ _ _-.__. _ _ ..._
_-._ _ _ _
• -__. _ _ ..-_
!
I
I
!
I
__._
o_._o
_PPP_ P
|
I
__.___l
1
I
I/'_I¢% I,,-• _. p,,.
•,- r-
0 _1 • •
._
0 CO • •
(M _'. F-
0 O_ •
0 0 (,_1
0 ,-" _-- 04. + •
O_Ch
_-- ffl t.'%
0 I".. O_, n') C_I
00_m-. _-- L'_J÷ ÷ •ChOmP
ChCh
C) O_O
÷ ÷ •
0_ Ch _r_• _'_
0_0_
"N_Q • •ONr-.
,... 0_ r.-
U'_ 0 _0
,-" _") n'i
t..-
_0_
f- i..
ZO_O_JZZO _
!
I
I
w
..,-
w _J
0k_
N
0
N
0
0
0
I =
ul
.o
ii
4J
In ,-,ILl O"
I,
_. 0I
_ I O0
= ,..= .J
Y){N IIIbJ rn I J=
L9 O )- m
•OF P00; _,
w
p-
f..
U%
p-
_OO
_OOO
Z_Z
ooo
I_
III
_P_O
_.o.
..- EEEP
O--.
P_OOOo..
_OQO
OZZZ
OOQ
P
z PN
III
L.J
r_
PP_PP_P _ PP
P PmPP_PP _
._o,_
PP_N_P_PP_P_
,_,,_
_P_P_PP
PP_N_P_pP_PP PP
p_PP_PP
PP_N_PN
PP_N_PN
PP_N_NPP_PP_PP
PP_N_PN
PP_P_PP
_ N PP
_NNN _
_. .o
_N PP
_, -.
_ N PP
O_PO
°_
U'_ I'-. I_ =r
PN_
_°°_
OPO_
OPt,
_,_
._,_
o._
_O_N
E NN_N N_
I
I,
_NPN_OOOO
I
PO-.O _.
_P_,_
I
__._
I
I
_ °_o_°
I
I
l
_PPP_P|
I
I
{....:_r
,P-I',.{Ml-_
i1_ • •
. r... r_
P-r.._
_300
,p-p.
_,0 •
r'-_:O
0 •
,P,.l',.
.p..l_
,r,..P.
r..r'-_"
o_r_u_
.r.,-_o(M_.-f,-
2; r-
_2
_ N
°.
_O
!
r", I O0
_-0 ,_
,00 ._ IIiv) IN ill II
_.10_ lie"L90 k.- {.0
I 261
L
p-
_...
ZZZ000
|II
J
0-.. _
_ • • o
QZZZ
000 N
Z _
|1|
262
._..
_._._
• ._ o___..
_o_o_
.__.,
_0_
_o.
_..
NN_
_0_o_
_0_ _._• -_-_.. o_
_-_-_ _.
__-_ _
0 000_
__Z_ _
p,-I _
CD,._r,_l_i-- e-
• o
0',0_
I-P-.
O0
p s.-
Uh_t_
00
_ed
p. p,,.
s-..
O0• •
s,.- s.-
Nt'Mn3_')p.. /...
000_0_
• o
ee) e,_
• o
I I
• •
¢- I,--
o0I
• o
I I
t_N00QD
i_ /..-• o
t%l ',_ _ ,.-
O_ ,.'- 0 .-"
eO_O 0 .T• -- _ N t%l
_'¢1_ •
• 0 I"- '_
0.00
_.
0_._
_._
-._
_o
oo_
0_00 •• •
_,-m-
o._
II
_(M .s. _.
/... p.s.- s--
.._
_o
00.----0.
I
O0 .... _.
__._
I
I
I
_ _ I
"'_._0__._._.
I
I
O_ U% t'..0 _ • •
WDO0
_o_r
O_ .QO
0 NO •
-e-_r
• ._" p...
p-- p_ p..
t%l U_ U'_
-CO
OJ P'- u'_o_
P_NO
p- o .
u'_ uh u_
• '- P"" t'N0
p... , •
_') OJ 0• ',,0 _0
00_f- e-- _4- ÷ •
0 e'.- u"¢
-OOJ
t_. o •
, i'-- P'--
aa
-4"
=
00
n_
0
0
o
0
u B
=
.o
II
II O"
_0IiI N_,-
I
f'_ I O0
3-".-I
-00._ iI'_ _l III II
(_ 01-" f._
_°°°
_ZZ000
III
_00
_000
1
0 ° • •
_000 ° • •
_0_0
000
0_0
III
r,-
iN
o
r-
r-
if%
0,I
.._,._
,-_-o_ _
,..°._
,._.._
°o_.o_o
.'._.._mN
° ._ ° ._
__Z_
0-_._._
0_..
_o°
.._
_°o
_0_0
.._
_00_
00.-
00.-
_PN
_ZZ
_N _m N
mm N_
_ °._
• . __ _.
PP_
PP N_P.
O0 _0_.
_ _.__ ._
_ _--_
O_ .... O_
O_ • • • -N •
I
I
I
_.....
NN _NN_
I
• U% _1 ed0 I'-. •
, I'-- b,-
+ I.,. p_.Npe-
.eOr_
,," ,r'- _'-
i_J • °
OOJO
L'%l r-- _'--
0 •
C'd ¢" _'-
t,') Od 1'_
r,_ ._ _.-._- _') _..
._.0_
_"1 _ ,"'-• _" ,"'9 ',_3 -
• _iD ,r'-
00
0
_r
Q.
n$
.?
r_
.Q
t._ ° •
° P" "_" _ "
• -- CJ e,'_ 0
_ CO r--
.°
,_ III fq_uOm
_. 0!
I_. 0 _,...I_ t O0
1.9 _N el I
I_lm II r"f._ 0 h. rn
2 63
_00W ° ° •
--000
ZZZ000
III
_00
m
.._O.o.
o • • •
oz_z00O
z
z _
iii
_ _ .._ == •_.__ _ .. _
II
__.. _ _ ._
__66 "'_ _" _ "_
oo
__ _ _ _._
__ _ _P _O-• ._.._.. ._ _N ..
II
__., _ OO --_
264
__0._
I
__._
I
I
0 NN_PN_
I
u_, • •
N f'.- _,0
° ',_ ',,0
O r"- ,-..='i'M
_- .'r ..-ru_. ° °
+,o-7(3_wooJ
°e- N
e.. _-- e-
u.._ _.',k/ °
.._ oe.-)
oJ _..-
ooJOJ
- U'_ u'%
0 Cs,,l.-
('+wlI"" _
,._"U"X u",,0 • •
0 .--'r,_rO 0", _
• u"x u"_{%1P e-
0
:io
_w
0
Q.
0
m_w
0Q.
0
0
0ww
w_
(30
= X.1,1
O
.°
I11
Ul
"_ 0
I
Q! O0
W m Ill"L_ 0 I-- m
L...S
_°°,
ZZZ
000
I_
III
Z PN
_,.,
3 • • • _
0.°o
OZ_Z
OQQ
Z P
Z PN
III
_ ° ° o
• , ° • • ° o , .o_
00_
_ ._0o°_° • _P
oooo.o°.
O_OQO_OO
oo_o_.-
NPPP o_
_P_PP_P_
.I_IlppP
Pp_PP PP
_._o_,
P_{PP_PP
_._._
_Iftt_P_
PP_PP_PP
PP_PP_PP
PP_PP_PP
,°°.°°.,
O_OOOOOO
p-
E
ZZ_Z_ZZ
°°_
OO_
°_
O_
PN
°_O_
°_
oo_
_A
I,_ °
.OI
rnn')
u_u%I !
.OI
(M{M
r_p-
_0_oi i
NN
,.nO,,
6"9 P.- 01,t_.,_"p.- {:_
e_1 'i,O o_I'9
PPPP
I|
_o_.,_
PNN_PN_N
PP _ N"
PP _N-
PP _N
I _,_-
I
_.o._.
