10:1 Pressure Ratio Single-Stage Centrifugal Compressor Program

UNCLASSIFIED

AD NUMBER

AD921137

NEW LIMITATION CHANGE

TOApproved for public release, distributionunlimited

FROMDistribution authorized to U.S. Gov't.agencies only; Test and Evaluation; APR1974. Other requests shall be referred toEustis Directorate, U.S. Army Air MobilityResearch and Development Lab., FortEustis, VA 23604.

AUTHORITY

Eustis Directorate, U.S. Army Air MobilityResearch and Development Lab. ltr dtd 18Nov 1975

THIS PAGE IS UNCLASSIFIED

i"D

USAAMRDL TECHNICAL REPORT 74-15

10:1 PRESSURE RATIOSINGLE-STAGE CENTRIFUGAL COMPRESSOR PROGRAM

ByWilliam 1. McAnUf, III

AO 194

EUSTIS DIRECTORATEU. S. ARMY AIR MOBILITY RESEARCH AND DEVELOPMENT LABORATORY

FORT EUSTIS, VIRGINIACONTRACT DAAJO2-70-C-0006

PRATT & WHITNEY AIRCRAFT DIVISIONUNITED AIRCRAFT CORPORATION

FLORIDA RESEARCH AND DEVELOPMENT CENTERWEST PALM BEACH, FLORIDA

DDSU $ Arm*v Aff 4*Utcr a

AC 2 19-i

L~§J Li

DISCLAIMERS

The findings in this report are not to be construed as an officialDepartment of the Army position unless so designated by other authorizeddocuments.

When Government drawings, specifications, or other data are used for anypurpose other than in connection with a definitely related Governmentprocurement operation, the United States Government thereby incurs noresponsibility nor any obligation whatsoever; and the fact that theGovernment may have formulated, furnished, or in any way supplied thesaid drawings, specifications, or other data is not to be regarded byimplication or otherwise as in any manner licensing the holder or anyother person or corporation, or conveying any righits or permission, to.nufacture, use, or sell any patented invention that may in any way berela-ed T'hereto.

Trade names ciced in this report do nor consticute an ufficial endorse-ment or approval of the use of such commercial hardware or software.

DISPOSITION INSTRUCTIONS

Destroy this report when no longer needed. Do not return It to theoriginac•r.

A

DEPARTMENT OF THE ARMYU. S. ARMY AIR MOBILITY RESEARCH A DEVELOPMENT LABORATORY

EUSTIS DIRECTORATEFORT EUSTIS, VIRGINIA 23604

The objective of this contractual effort was to conductthe preliminary design, detail design, fabrication, test,and evaluation of a single-stage, high-pressure-ratiocentrifugal compressor. The performance targets statedherein were derived from potential performance indicatedby tests of earlier compressors and are in no way felt to

be the ultimate in -,erformance for this type of compressor.

This report was prepared by Pratt & Whitney Aircraft

Division of United Aircraft Corporation, Florida Researchand Development Center, under the terms of ContractDAAJ02-70-C-0006. It describes the design approach,test equipment and procedures, instrumentation, and resultsof tests of the compressor. The details of the aerodynamicdesign are presented in a separate volume as Appendix IIto this report.

This report has been reviewed by technical personnel ofthis directorate. The conclusions contained herein are

concurred in by this directorate and will be consideredin any future research programs. The U.S. Army projectengineer for this effort was Mr. Robert A. Langworthy,Technology Applications Division.

Task 1G162203D14413Contract DAAJ02-70-C-0006

USAAMRDL Technical Report 74-15April 1974

10:1 PRESSURE RATIO

SINGLE-STAGE CENTRIFUGAL COMPRESSOR PROGRAM

Final Report

By

William J. McAnally, III

Prepared By

Pratt & Whitney Aircraft DivisionUnited Aircraft Corporation

Florida Research and Development CenterWest Palm Beach, Florida

for DDCEUSTIS DIRECTORATE . ...

U. 9, ARMY AIR MOBILITY RESEARCH AUG 2 1974AND DEVELOPMENT LABORATORY

FORT EUSTIS, VIRGINIA [!UD

SDistribution limited to U. S. Government agencies only,test and evaluation; April 1974. Other requests for thisdocument must be referred to the Eustis Directorate,U.S. Army Air Mobility Research an, DevelopmentLaboratory, Fort Eustis, Virginia 23604.

7 7.

SUMMARY

The objective of this program was to design, fabricate and test a 2-to-5 lb/secairflow single-stage centrifugal compressor that could be incorporated in a futureArmy advanced technology gas turbine engine. The design speed performancegoals were to exceed 75% efficiency at 10:1 pressure ratio. Since gas turbineengines for Army aircraft applications operate under part-power conditions amajority of the time, ar. off-design performance goal of 80% efficiency at 8:1 pres-sure ratio was established.

In the design of the compressor, parametric studies were conducted to select anoverall design consistent with optimum compressor performance at both perform-ance goals. These studies defined the compressor inlet corrected flow rate, im-peller inlet hub and tip radii, corrected impeller rotational speed, and inlet pre-whirl. Airflow selection and the selection of the hub radius were influenced bythe decision to design a compressor that could be used in a small turboshaft enginewith a concentric shaft front drive.

The tip radius was selected after determining the effect on axial Mach number,inducer tip relative Mach number, and inlet choke flow margin. The effect ofinlet guide vane losses, inlet shock losses, diffuser losses, and shroud frictionheating were parametrically evaluated analytically before selecting in IGV prewhirland rotor speed to provide optimum overall compressor performance. A remoteinducer designa was selected over an integral inducer-impeller cotifiguration so thatthe inducer could be designed using transonic axial-flow compressor technology.The work split between the inducer and impeller was selected so that the relativeMach number into the impeller would be subsonic. A pipe diffuser was selectedover vane island and cascade diffusers, because it has the lowest demonstratedlosses over the largest range of Mach number and because P&WAm has substantialexperience in designing and fabricating this type of diffuse%,

Demonstrated total-to-static performance was as high as 79.6"( efficiency at8. 192:1 pressure ratio and 73. 8(7 efficiency at 10. 03:1 pressure ratio. Perform-ance adjusted for increased losses from a damaged diffuser (10. 15:1 pressureratio and 75. 9"( efficiency) indicates that the basic compressor design would so'r-pass the minimum 10:1 pressure ratio program goal. A composi over-all per-formance map for the compressor Is presented It' Figure 1. Evan1ttion of com-ponent performance data revealed that excessive losses occurred in the inducerabove 95C of design speed and that a redesign of this component could produce anadditional perfoiviance improvement at 10:1 prossure ratio.

-C4

Mt

7 777

so

70

10

S7

MI4I( WEED100

uutA w SITTINIG

FO

'TWICALb. CIERATIN

80% $EED

Nding 0% SEED

0 1.4 ~is 0: Ov talPo 1.5~c LOp

....................................................

TABLE OF CONTENTS

SUMMARY ...... .. *. . ..... . ... . •. • •.. o.... •* .. iii

LIST OF ILLUSTRATIONS ...... vi

LIST OF TABLES .. . . . . ................ .* * xxv

S• LIST OF SYMBOLS ... o ............... . .. ,., . ... * xxxii

INTRODUCTION .... ...... .. ..... 1

DESIGN APPROACH .....* . .... ..... • • , • . ..... 3

A orodynninic Design 9******** *..*.*..**...*******.3

Mechanical ..si ...... 6Mechanical lRedesigm ................ . .... 16

TEST EQUIPMENT .............. ..... 21

Compressor Test Rig .... ... ... . * . 21Test Facilities 21

SINSTRUMENTATION ........ 28

SPROCEDURES .... . ,.. . ... .......... . ***

'Test Pocedures ..... ...... .... ... ... .. ... 44Data Reduction Proceduivs . ........ ,.9. .... ... .. 9 •.Validation of Test Data. ., .* 0 9. .. *. . * . ... , • • , * * • *

RESULT$ AND DIS( $SON .. •... . .. .. 4. •. . .• • • .• . ..... . . . ..4

Overalt Perform-anve . .. .. . . . . . 64 *********

C o p o n 9 . . . **l* * . **.* .* . . . . . .9 1

C NL ONS ...... ......... . .l.P..... 130

RECOMMENDATIONS .................. 131

A, PPENDIXIN - COvn- ft•fomm TAW tio . .... ,. t........ 3

APPENDIX 1 - A@ro4rmic Ds igii (a#,sifted CUIdeatIo -

,- ,ttt . .................. .i•II Lt" • LL l*99*i *99* 99 ll*16999999999999* 99 99 999*99909* 9

i iV

LIST OF ILLUSTRIATIONS

Figure haie-

1 Comiposite 10:1 overall Performance Map 0 ......... iv

2 10:1 Pressure Ratio Centrifugal CompressorStag.... . .. ... * ... *.***... .. .. 3

3 10.1 Pressure Ratio Centrifugal Impeller andRemnote Inducer.*. so .... o... so @*o*so**s oso**o 4

4 Initial Design of 10.0:1 Pressure Ratio CentrifugalCompressor and P&WAT" Drive Turbine .. . .. 7

5 nlucer Vibration Analysis.. .... .. ......... 00 00 8

6 Impeller andld ucrAtachettaohnient..... ........ 9

7 Tip Clearance ProbeIntalnstiom . ......... 10

8 Initial Compressor Drive and HecaringSupport D~esign . ........... * ... ** 1

9 Blearing Designs . .. . ... . ................. .1

10 Compressor and Turbine Rotor Assumbls o o .. o.o..o... 14

it In~itial Rotor Coaftguration Ptredicted Vrout

12 Initial Rotor Coutiguratioa Predicted Rear

13 Predite~d Hall Uearing MiAIM= Thrust Load

0 ~Redeigned 10:1 Pft*SUr@ Ratio Ovtrifugal

IT tde1idI -Rntwu.4 biy Predice n

thp ig-Lw' !lutor Awb*lAjy Prc~lc"o4 RearBearing L~adt* Spood . 04 see. bes -4* ........... 20D

. --- ,

LIST OF ILLU.STRATIONS (ContlitndFigure Pago

19 Front Bearing Compartment Carbon Face Seal * ... 20

20 Compressor Test Rig Nonrotating Components *. 22

21 Impoller Spin Tooling. *. .......... ......... 22

22 Impeller Installation in Diffuser Case ....... ,.,,,, 23

23 Redesigned Rotor Assembly Compononts o . ° o 23

24 Assembled and lFully histrunwnted Compressorlig *.*... .................. ,...o........ 24

P)&WVU•• VFRC fligh-Speed Compressor Test Stand .... 25

26 Purticle Separato,' With Integral Fast-Acting%3wtoff Valve. ,.. 0.,. o,, 0,,, 0 o ° ° o,9 26

27 Coimwessor Rig Installed in H-2 Test Facility ... 27

28 Compressor Instrumentation Station Locations ....... 28

29 Inlet Otifice hstallatioll "29

30 Ditffuser•Wxt To t Prctv Mikes . s............. 30

31 Location of Difftse" P~t Total PWsvre t t....... 31

32 Placoment of Total Pwssur, PtNvbs inDitfte Erxit PlaU to ......... ,.,........ 32

33 Collectr rttsnatatioa LocQatto .............. 34

a4 Amdal Locations of Inlt W and Int'rn nain . 36

353 Constrtwtiou of Travrse Cobra PtvbW S........ 37#06*r

36 LacationahnoftWUvr $r4~d StAtie Ptv~wv Taps... 37

37 taptllor Tip Static ePwo tv P g c............ 30

3$ mpeU .tr.Ito--arONw sullatta ....... ... 39

Cou~uvdw40

dli

C!i i i i i i i

LIST OF ILLUSTRATIONS (Continued)

Figure Pa_

40 Location of Diffuser Static Pressure Instrumentation... 42

41 Iigh-Frequency Response Kulites......... ...... 43

42 Maximum Front Bearing I Acceleration,Build No. 0................45

• 43 Impeller Ti)) Axial Shroud Clearance, Build No. 2..... 46

44 Clearance Probe Moasurements, Build No. 3........ 46

45 Shroud Clearance Measurments, Build No. 6 ....... 47

46 Inlet Wide Vane Exit Flow Correspondence,-- "Build No* G#ses 00 0 fe060 009 61

47 Inducer Exit hitegrated low Corres1o-oce,

Blldoe6

4S Impeller E'xit Integrated Flow, Corresponon±ev,!1 i•Build NO* 6.. • . • . . .. . . . • . • •

49 Over-all Nrforwance, Build No. 2 Shbadown

so high-Slped Centrifugal Comessr FlowC•qtartvt@4itk?8 .... • ... ... *,......, •........ 6%

I D iCUser Loss C0raetortstte. Build D4. 2

"5 ••Diffuser Thoat Blockago Charact@risti,

Build N. 2 Shkedowt Test .. *.*... *. .. . . ...

33 QXVU @rVl~ra0V. u~d No. 3.lQ-dcg 1V ....... 67

54 OQMU Klrfafn l tti.lld No.3, O- v ".......S

35 DW14 Nc. 3COvnll NCConmaw@, 20-&og 10 ....... 69

$5 ~ W @ Ctnuatn Ufic ruetsesem to ............ 74

ST Ca~kiwo C-ftvmu ve"u" ii z

-iITST OF I1l.UBTRATJONS (Continec

MEMO PAge

-• 59 JG 'thVTurnlng vs Settitg, Build No. 6 . 75.., ... 15

i•W' JOGV Exit k ttvalrl Distribution, 10-dog JOV......... 75

S6 1=1i4 &a-itf n Total Pressure Loss Characteristics..... 76

* 62 IGV Clrcumferroat'Ai Traverse, Build No. 6,100% Speod, 10eg IGV, 50% Span ............... 77

63 IGV Exit Radial Traverse, Build No. 6, 100%Speed, 10-dogI0V, Near7Stll................ 78

64 Build No. 6, Inlet Guide Vane Losses at

Approxlmatly 0. 46 M ach onr................ 79

65 Inducer Inlet Conditions, 100% Speed. 10-deg IGV..... 81

866 Inucer Inlet Conditions, 101% Speec4 -4-deg ITVO..... 82

67 Inducer Inlet Conditions, Build Ne. 3, 95% Speed,

68 taduer Performance, Build No. ............... 8$4

09 butuler Fixit Traverse, Build No. 6, 101% *eed,-4-deg IOV, Wide Open D iscarge ........... 85

70 niducer Exit Traverse. Build No. 6, 100% Speed.

71 IMdtWer Exit Travers€. Build No. 6, q&% *$mlI 0-dog I($Y, Nwwr tall ...................... 8

72 Indcer Ls ............................ *9

T3 hnpelbr Inlet Coatttos, 100% tI•stp Seed,

74 de-or Ext Air Angl Profit#, 0WN*Igc*w 10-4og tuv. ,, .,:,,. ...... 90

75 Ua~pello? Met Condltmos, 1011 Speedq-4-W ....... 0.e •.•.GB•. •.ses ......... s1

lad4 eet stit Air Ale PM'@@Profile, II% Si d,

ix

LIST OF I LLUSTUATIONS (Continued)

Figure Itg

77 W.V Exit Static h'vesstsre Profile, ioo% Speed,

78 Inducer Exit StAti I'rvssuru Profile, 100%, Speed,

79 tnxpolker Inlet CondItions, 100% Speed, 10-dog94

so 8 Inducer Extk' Air Angle Profile, 1900%t Speed,10-tcj by . 0 S CS C00 0 0 50 0@ 0* 0 0 0 05 00

81 Choke Point Flow as a Function of HotorSpeed, Build No. 3 .. . ....... o*o. 95

aZ Build No. 6j Choke Point Flow as a Function

sJ Build No, 6j Wimpeler Performa-ace DerivedFrom 1'ruvvLs, UWta. .. *.....................$

8-4 lziduceri-luinlk~lr Performanace Derived FromTraverse WData . .. . . *.. *.*.*** *********** 9$

$5 BlUild \o, 3 hwpdle1r Pe1frfiMzgwe MatfpPmrivokd From lintornAl F low Anlysis *.......99

M6WU'd N,-. 6 inue-upi oPrformanceMaptDrived From laternal Auat yss......... 1

s7 Build No. tnjpejl~rEfituynttw

88 Build NO. 6 idcrttle ~finy

69 Effect of Perowhidl eoi tNcer~uer r-foNwu-@ or10- 1 IN**v

90 ~Effect of Prowbidl ogti0c-aidi'-(otzagceo Niear 9,-1 P4eS*ucr3 ....... ..ea ...... 1 00 104I

9?a-1o tzj1exiSt sip8utor.pb .W g..... 6 , 10.5

t~iWý- Exi N~ttmrjv w~ca . e

LIST OF ILLUSTRATIONS (Continued

93 Impeller Exit Traverse, Build No. 6, 101% Speed,-4-degIGV, Near Stall.. to ........ ***6*****.**10

94 Impeller Exit Traverse, Build No. 3, 95% Speed,10-degIGV, Ner~arl tal............ 107

*95 Impeller Exit Traverse, Build No. 6, 95%o Speed.,10-degI1GV, Near Stall. .... . . ... 108

96 Normalized Radial Velocity Distribution, BuildNo.* 6, 1017oSpeed, -4-deg IGV ...... .. 109

97 Normalized Radial Velocity Distribution,Build No. 3,95% Speed, 10-de GIGV . . . ..... 109

98 Normalized Radial Velocity Distribution,Build No. 6, 95% N/V#t Design, lO-deg IGV . .... 110

99 Discharge Velocity Ratio Comparisons lit..... 1

100 Stall Transient Data at 1 Scan/sgec and Continuous

101 Static Pressure Variation Along ImpellerSheoud, 95% Speed, Near Stall, 10-deg IGV .... 113

102 Diffuser Static Pressure Profile: Build No. 3,95% Speed, 10-deg IGV . ...... ... . 114

103 Diffuser Static Pressure Profile: Build No. 3.100%loSpeed, 10-deg IGV, Near Stall .......... 115

104 Diffuser Gapwvise Static Pressure DistributionsAlong the Tangency Radius: Build No. 3,1 0-dog IGV Setttng so to .00 *#of ..... .. aa* 115I1.05 Diffuser Shroud Static Pressure Contours:Build No. 3, 100% Speed, 10-dog IGV... .. .. .... .. 116

106 Diffuser Throat Static Pressure Distribution:Build No. 3, 95% Speed, 1O-deg IGV ........ 117

4 ~~~1,07 Dam4aged Diffuser Pipe Leading Edges . . ... 118

108 Typical Undamaged Diffuser Pipe Leading Edge .... 119

LIST OF ILLUSTRATIONS (Contitio~d)

F ig'ure Pg

109 Post-Test Condition of Diffuser ..... * .... 119

110 Diffuser Static Pressure Profile: Build No. 0,95% Speed, 10-deg IG VNea eSall al......#. 120

ill IDiffuser Static Pressure Profile: Build No. 6,100% Speed, 10-deg IGV, Near Stall ... @b*.. 120

101% Speedl, -4-deg IGV...... ..... 121

113 Diffuser Exit Mach Number Profile: Build No. 3,95% Speed, N/Vt 10-clog IGV, Near Stall ..... ..... 122

114 Diffuser E xit Mach Number Profile: Build No. 6,.95% Speed, l0-deg IGV, Near Stall........ 122

115 Diffuiser Exit Mach Number Profile: Build No. 6,100% Speed, 10-deg IGV, Near Stall ........ 123

116 Diffuser Exit Mach Number Profile: Build No. 6S101% Speed, -4-dog IGV, Near Stall *........ .... 123

117 Diffuiser Exit Mach Number Profile: Build No. 6,101% Speed, -4-dog IGV, Wide Open Discharge .... 124

11b Diffuser Losses vs Corrected Weight Flow,

119 Diffuser Losses vs Impeller Exit MachNumber, Near Stall .......... * . 126

;20 D~iffuser Dum~pLosses Na eSalll......... 126

121 Diffuser Static Pressure Rise Coefficient vsCorrected Weight Flow ....... ..... 126

122 High-Frequency Response Data at 70% Speed4 ...,. 128 I123 High-Frequency Response Data at 78% Speed* .... 129

124 IGV Exit Radial Traverse, Build No. 6, 101% Speed$0-degIGV, NearStall. ........ * too**. .. 140

125 IGV Circumferential Tr~averse, Build No. 6,I

101% Speed, 0-deg IGV,10%Span.... .............. 141

xil


"Figure Page

126 IGV Circumferential Traverse, Build No. 6,101% Speed, 0-deg IGV, 30% Span .............. 142

127 IGV Circumferential Traverse, Build No. 6,101%Speed, 0-deg IGV, 50% Span................. 143

128 IGV Circumferential Traverse, Build No. 6,A 101% Speed, 0-deg IGV, 70% Span .............. 144

129 IGV Circumferential Traverse, Build No. 6,101% Speed, 0-deg IGV, 90% Span................ 145

130 IGV Exit Radial Traverse, Build No. 6,101% Sped, -4-dog IGV .......... 146

131 IGV Circumferential Traverse, Build No. 6,101% Speed, -4-degIGV, 10%Span..,............ 147

132 IGV Circumferential Traverse, Build No. 6,10I% Speed, -4-deg IGV, 30/ n ............. 148

133 IGV Circumferential Traverse, Build No. 6,101% Speed, -4-dog iGV, 50% Span............,... 149

134 IGV Circumferential Traverse, Build No. 6,101% Speed, -4-degIGV, 70%Span.,,,,,.,,,,,,,, 150

135 IGV Circumferential Traverse, Build No. 6,101'";ýpeed, -4-degIGV, 90%Span............... 151

136 IGV Exit Radial Traverse, Build No. 6,101% Speed, -5-dog IGV.. .. ,...,, ,,,,.,,,,,, 152

137 IGV Exit Radial Traverse, Build No. 6,100% Speed, 10-degIGV, NearStall #**a#***toot153

138 IGV Circumferential Traverse, Build No. 6,100% Speed, 10-deg IGV, 10% Span,, . .. ,, , 154

139 IGV Circumferential Traverse, Build No. 6,100% Speed, 10-deg lGV, 30%%Spn a........,. 155

140 IGV Circumferential Traverse, Build No. 6,100% Speed, 10-deg IGV, 50% Span ............. 156

141 IGV Circumferential Traverse, Build No. 6,1007o Speed, 10-deg lGV, 70% Span ............. 157

xiii

•..

¢L L.

MIFF


Figure Page

142 IGV Circumferential Traverse, Build No. 6,100%o Speed, 10-degI1GV, 90%o Span .. ........ *. .... 158

143 IGV Circumferential Traverse, Build No. 3,70% Speed, 10-deg IGV, 10% Spa n .......... 159

144 IGV Circumferential Traverse, Build No. 3,707oSpeed, 10-deg IGV, 30% Span ................ 160

145 IGV Circumferential Traverse, Build No. 3,70% Speed, 10-deg IGV, 50% Span...... .. . 161

146 IGV Circumferential Traverse, Build No. 3,70%/oSpeed, 10-degI1GV, 70% Span ................ 162

147 IGV Circumferential Traverse, Build No. 3,70% Speed, 10-deg IGV, 90% Span............ 163

1'1") IGV Exit Radial Traverse, Build No. 6,707oSpeed, 10-degI1GV, Near Stall ........... 164

1441 IGV Circumferential Traverse, Build No. 6,70% Spee-1, 10-degI1GV, 10% Span.. . .......... 165

150 IGV Circumferential Traverse, Build No. 6,707( Speed, lO-deg IGV, 30%b Span ... ... ......... 166

11 IGV7Crufeeta Traverse, Build No. 6,

70% Speed, 10-dogITGV, 507oSpann............... 167

152 IGV Circum~erential Traverse, Build No. 6,70% Speed, 10-deg GlG"70,7,SSan.. ... ,..... 168

153 Inducer Exit Traverse, Build No. 6, 101% Speed,5-iegIGV, Wide O pe ishren shag....... 169

154 Inducer Exit Traverse, Build No. 6, 101% Speed,5-deg IGV, Near Stn-11. .G *9. 0 .. .. 60 64 *00 . ..... 170

155 Inducur Exit Traverse, Build No. 6, 101% Speed,0-deg IGV, Wide Open Discharge ......... 171 !

156 Inducer Exit Traverse, Build No. 6, 101% Speed,0-deg IGV, Near Stall... . . ... ... .. ..... 172

157 Inducer Exit Travpr~ie, Build No. 6, 101% Speed,-4-deg IGV, Wide Open Dischaarge........ 173

x Lv

7~ r0 !

*1 LIST OF ILLUSTRATIONS (Continued)

Figure Page

158 Inducer Exit Traverse, Build No. 6, 101% Speed,N ~~~~-4-deg IGV, Near Stall . . .. .. .......... ..... 174 t

159 Inducer Exit With Coolant, Build No. 6, 101%o Speed,-4-deg IGV, NearStall . .............. 175

£160 Inducer Exit Traverse, Build No. 6, 101%0 Speed,

161 Iduer Eit Trvre Bul No167006ped

4 1 162 Inducer Exit Traverse, Build No. 6, 95%~ Speed,15-deg IGV, WieOenDshrg........ . 178

162 Inducer Exit Traverse, Build No. 6, 95% Speed,15-deg IGV, Niearpeallischarge. .. .. .... 179

164 Inducer Exit Traverse, Build No. 3, 95%o Speed, 4

10-deg IGV, Wide Open Discharge. .... .. ... 179

165 Inducer Exit Traverse, Build No. a, 95% Speed,10-degIGV, Near Stall....*. 0......... .... 180

166 Inducer Exit Traverse, Build No. 3, 95% Speed,10-degIGV, BelowNearStall ........... 180

167 Inducer Exit Traverse, Build No. 6, 95% Speed,lO-deg IGV, Wide Open Discharge... .......... 181

-168 Inducer Exit Traverse, Build No. 6, 95% Speed,

10-degIGV, Near Stall............. 182169 Inducer Exit Traverse, Build No. 6, 95% Speed,

4 ~~10-deg IGV, Wide Open Dischhrerge....... 183

170 Inducer Exit Traverse, Build No. 6, 95% Speed,lO-deg IGy, Near Stall . . . . . ................ 184

171 Inducer Exit Traverse, Build No. 6, 85% Speed,30-degIGV, Near Stall............. 185

172 Inducer Exit Traverse, Build No. 6, 85% Speed,20-degIGV, Near Stall....... ..... .... 186

173 Inducer Exit Traverse, Build No. 6, 70% Speed,30-degIGV, Near Stall . ........... 187

xv

ILIST OF ILLUSTRATIONS (Continued)

Figure Page

174 Inducer Exit Traverse, Build No. 6, 70% Speed,20-degIGV, Near Stall.............*... 188

175 Impeller Exit Traverse, Build No. 6, 101%o Speed,5-deg IGV, Near Stall ........................ 189

'176 Impeller Exit Traverse, Build No. 6, 101%o Speed,0-deg IGV, Near Stall,,,,,,, ... ... ... 190

177 Impeller Exit Traverse, Build No. 6, 101%o Speed,

-4-deg IGV, Near Stall . .. . . ................. 191

178 Impeller Exit Traverse With Flange Coolant,Build No. 6, 101%0 Speed, -4-deg IGV .. . .. . ... . 192

179 Impeller Exit Traverse, Build No. 6, *l01%oSpeed, -5-degIGV.#,, ....... 1 #*....... 193

180 Impeller Exit Traverse, Build No. 6, 101%6 Speed,

181 Impeller Exit Traverse, Build No. 6, 95% Speed,

95% Seed lO-de . GV.WdeOpe.Dscarg.... 195

183 Impeller Exit Traverse, Build No. 3,95% Speed,.10-deg IG VNid OenaDscharge. ..... 196

184 Impeller Exit Traverse, Build No. 3,95% Speed, lO-deg IGV,Beo Near Stall........... 1976

185 Impeller Exit Traverse, Build No. 6,95% Speed, lO-deg IGV, Welo Openr DSthal ge...... 198

186 Impeller Exit Traverse, Build No. 6,95% Speed, 10-deg IG V..Wide.Open.Discharge . 199

187 Impeller Exit Traverse, Build No. 6,95%Speed, 10-degIGV, Near, tall...............190

188 Impeller Exit Traverse, Build No. 6, 8%Sed

189 Impeller Exit Traverse, Build No. 6, 85% Speed,20-degIGV, N ea~arlal........... 202

xvi

V.


Figure

190 Impeller Exit Traverse, Build No. 3,70% Speed, 20-deg IGV, Wide Open Discharge ....... 203

191 Impeller Exit Traverse, Build No. 3,

70% Speed, 20-deg IGV, Near Stall ............... 204

192 Impeller Exit Traverse, Build No. 3,70% Speed, 20-dog IGV, Below Near Stall.......... 205

193 Impeller Exit Traverse, Build No. 3,70% Speed, 10-dog IGV, Wide Open Discharge........ 206

194 Impeller Exit Traverse, Build No. 3,70% Speed, 10-deg IGV, Near Stall............... 207

195 Impeller Exit Traverse, Build No. 3,70% Speed, 10-deg IGV, Below Near Stall........... 208

196 Impeller Exit Traverse, Build No. 3,70% Speed, 0-dog IGV, Wide Open Discharge ........ 209

197 Impeller Exit Traverse, Build No. 3,70% Speed, 0-deg IGV, Near Stall.., ......... 210

198 Impeller Exit Traverse, Build No. 3,70% Speed, 0-dog IGV, Below Near Stall .,...,. 211

199 Impeller Exit Traverse, Build No. 6,70% Speed , 30-deg IGV ... 212

200 Impeller Exit Traverse, Build No. 6,70%Speed , 20-dog-GV IGV...............,... 213

201 Impeller Exit Traverse, Build No. 3,30% Speed, 20-dog WGV, Wide Open Discharge,........ 214

202 Impeller Exit Traverse, Build No. 3,30% Speed, 20-dog IGV, Neat- Stall ............ , 215

203 Impeller Exit Traverse, Build No. 3,30% Speed, 20-dog IGV, Below Near Stall, 216

204 Impeller Exit Traverse, Build No. 3,30% Speed, 10-dog IGV, Wide Open Dsharge........ 217

205 Impeller Exit Traverse, Build No. 3,"30% Speed, 10-dog IGV, Near Stall,,.,,,,,.,... 218

xvii


Figure___

206 Impeller Exit Travenre, Build No. 3, 30c Speed,10-deg IGV, Bolow Near Stall ................. 219

207 Impeller Exit Traverse, Build No. 3, 30% Speed,10-deg IGV, Below Near Stall .................. 220

208 Impeller Exit Traverse, Niý.ild No. 3, 30-7 Speed,0-deg IGV, Wide Open Discharge ....... 221

209 Impeller Exit Traverse, Build No. 3, 3011 Speed,0-deg IGV, Near StaU 222

210 Impeller Exit Traverse, Build No. 3, 309, Speed,0 do IG lklowNear Stall220-deg IGV, Belot .......... .. .... 223

211 Impeller Exit TemperatureoTraverse, Build No. 6;,101, Speed. 5-dog IGV 224

212 Impeller Exit Temperature Traverse, Build No. 6,1011"f Speed, 0-deg IG...... . 225

213 Impeller Exit T''maverse With Coolatit, Build No. 6,101V Speed, -4-dvg IGV .... ....... ,.. . .o. .. . . 226

214 Impeller Exit Traverse With Coolant, Build No. 6,1011, Speed, -4-deg IGV 227

215 Impeller Exit Temperature Traver••o, Build No. 6.,101'. Speed, i deg1V ... 2

216 Impeller Exit Temperatuo Travorso, Build No. 6,94 V Spvd, 15-dogWW O ..... ,. ....... 129

217 Itrpoller Exit Temperaturo Traverse, Build No. 6,94. W' Speed, 10-deg |OV 230

219 ImWller Exit TomperaturpTt'avortsv, Bild| Na. 6,

95" Specd, 10-dog ICIV. .W.ide Oh . . . ,. * .31

219 Impeller E-xit Tempe raturvo Tr'avertiv, Bi~ld No. 6.95" SK00 10- CtV..... .. .,,....... •

220 Static Press'ure Pro[lto Alowg Flow Path. Build No. 3.1W00 Sjwed. 10-,dog IGV. Nour Stall ,..31.... t

xviii

L•1

LIST OF I I.LUSTRA TIONS (Continmed)

Figire

i21 2 Static Pressure Profile Along Vlow Path, Build No. 3.%%....eed, 20-dog WV, Near Stall............,. 316

222 Static Pressure Profile Along Flow Path, Build No. 3,9511 Speed, 10-deg IGV, Wide Open Discharge ....... 317

223 Static Pressure Profile Along Flow Path, Build No. 3,ar St, 31795 Speed, 10-dg IGV, Noatll ............... 317

224 Static Pressure Profile Along Flow Path, Build No. 3,95(1 Speed, 0-dog WIV, Near Stall ...... . ...... ... 318

225 Static Pressure Profile Along Flow Path, Build No. 3,70• Sp'eed, 10-dog IGV, Near Stall . .......... .1a

226 Static Pressure Profile Along Flow Path, Build No. 6,101UY Speed, -4-dog IGV, Wide Opea Discharge....... 319

227 Static Pressure Profile Along Flow Path, Build No. 6,10W Speed, l0-deg tGV, Near Stall ........ ..... 319

228 Static Pressure Profile Along Flow Path, Build No. 6,10W' Speed, -4-d&g 1V, Near Stall ............. 320

229 Static Pressure Profile Along Flow Path, Build No. G,ii ~M';" spoed 10-dog 1GV$, Nva Stw! .. 0.....320

230 Static Pressure Variation Along Impeller Shroud,Ouild No. 3. 1001c Sped. 10-dog 1QV, Near Stall .. .321

231 Static Pressre Variation Along impeoller Shroud,Build No. 3. 95f1 Speed, 20-4dg WV, Near Stoll ...... 321

232, Static Pir sure Variation Along Inmpller Shroud.

Build No. 3, 9%% SpeWd, 10-d4g IV, Wide CAW"•eag .... ................. .322

23. Static Pressure Variation Aloag Imtovslr Shroud,"ld Vk. 3, 958t Speed, I0-dog |IV, Near Stall...... 32.2

2,34 Static Prosaure Vatiatsea Along ImipelleprShroud,Wuild No. 3. WE Speed. 0-&Cgl&W, Ntar Stall.,. 323

S* 23 Static Pressuro Vat;atntt Along tmipelle Shor ,d,NUald Vo. 3, Sp% d. 10-dog MV, Nea Stall....... 323

• -71

LIST OF IILU,.TRATIONS (Continued)

Vkunre faIS Static Pressure Variation Along Impeller Shroud,

Build No. 6, 101(l Speed, -4-&.g MGV, Wide OpenDischarg. .. .324

237 Static Pressure Variation Along Impeller Shroud,Build No. 6, 101V Speed, -4-dog IGV, Near Stall ..... 324

238 Static Pressure Variation Along Impeller Shroud,Build No. 6, 00YtX Speed, t0-deg 1IV, Near Stall ..... 325

239 Static PIves.ure Variation Along Impeller Sbroud,Build No. 6, 95" Speed, 10-dog IGV, Near Stall 325

240 Diffuser Static Pressure Profile, Build No. 3,"95% Speed, Near Stall 326

241 Diffuser Static Pressure Profile, Build No. 3,10-degtIV, NearStall. ...................... 326

242 lHigh-Prequency Response Data, 707( Speed,Steady-State, PIDKI. 228

243 fligh-Frequency Response Data, 10• Speed.Steady-state. PS•DVKI ... . . . . . 329

244 l|tigh-rrequenry Rsponise Data, 7017 Spee.S - --ea y-t ate, P '$DVI2 .. .... . ..... .. . ....... 330

245 Hfigh--wequency Response Data, 7Q• Speed.

S2461 Hligh-Frequeny Response Data, TGLI Speed,$t rSitt,. Tim e -0 8, PT ID KI ............. 332

S24 Iligh-FreqU0ncy ReSpon.e Data. W7 Spe•e.

-2 T Itigh-Frequten_, R•spoe Vata, 7CV Speed,

Stall Trati@. Thu 434ttDKl ............. ,. 33-4

249 it-Veutwy Balio-se Data, TO- SWped,.llT t.im@ t, gTiti 43, PT ID I ........ G....,,s 33

LIST OF tL!.USTRATnONS -Wentinued

Vicvre Pagie

251 11ghb-Frequency Response Data, 70% Speed,Stalled, Time 440, lIf'IDK1 ... .. ,............... 337

252 111gh-Frequency Response Data, 7W( Speed,Stall Transient, Time 406, PSDVKI .............. 338

253 1tigh-Frequency Response Data, 7024 Speed,Stall Transient, Time 426, PSDVKI ..... 339

254 Iigh-Frequency Response Data, 702- Speed,Stall Transient, Time 434, PSDVKI .... ...... .... 340

255 11igh-Frequency Response Data, 701. Speed,Stall Transient, Time 436, PSDVKI ... ...... .. ... 341

256 1ligh-Frequency Response Data, 7014 Speed,Stalled, Time 43S, PSDVKI ...... ... ........... 342

257 1iigh-Frequency Response Data, 701 Speed,Stalled, Time 440, PSDVKI ... ........... ..... 343

258 tIigh-Frequtwncy Response Data, 701 Speed,Stall Tratsieot. Time 406, PSDVK2 .............. 344

M59 High.-Frequency Response Data, 70C Speed,Stall Transient,Tir•e 426, PS$VK2 .. 0..... . 345

26t1igh-Frequency Repolase Data, T70 Spe"e,RaU TratWs•t, Time 434, PSDVK .. 346

241 Uighi-Frequenvt' 1Rhtpaae bWt, Uf~ P e,Stall Tranist,P "Tm 436,P'D7,N ......... 347

2•2 Htigh-P•rquetwy Response Data, 70%T- Spevd,Stalld. Tim@43S P4DV . ... 49ftbf....9Sftftfttq9 a 344

2fi3 Hligh- Froquetiwy Respon~se Vata, TO% Speed,$944cd Tiwe 440, MSOVK2 .,....... . 49

W84 tlgh-Ftteqtenvy Responseo Data, 7&?, itwed,

A)A

$f5 Hg-ta q--e Tva4#ioa Titao 42% SEI ........

LIST OF ILLUSTRtATIONS (Continued)

Fku re ac

2"6 I1igh- Frequecny Responise Data, 70% Speed,

stall Transient, Time 434, PSDEEJK .... ... 352w

267 Hfigh-Prequency Response Jatoi, 7C,( Speed, .93* 5

268 IlighvFreqttetty Rtesponse Data, 70~1- speed,Stulled, Time 43S, PvSDEK1 .. . . .. ........ 0*4 0..0 3,54

26$ High-Frequency Ilesponse Data, 70P/%. Speed,Stalled, Timeo440, PISIWN.i. .. .. . .. . .. .. . . .. ... 355

270 hItgh-Frequency htespwsons Data, 78',; Speed,Steady-state, PIDII K I.. ............ 356$

271 Htigh-Froquveny Rtesponse Data, 7W-7 Spoed,Stvady-State, PSI)VNI . . . . . . 357 ....... s

272) ltiglW'Ftequtttty Rtesponse tVata, 78'1 Speed.Steady-State, P$*DVK2 . . . . . . . . . . . . . . . . . . . . . . . 358

273 hghFquvy eposData, 78KL4 Speed,

steady-State, PSU)ENI . . . . . . . . . . . . . . . . . . . . ... 35f)

274 ltigh-Frequetwy Respaons Da ta, 7C. Speed,Stall Trasienvt, Timew533, PTID)KL . . . . ,...*.. 300

275 High- Y-req(uenvy Hespons Jaa CSedStall Transietf, Timo 4 5 3.PTII'KII~.......

76 High-Frequencvy Iesonsew Data.7, $S peed,

27T HIIgh-Fre~qunc~y Respous Data. 7tK Speodo,Stall Tvanietu, .Time 563 TIPMI.......-. 33

274 fligh-Fr"sjucty Resonse Da'tk. tSe

34, A

s tal4'. i're U-6y Mt OMs aa I pe,

$taul Tatitentu, Ti-nw =3,PDiK , I.N.~.. 344

I..ST Or lttASThATIONS (Continued)

281 t~i~-Vequency Hleponse Data, 7C, Spoed,Stall Tn siwt, Tim: 53, Pv$DVI1 .... *.*...... 307

2-82 tilgh-Frequenwy Rtesponse Data, 78%4W Speed,Stall Tnmsient, Time 5611, P-*SDIVNI ............. 308

*283 fligh-Frequency ltcnpose- Oita, 7,Y7 Spved,Stall Tazusient, Time 5413, PSt)DVNI....... 3619

284 It~gh-Prequency Res~pounse Data, 781-7 Speed,Stalle. Timeo 56, 'SIM I .. . . . .. . . ...... 370

U53 Hgh-Frequoncyv Ilosponso Data, 78W Speed,Stalled, Timve 5417, PSDVNI............. ... .. 371

2$6 tht g-k'requenvy tt( poi'we Dat, 73 SpedStall Transient, Time. 533, P.SDV2...... 372

2$7 hIigli-1'requtnwy tte,4powsv Datu, 7W:, Speed,Stall Ttaset Tipmo 553, li$1AIr,)tV2.......

268 High-Frequency Rosponse Dlata, IC Speed,Stall Transient. Time 5611, PSDVKZ . .......... 374

2A 9 High-Frequenqcy ltspwris Data, 781, sveed.Stall Traqskn:. Time5461. P$VK2 ... . .. .. . .. . .. 375

290 llg-F'roquewiy ttespotiso Data, 748% SpeedStalled. Tiaxe SGS, PSDVK2 .. .. . .. .. .. . .. .. . .. 376

291 Hhtgh'-Frvqttoacy Uespouse Data. 78% $pood.tidTinw 5917, P$'IW~K2. ........ 7'

2$2 High-Froequeuy ttespoftse Data. 7914 Speed,$tatllTrnaientu.Tiue5.&U,PIl( ......... M

2-03 High- F mquieaey ttupoanto Data, 41i Spoed,

K * h~h-F~t~esw ttesnase ata, tw7LSedStall Traaimu. TttaS%1, P$VtW4 U.......

*~ ~ ~ ~ u 2STlg-rtqtwaey Tttcoas U3 t, 78$p~

IS" OFx IIlSTIIAfONS•WContlnuW

otu re Page

296 liigh-Vrequency Itesvonsie Data, 78Q(1 Speed,Stall Transient, Time 565, 1'SDEUI .. ..... ....... 382

297 lIigh-requcency Response Data, 78% Speed,Stalled, Tite 567, PSEKi . 383

298 SIoll Teansioeit Data, M0IT Speed, 5-dog Mv,,:r I~~~~~SPS Hatot ........ ..... . . . . . . . . ;

299 Stall Transient Data, 101% Speed, 5-deg tGV,Maximutu Itate . . . . . . . * * . * * * * * . . * . * * . . . . 3.

300 Stall Transient lata, 101', Speed, 0-dog IGV,Isps Rate ... .. .. .. ... .......... 387

301 Stall Transient Data, 10'tY Speed, 0-dog IWV,MAxiMuwltaIW ...... .. .,*4 ........... ... 388

302 Stall Ttmnsivtt Data, 101` Spymed, -4-dog IGV,s s R t, ..... .. . . .... . . ... .389

30$ Stall Trmannt Data, 1011 Spoo, -4-dog W1V,Maxitum Rato ...... *... .... ... .. .... . 390

304 Stall Tvatisetat Data, 94. V& Spoed 15-dot WUV,| S • R t . . • . . . . . . . . .... . . . . . . . . . . 3 0 1

305- Sta'il Tratrient Data, 94. S•¶ $ed, i5-d&g 14V,MW.iatumu Hato . . . . . . . . ........ a*******302

0 Stall Transit Oata, 951 $pet4 10-o4-g 10.Vtsp$ t @ .te .. * . .. *. .. S.* ., .* . .. . . . . . .

3 K *tall Trausioat Oalts, OVT Spee-d, 10-dog 1WV,

"iv

LIST OF TABLES

Table Pae

I Single-Stage Centrifugal Compressor Design Summary .... 5

II Bearing Features ............................. 13

III Test Facility Safety Systems ........................... 26

IV Compressor Instrumentation Stations .................... 28

V Compressor Component Performance InstrumentationSummary .................................. 35

VI High-Frequency Response Probe Locations ............ 41

VII Data Summary ............................... 48

VIII Estimate of Instrumentation Accuracy .................. 57

IX Typical Printout for a Near-Stall, Steady-State Pointat 101% Design Speed and -4-deg Inlet Guide Vane Setting. 59

X Traverse Data - Internal Flow Analysis PerformanceComparison ................................. 63

XI Inlet Guide Vane Circumferential Traverse Data - 100%Speed, 10-deg IGV Setting ........................

X[I Inducer Exit Traverse Data, 101% Speed, -4-deg IGVSetting ......................................... 87

XIII Inducer Exit Traverse Data, 100% Speed, 10-deg IGVSetting .................................... 87

XIV Overall Performance Tabulat!on - Build No. 3, 0-degPrewhirl .................................. 133

XV Overall Performance Tabulation - Build No. 3, 10-degPrewhirl .................................. 134

XVI Overall Performance Tabulation - Build No. 3, 20-degPrewhirl .................................. 135

XVII Overall Performance Tabulation - Build No. 6 ......... 137

XVIII Inlet Guide Vane Performance Printout, 70% Speed,10-deg IGV, Near Stall, Build No. 3 ................ 234

Xxv

LIST OF TABLES (Continued)

Table Page

XIX Inlet Guide Vane Performance Printout, 90,% Speed,10-deg IGV, Near Stall, Build No. 3 .................... 235

XX Inlet Guide Vane Performance Printout, 85% Speed,20-deg IGV, Near Stall, Build No. 6 .................... 236

XXI Inlet Guide Vane Performance Printout, 85% Speed,30-deg iGV, Near Stall, Build No. 6 .................... 237

XXII Inlet Guide Vane Performance Printout, 8/1 Speed,10-deg IGV, Near Stall, Build No. 6 .................... 238

XXIII Inlet Guide Vane Performance Printout, 8/1 Speed,15-deg IGV, Near Stall, Build No. 6 .................... 239

XXIV Inlet Guide Vane Performance Printout, 101% Speed,5-deg IGV, Below Near Stall, Build No. 6 ............... 240

XXV Inlet Guide Vane Performance Printout, 95% Speed,10-deg IGV, Near Stall, Build No. 6 .................... 241

XXVI Inlet Guide Vane Performance Printout, 100% Speed,10-deg IGV, Near Stall, Build No. 6 .................... 242

XXVII Inlet Guide Vane Performance Printout, 101% Speed,-4-deg IGV, Near Stall, Build No. 6 .................... 243

XXVIHI Inlet Guide Vane Performance Printout, 101% Speed,0-deg IGV, Near Stall, Build No. 6 ..................... 244

XXIX Inlet Guide Vane Performance Printout, 101% Speed,5-deg IGV, Near Stall, Build No. 6 ..................... 245

XXX 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, WDCV, Run 3.06, IGV = 0 degFlow Rate =0. 494, Speed= 19700.0 ................... 247

XXXI 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, Below Near Stall, Run 3.06, IGVSetting 0 deg, Flow Rate = 0. 302, Speed = 19420. 0 ..... 248

XXXII 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, Near Stall, Run 3. 06, IGV Setting =0 deg, Flow Rate =0.253, Speed =19537.0 ............ 249

xxvi

' i


Table Page

S XXXIIV 10/1 Centrifugal Compressor, Impeller Traverse,(.t -30 Percent Speed, WDCV, Run 3.05, IGV

Setting = 10 deg, Flow Rate = .432, Speed 19798. 0 .... 250

XXXIV 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, Knee, Run 3.06, IGV Setting = 10 deg,

te =0Flow Rate , Speed 19877. 0....... . ....... 251

}XXXVI 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, Below Near Stall, Run 3.005, IGVSetting 10 deg, Flow Rate 0. 246, Speed =.19647. 0 ..... . 252

XXXVI 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, Near Stall, Run 3.05, IGV Setting 10 deg,Flow Rate =0.223, Speed =19422.0 ................ 253

XXXVII 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, WDCV, Run 3.06, IGV Setting 20 deg,Flow Rate 0. 494, Speed = 19874.0 . ..... ..... . 254

XXXVIII 10/1 Centrifugal Compressor, Impeller Traverse,

`ý' z~30 Percent Speed, Below Near Stall, Run 3.06, IGVSetting = 20 deg, Flow Rate 0.358, Speed = 19912.0 . . . . 255

XXXIX 10/1 Centrifugal Compressor, Impeller Traverse,30 Percent Speed, Near Stall, Run 3.06, IGVSetting =20 deg, Flow Rate =0.269, Speed =19458. 0 .... 256

XL 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Wide Open Discharge, Run 3.07, IGVTurning = 0 deg, Flow Rate = 1.454, Speed = 45557.0 .... 257

XLI 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Below N Tear Stall, Run 3.07, IGVTurning = 0 deg, Flow Rate = 1.464, Speed = 45712. 0 .... 258

XLII 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Near Stall, Run 3. 07, IGVTurning = 0 deg, Flow Rate = 1. 452, Speed = 45542. 0 . . .. 259

XLIII 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Wide Open Discharge, Run 3. 07,"IGV Turning = 10 deg, Flow Rate = 1.463,Speed =45806.0 .......... 260

xxvii

K .Q,


Table Page

XLIV 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Below Near Stall, Run 3.07, IGVTurning = 10 deg, Flow Rate = 1.460, Speed = 45883.0 . . . 261

XLV 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Near Stall, Run 3.07, IGVTurning 10 deg, Flow Rate = 1. 462, Speed 45771. 0 ... 262

XLVI 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Wide Open Discharge, Run 3.07, IGVTurning 20 deg, Flow Rate =1. 448, Speed =45574.0 . . . 263

XLVII 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Below Near Stall, Run 3. 07, IGVTurning = 20 deg, Flow Rate = 1.459, Speed = 45699.0 0.. 264

XLVIII 10/1 Centrifugal Compressor, Impeller Traverse,70 Percent Speed, Near Stall, Run 3. 07, IGVTurning = 20 deg, Flow Rate 1. 458, Speed = 45913.0 ... 265

XLIX 10/1 Centrifugal Compressor, Impeller Exit Traverse,90 Percent Speed, Near Stall, Run 3. 08, IGVTurning = 0 deg, Flow Rate = 2.604, Speed = 58529.6 . . . 266

L 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 8/1 Line, Wide Open Discharge, Run 3.09,IGV Turning = 10 deg, Flow Rate = 2.898,"Speed = 61877. 0 ............................. 268

LI 10/1 Centrifugal Compressor, Inducer Traverse, 8/1Line, Below Near Stall, Run 3.09, IGV Turning= 10 deg,Flow Rate =2.902, Speed =61921.0 ............. 270

LII 10/1 Centrifugal Compressor, Impeller Traverse, 8/1Line, Below Near Stall, Run 3.09, IGV Turning = 10 deg,Flow Rate =2.902, Speed =61904. 0 ................ 271

LIII 10/1 Centrifugal Compressor, Inducer Traverse, 8/1Line, Near Stall, Run 3. 09, IGV Turning = 10 deg,Flow Rate = 2.818, Speed = 62035.0 ................ 272

LIV 10/1 Centrifugal Compressor, Impeller Traverse, 8/1 Line,Near Stall, Run 3.09, IGV Turning = 10 deg, FlowRate =2.895, Speed =62126.0 ........... 273

xxviii'4~


Table Page

LV 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 70 Percent Speed, Near Stall, 20 degIGV Turning, Build 6, Flow Rate = 1.279,Speed = 45548.9 .... ........ .............................. 274

LVI 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 70 Percent Speed, Near Stall, 30 degIGV Turning, Build 6, Flow Rate = 1.242,Speed -45528.7 ............................. 276

LVII 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 85 Percent Speed, Near Stall, 20 deg IGVTurning, Build 6, Flow Rate 2.125, Speed 55663.1 .... 278

LVIII 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 85 Percent Speed, Near Stall, 30 deg IGVTurning, Build 6, Flow Rate 2.071, Speed 55505.1 ..... 280

LIX 10/1 Centrifugal Compressor, Inducer Traverse, 8/1Speedline, WOD, 10 deg IGV Turning, Build 6, FlowRate =2. 823, Speed =61657..2 .................... 282

LX 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 8/1 Speedline, Near Stall, 10 deg IGVTurning, Build 6, Flow Rate 2.796, Speed 61819. 2 ... 283

LXI 10/1 Centrifugal Compressor, Inducer Traverse, 8/1Speedline, WOD, 15 deg IGV Turning, Build 6, FlowRate =2. 807, Speed =61605.6 ........ ........... 285

LXII 10/1 Centrifugal Compressor, Inducer and Impeller

Traverse, 8/1 Speedllne, Near Stall, 15 deg IGVTurning, Build 6, Flow Rate = 2.787, Speed = 61700.7 .... 286

LXIII 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 95 Percent Speed, WOD, 10 deg IGV Turning,Build 6, Flow Rate = 2.885, Speed = 62114.1 .......... 288

LXIV 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 95 Percent Speed, Near Stall, 10 deg IGVTurning, Build 6, Flow Rate = 2.877, Speed ý 62138.8 ... 290

LXV 10/1 Centrifugal Compressor, Inducer Traverse, 100Percent Speed, Near Stall, 10 deg IGV Turning, Build 6,Flow Rate 3.058, Speed = 65238.1 ............... 292

xxix


Table Page

LXVI 10/1 Centrifugal Compressor, Inducer Traverse,

101 Percent Speed, WOD, -4 deg IGV Turning,Build 6, Flow Rate 3.214, Speed 65934.3......... 293

LXVII 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 101 Percent Speed, Near Stall, -4 deg IGVTurning, Build 6, Flow Rate = 3.209, Speed 65987.7 . . . 294

LXVIII 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 101 Percent Speed, Near Stall, -4 deg IGV, WithCoolant, Build 6, Flow Rate = 3.214, Speed = 65779.6 ... 296

LXIX 10/1 Centrifugal Compressor, Inducer Traverse,101 Percent Speed, WOD, 0 deg IGV Turning, Build 6,Flow Rate 3.192, Speed 65834.8 .............. 298

LXX 10/1 Centrifugal Compressor, Inctucer and ImpellerTraverse, 101 Percent Speed, Near Stall, 0 deg IGVTurning, Build 6, Flow Rate 3. 185, Speed 65952.8 .. . 299

LXXI 10/1 Centrifugal Compressor, Inducer Traverse, 101Percent Speed, WOD, 5 deg IGV Turning, Build 6,Flow Rate 3.150, Speed =65974.5 ................ 301

LXXII 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 101 Percent Speed, Near Stall, 5 deg IGVTurning, Build 6, Flow Rate 3.142, Speed 66122.6 ... 302

LXXIII 10/1 Centrifugal Compressor, Inducer Traverse, 100Percent Speed, Near Stall, 10 deg IGV Turning, Build 6,FlowRate=3.058, Speed=65238.1 .............. 304

LXXIV 10/1 Centrifugal Compressor, Inducer and ImpellerTraverse, 101 Percent Speed, Near Stall, -4 deg IGVTurning, Build 6, Flow Rate - 3.209, Speed 65987.7 ... 305

LXXV 10/1 Centrifugal Compressor, Impeller Traverse, 8/1Speedline, Near Stall, 10 dog IGV Turning, Build 6,Flow Rate =2.796, Speed =61819.2 ................ 307

LXXVI 10/1 Centrifugal Compressor, Impeller Traverse, 8/1Speedline, Near Stall, 15 deg IGV Turning, Build 6,Flow Rate 2.787, Speed 61700.7 ........... . .... 308

xxx! ::'I,

• •Il K.-XX ..o1

71 .77-


Table

LXXVII 10/1 Centrifugal Compressor, Impeller Traverse,95 Percent Speed, WOD, 10 deg IG\ Turning, Build 6,Flow Rate 2.885, Speed = 62114.1 ..... ........... 309

LXXVIII 10/1 Centrifugal Compressor, Impeller Traverse,95 Percent Speed, Near Stall, 10 deg IGV Turning,Build 6, Flow Rate =2.877, Speed =62138.8 ....... o... 310

LXXIX 10/1 Centrifugal Compressor, Impeller Traverse,101 Percent Speed, Near Stall, -4 deg IGV Turning,Build 6, Flow Rate = 3. 209, Speed =65987.7.......... 311

LXXX 10/1 Centrifugal Compressor, Impeller Traverse,101 Percent Speed, Near Stall, -4 deg IGV, With Coolant,Build 6, Flow Rate =3.214, Speed =65779.6.......... 312

LXXXI 10/1 Centrifugal Compressor, Impeller Traverse,101 Percent Speed, Near Stall; 0 deg IGV Turning,Build 6, Flow Rate = 3.185, Speed = 65952.8 .......... 313

LXXXII 10/1 Centrifugal Compressor, Impeller Traverse,101 Percent Speed, Near Stall, 5 deg IGV Turning,Build 6, Flow Rate = 3.142, Speed = 66122.6 .......... 314

xxxi

LIST OF SYMBOLS

A cross-sectional area, in.2

A* throat area, in. 2

a speed of sound, ft/soc

B* throat blockage

Cd inlet flow coefficient

Cd* diffuser dischargo coefficient

C pressure recovery coefficient

o uncertainty of an individual sensor

G gravitational constant = 32. 174, lbm

h onthalpy/unit mass

i incidonco, dog

-1(7-11K f (Mach nuwmbr)o M11 + M2)2

M relative Mach number

SMabsolute M ach nuniber

N rotor speed, rpm

n N•unibr of sonsors

P static pressur, psia

prossurv I Atio

Pt total pressure, psia

gas costant = 3. 34. ft-lbe bm-•It

T total temperature, "H

T temporattuv ratio

T static tomporature, 9H

LINT OF SYMBOLS,(Continued)

U rotor speed, R/see

u overall uncertainty

V absolute velocity, ft/soc

W relative velocity, ft/sece

w weight flew, lb/sec

S.,absolute air angle, dog

0 relative air angle, dog

0÷ leading edge metal angle, dog

"ly ratio of specific heats

6 P /14.G940

* diffuser effectiveness

11 adiabatic efficiency

T T/518.68

SUBSCRI•P1'S

act actual

Cal calcUlate'd

cor coreetod tW staard day inlet Conditions

i ideal

4~~I axial cn~n

relatv

tS Stowtorwai

tWF totaal "gt 44ic t

1k tngIt a ~iWati

77

LISIT-OF SYMBOLS (ContWnued)

V vane

0 plenurtilJ

L1 IGV exit instrumentation station

1.i5 inducer exit instrunwntation station~

2 impeller exit instrumenta tion station

3 collector

'U P IISCiuprs

C) mass averaged

sonic (throat) condition

A

"XivI

'TION

Hi1gh-pressure ratio, single-stage centrifugal compressors offer the possibility ofAproviding rugged, relative)v erosion-. reSisto wt and low-COSt CompreMSS ion systems1for advinced gas turbinle engines of the 2-to-5-lh/sec airflow classi. If good offi-ciency can be obtainedq at high pressure ratios, this type of compressor wvill be A

* more us--ul than the multistage a.%iul flow comprossors now uoed in small militaryalircraft engines. The obijective of this five-phase programn was to designi, fabri- :

cite, and test n high pressure ratio centrifugal compressor with Uie following1

1. Pressure ratio of 10.0:1 and corre-sponding adiabatic efficiencyof 75i~ it design speed and. flow.

2. P~ressure ratio of 4. 0:1 and corresponding adiabatic efficiency Jof 8%at a speed and flow condition loss than that of the design

The five programt phases are (I) initial comprecssor design, (2) ftirdrlatlon tindassembly of the test hardware, (3) test of the initial confitgutratton, (4) evalrratonoft the test data, and (5) comapre-ssor modlIfication to include fabrication and assembly

of new hardware, test, and data evaluation.

ThsrlotCnansil'-n~to pertaining toalphose-s oif the program. it in-4wa jyalnsttoM dataie neecs

sayto the general tvxt. Apf~r'±dix It (separately hound) cont.in th dtie eodynamnic design and is clas~sified CON I1L)NTIA L.

The two test phasex of the program weore, completed during six test periods over a3-year span. T1he progress of the couri"act expe-rimental pr-ogratp was interruptedto peirmit rcdesign and modification of the te~st rig to achieve its ratetd rpml. Thefollowing chronological sequencc of the, tests conducted during this program willaid in understandinig thv design andc dntta described in thi~s report. Each suceetisiveconifiguractio of either the test s*tago or rig. IW idontifi~ed by a build numbler.:

11*414 No. 1: Thle initial build of the rig Was tested. to Approximately7C( of the, design speed, The itm-pellor MAbbd tho shroud]and the rma progtrAm was trminiated.

Build4 No. 2:. Tho *vcond Wuild contained imodl~fictatin to, pre-vent the

Imp/oller from contacting the shroud, This Wild was.

testing was ternuierned due to high rig vibrationt levels.Perfonuatiee data from this test were e-V4ttatcd and Pon-mitted a mintor erdaicredesign to beo conxitzkd. Acs aresulIt of the tv4*'Lgu, the diffuser wos rQ*WWe to mutawt thet

Build4 No. 3.z The vitbratiowonsewouatf.tt-d ifs tiuld No. 2 were diaapoosdaacrtic~al speed problem,. and the, rotor sys~tem sadu

impeOller 4ttae~huteut wvtv *titnW-ee to iWrt'as tho Critical

IA

speed out, hopefully, of the running range. While these "modifications improved the rig operating envelope, highvibrations were again encountered at design speed. Theentire test program was completed, with the exception ofthe design speed testing and special instrumentation tests.It was determined that the prediction of rig critical speedwas relatively inaccurate for the overhung rotor configuration.It was further determined that a straddte-mounted bearingsystem (impeller between bearings) would be capable ofrunning at higher speeds without failure, and the rig was re-designed accordingly, including rerouting the inlet nlow patharound the new front bearing compartment.