I
_ .... _0
I
I
*oNo_O
I
-.NNO,_N
_PP_ NP
I
I
• OOmO
,,I-.$- •
0,, 0'_ .="
p- p,. p-(%1 , I"".
ONO
_,. p.
i.-- _.- _wI4- + •
OO_"
.l_r .-
L'M i-- i.-
. p. p,.
{%1 * •
._p..
N _.- e-
{M o_'-r'-N •
Pc0C)
.u%u%
(M(7,
u%N
%0f"
o_e--I_e,-
PI • o
t_ P- P.
(M_P-
¢_) • °
_ ed P,-O'_cO
00_
-_- +
O_O_O °
{M
cr _._ X
_.cjp
o
o
U1",9
2
C_ 0 "0
+
5
= ._
o ._e_
o
I
I
Q,. 0I
h 0 {_--,
3-',J
-0 O a,_ I
wrn _ r-mo_-m
265
O
-__OO
Wo.. ._
--OOO
OOO
OIII
Z P_
• • . .
"v EEEP
Td2= oO • • •
_O00
OOOo
z--o_
|11
oo_._o_ .o___o_ _.
o_._ _
oO
266
..No-_po
_-_._PP_N_PPP_PP_P
.._.._
_._o_
PP_PP_pP
U_(MO
.P-OO 0a
p..
n'-_OP-u'_
p-
(MU'_
.::rod
cOcO
.7(M
(z}c_
u'%c_
-TN
P_
.='N
C_U_.='(M
•.- e,1
{I}_r
p.- o •m p.p..
! I
p- p-p.
Od
,-- (MN
• o
U'_ ,wD_D
• • •
O_ r.3r_
,,Or'3 .7,="
00_ P3ff'lnl{M ,--P
PG_ ur_. u'_
O.IN • •• _,3 tel
e-O'_ _.- e--e'l(M
t_lO_ rn_3_t_ _3 _'_
'.ON u'% u'%
• .=r_"
nlN PP
n')l_ • •• _r,_
e.-_ e.- i--
mN N_wI
• • o
a1N ,.- _--
t_-nl -l.f_
PN_N
_*_
N_PN
I_ &'_ •
u'% u'_ (M u'_
_.
._._
_PN
_OO_
_ I
II
E NNPN N_
E E_ EE_ EO_
I
_..o._. -
OO ....
I
.... _
__._
O NNO.N_
...._0N__.N
.,.._,
I
_ P_ PN|
_PP_PP
o..oo_
C3_
O •
f'... Owl_0
(MW3
t'M
_0_
p-t%l _'-
w_
p.
p..e-
_r
o:p,.
P- t_l e'l
_D r'.-
o
O
t%l _-- l,-
._. o •_3P-N
• ¢0 I"--Npf-.
_" .Nf"- 0_ •=r e-. _C_
• f-- I',.-
0._ o,.IP-
._p..
OOu_
•I- + .Ow 0.. P,.
o.. 0,_
{',.a',_ u'_
0 • •• {%i _.
,,- r-,,(M
EZ>>
=> O_X
p
: m 0
-n"
[.)
_. "-:
o
o
II
4JI
I O"
I
(1. OI
-._ I") I'%
1_. 0 I),"
Q I no:3-"..I
'_0 ._
I'_ t_l I111
I.,.I m 111"_ 0 I.- m
b
N
N
° • °
_QO• ° • .
__ ZZ_000
_NO0Q ° • °
__ OZZZOOO
= 3
-- III
g
Lj
r
d
-._.._ _
_O_O_ ..._
o.o.o..o
OOOOOOOO OOONoe-
p._r.-
m}
• _ ._.
_._._
o_ ._o
-_ .__ °__
6_666G6_
_Z_ZZ_Z
0o_
_o_
• ._
_.
.0__.
_O_
O O r'1 NC:}P
P-h-
_ZZ
u'% _'%
OO• °
I.-.e-
• o
¢'1 f,_i-- e--
e.- 1.-
e- e-
_{"% •-OI
_r_e'% iz%
• o
! I
o •
i.- i.-.
i.- i.-
_0w)• °
e.- i.-.
h-h-
e- I.--
_q4NN
W_W_• °
n'l e,1
Ne4P-r,-• o
i.- f-
f,% r,1
&f'% 0.OI
:z:i
_.N.._
_.
.o_
• • W'% (',If'_ --" O _f%
e'- _%1 e- •
_PN
a}_ -W%
e.- .
,.-COi--
_ONO
. ._:,__,-(%1 &¢% ,- _Z%
,'- 04-r" NI;% _r_ N -
ON_
O_N
.._
_.
_ChO •
_6t'-(_
i.- i-.
__._
__I
__-.
I
i
__-
I
I
I
_) i..- 0 °
op_p-
O "_" •I..._ re)
• e-- p.-
O _ • -
OO1/%e- _- It%
¢)-1_
_.- o
0 f.../_.
. P. P..
°r..I _¢'_4 f.- ,,.-
-h-P-
,--040,1
• h- r"-
_..._. ___._ ._
__._ ._
%
0
o
U
,../,
0
N
°.
• 0
tM
Ul
I r-,I.-- N
a
I...,1'9 m
"1..,_1
I.U m II _"{.0 0 I-- I,_
267
--OQO
ZZZ
000
_N
|1|
Z _N
0 • • -
_O00 • • •
O_ZZ000
Z
Z _N
I1|
o_ ._o
o_ -_
f_
o_ o_o
.__oo
-__o
00 _¢',1
_.
_m_m
._
._
P- I',-U_IZh
e- I--
I"
.=r._ur_u%
e,1 I_1m.- e-
OJ{M
.=r ,._
enf_e-_-.
t_atM
r_l i'Mi11{,1
re) n1
._" .=r
n'_ n')i-- f-
u'_ u'%
e,1_1
r,1 I,')
e- p-
h-P.-_O_0
_._
_ON
-0_0_00
I
I
_._
_o_
p,. o o{:_
¢',1
_o_
_o
.o_
_NO_
_d m_
_.._._.
i
_ .... _o
I
I
_oo_0oo
i
_ .... _0
I
0 _'M • •
m{M_
{_ .¢'Q
.p.p_
N ,"
0 0 •
{M_e-
r,-tM_o a') ,..-
. m- p,.N .--
u_ •
_NQ_• _11',-
{M e- _-.-
p- •
.enD..
_'M • •P- --_ .=r
o ou'% _' ,--
en f1") _
• •
--_'0OJmm
OOCh
e--p_4- ÷
• o_/'%
I|111111 IIII II II11 Ii1111111 III
GD
fN
,-.1U
L)
0
f
If'
_0
_ 0
I
h 0 _,,.,,
r"_ I O0:3 -.. J
•_0._ *'I") IN II _1W m W.£
re 0 I-- G')
268
__0 _.,
I
__----_ _
.o
; j
m ¢0
0°
_..p.
0
0I n-)
{._O0
m m,.i ff_
- . g_.,I _Or-
O40
QO
Z
C__QZ_
!
T_
Lj e"_
li_.i
!
o,_ _, _
II
illlllll II1| II IIII
_-_,_
_-_,_
_O_o_
o_
_0_
_..
_" Oa O (M
_-.
..... • .. o.N_00000000 _0_
_o
p-p-
_.. p-
• °
I"--t',-.
• o
p..I _• o
_-.f.
-0I
n') u'_
p- _-.
O_O_
_-_
_NO
P_ON
_.
_P
_0-
p-_O •
r- e,-
•__3 33_
IIIIIIIII III
_0_..
__-_
I
PP'*N,_N
I
I
I
. p=. _.-CM,'-
._J_
. /_. p.
_-='N
ma__4
• t- D-.