Bluild No. 4: This build was tested as a checkout for the redesigned rigbearing system; however, testing was terminated by anImpeller-to-shroud rub, which occurred at approximately85!" speed. Silver scraped from the shroud during the rubcaused damage to the diffuser entrance region during thistest. Data obtained during this test indicated the presenceof potential heat transfer into the inlet and also through thecases. The inlet and Inlet plenum were insulated and ather•uii dam was machined between the inducer and im-peller shrouds to eliminate these problems.

Bluild No. 5: The rig wat m'odified to Include a therrmal dam to limit theheat transfer wetween the inducer and impeller shrouds, andthe clearances were modified to prevent rubbing betweenimpeller and shroud. The rig mechanical redesign wassuccessklly checked out to design speed.

Build No. 06: This build concluded the contract test program, as a3l phase$of the modified test plan that coacentrated on obtaining datanear design speed were suecess-utly completed. Stable rigoperatim was draonstrted at speeos aw high as 72o,000 rpm4112% a design speed).

'It aetodrtamic and performance dat* It thil mort prImarily result from thetv of guidl N. 2., 3, and 6. T1w onaly change in aerody,=nic hardware from

"I.Iuid No. 2 to Build No. 3 consisted of rcmoval of 0. 020 in. of the, iMpeller exitdiameter to prvent contae-t with the diffuser shroud d(tirng testily. The Build

""o. 1 impeller did not havv thia O. 020 ia. of material removed and thus wastdetitical to the original des•ti. guild No. 6 did have a p--w itnet helimouth ar) aconverging inlet with tw" additil inlWet trutt duhe to the nwV rig mechanical-

t: anraag~ement roquit-d to achieve oeign e. The iatet gui4de vaes and Imgwer J,ere reunchanged front the ogtinal design, aed the diffuaer w#as repaired a*s mueh

as piwsible, W~t swtill conaind damage ita th@ ewntranc regiaa du@ to th@ Huild No. 4ImeU to,-*hrod rubh.

S.--

DESION APPROACH

AF.RODYNAMIC DESIGN

The aerodynamic design is completely deserxoed in a separate volumne .sAppetnix II to this report and Is classified CON1I•DENTIAL. A brief summary ofthe design of the stage, ihown in Figtre 2, is presented herein. The design con-slats of variable inlet guide vanes, an iwpeller with a remote inducer, and a pipedliffuser.A

11.004 In. DIFFUSER20 VARIABLE DIAMETER 32 PWES

INLET

GUIDEVANES IMI•LLER- F I •.12 FULL BLADESI rI

7.04 W SPLITTERS

24 SECONDARY

DIAMETE

24 BLADES

Flguro 2. 10.1 Pressuro Rtatio Veutdtfugal Com~preasvo Stoge.

The objective, of the program was to 4es1pn a 2-to-.54b/see airflow Ainggle-stagocentrifugal 4ompressor that ettid be incorporated i' a future Ary advanicedtechiology gas turbine engiae. The design speed pertormance gols were to exceed7&'11 efficency ait lOtt pressureý ratio. Sineo gas turbitle eagin-0s for Ar-my aircraftapplications operate utder part-power -voactitoas a majority of the time, aft off-

uesign aJ~l~il@~~ I Cs@W~W at #i pvura tiow was Vestablished

t4 the daestgg of the eoupv•ittor. paeametrie studies were co ed o s-lec-t anoverall dosign coNwsistedt with optimumt mrsstproraatt ohprformattergo~ats. These studes defined the compm*resriltcrcedfwrt,impeller tIlet hub d `tip va orrected impeller ota Qo lte noeed et a WEIprwhUrt. Airflow selectio and the aetectiu of the hu radius were tanf o dhy the decjision to diggign a votumptvtisr that coud e uued in a smatll turbahalft

ngi••e with a ionentr•i shaft froat drive. This cotg r•a reqirod a tar&re

iWlet hub radius than" that nrmally as-o•.iated with r•a•ch•--p-_•---- but it w..s

Mt -h* tho to*i-Aou

The tip rndlius wns selecte'd aIfter determliningl the effect on ::'dal Mach number,induceri tip re'lative' Mach numbier, aind inlet choke flow margin. Thie effecet ofin! et gu ilt' vaneu I osscs , inlet Shock 1 os sis, diffuser lossesM, a9nti shroud frict ionheatting werev paramlet neatly evalu11ated before selecting WV prewhini and rotorspeed Ito prolvide the optimum overal~l c'ompressoir performance.

A renolt', inducer (lesigil was se~lectedI ov-(e, n n integrral intiuci'r-impelIN't c (on-U 4j: rition so) that -theinduct'c cou:!d be dt'ý.iign-it usl hug t.m anoni eaxi al-flow corn-

pt't'55( WcttlhrWologV. The work spilit betwt'en the inducer' andi impelIler wasselectt "it)o that the1 vcrl ativye Mach nun:e ix' r:1W the impellIcr would he I.'uhsollicWith Iit oimn.' c t ha-l n aI :, findiue r prr'ss4u c rat in, The cesultUmt 10:1 pressurera1tio himpl! ' c~ with r'mote in1ducer. is Shown ill Fngrur 3c.

F i~ 3.tO Ptd'~ir t~iio (ti~ futl tnplle ,id teiotr Inducevr.

cct\ .iL-t!ti~t It¶ Wevv rv, 'iht'id with reajpelet to thecir poteti-fo r ni the %iflje-%tftt- A fttrS~ p3ip LiIfuIsC- waM selecýted

~r a~l dt~''L2Si~d d~tt~~eV, lc~t~eit hItS thet lowe.St IONSes oVer thte:rgst n h.~e'4Alteh umlrc tn hvcetmsr. P&W'AVM h~it had suhstnttial tt'q0ritntee

A Itti it4Ittiu ttit f pcvrt iarat cdtoinreitut' Lteiigtt intuntna*ticnit isi pr-setute hif Tableý t.

"14

TABLE I. SINGLE-STAGE CENTRIFUGAL (4 OMPRESSOR DESIGN SUMMARY

Compressor OverallPressure Ratio 10:1*Adiabatic Efficiency, % 75Flow rate, lb/sec 3.1Rotor Speed, rpm 65,300Specific Specel 80

Inlet4 Struts NACA 400 Series20 Inlet Guide Vanes NACA 63 SeriesNominal Guide Vane Setting, deg 10Specific Flow, lb/ft2 /'uc 34Hub Radius, in. 1.6Tip Radius, in. 2.6 4

InducerRemote Arrangement24-Transonic BladesHub- rip Ratio 0.615Nominal Pressure Ratio 1. 57:1Nominal Efficiency, % 87.6

i'-1

ImpellerNominal Pressure Ratio 7.34:1Nominal Efficiency, % 82.8Exit Flow Angle, deg 19.5Exit Tip Radius, in. 3.52.xit Blade Height, in. 0.230

Nominal Operating Clearance 0.005

Diffuser32-Passage Conical Pipe TypeVaneless Space Radius Ratio, in. 1.10Throat Radius 0.11Straightening Section L/d = 0. 5Cone Geometry, deg 3 to 5Nominal Loss (APt/Pt) 0.12Pressure Recovery Coefficient 0.789

*Unless otherwise specified, pressure ratio is the ratio of dischargc si-.Clc

(plenum) pressure to inlet stagnation pressure. Efficiency is also ',pecifiedusing this pressure ratio.

- "4-

5

rI v

MECHANICAL DESIGN jAlthough the objectives of this program were primarily aerodynamic, the mechan-ical aspects of the test compressor, drive turbine, and the test facility exerted a .•large influence on the course of the program. As is the case with all higu-speedrotating machinery, design emphasis was placed on the evaluation of blade anddisk stress limits, determination of rotor critical speeds, and calculation ofbearing loads. Tests using the original compressor-drive turbine configuration,described in the Introduction section, uncovered serious limitations to the saferig operating envelope and necessitated a substantial redesign of the rig rotor

system. Although the design and development of the high-speed drive turbine wasnot part of the scope of the work of this contract, the ccontract compressor stagewas directly mounted on the drve turbine rotor and their system dynamics wereintegral. Consequently, a c1scussion of the drive turbine design and its subse-quent redesign are incluc1,d in this section. The mechanical design of the 10:1pressure ratio centrifugal stage is discussed in the following paragraphs. Aschematic drawing of the initial drive turbine and compressor test hardware de-sign is provided in Figure 4. The design featured an axial flow inlet, variableinlet guide vanes, overhung spline drive inducer and impeller, and a doubleradial inflow drive turbine.

Inlet

The inlet section was required to support the inlet centerbody, house the variableinlet guide vanes, and provide lubricating oil to the front bearing and instrumenta-tion routing channels. It was fabricated from aluminum weldments. The thicknessof the four supporting struts was determined by the size of the oil supply line. Ananalysis indicated that the oil flow requirement could be accomplished with a1/8-in. OD tube. Therefore, a conservative maximum thickness of 1/4-in. strutwas selected.

With 4 struts and 20 inlet guide vanes, It was appropriate to place every 5th guidevane in line with a strut. All of the guide vanes were connected to a synchronizingring, which allowed variable positioning of the vanes from zero to 30 deg of pre-whirl. An actuator was used to position the vanes. The actuation linkage was de-signed with close tolerances wherever feasible to keep alignment errors to a

minimum. The calculated maximum misalignment of an individual vane was ±1. 5deg from the nominal angle. Sealing of leakage flow around each vane was ac-complished with an 0-ring on each vane shaft. Contouring of the vane hub andtip provided a clearance of 0. 005 to 0. 015 in. at all positions of actuation.

Tnducer

In the mechanical design of the inducer, a thin conical spliaed shaft was used totransmit torque from the drive shaft, The design of the remote inducer permittedit to be readily modified to a close-coupled inducer configuration at a later date.A stress analysis of the AISI Ti-6A1-4V inducer disk showed the maximum effec-tive disk stress to be just under 35, 000 psi, wh'ch gave a burst margin of 106%.An analysis of the 1. 04 aspect ratio blade natural frequencies indicated that therewere no bending or torsional modes within the intended operating range (Fig-ure 5).

6

NNJMW0

A I 4 zz 00 w

uJ 0ca <f (3LA 00

LU,

IMIwz =: 9L 0 z

4,w 1-- 0 jc

0.3)

(A W1___

M Z wqlý U. 0 cc I

ca 0 W oj ok S

DESIGN SPEED 50% 70% 8 59

I 10 20EIINDUCER BLADE 1 ,22NATURAL VIBRATION ~ - --- -- 12

FREQUENCIES I 40EI(UNCORRECTED FORCENTRIFUGAL I ______

20 -STIFFENING EFFECTS)-~ ~ ~ 1-

'18

16 -- _ _

'1 I 12E

S14

12.

ILL

N108 op____ _ _

12 _____4

C.0

10 \3___ 8_ E

Vi2EU. '4,

8 ~~~ ~1 -+

00 10 20 30 40 so 60 70 [

ROTOR SPEED. rpm (Thousnds)

Figure 5. Ind1ucer Vibration Analysis.

Impeller

Tbe impeller attachment design also employed a thin conical splined shaft totransmit torque and had a press fit on the drive shaft for alignment. The impel-1Fr and inducer were mated with an interference (snap) fit, as shown in Figure 6.Stress analysis of the impeller disk at 72, 000 rpm showed maximum predicted bore

* , ~effective stress of 133, 000 psi. Using AISI Ti-WA-2Sn-4Zr-Mo material , witha 0. 20/r yield strength of 153, 000 psi, at the maximum predicted bore temperatureof 220oF, the disk stresses are within acceptable limits. A calculated burst speedof 98, 900 rp-m results in an adequate burst margin of 37. 2%. A stress analysisof the impeller blades predicted maximum blade stresses at about 75% of the localyield stress (loss of strength due to increased temperature taken into account).

INDUCER SPLINE IMPELLER SPLINE

F jgvre 6. Impeller vaid linduce Attachtment.

Shrauds and Diffuser

Aoodymamic detdga of the Indtwsor had the tip diameter vvnvergo from the loadingto the trailing edges. This cnvorgonve resulted In a separato Inducer' and Impel-)or to allow assmbly. Theaivdcrynumiva dosign of the impoller and Indcer' alsorequired operation with vv~ small shroud, learances, mi the ardor of 0. 00)5 in.To Wainialize the Consequences 04L ruhbing a t-itanium blade on a stainlotis stoolKhoud. both W111eller anid inducer shroids were plalted with a thin layer of cailver.

0). 005 to 0, 007 in. thick, which providod a relatively soft rubh surfiace. P~osition-Ing of the Ahrouds relative to tim lmpellevr-"w~r was ac liohm bymva~o

shmseottion at assembly.

Stallf'seteel 410 waR Ovletctd ior mtIait tho 32-asbotgo pipe diffuser hecasoitti low coefficteit of themal e MAision iip roavhes that of tho titanium Impeller.Tim Insidoii of tho 44i1fusior pipes wov' nickel plated to mitnmlrze camian. The

W! FV

alignment of the diffuser inlet with respect to the Impeller blade exit was designed

to be adjusted at assembly with shims.

q To define running clearances and diffuser-Impeller alignment, mechanical clear-ance probes were designed to experimentally measure the running tip clearance(Figure 7). The projection of the replaceable aluminum wire Into the clearancespace was measured during assembly. On the test stand, rotor rpm was In-creased by increments, and after each incr,3ment, the probes were remove(, andthe length of wire wvas measured to determine clearance as a function of rotor

4 -3peed. The clearance thus determined is the net result of any impeller movementand shroud and diffuser thermal growth.

AMS 5613 STAINLESS PROBESTEEL PROBETI

DETAILS

CLEARANCE

CLEARANCE PROBES REPLACEABLEALUMINUMWIRE

SHROUD '

IMPELLERBLADES

Figure 7. Tip Clearance Probe Installation.

flaings and ~Sel

-rhe bearing support for the compressor and drive turbine consisted of two 35-mmrollng lemet harittgs straddl~e mounted on either side of the turbine rotor. The

compressor splined drive shaft was overhung beyond the front bearing, as shownin. Figure 8.

The forward Wearing, which also functions as a thrust boaring and, hence. Post-tions the impeller, is a split-inner-race hall boaring ftht can accommodate trans-lent thrust rversals. The roar bearing Ir q st raight-through-oute r- ring roller

beain tat can accommodate axial thermal growth. At the doig speed ofor~, 30-nm the Wearing DN (bore diameter in mmi x rpm) tit -. 28 wIllioni. Soloc-

ting a smaller bearing diameter would forve tuore tsevvr limitations an potential

10

front-drive capability in an engine application. It would also reduce the allowablethrust load range of the ball bearing (between skid and fatigue limits) to a pointwhere precise thrust balance control would be required. The bearing material,M50 alloy steel (PWA"m 725), was selected am tho best, available alloy for bothbearings. Bearing cross sections arz snown in Figure 9, and pertinent designparameters are listed in Table I1, Both bearings have one-piece, inner-land-riding cages that are machined from AMS 6414 steel and silver plated. Wide cagelands ensure adequate cage journal support area. Lubricating oil is supplied fromwithin the shaft to ensure adequate inner race cooling and positive oil distributionwithin the bearings. Total oil flows are 7 and 4 lb/min, respectively, for theball and roller bearings.

54

Ball bearing raceway curvatures for the compressor rig were selected to limitHertz stress and to keep spin-to-roll ratio (heat generation) to reasonable levels.The inner race curvature was increased somewhat above the optimum fatigue lifevalue to reduce potential cage problems and heat generation. The higher curva-ture tends to provide lower cage loading and wear by restricting contact anglevaiý'iation due to misalignment or excessive radial load. To provide additionalmargin in this respect, it was desirable to restrict radial loads on the ball bear-ing to no more than one third of the thrust load. Within this framework, thedesign analysis indicated that a complement of 15 balls with 5/16 in. diameterwould provide the maximum fatigue life.

The compressor rig roller bearing was designed to support radial loads (predictedfrom rotor dynamic analysis) up to 150 lb in excess of 150 hr with proper balance,alignment, and lubrication. This resulted in a complement of fourteen 7.5- by7.5-mm rollers.

DRIVE TURBINEROTOR

N - SPLINED COMPRESSORDRIVE SHAFT

:A PFRONT BEARING

:•:; ' •, -• •REAR BEARING

Figure A. Iniial Compressor Drive and Waaring Support Diign

, 11

- --

CiIw w,

-J -c00LzuN

- -en,U;IlU2 0 7

U44

Lcc

-cc

TAIILE It. BEARING VlkATUflbs( 1 )

Bull Roller

Nominal boxy size, mam 35 35OD, inl. 2.4408 2.4408

11),in.1.3177 1.3770Number of vlootiets 15 14

Elmn ie5/10 In. 7.5 x 7.5 mun( 2 )Pitch circle diamet.er, in. 1.9095 1.0095Outer raceway crature, Cl()52-Inner raceway curvalture, 1%4') 56-Contact anlgle, deg 25

(1Room teniporature

10loIn. radilus crown()(radius; of curvature/ball diatmeter) x l(00

Dute to the high centrifugal load1ing, the rollters were designed with more titannonmial crown drop to roduce the tetndoeny for end loauding and wear.

-,Rotor J~nttniies

A vibatioal aalysis of the compressor and turbine rotor a.4semby, sowtliFigutre 10, indlicated acceptable twariag loads and rotor defliuctioas. The boullceand pitch mode~s (11,000 awld 25, OtU rpm. respectlvety) of tlw rotor were inl tholower end of the! operat in speed rai anige and their response wot Id he roadily datnpvdby the oil film champers at the hllnt and roller bea-ring supports. Calculatioti of theft-st beatding mode rotor critical speed showed It was well out ot the runninlg ranige.Dyniamic rvsponse cakcuhatimiw of bevaritig loads are shown inl Figures 11 and 12 fora 0.1 ez-inl. unbalaance at the front and rear twarings, respetiveiy.

Thru-st Balance

Anl external system was inicorporated to supply gaseous nitrogen onl the rear faceof the comiprossor impeller to maintain a minimium thrust load onl the. ball beariung.Calculated load as a functioni of spee,,d require-d to preve~nt the ball beariftg fromsk~iddin is shown in Figure 13. The &04Q lb rvquised4at. eAii-tn speed wais wellbelow thm maxinu rcommeinded ball bearing thrust load capticity of 400 lb.Labyrinth-typ*e knife edge seals were used to contain the nitrogen in the thrustbalance cavity. ta addiinchel ancvt b twee the imupe-ller tip backtace

iand the thrust balaace cavity was in!orporated that tnirnuiaezd flow between theimpeller tip batklaee atud the thrws Walanc eavizy.

13

.0-4

cc*

DEI

4 1I

S - ----- 77777

1~4 3~ROTOR SPEE- -rp 4Taum

10 2a -0

ROO DESI-toGIhou~

Ficwv~~~~~~~~ 13SPEEDalHurniUnhiw b~uLa

} ~Modificatio~ns

The impeller attachment selheme was stiffened when high rotor vibrations inBuild No. 2 limited maximum speed to 58,000 rpm. The impeller bore diameterwas increased to permit insertion of a cylindrieal 8paeer, shown in Figure 14.The elVndrical spacer added stiffness at the impeller-shaft attachment point duoto the larger diameter in the impeller region. A spacer was also added betweenthe Inducer and impeller. This spacer, in conjunction with the impeller cylindri-cal spacer, permitted the axial load to be transmitted front the front of the lit-ducer, through the impelter, to the drive shaft in a more direct fashion. Furtheractio!ý was taken to prevent a rui from occurring in the event rotor vibrationswere in.' corrected. The impeller exit diameter was reduced 0. 020 In. to preventthe impeller tip from rubbing the diffuser vameless space.

SPACER f

; ~~~INDUCER I!LE' - = m l' ~ t , ¢ • J • . - C V L I N D R I C A L

!iI • •"•"•+•.++.,STIFFENER

ADDED

Figure 14. tmapelkr Shaft Attaehment o•4Ucatioa.The Impeller attuvhmtm tifieattot increased the maximum allowabl -rotor

speed to slightly in excess of desiLg speed (65,300 rppm). However, duringBttild No, 3, high vibrations at or near desig4 speed caused the front thariag tofail after 5. 1 hr of operation at 95 to 1W' of design speed, Rapid shtdown pr,-cedures prevented aay damage to the compressor, b4t it wag decided qot torasunte tetIag until after th@ r"tor •*#mby w#s complewtly" te igad for in-ret•sed dynamic Mtability.

MtfjCHANICAL REDESIGN

The approach taken In the mechanical rede-sig 4f the compressort!rbine rotorasiaemlAy was to sOgnifieantly inemease crtical s•peed margin and to ireýamse thefront beadnag radial load qapaityV to ensur eEodtitmtgL3 apemtiontieat deitj speedsaacd above. The redesitgned test hardware. oh- o ti Fkire S1,5 eatuires tumdeball and yoler bearingss a carbon face soat I roat of the itpdleor. "AdCumwC couplin to join the impeller kW Cte ri@.

16

* I -- r�. r*� .Wh�*I �crI.Yfl.rf pars* � W9%S -. ... rr WYI. -"ar ' r �.r - "'- r�. - ""�" % r 'r

I

'Ivi ii�. ��=a.s Is£13 �IiltfliF�Y C

" !I Fwc11Ž4

K ISJew

- j 4<¼

rf(7' 0.

4 16kM I 4 1C.

I I I..

I, �1 U,I .4�.F :4

I. If /� 04

r-� I

( Q

l� SN> :4 be

�i1 :4

�-''

1*I,I ¶.�r

4.'I

¾'

� N,N� -

H I-XL U U,11 3 '4

H 0.

Rur -

IY

Ii b S4.

I31 V .4¾'

J I�.- �-. S

a

VSn

I17

inlet

Placing two bearings and a carbon s-eat lit front of the Impeller reqtircA the flowpathi to he changed frow, a parallel flow Inlet to a converging configixwitio., to pro-vide the idditional sace needed for the bearing coprmn ndA1spoTito increased supply &f oil tneeded for lubricating the two Imurings &ýnd coolingthe earbon seal p!late, nnd additionnl scavenige and breather rquirenionts could Inot be acommiodated in the four original Inlet struts. Two ;; .. itlonal o.truts wereincorporated 'tor a total of six equally spaced struts, Bly tai ininating, the flow patconvergenee at the inlet guide vanes, it was possible to use the samo 20 inlet4guide vanes that were used in the original inlet.

Inducer

In the redesign of the rotor assembly. the attachment of the inducer was changedto eliminate the 4plinc. Fr~iction wsts used to drive the inducer and was trans-M~ted bv the tiel-10 ana the radial loducer-impoller interface. The axial tiobaltlond at room temperature was 184,.500 lb). A tstres.% analysis of the redesignedInducer disk Indicated maximuni effc( tive stress to be 06, 000 psi. This wasIequivalent to a burst tmargin (A 4CY', wvhich, although less than the original design,wab still more than adequaite. The inducer remuote spacing was maintained.

IMPOW r

j ~The attachment of the inipellor to the drive turbine was rediesigned to use aC UflVIC coupling for maintaining rotor assembhly concentricity and transmittingjthc torque fromf the drive turbine. The redesigned rotor assembly is Sdrnwn inFigure 16. The fotvward end of the original drive turbine was modified for aC VR VIC ooupling. A CUTRMI spacer was uaod to separate, the turbine and tin-polter. Vihe use of -a tic-bolt to hold the rotor assemb~ly together required changing

the origitnal Impeller bore georaetry. A stresq analysis 5howcd the redesigned

as thev original impeller design and provideod wk adequato burst rarW1n of 41%.

itt. odoosigned Au&~M setry

8earings and Seals

A dual front-bearing configuration was selected to provide a substantial increasein radial load capacity over the original design. This dual arrangement locateda "floating" ball thrust bearing adjacent to a roller bearing, which allowed theroller bearing to carry all the radial load and the ball bearing all the thrust load.The rear bearing compartment (turbine end) was the same as the original design.

Both the front and rear bearings were the same as the original designs and weresoft mounted on oil-film-damped supports, with lubrication provided through theinner races as before.

A vibration analysis of the redesigned rotor assembly predicted substantiallylower bearing unbalance loads at design speed compared to the loads predictedfor the initial design, as shown in Figures 17 and 18 for the front and rear bear-ings, respectively.

A carbon face seal was incorporated in the redesigned inlet case to eliminate airand oil leakage to and from the forward bearing compartment and inlet flow path.The seal nosepiece was graphitic carbon, and the seal runner was AMS 6322 alloysteel, with a flame-sprayed coating of chrome-carbide to improve wear charac-teristics, The seal ruoner was cooled by flowing oil through small radial holes.Secondary sealing was accomplished by a bellows, which was damped at the ODto reduce bellovs fatigue problems. The face seal cross section is shown inFigure 19.

5I

41

ORIGINAL DESIGN

•: 0 20 40 IO0N 1SPEED ESIG (NThuu. OG

Figure 17. le~dcslgned Rotor Assembly Predicted Front Bearin~g Load vs Speood.

•iicc 'elks

-.- *.---*K0

.5

-4

-a 3... a ORIGINAL DESIGN

I2REDESIGN

cc 0 20 40 60 N so

•, DESIGN

• ~Figure 18. R/edesigned Rotor Assembly Predicted Rlear" Bearing Load vs Speed.

•!i, GRAPHIC CARBON NOSEPIECE -- BELLOWS

i i

1.775 In.

SPEDrDIAMETER

1.945 In.DIAMETER

Figure 19. Froet Bearin Compartmtent Carbon Face Soul.

Although the seal rubbing speeds (590 to 530 ft/seco wore hi iter thai currentenghav experience, seal reqluirements in tetras of differential prestiu (loss than10 psi) and temperatu; e (23'•F) were relatively mowdernt A seal pressure Wa•cewas selected to reduce the conalhutia of pre•ssur-aroa forces on the seal faveto minhiize the wear rate.

The dual bill and rotlletr baring arratnigentnt and th- vari'ho Wace seal peformedsatisfactorily in subseqeuent tests In the copressor ria. md pOvle e ei-pertonce with igh-spoed bearing and seal dvsigni of this type.

)20

w *1

ITI

TEST EQUIPMENT

C-OMPRESSOR TEST RIG

The major nonrotating components of the compressor test rig are shown in Fig-ure 20. They consist of the bellmouth inlet, inlet case, inlt guide vanes, Inducershroud, Impeller shroud, diffuser, and diffuser collector. Prior to assemblingthe impeller in the cases, it was first spin tested with the tooling shown In Fig-ure 21 to 73,000 rpm (112% of design speed) to verify Its mechanical integrity.

Installation of the Impeller into the diffuser in its original spline-drive corfigura-tion is shown in Figure 22. When the impeller attachment geometry was changedfrom the original spline-drive configuration to a CURVIC coupling, the impellerand drive turbine became an integral rotor assembly. The components used tobuild up the redesigned rotor assembly are shown in Figure 23. The inducer, im-peller, spacer, and turbine were held together with a tiebolt. A "Z" brace wasused in the assembly to support the middle of the tlebolt and Increase its criticalspeed to well beyond the design operating range.

Iuring the assembly of the compressor rig, particular attention was given tobalancing operations. Rotor components such as the inducer, impeller, and tur-bine were individually balanced by material removal. Then the rotor wis assem-

bled out of the cases, concentricities were checked for less than 0. 001-in. runout,and the assembly was balanced to less than 0. 0003-oz-in. unbalance. again bymaterial removal from the impeller and turbine. This method put an unbalance inthe Impeller and turbine as components; however, the correction plane was near thetrue plane of unbalance, Aince these two components were the heaviest in the rotorassembly. Final balance of the rotor assembly in the compressor and turbinecases was accomplished within 0. 0012 oz-ln. unbalance with rivets and pins oneach end of the rotor assembly.

The assembled and fully instrumented compressor rig and drive turbine areshown In Figure 24. The compressor rig and drive turbine were Installed in acombination shipping and mount stand for ease of installation in the. test stand.

TEST FACI LITIES

A schematic of the Hl-2 test stand high-spotd compres or test facilitv is showt inFigure 25. The drive turbine was powered by eompressor bleed air from ap&WA"a '-75 slavv engino. Drive turbine supply air was controlled by twopnetumatic valves, one of which was used as vernier control. Since the drive rur-bMae was a radial inflow-type and poeticles in the supply air tend to erode theblade tips, a particle separator was installed downstrear,, of the drive turbine con-trol valves. The design of the particle separator (Figure 26) was a derivative ofthe semi-reverse flow separator, designed and tested under Army Con-tract I)AAJ02-70-C-0oo3. The separator eenter body was free to move and act asan erergency shutdown Malve. Safety systems built into the test facilities includeda rapid shutdown and an abort system. These systets operated automutically whenpreset conditions were eac~ountered. Table tH stummarizes the lntiatIag aortisrequired to activate .he safety coatrols aud thw action takem.

I

i in

• ~Figure 20. Compressor Test lRig Nonrotating Components.

A''

S 17

N I

I~igu~' 22. 1np4g1kd I toUt~o A* -tMublY COWOO t

,~q~23

Figure 24. Assvinb1tKd and 1,ully insti awnd op*o

-uj

uj 0j

. t. cc L

W• ID PfARTICLES

r rFL O W IN V A L V E - _U Q i-

•TO TUN"%&II

i PLOW COAiTA iM

StM9LATCH OPERA

Figure 26. PartIcle Separator With Integral Fast-Acting Shutoff Valve.

TABIIE IM. TEST FACILITY SAFETY SYSTEMS

Rapid Shutdown System

Initiating Action 1. Low thrust balance pressure2. Loss of control room A/C power

Results 1. Turbine inlet control valves closod2. C,•lpressor dischargo valve

opened

3. $lave engiue throttled to idle

Rig Abort System

Initiatiag Action 1. Low lubricating oil supply pressuro2. Rotor ovrpesed3. Matual abort by tesm egW•noer

Results I. xloasivo-uetuated tulain@ Wnletvalve closed

2. If manual abort, turbine ilet uaW-ofold veated to amabient

3. turbine, ihtnt coatrol valves clseod4. Compressor discharge valv aptwd5. Slavw engine throttl to idle

It

The installation of the compressor rig in the test facility is shown in Figure 27.The compressor inlet duct contains a sharp-edged flow measuring orifice, controlvalve, flow strnightening tubes, and a plenum chamber. An inflntable rubber sealsealed the plenum to the compressor belimouth. The control valve in the comn-pressor inlet duct remained in a full-open position during all testing.

i.igure 27. Compresfsor IUt Installed in 11-2 Test Facility.

The compressor exhausted to atmosphere from the diffuser .ollivctor through twoback-pressure control valves, one of which also acted as a .. ier control, flycontrolling the compressor discharge back pressure, it was possible to operatethe compressor along a spewedline with transients into and -at o" stall. A ,urgerelief system was designed and used to detect the onset of siurge. The systemused a high-frequency response pressure sensing L'ansducer in the diffuser col-lector. A control module allowed the selection of a rate of change of pressurethat would actuate the system. If a higher rate ofr change of pressIre was sensed,a fast-operating valve in the compressor discharge duct was opened. This re-duced tve compressor back pressure, thereby automatically allowing the comp ee-sor to move away from a surge CoWditton.

A differential pres.,mre control system was used between the impeller hub sidebackface ard a buffer seal dtn compartment (Figure 15). All performance datawere taken with this delta pressure equal to zero to prevent flow out of or into theflow' pa. at the impeller exit-difitsevr entrance. 1111s pres~Ltre difftrer-1týt -oUtthowever. he adjusted to either side of the zoro delta pressure point, if desired.

The requirod thrust load on the ball bearing was maintaintw by supplVing gaseousnitrogen to the impeller thtust -alanve cavity (Figurv 15) through an automtatioeomtrot valve. This valve was adgutabet to ay ,et pujut k-d ua4 changed khtle thv

rig was in pe-ratiou.

-. - - o~&~-r- - ~ -

IN8TIWMENTATION

Instrumentitioti wais provided to establish stage overall performance, stage comn-

ponent performance, and monitor rig operation. The instrumentation used to take

instrumnentation stations for the test compressor are defined in Tnhle IV, and thlenxial and radial locations of these stations tire shown in Figure 28.

TABLE IV. COMPRtESSOR INSTRUMEPMATION STATIONS-

00 - Compressor !nhyt-Plenum0 -Compressor Ink~it-Bellmouth Throat

1. 0 - Inlet Gluide Vane Exit1.5 - Inducer Exit2. 0 - Impeller Exit3.0 - Diffuser Exit3.5 Compressor Discharge Collector Manifold

INLETVANEIMPELLER

STATION, 0VSANE NSTION 1.5

INDUCER

STATIO I-- 0 -A

STATION 3

~J~'era~l c rfo anc cttiottitcttroi

ltftntrmatatioGi wa provdecd toi obtain Wetc flokw rate.. InpelIter 1peed. pres-.*r11atet eoaditioaa (Sation W4). and crpivcsior ctts hargo ewo itions (Statloo 3.0aaft .11. T hose data were ~onabine4 to dofttw. overtill per~ormaav@ in teR1aM of

-pn~--ttal-restwecanditions (Stationis 00 to 3. 0) and total-pro*urv-to*ati(-PtV**Ute voithiod (Satit6&u 00 to 3 )

Compressor inlet flow rate was calculated front data obtained from a 5.270-1.-diameter thin plate orifice installed upstream of the inlet plenum in a 12. 5-In.-diameter inlet duct, as shown in Figure 29. The orifice was installed in accord-ance with ASME standards. Orifice upstream statie pressure was measured bymeans of three static pres.4ure taps, each sensed by a 0- to 15-psia transducer.Three orifice differential static pressures were each sensed by a 5-psid trans-ducer. The uZwukI thw 43rifie1± flew W" nrdanmnTnl in the 1n14.- plenum.

-ORIFICE PLANE PLENUM

- F N INSTRUMENTATION

* ~~15 Iit . . .

F-_

DOWNSTREAM STATICPRESSURE TAP LOCATIONS

UPSTREAM STATICPRESSURE TAP LOCATIONS

Figure 29. Inlet Orifice 4tstaflatiou.

C'ompressor inlet total pressure was measured with three Mel-type total pressureprobes. located in the inlet plenum. The probes were each sensed with a 0- to15-psia transducer. Cottpresor inlet total temperature was measured with a

osemount resistance thermiometer in the inlet plenum. Redundant inlet teI-perature meoasurements were obtained with three half-shielded copper-constantan(C/C) thermocouples, also located in the inlet plenum. tieginming with Build No. 5of the comprpessor, the CIC thermoncouples were deleted and four additional Rose-moun r•esistatee thr-trteometers were Installed at various radial positions In theinlet pletun to accurately measure oay radial variation in teptroture.

Diffuser exit total pressure was meaisured by meats of th4 ~,ee total preslure rakes.each consisting of three elements, as shown in Figure 361. Each element was

etonsed by a pressure-seanualng systemti us ing a 0- to 150-psia transducer. rhera-kes were- istalled at the exit of the diffuser tubes. shown in Figure 31. Thelocations of the nine total pressurt probevs superimposed onto mo diffuser tubeare shown in Figuroe 3. it was nwowd that the centvr probe ou each rAke was in the

aiame rolative position tI each tube to provide rldundant measuremens at thi.poitt. Also showi in Figure 3? are the seven qal aras that were assignd to

fth totA prnsurs measurd by tOw rake fior mass-averaging pufos.or

A-y

0.0624n. DIAMETER TUBE

O0404&. DIAMETER TUBE

71ICENTER TAP

LOCATED ON(0) ~~ALL PARTS JOINEDDIUSRTEWITH SILVER BRAZE

0IT.1289In.

SUPPPORT STRUT -

TAPS TO BE LOCATED RAKE NO. IAT DIFFUSER TUBE

EXIT PLANE

RAIa MAKE NO. 3

RAKES NO . 2,Z AND 3 ARE THE1 SAME. EXCIPT FOR AOLES Sh4OWNI

Fiut 30 Mfw mt-oo r~ut ao

TOC

£30

2533

24x

RAKE NO.

2 02

3b

A.12S In.L

0.109 I.E

0.213O

V~u ;~. ~t,:±w ' ~00$

*-RV - - -

Compressor exit static pressure was measured by means of eight static pressuretaps in the collector, each connected to a pressure-scanning system using a 0- to150-psia transducer. The taps were located on the shroud-side wall so thatneither diffuser discharge velocity nor the collector struts interfered with themeasurements. The design and location of these taps are shown in Figure 33.Compressor exit total temperature was measured with four chromel-alumel (C/A)shielded thermocouples in the exhaust Pollector. The thermocouple probes wereaxially located midway between the collector manifold walls and circumferentiallylocated as shown in Figure 33. To increase the accuracy of the temperaturemeasurement, the thermocouples were constructed from a single batch of cali-brated special-limits-of-error wire. This wire was continuous from the thermo-couple junction to the reference junction to eliminate errors caused by connectorsand by the lower quality wire normally used between the connector and the refer-ence junction. A thermocouple was attached to the shroud-side wall of the col-lector manifold to aid in estimation of potential radiation errors on the collectorthermocouple measurements.

Component Performance Instrumentation

Instrumentation was provided to evaluate the component performance; namely,the inlet guide vane, inducer, impeller, and diffuser. The component perform-ance instrumentation is described below and is also summarized in Table V.

Instrumentation was provided downstream of the inlet guide vane to measare wall

static pressures and radial distributions of total pressure and air angle (and inBuild No. 6, total temperature). Similar instrumentation downstream of the in-ducer was used to describe its performance, with the exception that total tempera-tare was also measured at the exit of the inducer. The axial and circumferentiallocations of the inlet guide vane and inducer instrumentation are shc.wn in Fig-ure 34.

is Static pressure at the leading and trailing edges of the inlet guide vanes was mea-sured by one shroud wall static pressure tap at the leading-edge and single huband shroud wall static taps at the trailing-edge plane to define the static pressurealong the flow path. All three static pressure taps had to be deleted in the revisedinlet case, Builds No. 4 through No. 6. Inlet guide vane exit (Station 1.0) wallstatic pressure data were obtained by four wall taps equally spaced about the cir-cumference on both the hub and shroud walls. Builds No. 4 through No. 6 did notcontain the hub wall taps as the revised inlet case prevented their installation.The hub wall taps were located approximately on design midchannel streamlines.It should be noted that the shroud wall taps were located on a circumferentialtraverse ring.

Static pressure at the leading-and trailing-edge planes of the inducer was obtainedby single hub and shroud wall taps at the leading-edge plane and a shroud wall tapat the trailing-edge plane. Additionally, four static pressure taps equally spacedabout the circumference on the shroud wall were installed at the inducer exittraverse station (Station 1.5).

33

-,& .q

TUBE .- COLLECTOR WAL

NOTE: 1. STATIC PRESSURr ON FORWARD____ (SHROUD SIDE) WALL

2. THERMOCOUPLES CENTEREDBETWEEN WALLS

STATITRUT PRTYURTA

IF0~~~~ 0IFS ~ ~ 0 0

£0

0 0

00 0

0j0 0

CIFSENTE LIE 0TI.I.RDU

04 10 I

0------ -.. '-.~~..- ~~.~.~*

* r

TABLE V. COMPRESSOR COMPONENT PERFORMANCEINSTRUMENTATION SUMMARY

FlowLocation Variable instrument Type Quantity

Station 0 P Wall Tap 4

Inlet Guide Vane P Shroud Wall Tap 2(1)P Hub Wall Tap 1(1)

N Chord Angle IndicaLor 1

Station 1 P Shroud Wall Tap 4P Hlub Wall Tap 4(0)Pt, T Circumferential and Radial i(2)

Traversing Cobra Probe

Inducer P Hub Wall Tap 1(1)P Shroud Wall Tap 2

Station 1. 5 P Shroud Wall Tap 4Pt, T Radial Traversing Cobra 1 J

Probe with Thermocouple

Impeller Shroud P Wall Tap 14

Impeller Tip P Communicating Groove 4Shroud Side

Impeller Tip P Tube 2Backface

Station 2 P Shroud Wall Tap 4P Hub Wall Tap 4Pt. Axial Traversing Cobra Probe 1T Axial Traversing Thermocouple 1

Diffuser P Shroud Wall Tap 32P Hub Wall Tap 8

S1)Deleted from Build No. 6 due to relocation of front bearing compar'-,, mnt.(2)No Tt for Build No. 3.

A '3A

'A 3

A"i

4 COBRA

3 4-712 PROBE

STRUT IGV APELLERF O 5

i~.1 13 , 3

DEA C ENTRA OOINGUFOR W D

2-3 3.2 210i:•4.7 3.55 90 dog APART FROM

;• 47 dog

;I8-1F3.E5 90 doA APART FROM S~18 dog FROM COBRA

COBRA 3.35 330 TO 15 S12 3.7 3013 TIP INDUCER INLET - 0.05 135

S14 INDUCER EXIT +0.05 15015-18 4.701 90 APART FROM 15

COBRA 4.701 295S0 MEASURED CLOCKWISE FROM TOP DEAD CENTER.- LOOKING FORWARD

i• PROBES 1-13 ORIENTED CIRCUMFERENTIALLY AT NOMINAL(10 delg TURNING) IGV V#IDCHANNEL.

Figure 34. Axial Locations of Inlet and Inducer Instrumentation.

iRadial distributions of total pressure and air angle (and total temperature inBuild No. 6) at Station 1. 0 were measured by means of a cobra probe that wastraversed in both the radial and circumferential directions. The cobra probewas traversed eircumferentially at five radial locations corresponding to 10, 30,50, 70, and 901 spans. At each spanwise location, the probe was traversedthrough a 45-dog arc, which permitted measurement of total pressure and airangle data behind two inlet guide vanes and one inlet strut in the Build No. Ithrough No. 3 configuration. In the redesigned inlet configuration, the probe wasrotated through an are of 52 dog. The cobra probe used was constructed of0. 020-in.-OD tubing and is shown in Figure 35. The Build No. 3, Station 1. 0cobra probe was not constructed with a thermocouple. Radial distributions oftotal pressure, total temperature, and air angle at the Inducer exit (Station 1, 5)were obtained by means of a radially traversing cobra probe.

Impeller instrumentation was provided to measure wall static pressures along theimpeller shroud, impeller tip static pressure on both the shroud side and the hubside, total pressure and air angle distributions across the flow path at the impel-ler exit, and total temperature distribution across the flow path at the impellerexit. The locations of the 0.040-In. -diameter wall static pressure taps along theimpeller shroud are shown in Figure 36. The positions of these taps were&elected so that the data obtained from them would define the static pressure dis-tribution at the impeller entrance, the splitter regions, areas of possible flowseparation, and arrea of possible flow distortion due to influonce of the diffuserpassages. The circumferential relationship of taps were superimposed onto onediffuser tube inlet.

36

-t - -- ~~--

THERMOCOUPLE

ILET GUIDE VANE EXIT ISTATION 1.0) ANDINDUCER EXIT (STATION 1.5) COBRA PROBE

IMPELLER EXIT COBRlFA PROBE (STATION 2.0)

ISIXTEENTH SFigure 35. Conistruction of Traverse Cobra Probes.

SHROUDNO. Z.in. R I

1 0.02 0.1522 2.45053 0.3613 2.45134 0.4163 245235 0.11812 2.52586 0.9721 2.53387 1.0837 2.60618 1.2271 2.76219 1.4W4 3.1533101

10 1.4512 3.389711 1.512 .389

11 1.4512 3.389113 1.4512 3.389714 1.4512 3.3897S3,M

7/

9 SECONDARV'~SPUITTER

/ PRIMARYSPUITTER

R

Figure 361. Location of linwIpler Shroud Static Pressure Tups.

37

Ai

Impeller tip static pressure on the shroud side was obtained from measurementsof the pressure in n circumferential passage that is connected with the flow pathby a 0. 015-in.-wide circumferential groove between the impeller shroud and thediffuser. Pressure was measured through four ports that were equally spacedaround the circumference. This circumferential groove scheme is shown inFigure 37. Impeller tip backfaee pressure was measured by means of two0.062-in. -diameter tubes installed 180 deg apart in the impeller tip backfacecavity. Beginning with Build No. 5, one chromel-alumel thermocouple waslocated in the backface cavity to measure temperature.

FOUR PORTSEQUALLY SPACED I I ,I 0.015.in. WIDE STATIC•:i•• "- ..I I PRESSURE

SJ COMMUNICATINGGROOVE

Figure 37. Impeller Tip Static Pressure Passage.

Impeller exit air angle and total pressure were measured at an Impeller exitradius ratio of 1. 05 by means of a cobra probe that was traversed across the im-peller eý-it flow path in the plane of the rig eenterline. The probe was installedas shown in Figure 38; the cobra probe is shown in Figure 35.

Total temperature across the flow path at the impeller exit was measured with thetemperature probe shown in Figure 39. This probe was designed with an aspi-rated head to maximize the temperature recovery, and sheathed wire was used tominimize the conduction error. The temperature probe was installed in the samelocation as the cobra probe.

38

- . "A

a-a

LL6.

LL

00

oll

L- -

uP.

•ii 39

tzr

MIA

0 Mil

0 L3

z M

W U..

* Illm. id

aa

* 20

The diffuser instrumnenttation consisted of 48 wall static pressure taps. Thelocations of the diffuser wall static pressure taps are shown In Figure 40. Thisfigure also shows the location of the impeller exit cobra probe. Figure 40 showsthat composite static pressure data were obtained in the vaneless space, diffuserthroat, and along the diffuser length. The construction of the static pressure tapsis also shown in this figure; the 0.020-in. -ditimeter taps were made by the electri-cal discharge machining process to eliminate drilling burrs and to ensure sharpedges at the intersection of the pressure port and the diffuser passage.

Special Instrumentation

Total pressure in the plenum, static pressure in the inlet bellmouth, total pres-sure In the diffuser exit, and static pressure in the collector manifold were mena-sured with close-coupled transducers to achieve the fast response required todefine overall performance during compressor stall. The transducers used hada frequency response of up to 100 lIz, and the data recording system could recordeach transducer at a maximum of 65 scans/sec. These response characteristicspermitted good definition of stall characteristics.

Iligh-requcency instrumentation, capable of response between 200 and 100,000 lz,was installed in the c.)mpressor rig in the following locations:

1. Total pressure probe at Station 2 traversing location

2. Two static pressure probes on the Station 2 shroud walls

3. One static pressure probe at the diffuser exit.

Exact radial and circumferential locations in the diffuser are shown in Table VI.

TABLE V1. HIIG0H-FREQ'ENCY RESPONSE PROBE LOCATIONS

Circum fe.rent ial LocationProbe (CCW From TID') Radial Location. in.

Total Pressure 66 deg 5 rain 3.6DVaaeless Space Static 29 deg 30 ain 3.69

Vaiwless Space Static 315 dog 3.69Diffuser Exit Static 18 degThi total pressure probe that was inserted in the Station 2 tr'vrsing echanis-

was a flush diaphragm -nKuIte XCFLW-200A, the three static probes wore KullteCEL-115-200A probes. These se-nsors werempoutted in probe housings, as shownW Figure 41. The probes weiv temperature compenvated to 430W7 and had amaximum operating temperature of 521ý F (which correspds to an exit temper-ature ivached at less than 8W( of design speed).

41

A9

t.). .

29 30 DIFUSER TUBE MM.31 /

222

415d23

13 - O

0 4A

37 -. .A o :r ~ ru ~ U .

-STATIC PRESSUREIPROBE

-TOTAL PRESSURE

PROSE

IVigure 41. iIigh-Frequoucy IH pso Kulates.

M)ata Readout and Recording Systems

The primary data recording system was ,ui automatic digital 11eagnotic tape re-corder. This system recorded all compressor rig data except that from the high-frequency response transducers. The high-frequency data were recorded onanalog magnetic tape.

Information needed for safe operation of the rig and for setting data points wasdisplayed in the control room. Control room data readouts iavladed rotor speod,rig vibration levels, bearing temperatures, oil pressures, inlet orifice differentiidpressure, thrust balance pressure, rig inlet pressure, and rig discharg pres-sure. Impeller rotational speed was obtained from two electromanetie pikups

vmeiated adjacent to a G-tooth gear on the rear of the drive turbitie rotor sht.

When cobra probe data were obtained, the total prssur meoasured by the cobraprobe was displayed on X-Y plotters, located adjacent to the traverse actuatwrcontrols. This allowed changing the speed, or, if necevsary, stopping the vohrauprobe actuator to allow the priot to respoud to changes in the air flow magle.

43

A__

PROCEDURESS C

Shakedown Test

lrior to tie initiation of any performance testing, each build of the compressortest rig was subjected to a prerun (nonrotating) checkout of all rig systems andthen a shakedown test. The objectives of this checkout and shakedown test wereto (1) verify the mechanical integrity of the test rig and (2) check the lnstavmenta-tion and the data acquisition system. Any difficulties encountered (luring theshikedown test, such as test equipment or Instrumentation malfunction, weresubsequently corrected prior to the initiation of the performance tests.

During the shakedown test, rig vibrations and bearing temperatures were moni-tored over the entire range of rig o•pration. Areas of high vibration and hightemperatures were noted, and prolonged running in these areas during the per-formance tests was then avolded. A maximum allowable limit of 160 g's wasset for extended life of the bearings; however. 220 g's wort- recorded during theBuild No. 3 shakedown test. The extensive redesign of the rig bearing systemreduced these vibrations to well within the acceptable level. Build No. 6 vibra-tion levels are plotted as a function of speed in Figure 42. Vibration data weerecorded continuously during the main portion of the test program on analog mag-netic tape, too.

The thrust balance system for the test rig was checked ouL during each shakedowntest, and te thrust load on the bearings was ealuclated at several speed conditions.Adjustments were made to the auitomatic surge relief system, as necessary, sothat it responded to stall assoclated pressure fluctuations and not to the gradualpressure changes caused by closing the throttlo valves or rantdom pressurefluctuations.

The inducer radial, impeller radial, and impetller axial shr•ud clearances weremeasured with mechanical rub probes. These clearance probes were withdrawnand measured after a very low-speed rotation to obtain a cold clearanwe and thenwere similarly examined after tihe itmpeller rct•.ional speed had reached T0. S5,95. and 100F, of design speed. Figuzres 43 through 45 show the clearance ehangesmeasured daring the Builds No. 2, 3, and 6 shakodown tests, r'0*pCti•vt•y.

All insftrumentation was checked tor continuity from the test rig back to the datacecording system. Just prior to the start of each rtul, all transducer and thermo-icupte output voltages were calibrated over their operating ranges., and a set ofambient readings was recorded for all the Insitrumentatioin for eoappariWCs to truetambient eond~tioas. DrAing th shakedown test, overall performawe, componeatperformance, and tranalent data were obtained and processed throiugh the datareduction system !o dewtrmine if all the iastrumentation was ecording propertyand to chek the data atui}itona and the data rttioo systems.

44

I• •:.:

v .

Fi~urý 2.Ma'9-ml w kalgI covaiu U4N

'frm - ctAstoOvvivalt Pet aGi

F ach etwedlfte wa defind by a!italltvua~ivItadb t4d-wý,pit i -v

to thet Pubw iutti ilEro L valt m worett maetohldcioortm t"wd

aev.tl pinlt~ wasthe set.~ tut4t oa ~thd t Ivoltnt tor sy tatic t. c4ay fle 'elpztq.io widtaipu L ~wraigvLrd iao-tllwi I m t'o1tdiagda t a Wigt-esc. n r&*- . T~W-

revrk~j: taL- wa* 1t6 rwas'see durinig Build, Nk) 3 te-st, a~gd XL thLe .a~iaxul~i

454

low- ,

29

IsI

SKD- % ~ --.--

24"0 IMPELLER TIP AXIAL

I IMPELLER INLET RADIAL20 A •INDUCER RADIAL

C,wE 16I

C.-LU

z

4

40 60 80 100

SPEED - %

Figure 45. Shroud Clearance Measurements, Build No. 6.

Steady-state data points were distributed along a speedline, based on the collectorstatic pressure range recorded during the first slow-stall transient. A steady-state point consisted of cycling the pressure scanning units twice at a rate of 2ports/sec, followed by the recording of the rest of the rig instrumentation for10 sec at a rate of 10 scans/sec. A detailed outline of all data points taken isshown in Table VII.

Component Performance Data Acqu~sition

Component performance data were obtained by radially traversing air anF'*total pressure, and total temperature probes behind the Inducer and the hi., lerand by both radially and circumferentially traversing behind the inlet guide 'vanes.The data recording rate for all traverses was 1 scan/sec. Total pressure vsradial or circumferential travel data could be monitored in the control room onan X-Y plotter and the traverse rate governed, as required, to obtain good pro-file definition in wakes and near the walls.

47

- -

If --

d

I

it III

ii

I-- -- -- .4 -- --

I..ZlID

�eI 141111 .41411 - 4�1

-- -- 1404 DI - t, 14.4 4

(n II

I,

0. 0

� 1,11 -" .4- - - =�

0 X.. -

- .4

U

It TI

H -�

� .4-- -- -- - - - -"

�

a 1114..I .1.4.4 0

II H�

0 � - .4 '� -

� �0 �lID OlD 14 - ID�I IDID

� *010010101010 0400 10100 0 II ID ID

�

>10 000 000 00000 000 a o ee� 040 00000 00 y.¶ .400..I14 0.4 .4141410 I114 11 10 .4 I

�j.

48

j

Inlet guide vane traverse data were obtained at near-stall points for all speedlinesat 70% speed or greater. Near-stall points were set as outlined for steady-statepoints. The inlet guide vane exit probe was first traversed radially inwaidsbetween inlet guide vane wakes from the shroud to the probe's maximum travelat 2.0% span (from hub). The probe was then retracted to the 10% span positionand circumferentially traversed to include one IGV gap with a strut wake andone IGV gap without a strut wake. Circumferential traverses were then similarlyperformed at 30, 50, 70, and 90% spans to complete the inlet guide vane survey

4 for that point.

Inducer exit radial traverses were performed by running the probe in to its limitand then recording data while slowly retracting the probe back out of the flow path.

Impeller exit traverse data were obtained on each speedline at the maximumpossible back pressure with the traverse probe in the flow path. This point wasset by decreasing the back pressure in small increments until the probe couldbe traversed across the span without causing the rig to surge. Data were thenrecorded while slowly traversing the probe from the hub wall back into the shroud.A steady-state point was recorded to coincide with each impeller exit traversepoint. At the completion of impeller exit traversing with the air angle and totalpressure cobra probe, this probe was replaced with a total temperature probe,and data were obtained at the same rig operating points.

One special test sequence involved performing stall transients with the impellerexit cobra probe, located at approximately 15% span. Data were eecorded onsix speedlines from wide-open discharge valve into stall at 1 scan/sec and from

V: near stall into stall at 65 scans/sec.

Special Instrumentation Data Acquisition

High-frequency response data were obtained at 70 and 78% of design-correctedspeed. Data were recorded on analog magnetic tape during a stall transient andnear-stall steady-state point for each speedline. Transients and steady-statepoints were set as outlined in the overall performance data acquisition section.

DATA REDUCTION PROCEDURES

The reduction of data was accomplished In three steps: (1) reduction of overall

performance data, (2) calculation of component performance and velocity triangles,and (3) rndl: tilon of high-frequency response and rig vibration data. The arith-metic me.-r value of data from redundant instrumentation was used for all calcu-

£ lations except where otherwise noted, and all performance data except orifice andplenum data were corrected to standard day Let cunditions as follows:

corrected pressure = recorded pressure

corrected temperature recorded temperature

Under this system, Inlet temperatures and pressures are always standard dayconditions (To 518.688°R, Pto 14,694 psia).

49

Overall Performance

The reduction of overall performance data was accomplished through the use of anIBM 360-75 computer program. A DRIL (Data Reduction Input Language) programconverts raw test data into engineering unIts, ratios, pressures, and tempera-tures to standard day inlet conditions, averages data from redundant instrumenta-tion and from successive data recording cycles, and performs overall perform-ance calculations. Actual weigat flow was calculated from the orifice equation

SWac 14. 675 11 - 0.302 "To

0

where

P = upstream orifice static pressureaP = orifice differential static pressureTo= plenum total temperature

Dott weight flow and speed were corrected to standard day inlet conditions asfollows:

w-.-r aWct

andN

Ncor N

Overall temperature ratio, total-to- static pressure ratio, and adiabatic efficiencyare given, respectively, by v

Tr = T 3/518, 388

Pr P 3/14. 694

andideal (isentropic) enthalpy change - Ahf(Pr)

actual enthalpy change Ahf(Tr)

where Ahf(Pr) and A hf(Tr) were determined by fourth degree curve fits of changein enthalpy vs pressure ratio and temperature ratio data from Table I (Dry AirTables) in Keenan and Kaye Gas Tables. Two separate pressure ratios and cor-responding efficiencies were defined, depending on the type of data point. Forsteady-state points, the average of s' collector static pressure taps was usedfor the values of P3 in the pressure , quation. During stall transients, thepressure ratio was based on the valuo oi single collector static pressure tapread through a close-coupled pressure transducer. The value of this tap generallyagreed to within less than 0. 3% of the collector average during steady-state points.

A50

05

Inlet Guide Vane Performance

Inlet guide van3 performance was calculated by an inlet guide vane update routineto the main DRIL program, which includes the effects of the inlet struts, locatedupstream of the guide vanes. The static pressure distribution behind the inletguide vanes was assumed to be a constant equal to. the measured shroud staticpressure at each point, and the temperature was assumed constant across thevanes.

Guide vane discharge total pressure and air angle traverse data were mass-averaged over both one gap including a strut wake and one gap without any struteffects at 10, 30, 50, 70, and 90% spans. These values were then weighted,based on the relative number of struts to number of inlet guide vanes, and com-bined to give a single mass-averaged value at each percent span as follows:

t3P + 7tv

t 10

and3K +7

s v10

where: the subscripts s and v pertain to gaps including a strut wake and gaps notincluding a strut wake, respectively. A spanwise mass-average total pressure(Pr ) and mass-average air angle (31) were then calculated from these weightedva ues. The mass-average of any quantity (X) is given by

X- wX

s.- i=l

whereAW =K(AA)Pt sin a

Losses were calculated based on the values of Pts, rtv, and Pt at each percentspan and an overall loss was calculated, based on the spanwise mass-averagetotal pressure. These losses were all of the form

14. 694 - Pt•:, losslos = 14. 694

il 51

Weight flow was calculated by two separate methods for comparison with theorifice flow calculation. The first method involved integrating the incrementalflows, AWs and AWv, which were obtained during the inlet guide vane total pres-sure and air angle circumferential mass-average calculations across the bladepassage. The results of these integrations were then weighted to reflect the rela-tive number of struts and vanes and summed to yield an integrated flow. Thesecond flow calculation was based on the IGV exit spanwise mass average totalpressure:

KA (Ptl) sin a I

Wcal

where

A = cross-sectional flow path area at the IGV exitT1 = 518.688, same as inlet

Based on this calculated weight flow, an inlet flow coefficient was determined asfollows:

C calWcor

Inlet guide vane exit velocity triangles were calculated by an IBM 1130 digital com-puter program, Centrifugal Compressor Data Reduction Program (CCDRP).Again, static pressure was assumed constant across the span. The ratio ofspecific heats (,) was determined from a curve fit of "Y vs static temperature(Ts) data and involved an iteration on the value of T. to satisfy the following:

tl (TTs11) /-

The absolute Mach number (Mo) was calculated at each percent span by iteratingon the value of Mo until

(I= +'-Y Mo2

The local speed of sound was calculated by

a = (YGRT )1/2

¶152

a Y R sI

From Mo, a, and 5 the components of the velocity triangles as defined below ateach percent span were then calculated.

U

Inducer incidence (i) was determined at each percent span by

where + is the inducer leading edge metal angle at the corresponding percentspan,

Inducer Performance

Total pressure, total temperature, and air angle data from each Inducer traversewere first plotted vs percent span on a CALCOMP plotter. Up to 25 points fromeach plot were then selected and input into the 1130 CCDRP program. A constantspanwise static pressure distribution equal to the measured shroud static wasassumed. Values of I Ts, and Mo were calculated for each point by the sameiteration methods used in the inlet guide vane section. Total temperatures werecorrected for Mach number effects by dividing the recorded temperatures by arecovery factor. The recovery factor used in the calculations was a linear approx-

4I imation of the actual probe recovery factor vs Mach number calibration data. Themaximum error created by this assumption is approximately 0. 1% of the temper-ature reading.

Total pressure and total temperature were then mass-averaged across the span.The sum of the incremental %%ca-jht flows from the mass-averaging routine waschecked against the correspondir ; orifice flow, and an iteration was performedon the air angle values until the two flow calculations agreed. This air anglecheck was necessary since errors of 1 or 2 deg in the alignment of the air angletraverse probes on the test stand are not uncommon. Final mass-average totalpressure (PtI. 5) and total temperature (TI. 5) values were then calculated, basedon the adjusted air angle profile. Inducer pressure ratio and temperature ratiowere calculated as follows:

Tit

r(l-1.5) 5.

T3 '1 . 5

Tr1153 1.-'

Inducer efficiency was calculated by the same methods as the overall efficiency.Inducer exit velocity triangles at 10, 30, 50, 70, and 90% span were determined

by the same methods as in the inlet guide vane section.

Near-stall traverse data at 100% speed, 10-deg IGV setting and 101% speed,-4-deg IGV setting were also input into a streamline analysis computer programto calculate the actual spanwise static pressure distribution. This programsatisfies the equations of motion and radial equilibrium using input values of totalpress-ire and air angle, a description of the flow path, and a design blockage dis-tribution. The blockages were adjusted until the measured shroud static pres-sures vere matched. Linear approximations of the resulting static pressure dis-tributions were then input into the IBM 1130 CCDRP program, and new velocitytriangles and inducer performance parameters were calculated for these two points.

Impeller Performance

Impeller exit traverse data were reduced by nearly the same methods as theinducer exit data. Static pressure was assumed to vary linearly between themeasured values at the hub and shroud. Total temperature was input in the formof traverse data and also as a single value equal to the collector total temperature,which was assumed constant across the span.