¢0
D-- r'-
.cOt'-,
m _'Q r-.
00-_"
"I- 4" °
_J
0
U
e._ 0
0
°.
F- ,-_
g_ .-,
_o
oI
_ I O0
"-0
_ 0 1-.- m
2 69
or.,.
0,
--0(in
Z
Q
0
_0I cO
mm,_
<m
w0
o •J w0_
o o_,3Qr,,-
i'_o •
oz_o_
,.J re"
z
_o7" i_,'l
!
T_
_. o
u.j e-_
2 70
r,.,r,.-
P,z
",op..
p..
(M
p,.
{M
#.,.
{,Qp..¢Q
7-
f,..
0,,,,0
CO
'43
p..
PwI
,wON
_= "_ -_ _ • • -- o=o_
_C, ._ • -0 • •
• c'1 r,'1 • _:"_e,'}
__-.
Z__Z
XX_XX_X_
.._0
_._
_ _W_ ," •
_-_
_..
• 0 _0! o_
II
t'- t'--
t'- D'-
D.. f'.
m- mm
#'_ r,-
•_ ('Q (_ U'%
_t',w'% ¢',._ •
,._0
_._.._
_N.
_-_
.._
_.
_'- kO _'- 0
.'- w_ _" N_1 •
_--_
1
._._N_
_.._._.
I
-_._ON_.._.N.
-_._
_..N.N.
I
_ ,_ •
!
__._
l
_0_0_00_0
i
O0 ._0.
!
0 0 • •r-- _" _
-t'-_
0 u'_ • •
_,00d r--
("J e" m,-
Iw- e"} I",-
cO (_ _0-_r'-
00="• -- t- P.-
• o_,0
00_
'w_ • ,
'W:},'- r,.,.
. p. p..oj m..._
• of.- r,_ t...
• r'...r.,-(_j m..._.-.
._- . o
• r-- r--
='wow:o o
(:o .-..-:-
• _ r.,-
(,,,j .p-o,,,_'.'1 o
• _,--r.-
r-..(}_,_
Q0 o',,,,o• /..-r-..
. r..-r,.,,o
__X _
.X
eL
0
0_w
.-I
fj _-.
Ira.
iJ
L_
m
M
_0
'_ 0
I• ._I_ P4
¢3 I O0
"_0 4..=
• 00_ B
La(O _.C
i '
NCO
0
K.-r_
--0
0
0I
_0_0
I co
tO00
Ze¢'l •
<m
a,_,-r,- 0 ,,._0 °¢¢'t
or'_1"40 *
L._ _"
O0
e_lN
I|
N "* _
,-_**_.
N _°_'_
.*_.._
N _-_*_
°_o
_,_
._
*_
P_O
• ° _
I!
_ON
_ON
_ _._
o_.*_.
I
-_o_
I
I
I
I
I
_" ,'- u%
*cOl_
{M _-" P'
(MalO° °
_ e,- e-
p _ /',-
Od f'- _-
_OJP-
Q_QO
O_ ,'-
pp- •
U'_ n') P_ U'_ CO
t_l ,'-
oo_r
° °
o e,_ coo £_, n,)
u_ cO aO
c_ O_ _0
_... cO I_
* °_Or_O,'J"_ ,-- QO
.COP..
{'_J r-- i,-
¢%1 r_ 'WO
oO ..,-rQO
° o
_ _"_ ¢0
• QO P-,
• 0
0
• I0
a
o6,
o
n_
u_r_
°.
271
I
II_ _'_
i
!
L_ I O0
",0"0 0 ._ III_ IN II I
I.UCO I r
_CO
0
r.n
Z0
0
I c0
_o0_" r,') .
_c,n
• re)
t_CO !
0 •
," ¢> t'_
Qr'-
tXIQ -cO U_OZ_
OC_
Z
_0Zc'_
!
7_
_. •_,10 C,O
2 72
O
NO
OO
0J
N
t_
t_
N
C_J
_°_°°_,°
,_.._.
_._-_
._.._.__.._
_-__
• U'I /'.. ,P
;'.,. ._ ._._ r..-
r-,c_
_
_._
_ •
"-0_
°o_o
11 !
Nt_• °
nlr_
__. _.
0_._0 _-_• ._,_ _
_.o_. _
00
_r_ _0_
_-
_r C_ c0 o
i:% u-_ C_l •
i'M u-,, ,- t'Mu'_ _ •
rr) cv_
_-.m-
_-_._ _ .... _
*-_-._ _N_ _ _N._
__0-. _0_ _ ,._
_N_ _ _ _.
_.-._ _.. _ _
|1
• ._--_ N_ _ _.___.. _ _ ,0_
_-_,_ _.. ..
_ °_
_._
_.
_._._ _ ....
P_N_ NPP
I
_000_00
o_ {M ¢0o_lp-
Cl.I i_ P-
•=" ," N
001.'%
• CQ f,,,
i-- . o
n_ _J u_
oor,-
÷ ÷ •C_ O_ C_
ChO_• ,M_,
O_Ch
r,- _o (M
oo_0
° o
_ r',J CQ
• _:_ r.-
°.7
_cO0
• _-- 0_¢_ ,--
O
c_4
J_L
r_
O
O
o0M_
¢J
U r_
¢1,U
.o
C_r_
_n
O2=
OZv
IIIt
I_ 0 _
3 ir_= N_,-
n 0!
e"_ I O0
_0_
L90 p. _n
l
F _
--en
0
I.*.1 °I-- r'.-
--0
Z
__ 0
0
_0
I ¢10
Zn'_ •
V) CO
p- n__0en I
0 .r,3Or'.-c'Mo •
or',
,, _e"Z
O0Ze,_
I
¢"_N
,,J e'_
C._ ee"
f-
i,-
r_
0
0
0
p.,.
0
0
0
__0o
.,_.._o.
• o_ • ._ •
_0_._
_-__
_._-_
o_.._.
_-_o_
_00
_o
_..
o_
_0_
_0
oo_
_0
_o_
• •
e..-
r.- r..
o ,
I I
I-- _.
re)_
e.- i-.
n.)n_
cocoI.-
I.-.
O0
m_
gh_
i_,1 (N
• o
I I
_o_
_.
O_ .0 •
,.-f_
.o_
0_0_
_._.._
_0_
_.
_._.._
_0
000_
,'- u'_ 0 O,J
U'_ Od
• o_
_ON
e" 0'_ 0 •
__.
• _ __.._._.
I
1
_000_0_0
_.._-
_.o_._o
__._
I
_oC=;_', 0,,,(¢_
o_,-
0 0'_.¢,,,_
0_, r'-
-7 o/_-
n'} C_ e,,.
• e.,,,./..,,.Od --
t'.- •
_'._0
• _1 _.
o_-
_" cO e-
e- /'_. e--Ch (N O_
E_l e.= e.-
r.,.. _ n-)
. ¢3 r_.(%1 t.-. e-
=o=_N,Or"-
.=0_-IL'M e-. e-
_,Dn3 •
_',00oe-._
e- i.-
0
o
0e_
....4
e;
Q
r_
r)o
o.
m
fI_ O _
oI
" O Oq(N
(_ ! O0
1,1,1 _ lU _"C..,.,_0 I-- m
__ 273
--cOo
Oi_J •)-- r'-
--O
0
0
r,')
I cO
_00
3._<m_,J (_ z
o •
o .r,_0 t"-f'_O •
_.1 r-Z
O0Zr,"l
I
7_
-".t
_1,¢)
r,,3 ¢.._
L_ i.e.
I
I
• -_N _- N_-_
II
_._._ _..
274
r_
r_
oJ
r_
O
rO
cO
D'-
7_
__-.
_._,_
• _. ___o-
_._._
• ._ __._
_0_ _
• ._. __._
_..
• -N- _
• ._ __..
P_N_
0_0_
N_N
N_PP
PPON_.
-_-.__-_._
I
I
I
-_.__.._._.
I
|
I
I
0 _ • •
- _ I'.-
0 ---- • •
-Or'..
m- _.-
0 u'_ • •_ e,'),.-
I%1 _O I'-I,- IF-
00.,--0 ,-- ,-- 1"
+ + •O_O_N
O_ O_ _0
0 ,-- U'_ t,10 * •
0 .--- P"• t%l t_
_,- 0_, r.-.
0 f"- • •
I_. ° •
c_ CD w_
#etO_ .
(_ _ r-.-
t.,_ r-
.r,.-
Q_ ,'_ r-.-_-- r...
t_l ,-
m-. o .
,,o ,.- g'_
o: f.-.
CM r- _-
t_J D-- o_
U'_ O_ eOmOP-
Od r- f-
I
I
N_PN
PNO_
O
Q..
_J
_h
;J
--1(..)
Uo_
U *;-_
UI
i-i
II 0"
_O
Q,, OI
,,-_ ;") ,0
_I I O0
_O 4J-0 0 _._ II
W m ll.CL9 0 I-- m
0
--0u'J
Z
0
0
r_xO
mm_C.._ 0 O
0
_, • ¢'1Or',-
0--'_
Z
Z_
!
7_
en_L_e_
0
n"l
• ._ o_
_o.
_°.
• ,.. O0
_..
__. _ _ _II
°_
_o
• °_
_.
_.
°_
_.
_o_
_,_ _ _ _ _II
00000000 00_ .0 _OO ..
II
("J ,_, 0'_
',0 (",.I,,,0.p-._
,"u ,.-"
I
._
_ _PP_NP
(_,--
_ ._._,_ ',0 ",0
_) • o0_'0
..3" f_ •
rO
_0 xD P-
I
0oo_._OJ
PPP_NP
•"U ,'- p--p- ° °.._"0 rO
-QOP-O,I _" f--
'1
+ ._-
!
_ C} ,"0 ° •
,'- ¢0 _"
0
0
0
0
I 0 _
i
n 0I
U.O 0'_
_'_ I O0
1,0 _N W II
_ 0 _-- tO
275
(_'i
_wD
°
0
_0U_
ZO
0
',D
Z r_ °
W(._Z
U3 a:
r,3_3
.)aO !:3 •
•- 0 r'3C3 ._Or_NO
OZOO_
Z_
I
T_e,_o, 1
we-_
if%
N
o
,,7
• _ ._.__.__-OO_N
0.0.._._
_°,_
_P_O_0_°
°_._
_._
.._0_,.
.
_.__ ___._ _
_N_N
276
C_O_N(M
I"- r,-
I I
_0_3
OO
P-r,-
I I
rnfft.:" .=r
p.p-I !
O0r,')rO
P'- r'-i i
NNO0
, •r_
=3r_
NN
/'.- r--I I
_JN_JN
r_r,.I I
O%C_
C_chf- r-
• o
r'- r--! I
(M_J
n_nlU'_u%
• °
! 1
p-p-NCM
u'% •
.O!
i'_ O.l
O0
! I
_ICMC_OO0• .
0_00
0 •
O_N _'__ L'_,;_ •
__.:r
_w
,-__'- u'_ 0 •
r_e_
_00-
Pr.-o •
r,-_ _
Ne_ ul_
_, r',. 0 •
,'-0
i,- e-
o
_.
•-" r-
_',_ a'_ ._.,
rn,-O •
r,- ;',-
I
__0
-._,m__.,_._,
I
__-_-
.__
_N_-
._.__ ._._.
I
_PPP_PI
I
I
,.-m_or_oo
m_
00',, •0 Ch
0 .=" U%o_
•" O0
0_, (',;P-
o ,-:
0oo,_
POO_
00_'30
_.-_
¢"1 g'% _"1
0
0or--,,-;
," _ WD
•=" eO n'tt'_ ° •
•(- ÷ °
Ch O_ 0_
ChCh
(MOO •0 .N
0 _'--N.or'-.
_,'-Ow
• r'.- P.- -
•l'_J e.- _.
C>
4-
O
t"
"t"
O
?
• ° U'I
I;
+
,,q
I
"OO
"t"{a
r"
O
I., O O
• _
o _
e-,
o
o
b_ = m
°°
la
{I {:_b. ,._
.,.,0
* i_.,-n o
!
_ I O0:3"..,I
,.0 0 .IJ II
W(_ lit{3 0 I-- _
r
i_,:i' _ ,
p...0
I-- r,,-
,g0Z
0
o
r,. .0
t.,,,.I ,_ Z_r,. 0
I.'-
n "'!_.D
I _,0 .r',-
','- 0 r'_
OP...
(MO -
Or',nr
Z
,{,30
I
_m
__,_0 _ .. _
II
E" NN_N N_
EN_N_ PNZZ _P_
I
_P_P_P ppp
I
• -NN -_N_ ÷÷ •
m4J
tl
i-.i
Ill ,-.,
oI
• ,_ p') O-
b.O _r'_ ; O0
-0 0 _,_ •
P_ N II O;wm Ipr"
L9 0 _- r.O
o
£
0
°.
277
__ _ijf ' I•
APPENDIX B
STRAIN DATA FOR THE STRUT
Figures B-I and B-2 illustrate the locations of all the gages read except
KD FARI, KD FAR2, and KD FAR3. These three make up the vertical, diagonal,
and axial axes, respectively, of a gage rosette located centrally on the inner
surface of the mid-fairing.
The following graphs (Figures B-3 through B-40) contain maximum stress
locations calculated from strains read on gages located as shown in Figures
B-I and B-2. These are shown versus engine frequencies and speeds as read
from test diagrams. Each graph (with the exceptions of those combined because
only one situation for each was recorded) displays the maximum stress versus
speed or frequency for one gage only. Different situations are labeled for
each point. Note that almost all are labeled with either an XN2, an XN25, or
an XN48. These correspond to engine speeds as follows:
XN2 refers to the intermediate pressure compressor
XN25 refers to the high pressure compressor
XN48 refers to the Stage I propulsor.
r't_=C_u_l_:a PAGE BLANK NOT FILMED279
--L -J
.t-1
0
_J
0
4_1
c_
0
q-I
I
_:_
_J
A
..f
280
OF POOR QUALITY
v
v
\
0
t_4..I
U
Z
c-4
I_:_
_J
281
_3
X
00
I
0
_T
0
(3(3
0_00-.T
0
,=w
04
(ISd>t) SS3WIS MFII_IXVIAI
¢1
¢1
.<
i--I
U.,-I
¢1
Z
¢1
0
C_r-
I
¢1
_0
282
0
0
I=
_J
,-.4
.M
,.-4
-,,1"..-I ,00 Z
•,< Z
_. ::_ I
OOr_
O
,-4
,-till
,.._ c,l c,4ZZZXXX
I I I
oO _ ..D _:_
X O0000
I _ "_" _"
,-i-- ,-4-,1"-,,1" ,,
L}(J
,.<
[]
oO
Z
I
.z3
,, _'___ 00_
n
00...I"Z
I
,._,-4
-..I"
¢M
OIJ
<
4.1[.¢)I--I
_3
0
0_Q
0
¢,4
0
u _oo
c_ (II (J ¢J
¢n _.,<<
'i<
<23 q304 ---
0
I,-4
c5 o
.<
[]
©O
0- O
I'q
O- 0
(',4
0(3
r}. O0
OO
(,ISd>t) SS3W.LS 7,lnFllX'v'7,l
0
/',4.l-
>-
ZW
0Wn_b_
A
>
<
m°MN
<
m
.,.-I
N
dZ
N
.i..1
mO
.r-t
.r't
I
bO-,-4
283
°,-I
_"Z
_ I
__v
0
_ _ CJ U
_ _ m m
.t¢
Z
I
¢J
<
l.d
I I --O0
,.(_ O0 t'_ -,,T
O 0 I
U
0 0 0 0
0 o 0 0
oO
ZX
00
0
,-,-I
U
'L,
4--t
/"1
0(',4
0
I_o ot3
0--<0- O0
u'3 I
z .. 0_--_--- _
0
o 0',-1" 0
0 U
<
u - (M< ,,-
o
° _M Mr,l
, ,xg 0
§i--_ o __ " _' I
o u ("4
o 0•-- 0
0 o
¢'-1,
0
,""4
<
U"l-
ZL_
@
(ISd_) SSqa.LS FIrlIAIIXVlAI
(J
q#
r_1.4q_>
q_
0'3
<
>_<
0
Z
m0
I
_J
284
_DD-b_
Z
0 _
cD
oo oo co _D
•E -_ •
oo
Z
I
_D
0
ZZZ ZX*_ X
I I I IcO
0 CD .._ oO ,_-m"
r-.r_ Z XCD :_ CD
0 0 CD 0 1 ,,_ _ -,'t I -,_"
-.TZ
I
CD0
0
,--w
oo
,N
cO
I
00
01.1
[]
¢,1
I
00
.2CM
0
,--I ¢'4
o
,.--I!