Flow at the impeller exit was supersonic across much of the span at the higherspeed points, which would cause a shock to form in front of the traverse probe.For data taken where the local value of Pt/P was less than 1. 893, the flow wasassumed subsonic and impeller performance parameters and velocity triangleswere calculated as in the inducer section. For values of Pt/P greater than 1. 893,the flow was assumed supersonic, and a shock correction procedure was incor-porated assuming a normal shock to determine the upstream flow conditions.

The mass-average radial velocity (Vm) and mass-average tangential velocity (Vu)were calculated, and from these a mass-average air angle was determined.

a2 tan V /

V

slip factorU

The slip factor at the actual impeller tip was determined from the instrumentationstation slip factor by assuming constant angular momentum Combined inducer-i.,pcller performance was also defined by an internal flow analysis involvingstation i ftor byuasimn g constant ang momentum . Combned inducer-

sisted of overall total pressure ratio and total temperature ratio, rotor speed, massflow, impeller exit static pressure, configuration geometry, flow factors at theleading edge of the inlet guide vanes and impeller exit, and a calculated temper-ature rise due to shroud friction, based on rotor speed, density, surface area, andfriction coefficient for bladed disks.

54

A

Diffuser Performance

Diffuser discharge coefficient was calculated as

Wcor

d w

KA*Pt 2where W::

A*= throat area

P t2 = maximum Pt at impeller exit from traverse data

Throat blockage was then determined by:S~B**

B* = 1 - Cd

Diffuser static pressure recovery coefficients were defined from the impellerexit to the collector and from the tlrodt to the collector, respectively, as follows:

P3 P2

C

,= P3"pQ 2

P _ P*3

t2

Diffuser effectiveness was defined as

Cpi*

where Cpi* Is the ideal pressure recovery coefficient that would be obtained witha one-dimensional, Isentropie flow through the same diffuser. For the Cpi* cal-culation, a throat Mach number of 1.0 and a I1 of 1.4 wore assumed.

Diffuser losses wore documented by the following:

Diffuser loss (total-to-statio) Pt 2 " 3

Pt2

P _PDiffuser loss (total-to-total) =

*Pt2

55

p -pt2 3Diffuser loss coefficient =Pt2 " 2

Dmnplo p t_- pt3-3Dump loss

where Pt3 is the mass-average total pressure at the diffuser exit.

Diffuser performance was also obtained by separating from overall performancethe impeller exit conditions calculated by the internal flow analysis, describedin the impeller section. Performance derived from this analysis was comparedto that obtained from traverse data.

The Mach number profile at the diffuser exit was calculated, based on diffuserexit total pressure data and collector static pressure and assuming a V of 1. 4,Total pressure data from different pipes was first adjusted to match a singleaverage value at the center of each pipe to elimiinate scatter in the absolute profilesof individual pipes for this presentation.

HIigh-Frequency Response and Rig Vibration Data

High-frequency response (Kullte) data recorded on analog magnetic tape wasprocessed using a digital Fourier analyzer system to determine the magnitudeof pressure fluctuations, blade wake definition, and rotating stall characteristics.Resultant data plots included spectral plots (amplitude vs frequency), trackingplots (amplitude vs time), and various correlation plots.

Rig vibration data recorded prior to the installation of the digital Fourier analyzersystem was processed in a similar manner, except that a wave analyzer systemwas used in which the data were not first digitized. Resultant data plots weresimilar to those mentioneai in preceding paragraphs.

VALIDATION OF TEST DATA

Estimates of the uncertainty of the data acquired from the compressor test rigare presented in Table VI1I. These estimates Include both the uncertainty of thesensor and of the recording device. All uncertainty calculations, with the exceptionof those for weight flow and efficiency, are applicable to any reading. Weight flowand efficiency uncertainty were calculated at the design point. Uncertainty of airangle assumes no alignment error. Both bias and precision errors (precisionerrors are two standard deviations from the mean) were used in the uncertaintyanalvsis: these errors are also presented In Table VIII. Whei multiple probes,were availab)e for redundant measurement, the precision error was calculated bystatistically averaging individual measurements by the root-sum-square methodas illustrated below.

5

56

0 Lif-

*' 0 0 0 w 0 4

44 .

V 3 U -N~

00 0

60

'.44

C4c 4 C

SO4so

~~C can

o ~ 0o~0 I' 50

where u overall uncertainty uf flow v-11 lablee uncertainty of Individual sensorn number of sensors recording the same flow variable

Table VI11 also includes the effect of the Onta uncertiaity on the uncertainty of theoverall efficiency performance calculation. This performance uncertainty esti-mate was calculated by differ-entinting the efficiency equation and Inputting theuncertainty values of Table V11I in the resulting relationship. The uncertainty ofthe efficiency v~alue was 0. 55 point in efficiency, which was considered to besatisfactor Th uncetinty of the weight flow ws calculated by combining thetotal his land precision involved in measurigwihtlwwt h ne nj instrument~ation us ed for this program. The total uncertainty of +0. 059 lb/seec isbased on the design flow of 3. 1 lb/sec.

At each steady-state point, approximately 20 scans of data recorded over a 2-secinte rval were averaged. A typical printout from a near-stall, qedy-qta~te 'pintat 1W,(' design speed and -4--deg inlet guide vane setting Is shown in Table IX,A 2-sec average value, the maximum and the minimium value recorded, andI3 sigmia (three standard deviations from the mean) are ilso listed for each in-strument~ation. Those providle a measure of the actual scatter in the data dueboth to the uncertainty in measurement and to slight speed and flow variationsthat may have occurred while recording the steady-state.Integrated flows calculated from traverse datta at each istu naiostatowere checked for correspondence with the inlet orifice flow as is shown In F'ig-ures 46 through 48 for Bluild No. 4i. Calculated wveight flows, based on mass-averaged values of the total pressure and air angle, are also included on the inletguide v~ane flow correspondence iplot in Figure 46,. The approximtate 0,40 Ibmn/secshift in the inducer exit integr~ated flow shown in F~igure 47 can be attributed totwo fato. The first of these Is the con';tat.ý cipanwvise static pressure distrillt-tion that Ws assumed for the reduction of most of the traverse data at ti nstrumentation station, As can be seen in F'igure 47, the flow iesrrespoadeticeImproved for the two avar-stall points at 100 and 100~ speed, which were reducedassuniing a linear static pressure profile, which more closely approximated Elhecalculated static pressure distribution at thAt station. The Impeller exIt lnte-grated (low shown in F'igure 48 correlatets well with the orifice flow over Most ofthe flow range, hut fills off slightly at the higher flow rates. ThIs can be attrib-uted to 1age dfrecsbtenheaual static preteaure profile and theassumed haear static pressure profile for those points.

Prior to the. calcuilationt of component performance parm eters and Velocitytriangles, the in-ducer exit and the- impell;er exit integrated flows were adjustedto match the inlet orifice flow, and thus satisfy continuity requirements. Thiswas accomplished by adding a eonstant value to the measurc - airý angles sa thatthle Impeller exit airflow watched that Measured. It was felt tw t the absolutevalue of the air angle measurement was the least avcsurate of the data used inthe weight flow calculation, although the profito was considered accurate. Hoto-ever, as pointed out earlier, the assumed statie pressure profiles at the, Inuverexit and the impoller exit and the assumed total temnaerature profile at the tol-p~lter exit also contributed significantly to the- initial integrated flow variatid",especially at the higher speed, higher ptvessutv ratio paints.

58I

)P

0I- AI .0

-11

*. k o 4Ak IN .4

id~3I A Z '

~A t

Al Oflt422 u~yOS 00 0 W's 0N

At iiL It At f

.44, 14k

m7

[ tI ; I , , . , ,

I * .1

Ol "t0 a 0I,

4 . I .IRI

a ,

*4 ~~~~~~~4UU144 444.44 U.4

AS e r1cse•k I to4 • a 4ie 0 j V4 ftk 4 ak ,2 .2 1 '1 4, j. j 8

I:.ae*...a.8ueesjme..e....a qk•• eegL ed.

'A I 40 foist It | P%

C all 112 ..&*I

it-ILIU i;

HE , At

. .. .. .....

CVVifY.?7 JgL.l2Mt.d>

. 0

0

0)0N~ zJ. .

b0)

00

Nu

t4

00~0

0 Z

000

*0: uilv I NOU.WS

61

31 0 Tmvers Temparatui? Profile InputS-Copstant Temperatu Profile AU2l to / '

'.4 - ~CoUlctorlotaM T ,eattxe Asswed/

3.0

_ . . . . .....

1.4

t, //. .i

1.0

0.60.6 1..) 1.4 IA 2.2 2.6 .3.0 3.4

CORRECTED ORIFICE FLOW - 3bm/sec

Figure 48. Impeller Exit Integrated Flow Correspondence, Build No. 6.

Measured collector static pressure readings from all taps were within the estimatedI.6 precision of ±0. 34 psia and generally read within ±0. 1 psia. The measured col-

lector temperatures exhibited an observed maxlknum deviation of ±0. 75% about theaverage at the same condition.

Combined inlet guide vance, inducer, and Impeller performance obtained fromI

traverse data showed the same trends as the performance generated by an in-ternal flow analysts as previously defined in the Data Reduction Section, althoughthe absolute performanec; levels from the traverse data were higher. Table Xcompar-es the performance obtained from traverse data and that obtained from theIinternal flow analysis for near-stall points at 95c/% speed and lo-deg inlet guidevane setting and 101% speed and 0-deg inlet guide 'ane. Also shown in the tableare the diffuser losses obtained by subtracting the above Inlet guide vane, inducer,and impeller performance from the overall performance at those points.

62

II

C 4q

00 C

ri2

-- 4

00

IC-4

~ I c'

I63

RESULTS AND DISCUSSION

Initial Performance

Overall performance data obtained during the Build No. 2 shakedown test indicatedthe existence of a performance deficiency, which data analyses revealed to becaused by an impeller-diffuser mismatch. These analyses resulted in a decisionto bore out the diffuser throat to provide a better match for the impeller dis-charge conditions.

The overall performance data indicated that, if the compressor had operated at

design speed, the compressor would have been down appreciably in efficiency(Figure 49) and flow (Figure 50). The test data were used in conjunction withanalytical computer programs to separate the impeller and diffuser performancecharacteristics to isolate the cause of this degradation in performance. Thisanalysis resulted in two conclusions. First, the diffuser was setting the maximumflow rate for the compressor, as indicated by the diffuser loss characteristicsshown in Figure 31. Vertical diffuser loss characteristics, such as those inFigure 51, are commonly associated with a flow-limited diffuser. The secondconclusion was that, as a result of the reduced inlet flow rate caused by the chokeddiffuser, the impeller incidence values were higher than design, resulting in aloss in impeller efficiency.

Further analysis of the diffuser revealed that the diffuser was flow limited due togreater than anticipated throat blockages at 70, 80, and 85(X of design speed, ascan be seen in Figure 52. Extrapolating the throat blockage data to 100C speedindicates the blockage will be 1-19' compared to the design value of 85X. Byassuming a 14% blockage at 100% design rotor speed and that the impeller willmeet its design goals at design flow rate, i.e., incidence, it was determined thata diffuser throat area 7% greater than the current design was required. There-fore, it was decided to bore out the diffuser throat by 79 but to retain the diffuserdesign throat length-to-diameter ratio of 0.5 by aýso reboring the 3-dog cone.Changes in diffuser leading edge radius ratio, leading odge Mach number, anddiffuser exit conditions resulting from the increased throat area weore considorednegligible.

OVERALL PERFORMANCE

Overall performance data for Builds No. 3 and 6 of the compressor are presentedin this section. Build No. 6 of the compressor was tested with a damaged diffuser,and its performance was somewhat impaired. An analytical representation of thepvrformance that this stage would have produced without this damage was preparedusing diffuser porformance from Build No. 3 ; it is also preseated II this section.

Data for Build No. 3 at the design inlet guide vane settiu, of 10 deg are prerentedin Figure 53 and for inlet guide vane angles of 0 anti 20 deg in Figures 54 and 55.At the 10-deg design lGV setig, tile flow rate was 3.03 lb/see, compared tW thegoal of 3. 1 lb bsee, an indication that the induceri-impeller wai proftuing lesspressure tha anticipated. The maximum pressure ratio at aesign speed was 9.34

.-at ,an. .*A d e s5 o gn prmved a pressure

ratio exceediag the required 8:1 proosuro ratio at an officieacy olightly les* thanthe W(-7 goal, 79.6'(.

64

1 4

Solt

IVLt

Figure 49. Overall Performance, "uIW No. 2Shakodown Test.

6A

Fi~r4~58.llgh-Sp*dt Ceowtrtual Caupteaor Flow~ ChraIotodaiti.j

k4k

ViuN52 bfue TrotBlmýiut Carvttiiio Bil o.2 ~atoU l0t

too

so 0

40o 0

000k 20

10

44

4 0000

'C0

S0

IX

& MCWww~ Lwrl M

Mt~~S. Ovt't'att tlvrfttkrairww. Uttit V-.. 1, tvadwt B3V.

$00

020

200

XU SftaWUtrn

w~t74-4

IjI

I: ur 54Ivrl tfrwo ul N 3 -o GI

0 0

8000 0

400

w 60

200

£0~1. 0 tlyat

KHl St#l4 Tanuie

0 iuv5.HiJN. vralPmruiiý 0dcWV

F6

A?

The trends exhibited by these data lead to an interesting speculktion that the de.-sign conditions Inay have been attained by overspeeding the rotor approximately2% at the design inlet guide vane setting. High rig vibratimns prevented furtherhigh-speed testing.

The compressor maps obtained with inlet guide vane tettings +10 deg f-om thenominal were limited to a maximum speed of 95W of design due to rott. vibrations.The higher rotor work capnhility of the 0-deg M(V setting raised the maximiumpressure ratio from 8.2 for the design IGV setting at 95•% of design speed to overA. 6 without measurable loss in efficiency. This change in IGV angle also in-cronsed airflow by 4.5fWY. The improved performance at the 0-dog IGV settingIndicated that this setting may have offered the best compromise for attaining theperfonmanc, goals it high-rotor speeds. Testing of Build No. 3 was, howevor,terminated due to a bearing f•ailure associated with the vibratito problems.

lPoformawee testing of Build No. 6 after the rdesign of the rig bearing and rotorsyetem campleted the high speed tests initiated with Build No. 3. Data are pre-sented in Figure 56 for the various combinations of IGV settings and rotor speeds

SFthat were tested. The data show that the tm itnum pressure ratio reached at a-4-deg TGV setting was slightly above the design goal (10:03) at a rotor slpved ofI`( in excess of design, and the corresponding maximum efficiency was 1. 2 pe•---.. age points low. At the :1 prossure ratio point, the efficiency decrormit was2 points below the goal.

Comparing this off"Cieay tm•oss with the prvvious data lWd to the conclusion that thedamage sustained hy the diffuser was more serious than extected. Damagetwettrrvi when portions of the sitver plating on the impeller shroud broke off andSimpiged on the leading edge-s of the diffuser during an earlier impeller-to-shroudrub. A evinplete dicussioa of the problem is presentee .:a the diffusetr componentsection. The B)Aild No. 6 testýg also included a spwcial test in which nitrogenwas used to cool a thormal dam machitned between thle impe!ler shtoud fiange a0ndthe ind~u-t shroud flange to e'iminate heat transfer and to create a thersal[L-awirom-ont similar to that of Build No. 3 or a gas gmetrator with thin cwes.

i! This test generated a pressure rtio, of 10. 04 ,! and aP efficionv ;f 74. 3ý at 101,of desiga. spaeed and -4-(", 1WV settiog. TW*wi* 'ao anitp~wrovotu of (. 511 du toltho thoutual datu.

V*ting data ftom Builds No. 3 and 6 eompvsfior testg, a Vomposito pfwrfoiuaac-ýmap was f mulat i for thji com'resAor. This mtap, 3giro 57. uoes th@ BuildX'o. 3 data at rotr •eeds of 93• of dsign and below. At desiga spoed, vata fom

a u od . @ *tall traasieqt (n ta saedata po3intp wier oddhwrusdWht wore adjuSted to cn peaaWte for the Ledep fomanee of the dmagtA (If-ftvier uting Build No. 3 diffuser performawce dwta. ThI* itmpsite data ±iapoiftwfc that the ie mfpressoir achietVd It ao rskuwý ratio Pact efie."- g~i1.

Thy gffite thrOughout t)W *p00d rang MWa opimte I:W V-dly~fin tOW W@e PId&v aaes ai "o4d on ,he figurv.

* 4. 1 '"r~~49 ,~~.w% aq b _~~wh~ t . • .-' ." _

t - .

•i + 4

• . oiC: e.

t+a+, oa

i+'• • •+. I• • • • P'• • "

"p 24',

S. . •0,

-4

9 C+ 1 + LJ% :•+ + +

4m $+~ It m ~m•mm" .++•n

; ,.i, - _+ ! , +-3•2•5.)+.-U

t ,+++ .+ -• '+• -+ +m . ,, += , ,.,,+ -6

• • ?+ : +- -

I-

I

I;ue1�' I

0

S

S

C / � II

I.

I' // I� �1

-,

� @� _______

11II

I..

COMPONENT PER FORMANCE

A discussion of the performance of each compressor stage component is presentedin this section. A representation of the cumulative performance at various stage (.4 component interfaces is presented in Figure 58 to provide a perspective on thecontributions of the various components to the overall stage performance. Data

* •from both Builds No. 3 and 6 are used in this representation. It may be notedthat the inlet section, including the IGV losses, has a small effect on overall per-formance and that the impeller performance is not sensitive to operating conditionsthroughout the speed-flow regime. On the other hand, the ir. ducer performancepeaks at about 2. 9 lb/sec flow and then drops sharply at higher flow, suggestingthat a significant improvement in efficiency may be possible through a. evaluationof the cause of this loss and the accomplishment of corrective modifications.

Inlet and Inlet Guide Vanes

The inlet and inlet guide vanes were analyzed for the Builds No. 3 and 6 configura-tions. Tabulations of all inlet guide vane data are presented in Appendix I and canbe used to further define the characteristics of individual vanes and struts. Inletguide vane turning for the Build No. 6 configuration is shown in Figure 59 as massaverage turning angle vs inlet guide vane setting. These data show that the massaverage angles agree well with predicted turning angles. Although the averageturning matches the design prediction, the flow was underturned at the tip, com-pensating for general overturning over the rest of the span, as demonstrated inFigure 60. This figure presents mass average exit air angle at five spanwisepositions for the 10-deg IGV setting at three rotor speeds for Build No. 6 and onefor Build No. 3. The different distribution is evidently related to the Build No. 6converging configuration, since the Build No. 3 data using the same guide vanesand setting produced a more nearly constant distribution with slight overturning Aat the tip.

The redesigned inlet also produced a greater loss than the original design. Mass-averaged total pressure data from Builds No. 6 and 3, as shown in Figure 61,demonstrate this increase in inlet losses. A sample plot of the circlimferentialtotal pressure traverses used to generate this loss data is shown in Figure 62. Asample radial traverse is shown in Figure 63. The probe has been shimmed awayfrom the hub wall by approximately 0. 020 in. These data show the trend to in-creasing losses toward the tip of the guide vanes. This may also be seen in Fig-ure 64 in which the losses from the hub through midspan. are low, while 70 and 90%span losses are quite high for data taken at 100 and 101% of design speed. Thereis a considerable Increase in losses generated when the guide vanes were set toproduce negative turning. The increase in loss at 90% span with increasing pre-whirl may be a function of the clearance between the vane end and the outer wall,which Increased in any position except axial. At other spanwise locations thelosses minimize at 10 deg, which was the design setting.

The inducer inlet conditions generated by the guide vanes are described completelyfor every traverse point in tabulations in Appendix I. A typical tabulation (e. g.,100% speed and 10 deg of prewhirl) is presented in Table XI. Inlet relative Machnumber and incidence are presented as a function of spanwise position in Figure 65.

Pi

73

... . .....

4 .4I

*1$1

o11 Q 07 .4�)

I' -CrJ2

...

0

C)

I2

d � 0 IC in0

'0

�;i;� �

ii4

___________________________ 00

% a�aAw a�isiu� Ip.

74I 4

I

--- yaqqEw . -~'~

-- ' -rwrWV ~r~qnwvty-.qpp~fl .. _________4_

A 'a.... -- -- - -~ ~ ~~.~cfl ~ ----- - - -- - -

ib

ct4

A.A

.4NNN -A D

Sdd d 75

Ar

0 I04

* II

ewq I'd** e H =H TVI HD1AV)

U1 -sf .76

II

4W to. j.# 1

vtw4"

Fiui 62.&VCrufrnilTao" BidN.6 W PW 0dgI

t7

77777

.44

oaI-1AA44~

0 14.~M~4~

ig t- IC xi RdulTavroBul N.ii/Ner tal

-. * ' - * . r...' *

t�L - M

woCOOO U

0

U -U*4

�4

K

0L4

CC.0

I gp� IS

�1w U

U V

r

C*

'4

C9% 4

'3 1%ci I

0 ¾I

aa a a aa a a

)

I79

- - -.- ,."- -. .4- - - 'p0- .-. �

'tll I, ...-. o o

* .. * * . ,7777". S S * * S

o 0

Ca~~P- W4~0o 00

0000m0 N 0 N P4 N N

V-4 P o6 4 0. 0- t- 'o V0

P-4 W4~

-6 W54 -- t ý 4

Q~~~ 00 +- ol -4 M0 0k

-o m

-1~~~~ ~ ~ ~ a00 ' o 0-- týlCIc-6 - -d 0 M 0 .2 - 0 0

04 t~ 0~ 1Q0-P a o

~i . 4 5 5 S S S S S S S S S S CA

~~fg

- -'u: '.te5

W.5

.i~ ~.Scp C

10 f.

1.4

-0- T#" Deba3. ~- DuWa al 1011 PMSMWe VAUOc

9 O 6 g o

TiWA

*iur 64 a"oSo oiis 0%Sod 0JgIV

The inlet relative Mach number spunwise distribution and level are both in closeagreement with design; however, Incidence is more positive (stalled) in the outerhalf of the passage than anticipated. Increasing the speed and reducing the pro-whiri (for instance, to 101% speed and -4-dog of prewhirl) raises the relativeMach number and increases the amount of stall incidence, as shown in Figure 66.

Static pressure profiles generated by wall taps through the inlet ire presented inAppendix I for both Builds No. 3 and 6 at comparative conditions.

X1.2

09

7

VA4

WA

Figure U6. ladacet' Wnet Cwiilt~on, 101% Speed, e4-dog 1GV.

- . .. i.- -II~ - ~- - -

Inducer

Analysis of inducer performance was divided into Builds No. 3 and 6, witht amajority of the traverse data used obtained lit Build No. 6. guild NO. 3 inducerd -t a at 95"'~ speed and 10 deg of prewbiri were reduced using fiflet conditions frontlt -e only available inlet guide vane traiverse point during that build. Thlese data

-t. are shown in Ii-gure 617. Inlet relative Mach number distribution was itt goodagreement withi design values, hut lower overall due to the decrea sod speed (9515).The average inducer Incidence angle was about 2 degr abov-' the design value be-cause flow rite and, consequently, axial veloct~y were slighitly below design. Theinducer pressure ratio and efficiency for this point were 1. 67 and $5. G%0 respec-tivoly.

x

1.0

OA~

303

The inducer discharge conditionr: calculated for this point indicated that the Im-peller was operating below deMgn incidence over most of the span. Tile lowincidence in the hub region was caused by overturning in the hub region of theinducer.

Sufficient data were obtained in Wuild No. 6 to completely define the blade elementand overall performance of the inducer. Inducer pressure ratio and efficiencyare presented as functions of weight flow in -Igute 68. The inducer exceeded itsdesign pressure ratio, but at a lower than goal efficiency. The cfftcikncy of theinducer decreased markedly with speed above 95(7 of design speed; pressure ratioalso decreased for constant inlet guide vane settings. The low prewhirls at 101Q/Yspeed increased the pressure ratio and flow of the inducer, but efficiency monoto-nically decreased with increasing flow. Inducer exit traverse data also help ex-plain the loss in performance at high speed. A traverse plot at 101t speed and-4-dog inlet guide vane setting in Figure 69 shows that a severe idlucer exit airangle and total pressure profiles wore generated when the inducer was operatedat an inlet relative Mach number consilerably higher than design. The radialtravel increases from the hub to shroud, with the shroud wall being at approxi-mately 0.&2 in. Tabulated data for this point are presented in Tabe XII.

30

-Laker bat&

Ono

i ?:J.

I:. .

$�$� �*�ssJR�

Th*Ai.

w

-I. /

/'4

* U

* g $-� t .- .

. .

C. . -. . 1� . *

7 �.- .- ,.

-� *j�, w

£-. � ..- �. 1,.CA 4 C �J1� I

- I I II I.4-.-. 1.

� '� I I I ii

- I�

-�

� *1 - � �

I ' 4 1 4 I 4$ .4

- - 1 r�

=�i� +.AI�i- � t F�

4 C- fl4I�gm�i� ____

� � �Itd � �. � 4 tUg. Wi4�'� �

.1

s• - - •• =• ....... 4flf-l• ••W. W .=. r r•--IJ *÷ , •W•'7r •wrw - *- ..: 4 ... ....... r- . r.. s -.* .l • =•: •

Reducing speed to 100V of design ind increasing prewhiri to 10 deg, which a1l1wsthe inducer to operate at design inlet Mach number, improved the ni.spn prob-

em w~th the indu~cer. The majority of the inducer inefficienc, at the design joilntappears near the tip as tota, pressure loss, overtu-ning, ind excessive temper-tur 4e rise, as shown il Figure 70 tand Table XlII. A further decrease in speed to

95'T of design speed, as shown in Figure 71, continued to improve the perform-axtee of the Inducer', allowing it to retich efficiencies of over W., at pressureratios of over 1.7:1.

/r[

--Z

WALL - A L-

1~ OS

C; 0 E

* p

0 D 0 ; ,;.-ý1

- 00000' 00 (D

0 - o

W,.ý 00 cq LO00 j000V

U 2 to 4 - C 1ý 0 ý 0 0 t

f4 Q o )

L it- C11 cc

LO -l 0 t- .-v

ODC' 0 ) C'- Cl -

m00*4 CQ co0 - CII

Cl ' 0 m o- coo cotL-~-' t- .-4 -r ."

t- 0 m tn ý4O -C C

71 00 ý 4 Cl M

c' 4 co~ Id 0 ;) C1 O

m~~~0 Cto00tot

QC Q

0 D 0 f c 141

t- 1 000 --C ýC ýw o0m0 l

IV Ito IdA

10 1

-.. .. ~~~~~.. -. - -------- . -.... "

TL ---- 3

.. 70

Fiur 71 Inue7xtTaes, ul o ,9%See,1-e GNa

Stall

K8

"it =M=

Inducer loss coefficient is presented in Figure 72 to demonstrate the level of im-provement from overspeed (101%) to design (100%) conditions in conventionalterms. The effect of the higher inlet Mach number and incidence on the inducerlosses is easily seen in this figure. These data, coupled with the fact that thefalloff in inducer efficiency at high flow shows a dependence on inlet guide vanesetting, lead to the conclusion that these losses are largely ciependent upon thehigh inlet relative Mach number.

Inducer exit traverse data are compared with impeller design inlet conditions inFigures 73 and 74 for 100% speed and 10 deg of prewhirl. Impeller inlet relativeMach number distribution and level agree well with tho predicted design values;however, the incidence distribution shows the impeller incidence to be more

negative than the design value, as did the data available frcm Build No. 3. Thehub side overturning that causes this negative incidence is most apparent in Fig-are 74. Similar information is presented for the 101% design speed, -4-deg pre-whitl condition in Figures 75 and 76.

0 100% Speed, 10 dIesGV

0.48

0.44 00

S0.40

0.36

0.3

0.28 00.•24

(X20 0

0.04

02 0 40 11 8 10

Figure 72. Inducer Loss•s.

89

0.04-- ,. r'- '-'0

A.A 20 40 60 so.0

.00

0( V

I v

90

'.4

00

a up "V lu &V 10, 0, -

/C

'o 0 EC

c0W -40A

The bulk of the traverse data were reduced by assuming that the static pressureacross the span behind the inlet guide vanes was a constant pressure equal to thatof the shroud wall behind the inducer. This assumption was necessary due to thelick of static pressure taps on the hub wall or static pressure traverse data.Performance in these planes was also reduced for some points using static pres-sure distributions generated by inputting geometric and aerodynamic informationinto a streamline analysis computer program. The static pressure profile com-parison for the IGV exit (inducer inlet) is shown in Figure 77, and the inducer exitstatic pressure distribution comparison is shown in Figure 78.

Since the data reduction program would only accept linear distribution of staticpressure, the profile generated by the streamline analysis was approximated asshown by a straight line from hub to shroud. Use of these static pressures in thedata reduction program resulted in the profiles previously presented in Figures 72through 76. The use of the initial constant pressure assumption yielded the re-suits shown in Figures 79 and 80. There was only a small effect on impeller inletrelative Mach number; however, the change in calculated hub incidence was quitelarge. Incidence at 10% span from the hub was increased. Relative inducer exitair angle was, of course, similarily affected. This difference in inducer exitconditions should 1w considered during use of any of the standard reduction print-outs presented in Appendix I.

The effect of the inducer on overall performance, other than its pressure ratioand efficiency choracteristies, was evaluated in terms of any limitations it mayhave imposed on the impeller or overall performance. The effect of the poorhigh-speed inducer performance on the impeller was minimal, and is discussedin the following impeller component performance section. With respect to overallperformance, the inducer was initially thought to he the flow-limiting componentin the compressor. The correlation of naximum or choke flow presented In Fig-ure 81 is composed of Build No. 3 data and was used to develop this logic. Theslope of the speed-flow curve decreases sharply above 95% speed at 0-dog prewhirl,above 90 to 95" at !0-dog prewhirl, and above 901( speed at 20-deg prewhirl.This change in slope is generally a precursor to chokiug of the annulus and wasconsidered to he evidence of inducer choking. The vertically common ma4imumflow at lower speeds, regaw~ess of prewhirl, is due to diffuser choking.

The same relationship for the Build No. 6 data (Figure 82) did not exhibit shrilarcharacteristics. Choke flow at 7 0 1( speed was somewhat lower than Build No. 3.•ossibly from increased blockage In the diffuser due to the leading-edge damage.

SThe slope of the -urve did nt fall off rapidly at high speed and continued to in-crease with lower ptrewhirl at ma-imum spoed. The high-spowd choke flow hiBuild No. 6 exceedwd that projovted for BUild No. 3.

i ,>.•r I• •• '••,I ll = • • "" - .." • -'•' 1•- -: --:.....-•::: ....... .... I

* ****� .d�'fl%"yI -�;i-.r�-q �i�iJr.aYiww.j�.eu�u; *

I42

4)bOV a Lit41J A3M ItII � 1..Sn I

iiiRI� LA

U,*8 3kV k 'tin �III: �0-

IYR S * -1..