,....40
oO
Z
I
o,,=,--_
C:) C) 0 0 C::)
0 C) C) 0 0 0
(,ISd>l) SS:Ia.LS I/N_IXVI, N
L-
i_
(N
0
"Ontry0"--"
oL_r_J
- 1"-13_"--'0%
-
_ _-
0
_D
D_
to
Lo
c_
cD
C0CO
©
,dI
285
I
E
q;
m<
I
_4
X
-- I
00
0_.J
_J(J<
_.J
,-' ! !
>
C _--0 0
X <<
%
(J
¢0
Z
00
0
_J_J(J<
_J
_J_JcJ
c¢I
c4ZX
! u'_ oO
0 i i. I_O•._" 0 0 ..C,
.0 0 ,-_..,,_- _...,. r-,,..c4 0
Q; _ (J
_o
_'_
0
0O0
0
0
Cq
00w
0
0
0
UI
>-0ZW
0Wri-LL
_JC
_r
co
0
>
4_(Z3
<
X<
X<
g
Nm©
C
t_
r_I
=1_C
_iISd_) SS-g_J.S K4Fq#II_V_N
286
IZ--X
!
0
ii--_i
<
4_
[]
O r._
_._v
Z
E
v
._
c_ZX
00
0
_J
[]
I
00
('-4
0
z E
I-- z
"4 I
O
O
_ 0
<
ZX
I
0
0
O_U
<
_3
Z
X
, ,._
Itl r"ll,qIxvl,'tSS3a..LS
Z <.1"
I!
[] u
_r
0
O_
I
,iI
<
0
0
>
<
v
<
_q
<
Z
0
I
(D
287
3:(3
v
b
oz
>,,
oo
ZX
I
I-i
¢'4 _.1ZZX_
I I
oOO_
X
,-'_ 0 0I _,_ _
0 cj
_ cj _j
0 _._,_
oO oo
Z ZoO
Z__ I___ I.:x:.
_ o- _-¢_.
0 rm
¢xlZX
I
C
,-..t
4.1
ZX
00
z_
t_
_<
Z
_0 I.,_Z ,-_
0Z_O-
_00
I I I
4.1 r'_
_-_ -g_-[] _ oo
%, <..._<
(ISd_) SS3_±$ I/I r'11,'IIX'Clf,l
(D_0
O¢O
N"1-
>-
i,i
oi,irYI.i_
OO4
(D
t-
;:>
r,b
4.1
O
O.1_I
O
r_
0[-_
oqI
_D
288
Qu"h,2.,I
I
,-I
Q
0
O
I
0
01,.1
<
Z
I
oO _o0
,,'.4 r"_ ,.,,,4
F0 0 00 ,_ 00
,-.- 0 0
,-I
U-llJ
0
2 _Z
I_LI
0_I"jJ
n_
oo
¢ii
_.i I -
I I _ .z_
I_ I_1 I Z O
-,_ g..,- .... _.o.
0 0 0 0 0 0r'., _0 _13 "¢" _ 040 0 0 0 0 0
(ISd>l) SS3_I..LS IAtF'II_IIXVIAI
"4"
I
D
_qZ
I
r_,-4
g
"121
f3,
_-- -0
0
6
('4
0
cn_o_-_
_--'(,,)
¢'4
0
00
0
4
O..,r_
r._
,--I
vV
0
0
0
,---II
b.O°._
289
i
0
a_ _J=
-CO QJ--U
•_4 U
I
C',J
0
(ISd>l) SS3_IS I_lrll_llXV_
N"1-
>-0ZUJ
0UJr_U_
0
"0
c_
0
r_0o3
_4c_
u
I
0
--4I
QJ_4
290
cOCD
CD
CD--
_'_ Z
ZO ,-4
U
6
_J
u
O
I
o o O
I
CDCD
O
U•,-4 r-4
U
00
Z
CD
OCDO
O
Q_
I <
g-om
M
oN1
,,.4
0
_-- ,- 05 (I3 t'- _ _'3 _- t¢3 ¢',1 ,- 0
"- 0 0 0 0 0 0 0 0 (_ 0
0 0 0 0 0 0 0 0 C) 0
_ NIAIIXVl/i(ISd_) SS3WiS
c_
-_LL
d
d
_3
>
0
P_
0
P_
0
v
0
I
0
U
_q
I
291
r.-
CCC
a c
)
bb
I
FI
:0
DDD
D
J
w
J
J
)
4
I 4 i! !
1 ( i ! 'I iI
Z
OOO
6
¢'4
0
<
C0o.0
-,,4
(ISd)_) SS38J.S I_Cll_lXVl_
0,1
,"-4
n_C
O0
v
t_
<
"OC
DN e
N
)- .,_
Z :=ta.l
O _
_ 0_
Im
1,4 __ m
m _
0 mffl
°_-t ,i..i_ r..el
,-_I
c_0
292
oO-,,,1"ZX
§
O
,-.t
<
i ,..-.tm
,-.4t_
ON
._
O
3C
ZX
00
0
_J
<
0
¢D _=4
COO_O
O
rgg_Z
,-4 0 _
><[
(ISd>t) SS3WIS I,'IRt,_IXVI,"I
,--I
_3
X<
_3AJ
N
r'- r_, o
oW _r- W .,_
¢xl I-- n __.
_'- "_(.'3 >
t¢3
O
Iffl
t_ m
O ulr/l
_0
293
I I
!
Z
A
O3
_J
O3
<
>
0
0
<
I_:_
_0o_
(ISd>l) SSBbI.LS MIqMIXVM
294
I
0-.3"
J
0
_; cq
I
Z
I
t_
Z Z>_
I
.-.4
¢0",I"Z
I00
Z
I
'_ O- I
i-_ _Oo-,7 .,7 X "/-- r" °_e,l Z _ o l..
o_X_ I _>____ o
u- u-_- u___ u_._.___,,2 cl -_ L) -._ c)
---_---- _-_- _
oO-..h-E
I
O
0
_9
Z
I
-.1" .z3Z ,--4>4
z _ _::,4 t_ N
-- G-- 'J-- ',,) "
_m
¢'4ZX
I
00
0$.)
,-4
L;
<
[]
ZX
I
00
0
.-4
U
<
I-i
m
ZX
I
00r--
0 '
,--4
uo<
I3
c-.lZ
I
..o
00.4"
0
,-4
(.J
00-4"
0
<
0 0 0 (D "0 0
<0
"4-
%.,-.
0
,,--k. (h
"0C(n
or,,
- i-•.,.__,,
- (",1
4
C_L_
>
_0
L_
v
X<
>
0
,-H0
<
<D
I
_J
(ISd_) SS3alS I_tI']I_IIXVI_I
295
.,-IU
--0
0Z_
U
0"0
cO
Z>_
I
..Q
00
Ol.J
,--4
..K
r-. ,_
o (23
Z
xr- 1I
00
01.1
,-4IlJU
0 0 0 0
(ISd>t) S$:Ia.LS l,Nn_lXV_
("4
C)(',4
(,O
(',4m
- (_O
N3:
>-(DZW
0b.Ic_b.
r_
C_
0
v
.,.-IC_CJ
.H
I
0
0
I
D
.,--I
296
oO--.T
I
,.Q
00
0
,-4
U
<
[]
0
oo
ZX
I oO
,--t
00 I
.; o
0 0
,.-4 "4
iJ
[]
¢MZX
I
..Q
0--.I"
(3
I
..a
00
0
0
"d
one
p-
_D
cO
ZX
I
0
-- (M
0
Fo
0
r/l
>
r.,q
0
>
<
I
0
0
0
I
_0
ss a±s wnnl×vn
297
00
__.×c_
_=1
_o _]_ oo_=
U-4FU
O0...I"ZX
l
OO
O
u
Z
!
00
06.J
3
I
0
r-4
-r.._, o_ _
,-_
Q *,='1 __ X e" <
Q U _1
u'3 I:_
! I I I I I
0 0 0
r_ r_ _
"0 ' 0 0 0 O" "0
U _3 U c_l C_l
0 "- "" "" ...... 0 0 0 0 0 0 0 0 0 0006006660 ooooooooo
(,IScl>t) SS3_±S _rl_lXWl_
eq
O¢'4
N"T
>-OzW
oWn_i,
O0
0
CD
_z
c'4
0_4
0
,-4c_
0
m
_ m
o m
p_.._
I
0.0
298
,1_c_ti4
Z 0
0C _a0
0.4iJ
¢JoU
-<
°_
&J ._ ,..-.__iJ
m o o3
v'lo <
•t< -K
Z
I
0
0
0_ o
I _ I I I-or
. -.._r 0 0 0 ._._
0 I_ _, U
¢J _ 0 : 0 0
E± o
(,ISd>l)SSg_LS I/Ir'll/11xvl/1
eq
o
on,,_r-
c_
o
0
0
0
0
r_
oo_
m
_ m
o •
o m
o ,_
c4I
ORIGINAL P;_,GE iS
OF POOR QUALITY.
299
I ! I
(IScI_) SS_a.LS Hrl_IXV_
0
O0
04 N•-- "r
>-0
0 Z
0Wrf
CO h
¢0
,¢
A
¢)
b,
0
0_Z
u_
<
I
0
0
0
0
r-4C'4
I
300
OF POOR QUALITY
(!Sd_) SS3_JLS MnM!XVM
q_
©
P.
(,0
0
_J
0b_
0
.wI
I
.wIr_
o
o
o.lJ
o
C_l
I
301
3_0
m _j
¢-4Z.X
I
0
-3m
I
a
#I
0,i.J
Ftrl_lXV#
ZX
I
,,,,4
0
,..1"
0
_J
<
(,ISd>i) SS3_J.LS
I I
0 0
(D
© N
>-
ZW
OWn/b_
_D
0
A
>
r_
©
0
m
,.00
t4_0
cq
I
D
._IL_
3O2
I
CD
<
I
0
=f)
I I I 1 I
0 (_ 0 (D 0 0 0
oO,-.TZX
I
0
0
_j<
L.
04
©
O0
co
ZX
I m
ZZ0 XX
! ! oO
,.o,_ "_"
_ X.¢ M
0 _ I
<<
\t;s
rrl
0
co
c _
OrY.Iz
G
4
(/1
0
x_
0
0
!
D0
0
0E_
I_:_
CUt_
b_,_,_
(ISd_) SS:IaZS _n_lxv_
303
-X
I I
R-.T -,T
_J cJu _J
c5 c_ c_
0
0Z
P_O
41
Z
I
O
0x_
C? O6 6
(ISd)4) SS3N..LS l_nMIxv#
oo
D_
I
,.,o,-I
.0
,,.3"
.-3"
0,LI
rJ
O0'4"Z
I
,.o,--I
00.,.,1"
0
6if3
Q
ZX
I
0
0
O__JU
5C r
O
"-J-e,l
Q(N
(I?
NT
>-OZL_
O
U_
r/l
r/l
4_
O
(J
_J
.r_
I
CO
0
q_0
CNI
304
if3
O
ZX
,--4
-,.I"¢.4
O
-_---<
C'4Z
I
O-O
¢-4
O
_J
.<
I
C3
_J
_J
_f3
O
I
t_3
G
I 1 t
' I
QO
ZX
I
0
0
u
¢x_ ,-00 0
oO
.o
0
J
SS3WIS I,'_R i,_IXVI,N
0
0
_C
ZX
I
0
,,,.j
0
0
00"_"_"
•.U ncrr"r-j,._-
3,'-3oL.IJr" LI..I
_x--'%0
_D
>
0
,-4
_D>
I
_J
©
O
O
¢'4I
305
Z
i
,.Q
0
0
0
rl <
o 0
I0
0Z
&l
•-- ,-- C3 0 0O0 0 0 © C)
(ISd_) SS3a_LS
I !
Z ZZ
I "1 I
,-4 0_-_
0 0
_J
',3_ rj
C3 0 0 0 0 0
0 O 0 0 0 0
_n#!xvbl
¢4
['4
CO
"-- D-4"1-
>-
Z
04
"- 0
n,"!.J..
O0
C3
cj
0"_J
r.:.-,
:::;,,
(,9
r./3
0,-'4
v
4_
>
I,,-,
3_
0
I
4-i
0
0
0
0
I
_._
306
C4-Z
I
OO-.1"
",,,.1"
C
u
t I
© (D Q (:3
m,,,,-
G
e_
o
,to
I I
0 C) C) C:) C) 0 O
O (:3 C) 0 Ca ,:D 0
SS3aiS i,Nr'!l,N!xv!N
I I I
ZZ
"XX
I II
OO OO
u cc..<
<r'l,t<
£,4 ,,-- _(:3 O
O (:3
O(D
r.,el
r/l
>
_D
O
r-.-t
m
_ eN%O
I
_J
O
¢'4 o
O4-1
O
(NI
_:_
¢J
b.G(_ .,_
307
¢,q
X
I
--Q-
(J<
1.4
I
t,jl<
._._=
F't
C
0C
v
0
UCQ;
Q;
t.
0 C} C} C_ Q C) C)
LaO',.I"
X
I
,-I
0O_..
cd
0
<.I0
C}C-I
O0
tt=_
N"1"
>-q_Z
_- _.-10i,i
I I I I I
r,,,. r,,., r_0 00 0 O0 0
•-- 0
©
r/l
r/l
>
o
0
r._
r_
0
<
Chc_
I
D
lilSd_) SSqalS ian_lxvrl
308
eq c,..l
I I
C_D--CDr-- P.-
O O
U
< <
ill r
h.
C,
m
u_
,-4c..
.i..i
ix:
rn
u,j©
-K
,.IO
,.--,
q¢
I
',4-
El
c..lZ
I
Z
00
0
U<
1.4
E
I,'5 _
[]cO
X
I
,,m,-,.i
oo.,,I.
JC'M
0
U
C'4
©
,.n"u
oO ,'-m
_i3,_O '"
l--
I
"_ cooo_i---_ • W"
I
,j0
Q
< __'_.
mr- r
,-- ©
°
r,
>
ffl
c_
C
0
0
4-1
,,_
O
Cu
[.n
IZ
O
<
c_I
.+-I
SS a±S
309
I I
II_rG_1! )4_,t/ti
I I
310
<
¢D
0
r"
On,r"
t--
C-,i
rD
.r-.4×
<_
;>
0
U.-<
.,.._
I
_'iiSd>!.]SS3a±S I_IDIAII_,"¢I,°_
311
0
!