15 000N

Is . CaLi

a t 2 2 � 2� �ma uaazs

4

4)4)1Li

0

4)4)"4)I

j� I,

IIi �

I =

_____________ _____________________

Ca

� * I: &

0 = = - -- �j�gis4 SLUli tAt

b

93

-- '-�-4� -- , 'a -- 5- - - ------ - AM

-~~~~~~" .--9.' ,

-- - _MT-~--.----~

34

JI-

LIo

* o4

A * ~ .-

0o

I ~cu

ii

Impeller

Impeller performance was well defined from a combination of Builds No. 3 and 6data. Low-speed definition was available from Build No. 3 impel er exit traverses,and Build No. 6 impeller exit traverses defined its operation at high speed. Aninternal flow analysis was also used to analyze the inducer-impeller combinationand was available for every data point. The impeller pressure ratio, flow, andefficiency data from Build No. 6 are presented in Figure 83. These data showedthat the impeller produced its design pressure ratio at an efficiency approximately5 percentage points higher than its design goal and at a flow rate slightly higherthan design. The consistent h'gh efficiency of the impeller and the smooth pres-sure rise characteristic showed that the impeller continued to operate well, evenwhen its inlet conditions were being affected by the large inducer losses at highspeed. This good high-speed performance was duet in part, to the separateinducer-impeller concept as the impeller performance demonstrated its capabilityto work efficiently at conditions that would have severely affected a conventionallydesigned impeller with an integral inducer.

The tabulated data for the impeller performance can be found on the impellertraverse printouts in Appendix I. Since insufficient data were obtained to properlyseparate the impeller and inducer performance generated in Build No. 3, the com-bined performance is shown in Figure 84 with comparative Build No. 6 results.The Build No. 6 inducer-impeller pressure ratio was slightly higher than that ofBuild No. 3, possibly duo in part to the Build No. 6 impeller having a 0.020-in.larger diameter.

Build No. 3 inducer and impeller performances from an internal flow analysisprogram are shown in Figure 85. Similar Build No. 6 inducer and impeller per-formance are shown in Figure 8d.

Inducer and impeller efficiency as a function of incidence was also generated bythe internal performance analysis and is presented in Figure 87 for Build No. 3data and in Figure 88 for Build No. 6 data. A limited amount of data that wereobtained in Build No. 2 with an undersized diffuser is also presented In Figure 87for comparative purposos. Trho lower efficiency was due to the inducer operatingin a stalled condition. Low-speed characteristics are essentially identical forBuilds No. 3 armd 6. The inducer-impeller was operating ou the stalled side of itsperformance ,haracteristic and efficiency decreased with increasiig incidence.It appeared from the Build No. 3 charactoristics, that the inducer-impollor anddiffuser were matched at 9V speed, and at 1007 speed and 1O-dog prowhirl, thematch may be on the negative or choke side of the incidence characteristics.Build No. 6 data could be interpreted similarly if only 10-dog IGV setting datawere obtained at 1001 speed; however, other 101% speed data at different pre-whirls behaved in a manner no different than that observed at lower speeds.Build No. 6 data at 95V speed also failed to exhibit the peaking effect demon-strated in Build No. 3, hence it is probable that Build No. 6 operuted at slightlypositive (stall) Incidence at 95 to 101% speeds,

I

il96

001

'33

01

Lil

- ---- k.--:->.-

Cl 0

0 A4-hClp 0Ca

- .4Ca4.

0 C.''4

a0

t.4

Vo Ca

Cl 4 Lt

020CaCa

5';

I U'.40'4

h1

t U

'S

Ca

go &

- t4.'

2 -

& Ca

d � 2a a aCl U a

A

* b.�tS4At *ISW Iiii�.h �U�tI'4�i� �4i4�&-win

t

- -. - - � �? * -

Ile

*w

S'i41l0

-+ .. ,4

* -~W til

S•. +..,I

4 4 1

wq1

ai

cw~im tg" "4W~b

z z

va 00

/101I

0

La:1mNN, m ;ý X

00

00

102

The internal performance analysis was also used to demonstrate the effect of pre-whirl on inducer-impeller performance, as shown in Figure 89 at 101% speed.The increase in efficiency with prewhirl was due to reduced shock losses as theinlet relative Mach number was lowered, and also due to reduced profile lossesas the incidence was improved. The 100% speed Build No. 3 data showed nearlythe same incidence and efficiency.

Comparison of the two builds at 95/ speed is shown in Figure 90. The Build No. 6inducer-impeller was operating at a more positive incidence most probably be-cause of the reduced airflow at this speed caused by the smaller effective diffuserthroat area in the damaged Build No. 6 diffuser.

Slip factor calculated at the impeller tip (discharge) from traverse data of BuildsNo. 3 and 6 is presented as a function of corrected flow rate in Figure 91. Thegeneral trend of both data sets was similar and the level at near design was alsosimilar. Low-speed data exhibit an indication of higher slip factors for the BuildNo. 6 testing. Impeller exit discharge coefficients calculated by using impellerexit mass average data to generate weight flow which is compared to orifice floware presented for the same data in Figure 92. These discharge coefficients wereused to generate input for the internal flow analysis for both builds. While levelsvaried from build to build, both exhibited the trend of high blockage at low speed,which decreased with an increase in speed and flowed to a low blockage value atflow rates corresponding to 90 to 957. Higher speed and flow conditions againshowed a considerable increase in blockage.

Impeller exit traverse data summaries for every traverse point from BuildsNo. 3 and 6 are presented in Appendix I. An example of these data are presentedin Figure 93 for 1017 speed and -4 deg of prewhirl. Examples of impeller exittraverse data for 95/ speed 10-deg prewhirl from Builds No. 3 and 6 are shownin Figures 94 and 95, respectively. Impeller exit total pressure data indicateda larger boundary layer development on the shroud side than on the hub. The airangles shifted toward tangential on both hub and shroud sides, correspondingroughly to areas of falloff in total pressure. Nurmalized distributions of radialvelocity for Build No. 6 101W speed, -4-deg prewhirl and Builds No. 3 and 6 at95% speed are shown in Figures 96 through 98, respectively. Builds No. 3 and 6show essentially the same characteristic as the higher speed Build No. 6 data.Generally, velocity was less than design on the shroud side, with a peak greaterthan design near the hub.

Temperature traverses were also obtained behind the impeller. All of the tem-perature traverse data are presented in Appendix I. These data were input intothe traverse reduction deck and a mass average temperature was calculated, Thistemperature was considerably different from the average temperature measuredin the exhaust collector; for example, at 1017i speed and 0-deg prewhirl, theaverage impeller exit temperature was 40-deg hotter than the collector. Since theimpeller exit temperature was measured with a one-of-a-kind experimentalthermocouple and since the resultant data differed not only in level from theaverage collector temperature but also In profile characteristics from the check-out of this probe during a previous test, the data were considered faulty.

103

-. F W t.

.. ..

Si

0 0

00 0 0 0

~Y1

I

0. " .- E I ýA

L4

Q c

•J- .÷

0104

i• !V-

0 0- 0.

U C_

si~r•,•,•,: ,, ••_ ,-• ,-,•m•,-•-j•..•-•,•-•.•,•,- -,:•raT~i_-•_•.••.•.•.•--'-•-'r•';'•'.•..• ....... •-•---• ,... •:. ,., •. •-

70

0/0

4190"41f.!)~c Uta0twwOF4 w^

. 41 4

~as77,-. 71

V 1 igUre 91. InIpelor Exit Sipw~ actor, t.:4

A144 Af~.&

in

S. vi

Fiuo9.w aelo tTuo~,H W o ,II 00,--kgI#.Nastall

"4~

��IWW5UY VE� I - -- -v-cr �'.w�p�.-iL r'-w t*�. .w� r1'.n.r wrhruu5�rff..arr�

J:1

'St

I F I

A I'I

V

* r�*' �

$

it.4t\.

F

'I. I

ti

U

I-

fljW. it-Ii

� I 4I 'I,taAAikr'4'

V t.

�c. N*1� p �� 'S

I.'' 'S-

r r -

�F '�

- 4 pMt�aw�t. -1

I--i--- 1 -�

Vlgtwo N. Impeller tAt TrtwerMo. BuIW No. 3, 9�% Speed. 1O-do� RW, NearStall.

107

- - �4�-S C '� �.. -

4-L

-F -j

-1 T-

Figue 9. Imellr Eit Tuveso.MildNo.6. Seod 10dog GV.Noa

N *:~--~----mail.

1.0

0.4

3 0 0 20 30 40 50 60 70 60 90 100Hub SPAN -

Figure 96. Normalized Radiacll Velocity Distribution, Build No. 6, 101% Speed,-4-deg IGV.

'Ti'.

ul 4

I~ez7

l0t2- IGV.

109

-- 1q7jieyr.3 , -j 7---, .- 7

RIP

0

:,,,sgn 10 de I.GV> .-

' 'U

ia:

• • A boundary layer analysis program was used to predict the impeller exit velocityprofile for" the 95(,( speed 10-dog prewhirl near stall points for bothi Builds No. 3and 6. Figure 99 contains both analytical predictions and expqerimentally obtainedvelocity ratio profiles. The predicted boundary layer thkeswsto s malthe shroud sidie, implying greater losses along the shroud wall than were assumledSin the an~~~alysis. There, high shr'oud side losses also pointed to anarafrotn

in the an, 10 0) o0 60 on5

tial Impeller performance improvement. The predictions allowed an accuratepotential flow calculationi whiR i mathel toi experimental profile frora 20 to N/V

span extremely well.Stall transieyts were conducted with the impeller exit traverse proel positioned atapproximatelf 951 of the span from the shroud walp. These tranbsients were ex-pected to exhibit evidence of an air angle shift toward tangential flow imedibatelypreceding surge. An e. ample of the data obtaioned is thikesented wa too 1s0mwithhe reouainder given in gppendia 1. Impeller exit total pressuret angle, assumdcollector anls-isT pressure are presented vs tsidee. A ts plation of speed flo wvs timpe is alsor gver . Data from two different transients at scanning aates of Indapproximately 65 scans/see were used to describe overal triends ad t w

surge behavior in great detail. No change in air angle could Wt dita•fow ined fremthe typical searching psatteug observed In these data. In nearlet everdi iase. therewis a drip In AmpxllImr exit total pressue ,n anle, a nd

was adropin ipelle exi totl prssure accompanied by anl increas( in pe

110

and a slight decrease in weight flow before surge. Collector static pressure con-tinued to rise even though the impeller exit total pressure decreased, which indi-ented either a shift in the impeller exit profile causing a pressure decrease at 1Wspan or that the diffusion efficiency increased faster than the rate of impeller exitpressure decrease.

Static pressure information was recorded along the shroud wall for everV steady-state data point, Figure 101 shows Builds No. 3, 6, and design data for the 95%speed, 10-deg IGV setting condition. The trend of static pressure showed nosignificant change from build to build. True evaluation of the significance ofthese profiles was difficult since it was not possible to separate the contribution

S due to iP, peller work and diffusion; however, no obvious discontil uity showingareas of separation was visible. Most of the static pressure rise occurred afterthe transition from axial to radiial flow, with sonic decrease in the magnitude ofpressure rise near the exit of the impeller.

The five static pressure taps near the exit were located in such a way as to allowdetermination of the effect of diffuser pipes on the static pressure distribution inthe impeller. No effect was seen, most likely due to the large vaneloss space.Other exnmpleau of this static pressure dlistribhution are presented in Appendix I.All show tho same trentds as the exatmples. The distribution and level (4) riotchange from near stall to wide openI discharge; however, at a constant speed thelevel of static pressure is highest for the inlet guide vrkno setting that allows theinmipller to do the most work.

7I gLP

I00

r'iguiv 99. tyt*Vhatge Vvlacity Iatia(olqr,•t,* II ~ a

- • r~ m wn W~7 .t -** * * **ai

"I I i

41 4

-91

of-t

04 tA-

kii

ti LP* - ~ .* r~?~:P & -

) tdntfl

J~

vi v Anq 4 ~

Difsr Perfqrta =nm n4

COS ilI oo m

BIuild N01. S tatIc Pre~ssUre Variation Along lhmtogth o Shroudffuso% pip", Nea

toDiffuser perforrataee cvmum ack-trn, ncuding. anw- twaiatonm of the thfuioroctetsandhtl anascsswntOA the tilotalo#- t thast reosssincurted uring a h difsion pro-ces, will The. pereseteailtefolo static pressure ditjbtai pasatribut-ios thr00ho7 thedn iffumser, toal 103 essun prfihat theo diffuertv e~throat varous h toss dorrelatgnsSinC-es thediffser Wa Carnage'd apriorto thel Buil No-. petrcoforac tho sting, the

-ics oat will ineg d a twvthe Build -No. a an difue performance, th naur of* thdamage tvz or the diffuserendtheg rsltiong erfnn -ilWtanced los seenk Inave-NoAll diffuser@ t eranuaace adat ar s avifbl frow m tlndtlocl I tha perfomac In IitoBuv id Nio. 3 I~v14Peowti staticpecr ithuln ln h tegthut of ithdiffuse pipe tohuandth*hvtuct id avtiA sown at near stll anat Ow~mt hso- rU acpesrefr 9Orl WW

illustrates thle complex nature of the flow in this region. T103 is also evident illthe diffuser shroud side Isobar mnp in Figure 105. The representations shown inFig-vires 104 aind 105 mre comiposites generated from stntic pressu:• infornmaitionfrom several locations and. may, therefore, contain scatter ns n r-. suit of pipe-to-pipe vatriation. Dfownstreall of the effective thront, where esse. i'-rly nll the. dif-fusion takes place, no evidence of separation is noted. The circunift-reitial 2'qticpiressure dHstributlon at the geouwUiL thront, as shown in Figure 1W6 ni" 95"Mspeed and to0-deg WCV, is fairly uniform ucspe-inlly near stall.

C. -C

i °ito 0

0Wa~w kSiNuwMO

U,•

U

* 14

0 6oo I0 0

0 0

U 0 ~ 0 " 16

* A7 r�r��X � � � �

K - I'I

I as.- -- z*1 *

� /

f-s� .7

L �h4 1'

�

4f-s

/�'�(" '

/�//�*

V /

£ *1-- /

�- 14��I1; t'� Q

"i'

4•Ii

4

1�

31 iti

A, 0. ItaI~.I r. ?4q.(t7• -9i0 &C

L0 aw T\a a

Ito

20I IN

Figure 106. Diffuser Throat Static Pressure Distribution: Build No. 3,95% Speed, 10-dog IGV.

During Build No. 4 testing (mechanical checkout of redesigned rig bearing androtor system), a rub occurred between the impeller and the stationaty shroud,spewing pieces of the shroud silver plating through the diffuser and causing dam-age to the diffuser pipe throat leading edges. Photographs of this damage fromseveral typical pipes are shown in Figure 107. For comparison, a photograph ofan undamaged diffuser pipe is shown in Figure 108. Prior to the Build No. 6 per-formance tests, all the diffuser pipes were polished to reonove any burrs extend-lng into the flow path along the damaged throat leading edges. One diffuser pipefound to be slightly oversized was nickel plated to the same size as the rest of thepipes. At the completion of the Build No. 6 ttsting during the rig teardowll, itwas discovered that a considerable amount of rust had formaed in the diffuser pipes,as Is showv in Figure 109. This rust probably also contributed to the loss in p.r-formance noted In Build No. 6, ,alhough it is impossible to predict the degree towhich the diffuser pipes were Itsted at any given paint in the test progrmi. Thediffuser moaterial, 410 Stainless Steel, was chosen because Its thermal Ie•-mtnsionclosely matched that of the titanium impeller. however, it appears that for avro-dynamic reasons a corro~tom-resistatg material would he bettvr.

The. effects of the damaged diffuser can be seen in the Build No. 6 static pressuredistributaon plots for V5 and 10r, speeds and 10-deg WGV in Figures 110 and M11.

SThe effective throat region has moved downstream from its Build No. 3 locationand nearly coincides with the geometrie throat. Also, rogiom• of flow separationare now evident in some of the pipes toward the diffuger exit. The individualpoints shown on the plots are not necessarily from the same diffuser pipe, so thOinaosistvtu ture of the sta~lc pressure dati tioar the diffuser exit indicates

117

iI

varyving amiounts of Separation from pipe to pipe fox those pipes in which Sepa rationN' does occur. Figure 1M slhows Build No. 61 static pressure distributions for 10l1

speed and -4-dog IGV at both near Stall and minimum back pressurc'.

M-11

.. .1 .

7",t

-tll

NA

Vtgutv 10)7. Damuigted tDht~tor Pipe Lt'adiag Edge~s.

- 7-

Figure 108. Typical Undamaged Diffuser Pipe Leading Fdgo.

rVlgur@ 109. Pact-Test Coniin fDffsr

119

L. ,�II�,Q . ><..

411 I.'

� 0 . UI �w �

I 0 �

-i �

0

0-.-�� I-.- cJ

CgI

(o4� . .

'4 �.

1 0U

0 pI 00 �

0

e

I �I t a 4

ISi

.......................................�--.

140 000

Neat S4JII~ 0 0 CoU.,Me

1200

01

lot)

lOG toa rssle3

Coltkctoe

I ~Wde OPn R

40

00C, . 2 3

DOCANCM ALOWG OWUWSR MR b.

FigurV 112. Diffuser Static Pressure Profile: Build No. 6, 101% Sp7-, , -4-degIGV.

1)Difl~tser exit MaNh wuhber profiles wore generated, baset on diffuser exit total

pressure rake data and collector static pressure. A resultant profile for BuildNo. 3 at 95'; speed, near stall, and 10-deg IGV is shown in Figure 113. Theprofile is marked by a large low-diffusion region on the shroud side and by a low-flow regioen on the ht:b side, which probably represents a large bouadary layerand flow separation. The Build No. 6 profile for the Pame point is presented inFigure 114 and has the same high-loss region on the hub side, but hIa a commen-surately larger low-diffusion zone than Build No. 3. Plvofile at 100 and 101"speed from lluild No, 6, shown In Figures 115 and 116, have the same general

. iharacteristics as the 95, spced profile. Figure 117 shos the iffuser exitMach uatmber profile at minimum back pressure at 101( speed. whlich Is ehatac-torizetd by very low diffsiou with large high-loss re•glos.

121.

__ __C5i, -a.,

***** *Q%� ****.. ** W� W 9a.. ¶ ml..!,

IA

a

Ac.p4 -I-

1.404)-

C)

LiaI 1 1.4003

.5'-p0 $4

Ct

-1�

aS :2 $4

£z4

4)

1..

9* -

UC� &�

I

'v-el

S

C,-4-4

5 a 5a b.

iC'

122II

-a

0 (M

- -

* iiv12

123-

It A

' IFigur'e 117. Diffuser Exit Mach Number Profile: Build No. 6, 1011T• S|)ecM, --4-deg IGV, Wide (Vonl Disoharge.

D~iffutser total pressulre -Uld diffus~er dump losses arve plotted as a functionl of cor-rected weight f'low ,and impeller exsit Mach• numbe~r for' both Build$ No. 3 and~ 6 inFigures 118 and 119. The tat~al pressure diffuser" losuses were essentially idea-tiea!l for hoth buikl(I, ,ilthough the Build No. CI dump losses were higher. ih.

ca edirectly attr~ibutetd to the lower" diffusiton levels antd r'etltnhieri-fuser exit Mach itumbers for Build No. 6, as isl shwn in Figurtte 120. The Bu.ildNo. G .static pressure rise coefficienlt values wr'o vsioc$deraby lowvrc t ham the

) Build No. 3I le-%vlso as shown i.n Fivgutv 121.

a ]

t--

h4-dei V,-WideOpen Dicharge

00- - - -

i aQ1

U0

S~~ 2 . ,? t A U iM

Figure 118. Diffuser Losqcs vs'coirreetu- weight iFlow, Nvar Stall.

Wt .zw -iwr No -SW

ga U4 w am kid IM: %3$ am @A-W

60OMS"~

Figtwo 120. ifue upLsoNurSal

10AM-

Al

IN 1

scto Pt*uvP~ ofc"- ocr, Wh tk

1 H-1F11- F REQ ENCY RESPONS E DATA

Ifigh-frequency response data were recorded at two speeds, 70 and 7847r of designspeed, during Build No. 6 testing. These speeds were selected because theywere the highest speeds for which performnance data were available and which didnot exceed the temperature limitations of the high-response KILITEW" pressuretransducers. Data were recorded at steady-state conditions, near stall and duringtransient operation into stall for the followig four pressures:

1. Total pressure at the impeller exit2. Two static presstures In the vancloss spacea. Static pressure at the diffuser discharge.

These pressures were recotrde using KU!,ITEF pressure transducers capable offrcfueneies from de to 100, G00 1iz; however, Atet to the high magmetiv tape speedrequired to attain resolution at the, high frequencies, data below 200 Ilz are notreliable.

Typical data are presented hi Figimres 122 and 123 for 70 and 78KI of design speaed,respmetively, and show the respumse of the four measured pressures during a1-.ms periodt including the surge event. Typically, these data define surge both

in the form of an amplitude change and a frequency change. The amrpl'tudv dif-ference between the two vanless ,sace static measuroments is believed to Ix( d(tUto a faulty gain setting on one ehamnel (PSDVKI); however, the wave form is unaf-fectedi.

Additional dat' from each of the high-respmnse probes during the stea4y-stateintAt and at 2. -ms increments into the stall transient are presented in

ApIpadix I in the form of pressure amplitude vs time and pressure amplitude vsfrequency. These data indicate the presence of strong discrete fluctuations atfrequencies eorresponding to 12 E, 24 F, and 48 E - rotor speed). Thesefrequenwies correspiond to excitations from the impae*!r blading, which has 12adll blades, 12 primary splitters, and 24 secondary splitters. Prior to surge,the dominant frequency ohserved was genaerally 24 E which seems to indicate adifferential work input hetween the secondary splitters a the ful blades andprimary splitters. Corrlations which were iprforimed oa these data, however,failed to Idtify the 24 L sigmal with the vatous types of blades "- splitters.Foll owirt surge, the dominant frequency became 48 £, which appears to be the

presstwe field associated with the sinled hWtading. This patteor of frequencychange -,4 abserved for bath the TO and 74 spoed. No evidncvee of rotating stallprior to surge "Ii noted; however, the local distortioa of the impeller dichargeflow field due to the itnertion of the relatively large Ngh-frequeney pressurepnohe caused •urge t occur s"ignificantly below the unadstmoied suirge l•i. This

assytUnwtry aly have Prew•,,ed -,, fora.tion o•' r"at-g stall "--s.

- -L- -- ..-- r- .. ~ tq p.lt~ n.

)L

rit.

-47

t~ 7 -

117

T~I I. A;

1 l~T,.v441:ii '4

I I-

., . . ...-4-

.1 .4.. ..Mt - -L-Hirtr

1T I ::~.Iz129

1* CONCLUSIONSTle overall total-to-s~atic pressure ratio and efficiency Gf this compressor camevery close to meeting design goals and, for all practical purpose is hompdemon-

strated that high-pressure ratio, high-efficiency, single-stage conLrifugal com-pressors can be developed. If the diffuser had not been damaged, the 10:1 pres-sure ratio 75V efficiency goal likely would have been surpassed, as shown by thecomposite compressor mnnp generated assuming undamaged diffuser lossos on theoverall compressor performance.

The basic approach taken for the aerodynamic design has proven to be sound. Theus,; of variable inlet guide vanes has allowed obtaining optimum performance datathroughout the range of operating speeds and generation of an inlet guide vane

A schedule for potential 3se of this compressor in a gas generator configuration.

The inducer operated well until it was run at conditions beyond its intended design.When it was discovered that the overall pressure ratio was low at design speedand prewhirl, it was necessary to produce more work in the inducer-impeller.

This involved a decrease in prew- Irl, an increase in speed, or both. Either con-dition increased the inlet relative Mach number into the inducer and forced a morestalled incidence, generating high losses and resulting in low efficiency. Theremote inducer, which was included In the design to accept shock losses shouldthey be higher than expected and to allow separation on the inducer without de-stroying impeller performance, also functioned properly in this situation as theimpeller operation was not influenced by the poor inducer exit characteristics.

The inlet guide vanes were set to produce less prewhirl, and rotor speed was in-creased to produce the required rotor pressure ratio. The impeller efficiencywas above predicted, indicating that additional tip diameter would also have ob-tained the design performance.

The diffuser, including the entry region, is the most critical element in the per-formance of this compressor. The mismatch caused by the undersized diffuserin Build No. 2 and the resulting improved performance obtained in Build No. 3 byonly increasing the throat diameters from 0. 228 to 0. 235 in. demonstrates thesmall latitude in dimetisioning the diffuser throat and predicting impeller exitconditions and diffuser throat blockage.

The losses and static pressure rise generated by the resized diffuser were quitegood and would have allowed the overall performance goal of 10:1 pressure ratioat 75% efficiency to have been surpassed in Build No, 3 at a lower prewhirl than10 deg if the rig had functioned properly mechanically. Damage to the leadingedge region of the diffuser, resulting from silver impingement when the impellerrubbed the shroud in a mechanical checkout of the redesigned bearing support androtor system, resulted in enough change in the diffuser characteristics io causethe performance to be short of the goal. The change in throat blockage causes theflow rate to change and the static pressure rise to decrease. This behavior alsopoints to the critical nature of the diffuser in a compressor such as this.

130

A-7

RECOMMENDATIONS

Analysis of the data obtained during this program to design and test a 10:1 pros- Isure ratio single-stage centrifuigal compressor leads to the following recommenda-tions to improve its aerodynamic performance:

1, The inducer should be redesigned to operate efficiently at theconditions required to produce 10:1 pressure ratio. The in-ducer redesign would be based or the optimum inlet guide vane

exit conditions and the desired Impeller inlet conditions andcould improve the design speed efficiency up to three points.

2. The modified inducer should be tested with a new diffuser toeliminate the detrimental effects of the damage leading edgesof the current diffuser.

3. The potential performance increase that could be attainedthrough reducing impeller blade-to-shroud friction heatingshould be evaluated. This friction heating could be minimizedthrough the use of a shrouded impeller. The shrouded im-peller also would eliminate the problem of impeller-to-shroudrubs encountered in this and other programs as a result of thesmall clearances required for conventional open-faced impellersto achieve good performance. A shrouded impeller would alsopermit extension of hub and shroud walls to form a mechanicallypractical rotating vaneless space to reduce diffuser entrancelosses.

4. Means for improving the range characteristics of this compressorshould be evaluated. The useful performance of t, o compressorin an engine or gas generator application would be substantiallyimproved if peak efficiency could be achieved off the surge line.Several approaches have shown potential for improving the near-vertical speedline characteristics of high-pressure-rise, single-stage centrifugal compressors. Particular attention should begiven to the impeller exit/diffuser entrance region, which, to alarge degree, controls range from choke to surge. Specificrecommendations at the 10:1 pressure ratio design point includereducing the number of diffuser passages to increase range, aswell as using sweptback impeller blading.

131

-i(

APPENDIX!IOVERALL PERFORMANCE TABULATIONS

The tabulations of overall performance data used to construct the performanceplots in the text ire shown Wu Tables XIV through XVII. Table XIV contains datathrough the range of operation for Build No. 3 zero degrees of prewhirl. Allsteady-state data points and the incipient stall point located by stall transients Pare included. Similar Build No. 3 data for 10- and 20-deg inlet guide vane set-tings ire shown on Tables XV and XVI, respectively. Table XVII contains similardata for all prewhiri conditions during Build No. 6. Data points are listed insequential order down the speedline from the incipient st'll. point.