L_
0
.QCxl
0
_J
C 0 CD 0
©C_
qD
B
(Nm
- O0
0
r.
r_
N"I- ._
cJ>-
E--
W
ILl Z
c_©
c_
C@°_
I
.,_-4
(ISd>_) SSS_J±S wn_xvw
312
Z
m!
mG _
i I
_ _ C _
0
COt-',.
r'-_
rl
On,,"c"
qDl--
0
,g,,.,
I.-i
v
r-'
E'-
Z
©r,.b0
0
cg
o,_
cmI
o£
313
Q,
XII1
I
Xf
o! cC
°l "
I
CO
-.'rZ
X'
O
O
tIIIt
i[]
Z
X
f
0
0
0
<
I
III
t1
II
I
I
O
,---t
U
<
9i i ,if3 t¢3 _'3
© Q
Z'
X
I
0
0
<
i-1
m[,
04
_DZ
X
I I
OO
0 0
-4 ,,,,,4
QJU _J
O
X _
o o_
<
l.J
m
m I---_,_ ' ,
if? ,-
I
II
I
,I_p,,,lC_l
_ Z Z _-XX_
I t I
0 :' O__.' ._,J-..T
U r _0
< - < _"_--
I ©
(D
O
(13
N
)-O2 zW-..._.
O
r_
A
r.
_A
C
c_
e-.
v
I.)
4.J
t_
i
_2
-.,c
c
0
i
314
[3
u_
I
__,z__i
00.-.1"
0
,i4q_
{3
0
O
c-jZ
I
_m.-4
oo
o_a
rim
_3u<
cO
Z
I
,._
C_0
0
(U_J
Q
ZX
I
00
0
,-I
ZX
I
0C_
-,:r
o
QJ
[]
t¢3
Q
CMZX
I
0r_
0
,--t
4_
1.1tel
3
_ Zc_ X
X X I
I I 0O
O O r--
&a0 0
u U
m 'I m
C_I 0 "-C) C_
(ISd_) SS_±S V_TI_I×V_
,rI-
C)
Z_I un
I--Z--
,.0,--_ I
-,,I"
eq
0,,_ od
Z
I
Z
0 I
_a 0
Urail _:
rn
_0 C3C_
0
0
cC)
OrY¢-
q_
_r
(U
C.
m
_J
4-i
-.T0
_9
v
.-4
.-44.a
q)
_a
0
I
C-._9
©
"_Z
I
_J
,_..-I
315
Jc%1 u_Z e,__< Z
XI
Im
0_ 0
it-._
0- .i,_ - 0
1-1
_ u
<
_ u"3
_XX
I I I0
0 %1 00_
U
0C_
0O
_O
N_- "I-
>-r_ <.2__ z
0_2
OO n_b-
_3
u'%o_ c_ cq
I I I
c_._--- _-
I_ _- --_-_..T _
',.1"Z
Z' Z,iX I I :_i
i
I I _ I
."'4
--ii .i_ _1 t_
:_ _I _ __ (,,4u _IO
"'1_I_
• 0
(!Sd_) SS3_±S i_',lrT_i×vl,",l
>
r,,/)
,l,.,i
0
v
.i-I
0N
,+-t
0
<
0
I
k_0
I
316
z _[1Z
I--_-- I' Q,I
,-_ e'_ c/3 _1 ¢N ¢N___ ZZZ
< I I
Ci_r-- r,--
-0--0 _ 0 O0
(J c; _J 0
l.,i _,J .i.,i _.._
O_Cn<
-K Q
m
0>
° _,_•,-- _J
u3
,,00
M
m
©
<
0
I
m
0
_,_I
317
r_
,4"Z
1:: ,.-
_'_ O_
rv 0
,-_ 0 _
tl _JcJ
1I I
....., (3' C) (D 0 (D 0 0 0 0 0,-_.,, cD 'D © 0 © 0 0 (D 0 © 0 0 ©
I'0
N"1-
>-0Z
0
t---.
0
0
"0
cO
,--H
0O_
cU(J
._--I
0
_ u
I
co m
o mm
I
318
,-,j
0
0 C' 0 C)
i
0
_00
0 ,-4
.,-.II..i0 "_
I
0 0 0 0
0 0 0 ,:D 0
I
0 0
0 0
I,_n_l XVi_
I0
_z.4--._4 ¢-4 _
".-1"
_ N
_ U
I
rQ (',4 ,.- 00 0 0
© 0 0
'e-
I'O
/--,,.,.
ic I/_,
on, -¢..
0
0
"0
,-..I •
,-4 t._
u :>
::>
._ r_u_
I
0,-,
_d
JI
ca
.,.-I
319
DISTRIBUTION
Government Aqencies
NASA Headquarters
600 Independence Avenue, SW
Washington, DC 20546
Attn: R/R.S. ColladayRJ/C.C. Rosen
RP/G.M. Reck
RP/J.R. Facey
NASA Lewis Research Center
21000 Brookpark RoadCleveland, Ohio
Attn: J.AR.J
P.G
G.KM.J
R.E
L.J
44135
ZiemianskiShaw
Batterton
SieversHartmann
Kielb
Kiraly
J.F. GroenewegJ.C. Williams
R.W. NiedzwieckiD.W. Drier
B.A. Miller
J.J. ReinmannL.J. Bober
A.G. Powers
J.J. CoyR.E. Coltrin
C.L. Ball
E.A. Willis
L. ReidE.T. Meleason
R.D. HagerLibrary
Report Control OfficeTech. Utilization Office
D.C. MikkelsonAFSC Liaison Office
Army R & T Propulsion Lab
NASA Ames Research Center
Moffett Field, CA 94035
Attn: R.P. BenczeR.C. Smith
MS 86-I
MS 86-IMS 86-I
MS 100-5
MS 3-7
MS 23-3MS 23-3
MS 86-7MS 500-211
MS 77-6 (2 copies)MS 86-2MS 5-3
MS 77-10
MS 86-7
MS 86-4MS 86-4
MS 77-10
MS 77-6MS 86-I
MS 77-10
MS 77-10
MS 86-7 (3 copies)
MS 60-3 (2 copies)MS 60-IMS 7-3
MS 6-12 (2 copies)MS 501-3MS 77-12
MS 227-6
MS 227-6
320
NASA Langley Research Center
Hampton, VA 23665
Attn: D.G. StephensResearch Information Center
MS 462
MS 151A
NASA Scientific and Technical Information FacilityP.O. Box 8757
B.W.I. Airport, MDE 21240
Attn: Accession Dept. (20 copies)
NASA Dryden Flight Research CenterP.O. Box 273
Edwards, CA 93523
Attn: D-OP/R.S. Baron (2 copies)
Department of Defense
Washington, DC 20301Attn: R. Standahar, 3DI089 Pentagon
Wright-Patterson Air Force Base
Dayton, OH 45433Attn: AFWAL/PO/Co]. J. Radloff
AFWAL/POTA/M.F. SchmidtAFWAL/POT/H.I. Bush
ASD/EN/CoI. L.G. vanPelt
ADS/SR/K.I. Collier
Eustis Directorate
U.S. Army Air Mobility
R & D LaboratoryFort Eustis, VA 23604
Attn: J. White
J. Gomez, Jr.
Navy Department
Naval Air Systems Command
Arlington, VA 20360Attn: G. Derderian, AIR-310-E
Naval Air Propulsion Test CenterTrenton, NJ 08628
Attn: P.J. Mangione, MSPE-32R. Valeri, MSPE-34
U.S. Naval Air Test Center
Code SY-53
Patuxent River, MD 20670
Attn: E.A. Lynch
321
USAVRAD Command
P.O. Box 209
St. Louis, MO 63166Attn: R.M. Titus
Department of TransportationNASA/DOT Joint Office of Noise Abatement
Washington, DC 20590Attn: C. Foster
Federal Aviation Administration
New England Region12 New England Executive Park
Burlington, MA 01803Attn: D.P. Salvano
Enqine Manufacturers
AVCO Lycoming550 S. Main Street
Stratford, CT 06497Attn: H. Moellmann
Detroit Diesel/Allison DivisionP.O. Box 894
Indianapolis, IN 46206Attn: B. Wallace
Garrett Turbine Engine Company111 South 34th StreetP.O. Box 5217
Phoenix, AZ 85010Attn: R. Heldenbrand
Garrett Turbine Engine CompanyTorrance, CA 90509
Attn: F.E. Faulkner
General Electric Company/AEGI Neumann WayEvendale, Ohio 45215
Attn: T.F. Donohue
General Electric Company/AEG1000 Western Avenue
Lynn, MA 01910Attn: B. Weinstein
322
General Electric CompanyP.O. Box 88186
Cleveland, OH 44181Attn: M.H. Rudasill
Pratt & Whitney Aircraft Group
United Technologies Corporation
Engineering Division400 Main Street
East Hartford, CT 06108Attn: R.W. Hines
G.L. Brines
G.L. Bywaters
Pratt & Whitney Aircraft Group
United Technologies CorporationMilitary Products DivisionP.O. Box 2691
West Palm Beach, FL 33402Attn: R.E. Davis
T.R. Hampton
Pratt & Whitney Aircraft Group
United Technologies Corporation
24500 Center Ridge RoadSuite 280
Westlake, OH 44145Attn: A. Leiser
Teledyne CAE, Turbine Engines
1330 Laskey RoadToledo, OH 43612
Attn: R.H. Gaylord
Williams International
2280 West Maple Road
Walled Lake, MI 48088
Attn: R. vanNimwegenR. Horn
Airframe Manufacturers
Boeing Commercial Airplane CompanyP.O. Box 3707
Seattle, WA 98124
Attn: Dr. C.G. Hodge
F. DavenportJ. Farrell
161-04
162-25162-23
713-11713-07
323
Boeing CompanyWichita DivisionP.O. Box 7730Wichita, KS 67277
Attn: D. TarkelsonC.T. Havey
Cessna Aircraft CompanyP.O. Box 154
Wichita, KS 67201
Attn: D. Ellis Dept 178
Douglas Aircraft CompanyMcDonnell Douglas Corporation3855 Lakewood Boulevard
Long Beach, CA 90846Attn: M. K1otzsche
W. Orlowski
Gates Learjet CorporationP.O. Box 7707
Wichita, KS 67277Attn: E. Schiller
General Dynamics ConvairP.O. Box 80844
San Diego, CA 92138
Attn: S. Campbell
Grumman Aerospace CorporationSouth Oyster Bay Road
Bethpage, NY 11714Attn: N.F. Dannenhoffer
Gulfstream Aerospace CorporationP.O. Box 2206
Savannah, GA 31402-2206Attn: R. Wodkowski
Lockheed-California CompanyBurbank, CA 91502
Attn: J.F. Stroud
R. Tullis
Lockheed-Georgia Company86 S. Cobb Drive
Marietta, GA 30063Attn: W.E. Arndt
324
D/75-42
D/75-21
D/71-01, Zone 13
McDonnell Aircraft CompanyMcDonnell Douglas CorporationP.O. Box 516St. Louis, MO 63166
Attn: G. Phariss
Rockwell International
International Airport
Los Angeles DivisionLos Angeles, CA 90009
Attn: A.W. Martin
Airlines
American Airlines
Maintenance and Engineering Center
Tulsa, OK 74151Attn: C. Wellmershauser
Continental Airlines, Inc.
2929 Allen ParkwayHouston, TX 77019
Attn: Jo Arpey
Delta Airlines, Inc.
Hartsfield-Atlanta International AirportAtlanta, GA 30320
Attn: J.T. Davis
Eastern Airlines
International AirportMiami, FL 33148
Attn: E. Upton
Federal Express CorporationBox 727
Memphis, TN 38194
Attn: J. Riedmeyer
Pan American World Airways, Inc.
JFK International AirportJamaica, NY 11430
Attn: R. Valeika
Trans World Airlines
605 Third Avenue
New York, NY 10016Attn: K. Johnson
325
United Airlines
Maintenance Operations CenterSan Francisco International Airport
San Francisco, CA 94128Attn: J. Goodwine
Hamilton Standard
United Aircraft CorporationWindsor Locks, CT 06096
Attn: B. GatzenS. Ludemann
Massachusetts Institute of Technology
Department of Astronautics and Aeronautics
Cambridge, MA 02139Attn: Library
Penn State University
Department of Aerospace Engineering233 Hammond Building
University Park, PA 16802Attn: Dr. B. Lakshminarayana
Rohr CorporationP.O. Box 878
Foot and H Street
Chula Vista, CA 92012
Attn: Library
TRW, Inc.
TRW Equipment Group23555 Euclid Avenue
Cleveland, OH 44117Attn: I. Toth
University of Michigan
Gas Dynamics LaboratoriesAerospace Engineering Building
Ann Arbor, MI 48109Attn: Dr. C.W. Kaufmann
University of Tennessee Space Institute
Tullahoma, TN 37388Attn: Dr. V. Smith
MS 1-2-11MS I-2-11
326
Cessna Aircraft Company
McCauley Accessory Division3535 McCauley DriveVandalia, OH 45377
Attn: H.G. Starnes
Hartzell Propeller ProductsP.O. Box 1458
1800 Covington AvenuePiqua, OH 45356
Attn: D. Edinger
University of WashingtonSeattle, WA 98195
Attn: Dr. R. Decher
32 7
"1 Report NOCR-180869
4 Tree and Subtitle
2 Government Accession No 3 Recipient s Catalog No
5 Report DateDecember 1987
Full-Scale Technology Demonstration of a Modern Counterrotating
Unducted Fan Engine Concept - Engine Test
7 Author(s)
GE36 Design and Systems Engineering
9 Performing Orgamzatlon Name and Address
GE Aircraft Engines
One Neumann Way
Cincinnati, Ohio 45215
12 Sponsonng Agency Nameand Address
NASA Lewis Research Center
2]000 Brookpark Road
Cleveland, Ohio 44135
6 Performmg OrgamzaUon Code535-03-01
8 Performing Organization Report No
10 Work Umt No
11 Contract or Grant No
NAS3-24210
13 TypeofReportand Period Covered
Topical
14 Sponsormg Agency Code
5 Supplementary Notes
ii6 Abstract
The UDF TM (unducted fan) engine is an innovative aircraft engine concept that is based on an
ungeared, counterrotating, unducted, ultra-high-bypass turbofan configuration. This engine is
being developed by GE Aircraft Engines to provide a high thrust-to-weight ratio power plant with
exceptional fuel efficiency for subsonic aircraft application.
This report covers the successful ground testing of this engine. A test program exceeding
lO0-hours duration was completed, in which all of the major goals were achieved. The following
accomplishments were successfully demonstrated: full thrust (25,000 ib); full counterrotating
rotor speeds (1393+ rpm); low specific fuel consumption (< 0.24/Ib/hr/ib); new composite fan
blade design; counterrotation of structures, turbines, and fan blades; control system; actuation
system; and reverse thrust.
17 Key Words tSuggesleC by Authorls))
Unducted Fan (UDF_); Counterrotation;
Ultra-High-Bypass Fan; Propfan
18 DlstriOubonStatement
19 SecJrd_ ClasS_ ,Of this reporti
Unclassified
20 Security Class_f !of this page)
Unclassified
2! No of Pages
326
22 Price
I
_1 i