A1

Ni

1321

TABLE XIV., OVERALL PERFORMANCE TABULATION -

BUILD NO. 3, 0-DEG PREWHIRL

16V WeightSpeed, Setting, Flow, Pressure Efficiency,

deg lb/sec Ratio

30 0 0.213 1.257 0.7470.253 1.272 0.753C. 302 1. 261 0.7580.389 1.259 0.7040.425 1.218 0.6060.466 1.168 0.4870.494 1.142 0.383

50 0 0.620 1.953 0.7510.638 1.942 0.7360.667 1.959 0.7420.737 1.907 0.7210.769 1.877 0.7040.794 1.749 0.6170.801 1.301 0.276

70 0 1,370 3.603 0.7641.369 3.539 0. 7561.412 3.515 0.7481.414 3.505 0.7471.452 3.196 0.6861.464 2.999 0.6381.454 1.878 0.346

80 0 1.879 4.904 0.7671.873 4.872 0.7691.897 4.878 0.7711.932 4.812 0.7451.936 4.202 0,6691.934 3.861 0.6231.938 2.53i 0.402

90 0 2.597 6.935 0.7612.600 6.921 0.7612.624 6.926 0.7602.650 6.865 0.7612.597 6.714 0.7512.604 6.296 0.7192.605 3.496 0.446G

95 0 2.954 8.661 0.7912.965 8.563 0.7882.969 8.555 0.7862.989 8.520 0.7843.019 8.376 0.778

* 3.031 8.019 0.7553.026 4.198 6,472

133

TABLE XV. OVERALL PERFORMANCE TABULATION -BUILD NO. 3, 10-DEG PIREWIIRI,

IGV WeightSpeed, Setting, Flow, Pressure Efficiency,

deg lb/sec Ratio

30 10 0.223 1.200 0.5260.246 1.223 0,6620.224 1.251 0.7410.269 1.261 0.7560.301 1.271 0.8110.337 1.251 0.7460.360 1.241 0.7090.407 1.226 0.6530.432 1.120 0.372

50 10 0.637 1.950 0.7460.625 1.941 0.7470.655 1.941 0.7480.663 1.955 0,J530.649 1.928 0.7430. 646 1.947 0. 7590.698 1.919 0. 7370.777 1.8'71 0.7170.776 1.664 0.5670.790 1.305 6.291

70 10 1.355 3.555 0.7721.406 3.503 0.(631.417 3.480 0.7621.453 3.408 0.7421.462 3.202 0.6971.460 3.006 0.6531.457 3.152 0.6881.463 1.884 0. 352

80 10 1.874 4.943 0.7791.898 4.851 0.7691.910 4.836 0.7661.930 4.720 0.7521.914 4.549 0.7341.933 4,384 0.7081.903 2.500 0.401

90 10 2.562 7.137 0.7972. 575r 7.046 0.7902.595 7.046 0.7892.605 6.973 0.7862.607 6,882 0.7812.611 6.258 0.7342.586 3.496 0.461

134

: l

TABLE XV - Continued


1deg lb/sec Ritio

95 10 2.778 8,192 0.7962.818 8.261 0.7912.816 8.176 0.7912.819 8.174 0.7912.854 8.,145 0.7882.873 8.083 0.7832.902 8.006 0.7772.878 7.610 0.7572.895 6.927 0.7102.902 6.825 0.7002.898 3.990 0.4672.895 3.978 0.467

100 10 3.005 9.338 0.7703.025 9.288 0.762:•i 3.037 8.885 0.7423.039 8.575 0.7233.034 8.135 0.7033.033 6.594 0.6113.025 4.287 0.451

TABLE XVI. OVERAI.L PERFORMANCE TABULATION -

BUILD NO. 3, 20-DEG PREWHIRL

IG, V WeightSpeed, Setting, Flow, Pressure Efficiency,

CV deg lb/see Ratio 17

30 20 0.275 1.253 0.7850.269 1.259 0.7470.358 1.272 0.759

0.394 1.235 0.6740.419 1.229 0,6640.484 1.168 0.4970.494 1.119 0.367

50 20 0.614 1.976 0.7800,611 1.942 0.7500.664 1.923 0.7500.718 1.918 0.7460.762 1.842 0.7010.774 1.686 0.5880.790 1.289 0.2740.777 1.297 0.278

135

,, _- d : l• " 1" .. . 1 I I •" I : t ''" : •"" • • "- • " ' t • ,':''"• '• "• -'' "• : •'""" -:•"•:;L-•:"•••J -•:''

TABLE XVI - Continued

IGV WeightSpeed, Setting, Flow, Pressure UEfficiency,

% deg lb/sec Rattio

70 20 1.34,5 3.508 0.7581.372 3.506 0.7641.350 3.498 0.7641.409 3.488 0.7641.1458 3.230 0. 7131.459 3.004 0.6601.448 1.886 0.354

80 20 1.818 4.818 0.7801.841 4.817 0.7831.848 4.809 0.7841.865 4.817 0.7841.841 4.752 0.7741.910 4.393 0.7271.919 2.483 0.409

90 20 2.389 6.849 0.7932.388 6. 805 0.7922.402 6.772 0.7922.436 6. 821 10.7942.501 6. 712 0.7882.510 6. 232 0.7512.520 3.385 0.456

95 20 2.664 7.882 0.7922.697 7.749 0.7832.696 7.732 0.7832.691 7.726 0.783-2.721 7.593 0.7752.733 7.236 0.7522.740 3.740 0.454

136

TABLE XVII. OVERALL PERFORMANCE TABULATION -MIILD NO. 6 __


deg lb/sec Rntio 17,

70 30 1.227 3.405 0.7351.212 3.320 0.7221.441 3.255 0.7171.357 2.952 0.6181.346 1.874 0.349

70 20 1.288 3.500 0.7341.279 3.346 0.7131.35,3 3.04.1 0.6501.353 2.598 0.5441.359 1.829 0.334

85 30 2.044 5.624 0.7672.071 5.425 0.7562.065 5.168 0.7332.097 4.040 0.6002.101 2.866 0.426

85 20 2.078 5.690 0.7512.125 5. 553 0.7422.U17 5.233 0.7192.113 4.692 0.6572. 088 2. 925 0.422

94.5 15 2.734 8.079 0.7752.787 7.957 0.7682.797 7.913 0.7682.823 7.298 0.7282.819 6.703 0.6892.833 6.539 0.6762.80' 3.945 0.461

94.5 10 2.792 8.169 0.7732.796 8.037 0.7682.827 7.808 0.7532.841 7.414 0.7262.828 7.273 0.7172.837 7.015 0.701

4•059 0.471

95 10 2.833 8.388 0.7784 2.819 9.238 0.769

2.,883 8.122 0.7642.894 7.487 0.7252.894 6.897 0.686•i 2. 881 6.-792 O.68

137 i

TABLE XVII - Continued

1(V WeightSpeed, Setting, Flow, Proisure Efficiency,

C1 dog lb/see HRtio

101 5 3.129 9.661 0.7353.136 9.693 0.7373. 142 9.444 0.726

1'.147 9.360 0.7243.144 9.119 0.7123.142 8.520 0.6833.150 4,679 0.448

101 0 3.183 9.882 0.7373.185 9.649 0.7273.192 9.552 0.7223.197 8.993 0.6953.194 8.516 0.6723.195 7.830 0.6393.192 4.743 0.450

"10 -4 3. 196 10.034 0.7383.198 9. 965 0.7333.209 9.698 0.7213.217 9.672 0.7213.221 8.786 0.6823.222 7.398 0.6103.221 4.728 0.442

136

THAVER1SE PLOTS

This section contains plots of traverse data obtained during Builds No. 3 and 6testing. Total pressure and total temperature are rattoed to standard day con-ditions, but are othewvise uncorrected. The keys on the plots identify the correctscale to the symbol from right to left in increasing number. The temperaturesensor on the Station 1 and 1. 5 traverse probe is offset 0. 045 in. from the pros-sure sensing ports. Both the Station 1 and 1.5 probes were stopped prior to thehub wall and traversed into recesses in the shroud wall.

Fi,;ures 124 thromgh 152 contain Station 1 (inlet guide vane exit) radial plots oftotal pressure, air angle, total temperature, and circumferential plots of totalSpressure. The radial traverses are from hub to shroud, while the circumferential

traverses were made at five discrete spanwise locations. Figures 153 through 174contain Station 1. 5 (inducer exit) radial traverses of total pressure, air angle, andtemper�ature vs travel. Figures 175 through 210 contain Station 2 (impeller exit)traverse plots of total pressure and air angle from hub to shroud. The probe wastraversed from a recess in the shroud wall to a recess in the hut wall, and datawere recorded on the return traverse. Figures 211 through 219 are impeller exittemperature traverse plots taken in the same manner as total pressure and airangle.

1

,:9

2 SCAN3 (f CAN

ia

v-; 4"

10

- ia

-. ev IC"Na SalU

SCA

1ICr"rttta ev-o.P0

lo" PLo,---gI;,11 PU

L

T I~

F$, Wu ' IVIr G icu~vtilTtvs.BidN.-itC **OA O-~ V%- 0

r r A

7 1.

I L

T-t-2 4LT

.44T..~e~

6L q

. . .. r 7- . 2 1 2 j 2 C . V ~ ::+47~ .1774J21

KPII I'll,

Figure 127. IGV Circumferential Travertse, Bvild No. 6,101% Speed, 0-deg IGVI, 50% Span.

1143

-44- -

1w.

-4

7K.I

J- . _L.

2v I 4 -T7 .

.4Figure 128. IGV Circumferential Traverse, Build No. 6,1010/4 Speed, 0-deg IGV, 707( Span.

144

J .

0. 1-, -- 1

I: 21JI7

7-T ~ f

7- I T

T iL:--- t-7

7. -- ~ ------ i +~ i------

o-1

-l '--i-'-- -*-- -----

Figure 129. IGV Circumferential Traverse, Build No. 6,101T~ Speed, 0-deg IGV, 90r/( Span.

145

LI

I SCAN

? SCAN

I 71

7 7-~- f 3I:_ - --

Ali

- ~ -- -- - --- -. - --

-7 -17

JL tI 7,I .0

717~-.* -- ~ ~-

Fiur 130.. VEi ailTaesBidN.6100(1 Sped -4de 1

'~~1 'I FI146

UI

1

ý7 '1 t "

T7 I

1017 Sped -4dgIr1% pn

1147

-- 4

4.1. ii;ii

itPCL.7

7-.It' I: -r *

:tft"

7_ tt

I IFL , 41...

-. IT

7T- - I- Iit 14t-rt J.

Figure 1.32. IGV Circumferential Traverse, Build No. 6,101(?(' Speed, -4-deg IGy, 30% Span.

148

':1

4~4

1Z.20 .I r I I

4 ' L

I17j JI jjfl .t .j . i o I.-

-J; tf I U1

Fiur 133 G CrufenialTaere uldN.6

10 1 peed -4-dg IG, 50%Span

149~1 K -4A

*T-

4 -

-A-

dii .1.: 4ii 4w.

Fiur 134. 1 iru frnilTaesB ldN.610[ [ped I4de 1V 707 pn

Cv' ~~ -t 4 4 ~ 150..L.~---

I ~ KJL

I Lii r VK I 4Fill!

~2A7 nI 4+. -nI

I 10

J--

I . if

* ~ I, ~ ~ ~ ~ 1 -;7I~ 1~ ll I

ii * II

I;7z

Figure 135. IGV Circumfe rentital Traverse, Build No. 6,1OV7( Speed, -4-deg IGV, 900/ Span.

151

Ir SCAN

J.4'

I ' .

--A Inn

417 I 1

IT . -1-F1 gur 3. IVEi Rada Travre Bu N.6

10" pe, 5dg1V

- .-

-- ,

p *3. 1 *

* °

- • ?•.. • , . •.......,. . . ..

., ..-I . - -r • . . . .. . . . - . ..

E.'.SFili• 13. ;V }xi Raia Trve•seI~h-!No.p I'J ,' ' N.,r S4l. .

4O'Sed O-e t\ *•

in 5t~ t):3

tit T. .- -

s-.-- -. I:

4 ov

-,4-

I 11 3 p

w.". oi -- 1!Fiur 13.JVC m~ern -lTaosBildN.

1001(T $1.I-o G ,1" pn

154-.

17 j

It4

7

711

4A

Wg -Z

Fiu 4.fVCrupovtilTaeso ~ N.i-

sr156

A~ ~ ~ Z 4 , L4 4I

II qi

1'7 1

1i 1JT I::-l

IT id i

;.I r

ii:ý'4,ý

1-: 4A J . lit

Fiur 14. IG Cicmeeta Trvre uidN.6

10/ Sed,1- de: IG, 00ISan

~tFI I 471- 4 T T157 -

I 44

4 .- -t7

4.4

100,,' Sed 0egIGV 0 pn

L4~ I -'1-58

* .-77 - - ..~lr .~wU~~ .. . )*.-r

A;j

~-4

70% Sped to-do IV le Span

F .- - * ~ - -..............-

I I

I 0.

dcqIi (L o4i

__IV _icmcva Trvre Bul N.3701 Sto.1--i 1;.3" p

*1

.............................................�- I SCAN

I . .

�21 .1 -- I..

I . �

I .4'.

I... I-' 4

771.1

-�

.1 rH-k-

t e) I I

.4- � �I i I ii.I . . I I jII� I I I I

- . VII'4 .4 �IL

*1* .� I .�

- .-- I j I

F ---. 4- �1II

_____ K' �-4

4 . PL�

Vigus� 145. jOy Cii�curnferenti�d Ti � Bui'd No. 3,

101

.. -. .-." -. . . . ... . . . .... . .. .....-.-.. ... .... ,

II .

-... --- --.--

'a-

6 S•d I -, ' # 7 S

.i . . ,.. .

!" ........... * 4.... ..... . . .... ...-I...!..

I ,."I . .

I! ' p

• .•~I~ . :-A : -• • : -,

-.~ f . . .. ...,. . - • •' i ; '

2... I4t.

........ "~ i ... .: - ;I:" •,', .• '.. I .tr" i : !..II Fiue I, G Clo~'•ota raesBidI,3

0 i

I I

i i

7-~ T7

F:- 1, ;7.

I

-- 17 L

77 T.7.

T} I

it.---4-j-

~w~ftj

Figu~ re 1 7 G C r u f r nt a1rv rs , B i d N . 13IttO:HH 'IH LIV901"Sal

4t_

it:

3 :71,3-7

0-~0 sped o-de .-- Nt Wt

IF.

.10 I.

4 4

4D f.

!GV Ci - u e m ii f m v, -. b id N . 6.

'OPT~~.. I" 0dg!#V 0,sak

-- . .. . C - * - D "�r� qi*2 .. - �

� - ..�' -

4

ii'i'i�

�; I I

4 J - _____

:1 L...2� 45 - .

.4 t i.s �

/ K - *I-'��-- I..

.*.4..,. I *

- r4..---.-

¾4� *

�.Hrt.

* I�t

N . I

A;'A

K4�. A-i.:1�-.

* __a

.19 --. .-

a'A _

ri�ttrt, UiO* ILA L&rct±mftrttatial 'Vt tt'�* aAKI N�. 4

I"

4u

7

I4" o 14 Who"

V 14 IG *"wt'-att rwro '~dN

£" . ýPa

-1

1�- ., --... a $CAM

- .- -

'-1 -4.

--4.

-4

* 1.

. .

'4.

.�r>.,. . .* . -

. . .

.

I -

7L1� -I..

- � -�- -. -

< 1 -.-...

7

-1----- 4

-- -

a' C. Kc SCN-

---- 4 -4-

.. I- 4 ------

I A1- -1---+---- - .- A-s.--- --

t 1---------> -

~7 714

-4 - 7 --- H

-~ -A 1

ZILE4L

-.. .. ... .

07t 7-~

-1 4

'Sý 1 WAL

1.8 so0P.r A.

__de IV WieOe Dshre

I 7 + ."1

r2o o

9m o

iJR

ItI

Fl! 1 *Amr71

L -J - -.. t.

5-e IGI Ner tal

9, ~---4 ~ I~L.L.;.i.~t1701

C..9 C

.'',

J~

Cc -4-

C: CL CA V S

Fiur 155 Inue ExtTaesBidN.6 0%Sed

0-degF CGWd pnDshre

171

If , 1 I SCAN4v-~-1iij .. A SC '

k--9

:.a.

I 4 L .4-

V dt

~-7

-1-- .---- -

O-e IGV Nea Stall.1

172'

S.

t'Ig• r 151. Indur Exit Tr'avers.. Build No. 6. 1011 Spotd

0 173

I

.'1 n : , jr1

"• . . . .. ..

V . . . , - . .. .. ... , .. .

T~ ' F

" a.. " ... . . .;.. .• • I .• " " • .. •.. . . . . .

Mu 159 taue ExtTao ulI

.: . IS

I -I.-.."

t. -' -. b. flt IiF*-. .•- r ? 0 lf asr'~ .y' r- - - •

I .. . ; ! "• I "I ' " '

K.... I i i-•' ) i

V!• ... . ... .

£ I

I! .

""I I i • .I Il I .'i ,s-A . .l I i-'--P ... t -• - - - - - -b •f.. t . . ..• ---- -!-j-.t . ....---

.t t • : • • l i 1IJ -.t I 1 I .

S] 'I

4 4

-- . L. ... ;. ....... L1 ... 4.. . .'..

- -- j - - - i i t . I 1 .• , " '

~ '- : -> 1 9L .I• - .' • -. ; •+ -". ..:. . .•',...• •• f . ..

[[

rtr7;...•h.}•'-, 2.••i• •••• '.Z =•-- : T •£. •-2 W• •• •_ • .• • .•T 'i•'"•-i• •.w'*-1 ••,

- -Age. U7;.*-*- i T

f7

ITT~ L.T A7

-r-FI21

714 T tLtT

-7~

-rF l U1 111 7 -

~~7-F tI I~~~ .ILL 4 L_

............. ..4 ~ .. .2 ..__..

... I ..14. - I

-5Pak3 ITy

176

-AIR ANGLE------- TOiTAL. TcmpcRmrUR

-Fr

1tC; am ..so- I

.4 0J I

J.L

L . 0 !I .

cc I l

F it

K4 J T- i~iJ1- :jI

T` ! - .. . lffFigure 161. Inducer Exit T'raverse, Build No 0 , 1 00c SPV~d

177

IICO

:11 2,:17

g, g -

ciA butTr *

1 178

II

.... e4..4ItI¶.g

j4j

SS

ý7 !-WY7

........

* A

.1.

4 z co.1u

fff

42

* - -. -, .--* -- *,*-- .- --- ~--.---*------.r'-c

I-,, '~Z

1~z

-t 1-- 11i'4 44

~I ~el

1'vi

4~4.

0*t'' VV,

t A-

ini

71 7-

Fi-r WT 1awe xtTav~. idN.4.W PO

Fr �

I tU- 24�GfAN -4---- .

9p - - '-4 ..-�- -.

S �- 2.4 - *-* ¾--'

'-It .

4;a

a. BF C F

a

IC F ,fl ''¾ �. 7I'A ¶ 0*

� . '�i' -

- .I.

FtS� . - -

w �. c' F� A �' Is'5 5 *41 ''-,-,.'--.'---- -

UF, ..11 - Ii �

- � a - r '----

WI '---=-t-- -� --�--'��-K � - ¾4 / - .

t� ( -�

/ I ' "''"' --

F - - ,-F' F

ftF' F

1 VFr I;LLL_--�-

asb4�, ?ti$L dEc46�&

I

IC

w~ ip

raa li

.1.7

-- 4-

.r7 -7 Iv

~... .. ......

ft. a

_ 071 i10dgIGWd Ope Dicag.

4- -1

... ... . .. .

~ :.., ft I ~ ' ~ jc-

in1-±~~:00it Ž- -

7 0 . M, ..... ..4 ~F,

-~. .. -7I -

.4. - - z--

"17

I _4,

m-LI -7AT--`0_ _ jj...

UjI 01 4MI L.2~ijtz z

J~tl. M.~224

~ iT ill O1ol flul,~ t 17 E =1 2

Figure 170. inducer Exit Traverse, Build No. 6, 95%/ Speed,10 -deg IGV, Near Stall.

184

I !9F4 .1- , 1~ 11 -77!i.7II

.17 ---l

Ir .

dO3

4-3

in'3

F igur 171 Inue xtTaesBidN.6 5,Sod

30-dg IGI erSil

W4 .~1.5

-.;X•''.. . .... - rw a .......w-q W WF W W :•''"•"+" •:•••--' % •,•.•'"•••°'? •"-<:"rmD<'; -?'•.,--:-''',-''¢•'--•••..• • . .. '• -. , .-. • .. ,wP.-•••. .- ,.•• .1-_ :, • •,

&.i

:1

, . , I

A -LL • f/..

I.. .. ,: ?. .. . .. .

tI

7- -

"10p

. . - .-.

,V • . . /, -

f 1

* "

Il

Figurel~ 173. Inducer' Exit T r'averse, Build No. 6, 701 Speed,30-dg IGVI Near Stal!.18

aa

* n IN

ID

4

I.... re 114 Indce ExtTaesBid o ,7%Sed20,'-, 1GV NerSal

~. .- t I ~ ~ I I ~I188

'F~ SUN.1>1*.2., UrAN

1ý Vs

-1 - - ---

'169

______________________________________________________� � -

t.

I1: - I

U, -U I -

I i --- I

I .i

I, II. I .

* a '. .-

I....I. .

.... Ip �

* �J.........................................................................................................I

p* 0

C I . .0 0

- S

* C .,-.., .9 9 . .

kB �'I*'�.-'I.�

- . .

I1M� WALL

I. a�AI�Isi

O-d� Icy, � II.

I�4)

- **� .. , � �**�*�** .-......-

'1AL, PIU:3U

AI AtJ4

441

Figu 17.Iple *, rvre.UidN.6 0%Sed

-401IG 'NarSal

SCA

1 SCAN,

43WA

19. .

SI .... . .. ..

4 *4

Wa ! a 00

"00 - 0 ;

:000 1" 44,6 W196 04-0 a W WO

.. .'

I . . . .

I- . I

: : - - '... •:;; ... • '" • • ' "; ' " • •• "• • '• "• • ' °'' '- ?•r ~ -"V •= =. , - --• -- " . .= "'- . ... . . - • av"'., -

I-

-1

II I - -� f---1 I '� �C�N--� I j I�! I

L

,.- � I�-�----------�--- I

..-. 4�.-�-----.-.---4-- -

p - -t t,�- 1,1 -

S- - -�

I* __-

I *I 1�� 4 L.�-.... 1 � T � F

-i �

-iI I-

'-.' I --I-� L

17 �--

-- #5

-4---' -�w �2 ''�

if - -- - - -- -- - ' I� -� - -�.. - - -

SV - - - -v

A

-I- "��1-$L--- I - -

- - - I - - ��-�*- --

� -

� S I -- ,-'-----.-- -� �-

-� �-t---4!S -- -------------- --- 4--

- U - ��ALL�

F- -� � _________________

Mi- I. - a.'.-A�4 ---r- M�-t�d� I�*U, tt�i� ___

�Ii�ui'� �t. tmp�t1�t Exit h�w@i�,. �uiW N�, 4 �

- � - -

tl -1

VIA

TI a. I7

e-ut 4.Imp~ exi -+-vre ul No ,wj w

- J

1.

-4I

' -r--t �-

-- 4 I

- I-n--I-. t I I - I\ '��'A-

I Nit .:y .

a - 4tt444 N,V C tAt.f41

I

I;[�� I ':N\..

I. .� I III $'--1 f-- I -' . ., I

I- NI-

4 It - :� - - 4 ; �2it

4

r-�(E 4K .'

41-r- '4 . It1 1 t

-i1-4- 1 ' . I

4 - 4-I . -± jJ- -- A

4-

[+ 4

h�tw@ tEl. lrnp@tlbr Ezit Tn��rse. tttt�l No. 3, %�I O-4t� MV, N�arS�I4.

-2'

.

rOTAL Pff=&JI E

AV

T*.42 -

Ld/cc

AIR ANGLE I

Ln 0

10de IGBlwNa

~191

My.._ _ _ _ .........

2'1 URN:

. ...........

* C ,

-'A-1

Co C;

z ci

a'a

Figure 185. Impeller Exit Traverse, Build No. 6, 95%o Speed,lo-deg IGV, Wide Open Discharge.

198

- -. - -. ... . r n..... .. q l k. w . n. r w

-, .__ ... S C O W

'1_ _ . . . / • •

". "al

F'gu ro 186. Impeller Exit Tra•verse, Bulild No. ra. 95i Sed.i: tO0-dog 1GV\:

ak a

!(4-

I..-

�-c-�.- - ---- �'-- �

A&�L.

a0

R* a *.*p *

0a -I.

(.

*

� *�

I�

L�. - *.� * * * I* 0a

1�** 1>

I 1***

I -I4�-�- -& 1.

I:

� �l i� - - 17.K � -

1 � I � V

L.� 4

I * * *

� ** 4

- - 4- 4��-�-*--I I�4 �

14 -�

I �- �

I�I 4* Ir-.f �I!I 4 i I

---i-� 1 7 7�'

3 _____________-�

�r�ur� 1S'7. � � � j�4�j \�. �, 9�

io-t� TQV. X@a� Se8fl.

�M)

... . . . . .. ,' - - . . .... . ... . . .. . . ..r"w. .. .. T.. .1 N I 1 1 1I-, 7: ,t ! 7"T, 77 M ýT`17,.71

"I" *i' " I -. . . .s-'.-c'

i - I ' -

a -. 1I I IV

2111I 4i

7k..r". .

-A -A s--' -

F'laur. 188. Impeller Exit Traverse, Build No. 6, 85% Spoed,3O-dog IQY.,

201

zo 0.

A 1MRML .B E (ICHS

20

-I 1154'7r'`?

I . < ~ ?777'

-IL

U.)

uJ

0200

- . I

I �----t*-�- I

I VI - -- I -I---

. - S

I I . I -

I. -Ii.; IV*I 1 4

I. - -. I

t I

I 13 I I

H -' � . . I -

j I I -

0 1 -�.9 I. . I-w I

I . ii.

1... 1 .

K �* i� I* ia

�

-- I

I

n I

A1

o 0 -!

WVI.� ii *�I4

- - - - - � <-' - -

g :1!

1 . 'I

* ,*I I

61

* Ip . I..

C

p

a.8

* C? I)'* r.o

- I

C) /C) CI

* � I >1

'IA II **

'a'

I / I .1

C) �I / .

- � 'ItI-IC. I.

'IC)

.* /I

o p

CI

* **� 9.I� 6 �6 ��I&

Vi�urt 1�)2. Impellet �'w� Travor�ie, Build No. 3� 7(1% Speod,2i-de� I&W, Below Near Stzdl.

2(15

<I

1 7.1 G

10--, IV WidOpnDshte

20

I~ I

I. C!

tot

4L-

Ci-

F--: t I Im elrK it'L-v rw, R Np d

I *

- .

S

*1S0

As U

lii .. - -'&

- �

I

.�I �i -L� (IF

09

t�a iit, j.

- � - - � *

7 I

0£1o Wv io 4 M*hayo

4' 7too

S....... ..... _....

1.i

44

•0 4

i. Ei -

W -s r S all.

t . .

*15CA r

p33

r'. 00 0.G 00

A0ILY' f S

0gr 9,Iple xtTaere ul o ,7%Sed

O-lgIV eo erSal

I21

I O'AA

j:. C --.--

I--

;F jF 0 0

C C

cc IJ

j.L~1

F

7'

shic. - -

122RiJL,19+ IR

Fiur 1f9 melrEi rvreBidN.6 0!Sed

30de IGV

K4

C - -- --- -- -- --- -- -- ------- -212. -

.1 .~ ..

IT

ii~~iI I07

CIO

20-d-- IG

7213

I IA

A 14

Sli-m . ~ . ...

. I

4 .4

14

6ul o ,V

Fimr 20.IWIe xtTaes,1 -

'oa Stll

1 .0 0 *1i4 Oa

*IL TRAVE (INCHES

*21

I7

7-1 '.CAP/

9A

u,

LID

99

o uV

V. D

P.'

:21

* - *aSqh9� � � . � n Fqr.re. 'inW�

-- rrnzrrr�r�

A

I.

C1 %tAN

- I 1

S.

ii /

4Ž'� (9* C

If P A

"I-� I- ivii,! -'

�tiJ

wp

2j C:

a

4 aa aaa U4&OUO WALL

a a p5- V� - W � -r-w-----'-�r� -r--��-

0.80 0.04 0.e eta *.�aR*ONJ. IMAVIt tt�CWt�$

Figurt� 2O�. Impeller �It Traverse, Build No. 3, 3QL� S�@d.lO-deg iLA', Near Stall.

t

�

II

SI I IN ,1

S i e*1 pllrEi rvrsltl o 3 0,-S I

lO-• |GI BI. et"S l.

i• #>

•-:•' • ,•= • •' "~ I fs•i I I •: • ••• • • = - = ' :--,. .. . . •• :• - "" :: : : : -- : :-

-. - - r 1.whw.YiflYr.YImniYwIur �

I

'-I1 - I

.4 - -F - - �. 1�'��

. I I

4 4

I -

'WI

0 IC

C-¶ C

r�J43 �1-j Ir r

¶ I I,A'

-; 1 . . I

-- 4 I -

p4 S ' - - 4

4 1 - I 4

1 S f 4 'C* Vs t 4 I

* Ii �jI 4I I , .

'I r 4r '1

LA' 4

a--- V

� ½

t �

* - - - - 4

C,

pat 0�4�

lfb�Kifl

U�ut� 207. tmpclkr �t trztv�r�e. ikijid No. �, �Vtti-dc� tGV, IMaw War stat).

-inz,�-t �a- - rwrr-rr�r- �- � �-> 0 -- - - -- *

----.

-1-'I

�- SCI'*J

-it2 4)2

B

gJ

C-

w

'-pC- w �-

C-'

C?

� C�.

* I

I L ? V

-� 4 �t' 1� _ �

I r

- � - � - - C�.)

-~' S

W-:iw NO f4

ft ttwc .

Fic-' 20.- ilie xtT vto adN.3 0 q~d

p-o ttI oir'al

0A

IJI

41

Q

04-- WALL

-ýt! r I C. I-, vv sr Lit I~I Ni . 1, S r

-=4

I �CA�

-I

,..�4 '.,

I .4-.

-� .

¶ I *** 4.

*I - * -

*� . * - 4

77j-- 1� I

* �- -4 .�

-I.* . - I-

-. ---- * -�" ±

-. - _ _ __

-� -- �-r-�- -I 4

* I- i.- -�---

.di e.g e.bI e.m�

- ---. --

II ;' 1 v

2-4n 4--. ---

4'~

In-h

L~ I- --T-

F A

4:7. 7- - -

-- T-~~ . [7r

t; ~3-~ - -

- ~ ~ ~ ~ ~ ~ HO -- .I sioiWALL

(0.97

F igure 212. Impeller Exit Temperature Traverse., Build No. 6,1W1' Speed, 0-deg IGV.

I, 225

-~ ~~~~~ 1---------4---- ---

-I

-i~r 1 3 . .mee ExtTaes .ihColnBidN.6

tW,( Speed 14de I I-

-22

'11

A :...r - I

r .... - 7

I- ~- 71

S----

4 -4--A~

-01 Sp ed -4 de I

227 i

-: 117-j-4.

- 4A-Li~

A- t, a;4

-- --------.

4-r 4

4.I 0 31 03

K... i e 215 Impelle E it Te prtur Frv o Build No.-- 6,

-I--, Sped -4de 4--.-

j. 228

-. ~~~~~~~_ -------- .t.A ~ s * - -

...4...................................

8 -4

Figu re. 216 Implle ExtTmeaueTaosBidN.6945 pe * 15dg IG

* 229

ILjJ

945 Sped 10de IG

236

74.

L L . . .. ....

K] ~ ~ ~ -.7 : .:. .. ...1 *. I

t-- r' -I

-- 4

I A

',V,:, r j. u 11

I 7, 7 .

4 4--

LI C

FiIgure 218. lmpeller Exit Temperature Traverse, Build No. 6,95% Speed, 10-dog IGV, Wide Open Discharge.

231

e T

II

I.- -,j,]+'.4

Sped .0da 1U I

.1 i~ II 232)

INLET (W[hE VANE PE.RFORMANCK PRINTOUTS

Summary printouts from the inlet guide vane circumferential traverse data reduc-tion program nrc shown in Tables XVIII through XXIX for Builds No. 3 and 6.

Parameters pertaining to inlet guide vane gaps, including both a strut wake andan inlet guide vane wake, are denoted by an (S), and those parameters pertainingto gaps that include only an inlet guide vane wake are denoted by a (V). Values ofthese parameters were then weighted, based on the relative number of struts tonumber of inlet guide vanes, and combined to yield the "weighted" values at eachpercent span. All other pnrameters listed in the printout are self-exp.anatory.A complete discussion of the data reduction procedures can be found In the mainbody of this report.

/2

2333

It P4 V6 N t1j 04 f"N 1~ 0. U 49 '414V

4' fp gor W V 04 CV V V N f" tv 4 N 04 In

No 'oW $N 04 ff0 lp IV- au (P A 040a P444 4p q -w% b v- c th op *

W4 4 0 4 vo"b 0 oo 0

CA 00 00 04e~ ca JO0

01 .)p~-om

0 P. 'r r- OO 0

N N V -4 4P4 0%n Q w 4- ju

cv ..J

w C) .4l 4k 0L 04 10Q aw J-6 1 l 6

Q4 w .4 VIV A-N. "q )u

.1 w wUUAýrww vw J T LiU P4 AZ ~j-6 4-4. .0 W 1- w A U UWd vi

X4 V;V 2M 4

L- w4 ' %L

~~~ 5-..

at- ki .- 3- 0 2

4K 0- A A ww J~W40w0r d d4

4L P &WW

Nip'A ~A a~Z~DS iUI~

I . .q

- I234

* ~ ~ f **O~ 06400. * @

00@0A

~~z4-40.404@ 040@~P*-~.V

a4 ~ ~ ~ ~ ~ ~ ~ ~ ~ . @ @ -0 I C n80 -41 ASA0

E4 111 o0'PC~. 0A f9 N

000 4. @

- * 4400 a

:8 6%i 4 003 Wd 402 ft 0 0 0.4

:ellw . . *s. ue 0000,M ~ us44V~I 0*F~* O usel

000

z4 in. ~0 5

us E IA L

W #A 0I.d~d V 40302e~~

a4 1"S 4464 a"ad#

00I-s-~~U.VIUn 'Jn.3 a4

1 4 4l .VU000@w 000 Ww

"44 6~nWS~~ W S 0AS

EIUl@ I0

235

w 0 4NN0m"m -N0fbP

-@~ 0 000 0r.N44P4e.4

I- Lb 0 0 16 0 0 f k" a r- r- Qa -000 aN N r- r-..4.. 0 ho N0

00

wl~~~ ~ ~ N o o 41-.NCr

*000 -

a~~~~ ~ ( 0N100Q00~.44 4C4Cz0 - D o 1* -

LA

C4 GO J-.4 b

0J 1.41~ $ 1.44b >

00 , *. '... Z..0 C.) 0 39~ 2!-4 -04. d 41 0 0 6- ChIlZ

-4 At 4 & 4or

ul A4~ *44

&a .1 0

b- Op IA. U.U.04A xA -A

.9 .J4 gf 4

4M 00 Mob- 220

wo w'

a. 0 C~ kj ig

0-44E if % &Cwt~ U4A2 ~ 4E&444.4.4us

W0-~2 f;~W~

op00oo0No Q@QNNN

-40 P N * 0 * N tP h****#*~ P.. . ~ N 4N

lz ;8 AAA0001. Pe4. 41O4O f* 10 Wo

M 04 l 40 f" 4)

* * **too SS*0.060 *i00

P o- 04,4' O4 0 * a oj-

o ow

fli0 a0 ON 4W (-0 4"c t "N

UU

00046 0 '~. O0

we- 4% -a * Q *

6 66 ** se so 64 6* ~ ij(1 e

hi 4A *4.6

44 ti-

p.. ~4%04 P.

F-~~V4~ el* ( 4af

-4~ 44 %A.. 6" i* ý AU00

6- 41 it 4 91Qc

t090@000SP a ON 4 Fb AIN

4U000

it . 1aA eo94N

F 1 p4~~ 000 60de

000ONNNO0*4eS.O00 p 0004

U wuZ4b0S~e b~SSb

oofto~r'do"f~e. 000W~w 0 0b

00 w@E ~ ei04e~ ie 4 r6AW% 04b

-l 0 @dz0f 4e. x~ x ~ 44 a32ku&

4 000 312 V

14 44Wb- u C, 0

~ WIA.UI4 44W bi 4

ae %& 4WI a , w h-*Jb ofs todon ahU4:99'MUwm l

Its 19900

(IN N N in 'A C 0 co 9% coA CP in ItI, t Q%'00

An00OnAANN mo 4 0.t0 0000

0000

V4 4 4 O0 04 P4 0NNO

pq ~ ~~ 0 0. P4010 0 0 0 0 0ý 0 a, 0 0 0000( 4- 0a mrU) in r- NO0' (O00OV ý A00 OOONNN044444t40000 OONO 00 *en

o4P V1 P4 OO04 1ooe# e *40 'A00ANC ~ 000

Z 40PU

ý'0O @g 0 o -P-f4 0 0 0N O0

o en 4 ft.i.-- Aoo 4 Nz 0 O Q 00, 0-rN 0 ,. ' 0Nir

cI ~ ~ ~ ~ ~ ~ ~ ~ P P4 40'''IC N 0'.4N~N O 4 0

W%~~~~ ~ ~ ~ enM, 0w 0 P f r % .

(1 00A 0NN n l U% 0ODN000 In N V) -,-~

0. W%44 I.4

Q~ LL 0z4 P4 04 04V UP 4 .

UU 0 0 ZQW P-w~ ol W40 0 %" y 400- i *00Q4.. 08 Z "~

4% I-J0 0 >>Z Q

>~~~~.- OWL(Uj. O O z000 z 0 V) w -W1uJC1 Q

H Z ~ZKZ44W440-Jmauu1 L2'fL CIDw

4.)U 0 400a :

LL 1. 1.4 c239

____________ (.b .' - .n--.,- . D..- w - -

co to nP-f.f. a 0 1 41".Ok a4 m

4;4 4tz4O 0 0 0

~00~0~000a0AmV 4P4a 44Nf 0 ~ O j0 I m 'aI N0 -

Q00 N40~4v~*aC000 0 0 0 m a. 0 01

ooNN-VN..4ffltfl MQ~Q%N00N000N 40b

04 4 P4U v- - - 4 D4 O4 0 0 09-0 000

m ONN f00 - 4N0 0 t 0 OD0 V4 P4 0cy N r- 1- 00 * 0PW i

W 0000,. mmmf-**** 4 00 mo 2M M~ P-4 v4ý A-0U , 0 00- "d M wU

O O N O # 4 t oooo aK -a4 0

0 24 -Kwi a 0 9- a

Ok P4 JA "40 - ,, -q ý u

. 0 a00a00 a* * 0 0 *0 0# 0000 0 43>>i 9 w~J GOONNN00#~4iA40000 W A4 m2 z

A000 2 woo aj0.4"z

U4 A0 .4 L~.U Loz

-Ac Kzwuwi--wz

E- 0 Lc

rzx -i at 3c i k LUzi U V 4.4 9-x z z ).xz~ nIF LU w V)4

U1 0- LW - -Uu.

44 9-QOO ZZQ

U4-4l..~flVWWU W LLU

,I WZ22D000000wWWO

LUw W x I& w m

240

0, N N00 940m wo "443k .044Ol

Ok P4 4 P4f f 0.4 Go 0 40Vha

0 t*- 00 ce N ot Mt 0% QD 4

(, ON NO *44 4000000

e.49000 94400

*O e****@ea.900 00 'OP0 04

in V4- 4P P4 P4ft-9IP .P. . '4 0 00-000 z.P

oo0 0N"4 m NNN rl"0. 4 n-H~00 In 00p4x4fJ -IZ~ 000fu0~O 4c~9 CL 049

M P4 04 .99099990 f*9OW f 000CKO 004-40000 a C -

0 XJ 49

z 40 10 % MI 0 U Y0 L61- 0- LL

ZH 4c > > j I- u- n u.jJ 04 ~0 094N N N fw 0 u44P-f 1-04f - in 0 z

> 0 0.0 00 0000 z W 00 9-J b Z

ac m W4 u V)in U. WW i

U.WUO

W 0i w V)lO O~

of .>>~ WOO~~zA

ofW 4,j CL fi 1.0inWin~0 0ao0~.

0-. u W *4 -1 -j -us w -0 w us

4c a: ge -C P CgwCA (b vw wU.

WI. Z 14ZWw ZWo moxwwWOxm

241

Tom"

00

o- 0 nl4 PO

&~o 0Nf40t@l§0

40 o Vil C)+Q rot No (ft%

00N N 0w

mN 00l 4 NANuM ~~ ~ ~ ~ 1- 4I-ptt

0Z o 0 ý." -tfil l O0iOOO~I VI WWo ' n w9 0 0 6 40 4 ~ mm *ft 0 W 000 0> iW -4#4W lei ~ ~ 0~ W4c. P-a II- u

Is 000 0 *.4c 14

2u w U..

4r -4 "43 -xj-? 0 . -2j&

04 000 2

0 C3 2 z

U . Iz 42 ocW210

U) LU

b44UN. WrUI 04W MWWVIt

242

MOO Foo*4 MM r- M-4 QOp~m0

OONNNc04'tm~mooA

o' o. 0.@ *~ 1*.m*9NNN Q N~l.-O

000

P40 r4N NN N ýflf N fl 000w m 0 # N 10 mN 0~4Q 0 0004P n( - nW l "7O 0 4t 4b N

0*000 -o)

U O400004444f*4@000

uui"-O@"@ 0 0 0 c

OOc ?JONNNN.4NN ~ m oo -CNI CL0-.

.Q0,4~zoo. 0f0M --0 1- 00c

A 00 0k .J4 0 4fk 4

0 z

0 000. Q 0 A > > j I.- w

000 2 .

W tfl U.. coZ e I- w44f w 4I-w

LU 0 D- VI ) C m

- U. b-B 44 uz3cZ -J 31 3 0 LU..j .. VI 1.- 44Z49 .-.9 U0-

ac U I 4 >W

44 6-000 ZZo'-4~~~I 4J UJ fl uW

IL WwIL j

00243

4.4-'o@NNN4 4D p4NONO0

pq P4 ~ P4@ Q co 040 0coo

fy00W a01 Af 40 Ck4In 1

P4 0 04* P4 V4 pi o 0 0 00 oocoo 00 O

o P Q 0.4 EpN N N 0 4 r-f q0OL vw.4.N4NNOi.4 V %*MON NftC4OONNNOO #0#oIW Coco~

.44

I- OOONNNOO*40040O0000IL(1-

fn V-. m1 N N p 0N 0 0ý w0 -

A j

o w Lcorez Iwa- u -ww-

Q1 09 3 j000 3p u0 M Z q=

de of- 449w0204 l

.e-ir-o &fl

-of U ILN 0 40~ Q0 C W :l w 2P

44. 49 1 v4.z t

4-' Opt 40IL 0" WWU.O4z 49 0944 49 w oo oWXo04-4e~ W-J#f o o 0

OO 4A~ .4 4 I.. JR3.JLJ

.J.J Ou 0 00 wo4

.4- ZiZl4W 3: a

Hum. "ll

~4~44..)0O.O2444

N 40IA Nty NN a o InW% 4 m

Ok P4 0 P 4 P4 0 a V4 0

a- 0 tb00 t. f0mm N a

* N~P 000000W0.0 0000 0

A4 aQ'.,00 -

0000

z 4A

NzCI m fn0.' fn m fn0a. -f a1%. - 4U LU00 NN '0tU0N000 0 -A

0-t 00 O0F-0f49 1 .. V4 P4 " 10 W V4 400 z w

oz~

0%4 lc.04Q 4 LUlf~UI.

49

pj -q UU.JW 9

441-UC~ 0 z.00F..AnII*l.~I4IX ~ ~ -n WWL4I

4gj 1 0 uj 4JbI'4

I-4 WY1 M ý ft w to 00

41 Z4r41-4JUJaWAt- 4IA 49 w 222 J2 j4

wg 4 5 u. C, ld- &. or t of z 4 z (9 IA -IL #A 44E 44IIL44 4C (L4CU.U.U.*

K z A-1 - 1

TRAVERSE DATA REDUCTION PRINTOUTS

Component performance parameters and velocity triangles calculated from tra-verse data are presented in Tables XXX through LXXXII for all Builds No. 3and 6 traverse points. Tables XXX through LIV include all Build No. 3 traversedata reduction printouts, and Tables LV through LXXII include all the Build No. 6traverse printouts in which the data were reduced following the basic set of as-sumptions and procedures set forth in the data reduction procedures section of

, IA this report. For both builds, tables are arranged in order of increasing speed.Information in the title is self-explanatory, with WOD (wide open discharge),knee, and near stall, referring to the back pressure condition on the speedline.

In Tables LXXIV and LXXV, the inducer exit traverse data were reduced assum-ing a linear spanwise static pressure distribution instead of a constant staticpressure distribution for near-stall points at 100% speed, 10-deg IGV and 101%speed, -4-deg IGV, respectively. All impeller exit traverse data in Tables XXXthrough LXXV were reduced assuming a constant spanwisc total temperatureprofile equal to the collector total temperature. Tables LXXVI through LXXXIIinclude impeller exit printouts In which the traverse data were reduced using thetotal temperature profiles obtained from impeller exit total temperature tra-verses.

Component performance parameters and velocity triangle components are given

in the summaries for each traverse station. Velocity triangle components listedin the tables are defined per the following.

w V0VM

/ VU

Other parameters calculated for each percent span include incidence (I), absoluteMach number (MO), and relative Mach number (MREL).

246

0 1 .- 4 t 0. 4 in p - P A ! '

O, (- .*0.4* (f MAO 44 -* ON

00

Q 0 N

0~~~ V 4 0. -

a~N In 04a O -ip-) mO M m ou i'

In 0 . 6000a0-0.

00 04 A 4.

a 43tU 0N 0 P P U 4 414W - x -4 N4 Pr- m0 1-. I' 4,u !

4 0 or- 1170 0 - -4 Wo 0ýi m -0 0IafL04m -

o 0 6., *4 . 44 . 0

'DOOO 0 O.4 0100000 W0. M 14. C

"N a 0 a4m4 . N ,3 - o-.N ~ ~ ~ ~ '~ NNN.4W -OUC

0~ ~~~~! 0U.0~g000U. 4or- I- ;- P. p-I iN!e%9-3m O0 . 40 010 tN 0044 P 44-L L ý

S 0 ~~0 00 w, W% a- 4.u 0. 0 x' Q. P- 0 0 . .-.

oj oýo o o.o o -i L I--4

a"ý 0-0%4 .0 I

40,M - 0 nln0.440U , , 0- C%-0 M~ C.-I2 .Of4F-A UU1J *UU I, ;IL t.. . .%. 9. t. 74 T ) -C t.J

d% m m1'4 44 14 mU W 0 . CL d

Li 0 J A4 UA-m w 0 N4 4 CP. InN 0 mp1- 0! 4 CU 0 t 4P- 4 " a4 4PN 0 4P- Oa ,o - I-

0 0 PON- Q a, 4N CI Q--) 1.'W * 1,-^44 P&N N A .#.?n 4444,U A IQ M 'A ZOO C4.

9lPU A-. 0 aij m..-

W- 01)P-P - -PP - 1 1 - - 44 '00 4

at l 00 01) .4. -4-4 44N 4 9 U A T -.- 4Vl-44411 O 0 .104.1.0 0 0 1 W4~ O I- 44U'44.

00W* 0011I-----. P, l-C 04- *U' SCL .90.-P41'..P m0' 4-4,. t Wl-U4LW

C4 Qý 4 0 -I A4NW W L. r" PQAw ~ o ni -c 41. .o 4& VIPJ. In

-4F W4. 0- * U,4 C200n 330 l.t

1 4

2~47

- 2 t

A "N44 N N P4O 0

046 .44. .4 .4 .4 .4.4 .4

(13 ~~ 9w 3 N 24 k M. x 21 43.4 (AA

Ow W

4L - O E N InN AN a 'A .4 N

.4a *. or-. a a*M .a .

0 48 -4044N0f a -0~~ J~ V n-

*i 06 0- of

t4 et 4;ot

I-~~f MO* N. A In ** W

No-aa nC a fn 5EW.6- - C 1

(El wit 4' * O OhE7 ( 3 U I. -

N .4. 0 4!.4A20 :3W1

0 ~ ~ ~ ~ ~ ~ 4 %L IL* * * ** 45 49 O 0 106 N -

w f..4 NPS O c# ft" U3

040.2 ., N a r.A *"A 0O0 P.A1 a. P11 . 0 Nk N.NNf - 1~I

0~~ z- y a,1O. N N10 ' .4 j dl0 *0099O000*"** 0 'a. 4-

0ab oat N 4- 1.-1- *1- OS- P. 0q 0Uý4b A-4 a. M. 0.

* 61- 6sob0.~t~o.*(. e~l40 P-. a~0 4 4 1

4%n 6- 6 46 1&..4 4 * 0 b'p .--

O ~ ~ - N alQGO O .4~ 0 0~w1.0 w* u I,. 'A0*1K

N ~ R ~ t N N * - . 4 ~ 1 - 0 1 - 4 9 - a .x -. 4 N99*99900*9*.. 0 A'- A C P

4l U -3 w 4

tv if~ WaIw IL ki A A A41

"C',1 W fWIi 6

-4 *441- 1-P-A~p44.40 Q249k

.14

cit - ~io4 Ailp e k

I,. Mf i i* a .,

A :,I ES ~ W0.404~ e

'C~t in . j - 44 O@ . P. IPs.'r

A 4.. -l

.4. 044. .ou004404

~j ,.. .. *** 0,C) 0 0 .

4~' a.~.

5.1 '0 i

o 0 Q!oWAs1 0 -ft .4 0 f%00O v.e-4 045 4 9 *4 N0'45- 0 ,00%, 4. W.4

fw P 0040O O.0 n-a w~~~d ILe a 0sý-.:4 C6 IL 9 -

U-. ... I.5.4 F*f4 .4A

~~," .4w 0145 10 0 P4 4e -E4 5.4

444 .4 .. 4 5% 5

TS .4. S-4 -6 0.a,0)

-. UA w b*-r- 9

114 0QQ0 4C.0% ~ 0 .

so4 .44 04 42A. sob-- .0 a.

ct l * * 0 * . 0 -A * O ~. 9

0 : 01 0 - 90 .5-4 ! 4 .4? S -lA

0 0 a-Al 0 0%94 d0 aU a l 4A 00 4

IF. 6*,9 a9 oft 99 WWP aS.0P I 0 0 U'.,e

a -4 0400 sb tft 10 a 04 640 0A ki-4.

2494

* 60 dO400444 *0.4 40.4 " I d60000 13.

1- 0 a a 4 0 -0P

" "*00400 00 aftU4

4F 44 4 ^*-A 44~

404440 0P 4OO4 -

Q4*4

-o~~~~ a. .8 P.4 p q.4 %~4a

CL0 4 04404SEm 1- 0us A 0

at., I a.

U64 0- ~ 4U

#x4f .6.6 0 u

4 *. 0 C. 490

0I4.60P4 *4 44. 4.kaU*S.O44 u4d 4 4.

m g- a *4 0 0 ao4.P.* .4 0 0

4m *&W% .4 Oa lb

-0 "4.6 * 4~~ " A 0 40 to

.1 L

0- 4k 6- %

hi~i 446 !W..4 - X. 5 0 46

"a at Po 3- e. 4 9 igQ

~~go

14 .4 4 A t4.124 to44

*. P044 N-4 094 a

~d~-0 14 - 44 . 0e

40 4--040AM*A A .0 A 4 (A

I4

(L 0,4'11 *0000

4~ft644! 10 'a r4 '4 U

*~~4 '4 4

944 4 . 44Is~

"-~' R.) 74S

4~~~~~U L ,~i4 P- p. 3 4(

V.B Q 1` 4 -a w-4 :!

.4

fm w 444

fi fa 1% -'. .a 4 4 4 ,j

10 ta - 4( 0 .

P464 0% A440 4Q

W44 04 Q 4a A 9 :* * *

(1G ýk, -. 4(

Iz Q

Q,4 04- Q, .0coQ , -uw c

VA44A .4 4ý A I c5S~~. 4* A .4 *'

OL 4.4 '44i0LA 4NO A" 21 ft4444a4

P, 0*4J.~

*.4" .4.4o @0. O*5 so

:::*a: (p,.

fft IU"Ot ORA .4

464U ~~~.4O.I -46*0S* ow4

o~ ~ 0 *.L40.0~.0U swol

Alm444

0~~i 00 ....,.p * 0"0-M ilf 40 4 13.U*0 t8hit Ift *£f- @d, 414h

*c~e@@s~s@O 0 -i00

V4V

-. f 1*. -6

A4s 4% an i.44 .5. 5 6A

.M

*I 5*@SSSS*04a .4 2E&~ *~SQ@@ @.4@0 0 .us

bAll.4'. :4;A A

0~- .0 4%"~ 4 0, 4

04.401ý ý O'0*04'.4 -

r- N A In 'A 040 A N t.40aMA.4 0044 NN.4.4.4-4b(

N4 4l 00* 0 4 N N- 0 -q 000- i.4 A a a0. A .4* .

eft N'040Ff1 .1 1: -4,4 N : ll* '4 t0It

0

23 C~O t A-t4 04 9- (tr- f.4.04 ,0.00-.4 * .4 0 ftlSI*4r4 44*0%NNNN'0A

1. rU N4- f -1 -t ,

U) 0 ** ** *. I-

9L N 00 0 0 00 0 0'

'0 0 0,Q0W 0 0- tCL 0U -

4C *0 Ct410 0 0 -4 0 Uýo1 - 0000400I44*'

t-* tlo. .4..4N A a'0"uk 0. j . 4 e010i, z . . J U

400 N4 0C N*' 0 000,0 W-0

4L 00 10 ** 4 ** . ** 43 UAa0-I'- NA 863 tifi..4.

I-.

CL 44G! O a 0N 40e.A0NE. 0 0a.

at0. w0

<. 4 4 00*100 0A or-a 0 'ae

a -. > NNNN9LV0000440'00.

*4N 0 0 *4 1 - N in 4U N x5

U NNN04N..4p-4 4 IL..4 '0I-WU i> 0004 .4 0.4,440 4 A *Z "a0

.4 .. 4..4. NOO 40? 11 . 4 Nn O L-4LJ 1- U) 0 M.2(a

*44*4 04 0* 4 âr- M.4 0 0 W U.IA Ok 4s'0'0AO4 m 0 m n 1-WU

*~ ~ - A A00-N'''NO 44 0 w *00- Z

* ~ ~ ~ ~ X 00000000 0W4 U4~ 5

5-~~ I.-5 4 04.

a"40 t- OL.4A A - 0,.4 A 00.4 CL c 3-40 1.- 0 vm ft£ 0 a0 a 0 4 4444lN 1.-1- 91 ! L C4A1

44 O4 11 C. IL w~U I-J4W M4 S (L 'd4 0

.49411.408-A O Cq44.0IL.-.- inUUin 'A*~~~~~ 1.- W04 . A~- 0 *. U.--

I~~0( 4,,* *x~* CL <* 0t-tIL U.

US a~o~o. * IAGOO4A

253

o- 4

N 4 a a 4

U1 04 .0 144% 43 * a-,-e. r

r- 40ca (ato r- 0 -

ID3; -- - ) N A 0 M MaDC4 * 0h' A4. 4 -t f. A

OldO0

OI(I 3 P P 0 3 k(0 0- -U 8o at q. aoooo

CI -o W L4A.* 0,ri

r- co .0 -N~ cc 44-A 8 mA . Lamm 0 *9 ***.

(L -(Ias~.r'. LLC

;: LUL

M ti ý4 4 t 0 00%.11,4uwi

7--*4 1-4 ,_jt

0- 0ILL a>-(

U. 17 . , . . .- 0. O> *M-t0M01'%-400r. A~ 1t N ~ NN N N 4 -m( .r

0A La 0 iy

I - N .- 4~0 0)11 NO #N N N N N N1L ~ 0a ID In D4N(0. .-A .-CO .e.-M( of-

01 A M3 .- t '- 4, A0-L

3 0 fl * '(ICILRz.~- Z> oQa> I 4 N N..- N* c.. N N w

uj l'i 0 j . 1

< rx 0 aN .t-N M0.Nm oiQ. N O.- X C41

w) C4 caA ) -

0 %

oK N' 0.- .1r N4 r 10 Co 1 N4'A -LL%0040 14-Li :A A .10') ON-.4 0O. 0 N II04)i-j

La a. ' WtIN.4 00 O ,,.C - 0. 0 0

ul > "NN 1 : 4* 00 LN * * ** * * Lat 0 5ccN 'o' IL0I0 00 O~ A o 0.La01 5! 1-

;;NNNM1r 4AA4 M a

0. ll-..M . U.41) po -.. 4 --

x 00 h % 1 4 % 31 t- < A 0 w0 0: 0. .1'f

O4>"M or okl MAI--Ja, 4. .qM ý 0. r4 'l9 IX W 1- V

LU W W'0Vu t -0 6 accw-ic t 0 U. 4111

*>0 I-. a ** * * j~ * 9 9 9 9 9 * * 9 9 9 0 'fl U'~ U * *. 0 1- U. U. U

* J La 00Q000000000 44 ', 3 ,0.. * L <LL ILA 2

1* 0.~ 1! U..... to 0 4000 f'

254

MtAN-# .r-O.O Ný4O !4-4 t.

AM 0.0t 16- 10.04 tn 10

In

.41-1. 4 4ýz 0101 00. a 10 a 4)10 N IL

I'.3 2~ w1N N -. 4co 0 -00Z .

\~~V.~ W 44 44t44 *cm 04-.. 4. 4O

2~M' N4w4N 4 e W

.00.

)'ý, I JId.U

A W 0 4 citn30O O0'.0 t- Of

0 L 0 Ot. t .-. 4 0~ 0 0 at .0 Id Wd'0N

1:U.':j

>o ýf ?-rc 0 . N 0 [-r--.0Oý 0' a - :xr4

44 0'.Om0-- ,404 -7 01-0.U. ~ ~ ~ ~ t 0 W US 44 4*41

5, 0. Am '4 0 0 4 ..4 0110W4I

.0 ~ ~ ~ ' _j 1O- W~IT 0%4 .111 01001:0

r-0 NN a- t -T13 4, - 0'

0j -0 t-

4 : A t- A W% 0J 0- 00,01 00 "0 )'

4k . . . <4- .- 1 4-

>0< - . i 010. "I 0 . C, (1 01 t- ax IS

a 0 00000000.0.000 4N W7Id ttjl C;

.. 0 a. T

N 10 A<. 0 0 In A 4xA0

> ~ A .I-N0- O0 Ar--0% 4 -Z, - 0

NN M 4 ^n2tn24LI( A4 LUUO-

W t-> V)0 DV

QD A 0 a 0 014 0ý 0 OlU. 0 t

N N A 0 0. 0QIZS

0 *130 00 0000000 fe 1W-ab (I. <-Z0.4 o- - -L *" U. W Z- -.

.J4* V•) i* I-.-.0 wx4 0 0 0 .7 ) -O Ol r- a 000 142 01U W

o 4 nt-0 00 0 -.-. 44N 1- 1 0U) 0• tnCL a:u•o-tW4 Y 000 0 0 000 WU 01. 7 Cdt)V) L Tc 101)il

ui. . . . . . .c >0 0 w0.- #- Z1) 10 v tJI WW>0f * U OWWMO

C0.L z 010 1-4/1. U. (/0 T

U)J '00424 1 ' I > > 0 V) U.0 U W WI- W -

Ul 1 * * 4 4*11I14 *Na'oM n* 0 l 01U ..LOWU.)

255

A~ m 0 0 tl 11 a *00 r- 0, N

NO4 0 vO 04 -4O

N-0OA 0 00-444 .Qk 04 t-.0 r- 0-I N Z ON N N 434

03 -t %0*4 Nn'tN 40 -t N '4 Q40 *** ** ON **4

A1 0 z' ic0 -

A N4U00:O.~ 1N t-4400O..-0.O..ý 4 0w

Ul .() .NNNN. . .. . .

:3 OWM MO O =C O

UJ0 U.

0I LU

0 mq. .40 -M% # 04 *-Z LU

Q sa- Ui. D.

(4)A 1.4ý O' NO N 000m 4.* a)0M uj Nn mn N00.ON. %.4)N U

', 0000000.4000.,j ti'*0J LU Q..r.. . JU 0Cy Ca LLg aU0.-

>U *I-

j 1 t .- mf N% O 4 0 4(" M(0 f.4U. 0 .. .....

r.4~u 14 0I(4 40g-3. ? u

10 0i * 0.

-4 Ii OS .0'- 00 0 n0n0 nU

.4 0-

-. _- . .0 S .- Of-. 00 2 0 00 0.0 0' - ~ ~ L.((C (I' 0,0N 01naw

1 0d" 24 1*0 0 a .- D(,43, 0Nmo (0 ) (If t4) 0-.0- (3 i O.'4''.0m 0-0 x40U (4 .

LUL--

9N ( 0 I. -0

tn) m0 01 r- 40-N 0-4 .41 4 it. *'nm'N N (4 4 4 4 (00 00 N4 w Vi)t

. I. > .) .

mnI-.4m 0-. MI L.J. -.in- r-4 m <6Aýt

0* 0*. 0,) * I-.'"--1< ~ ~ ~ ~ ~ C Li)'')'''0~0r .1. (v0 Li (Y 1.

2.4(4)0.01' M1 .-4 0(4)00- 20 alp (0 0' 0 .4Lt0 .0 C - i)

.4 ~ ~ .00 00 0 00 0 " (~ N.' U) U)A

Ill 2r th4>U *) U. WUI )0.~D (Y 0& I-U.. .

Mt0 4..O 0-4 . < .1 > .- tL .ui u -i -J,N a 0- 1- 10NIn 4 N 0 -m VO n D I

> > xL 4 IU.4>:0

! .. -0 0 .... . .A00I se C _L

-C

ii I4

In

00 C 0 u

4 4 1 1-4

0 0 U

0

ItW 4 C N .4 o a N:_7 n n oO'N r- LL1

>N IL.~a

~ixn U] m c4aN 0 in 4 N N 0Lm ul t 0 0 ON C1 N4,00 - '4 'cMr 0 O1 .J 01M IW; Z tr-m0 0' 0 a, 0 r - O'

c. 0 0 a 000 03000000 C 00 0 LUJ-.0O

4.7

4 o 40 W.-i N0Qm ýo ) 0 I N4 I- N-i m 0n 4 CnI40 -''N4 0n 0 i-

In. a. Nm 0 N

U)0 xn 0N

xi uJ 4 C; < 4> U~

X 0n LLiI U.nmAO~

UJ0N* .0 a000000 80Om0 0'.0 0,* a* _

J*0 N W INN INNNNNI 4u iz go4

ILO 39 4 00 0 mýANt-4 QI D0't-4N W0 01 I 0 I.N 00.IL1 0 ?Z !.4C *

I!:z LL .. . . .. .r

zWz0 0000000000ý0000000000000 0)' -U T 4 -1

UI-0~~I I N~

-0 N1, 0- a ' fn a .4 in tAG' UAN mr -o 9 r .fn N. N N- In r- I0 L In.44 o .N4 P..94 C) -4" ap ON Z4N'- i

3' 4U NNN4N m M A M M N0NNmANQ904N0 Z .4 .0LU1, W0 44

4~~~ 6-"nItD >~ 0 0

In 10 :3I m1xA a 0 V

0 IN00 Wý iu

d 000000.- 444 .4I-000000000ý WIjJO1 -. I.J

x I a 21a 0,0'''O'm 'Q Q0' G'0 ' -44II000 111. En49NNNNNin N NNNNNNNN4 I-,~ I- W0.4NA 4'Jlam0 c)ý £O' l 0m la. V (PLO-4 Nin 00M -ua ' w

af0 0 0 00 0 00 04 ..4..44in.4NNNN %n41~:I~44 000000000000 0 Col 0000 W ' W

k! Ila

a a- L''0OP ON*r. QmN >t Ix0G -4

r 000000o D0000 O0 0~~

257

r 4J a

a 'a r1

n 0 a.44 N 2

ol C I

4N0 C,4

I~ -4um4 144 '.4, a ITC a -Q..

---4 c U- Mac

N- m en m ' N' c 'a

30a X 4 C 9 300 00l0 71 3 2 N.Nu

x uj M4N m 1L O - .11,1

4:: :1 C; 1 jý -> U.

U. 1 4 0 a x4Q 0 4 4 4 0 4 4

y LI 031O 4 4 NNNN N- 4 14 a,~ .'.L

0. z44

r- 44- w

Ii ,LI LL

a"-N 1- INN0 03' 1 Q 00

0I'

20 'I0

ý I.088 .0'@ I -00, 4j411411 N 00 0 N IN NJI mJ1 m' INNN'4. 1NN I-H 4' 4

4.N M 0 4 D M i4~h 0 o 4 1-1

,0 O'.-2 4%1

InQ)O0I ~~ 4 rI-i Nom .4jNNNN 0ifl

4 1

*~~, 3-0, .j0

01010 14

Li it

U* ol mU~p.' 4r

0258V~

C)C C ..

o14

41f2OP. A4 .4r. m.

"A 4U4 N' 4 0

oPý. 1 44 *P

4.4

A 4 A' 0. AO4

1 0S4 .4 P0

~~~~ %-.

0

A Id

no 00 0 0

a 0P

op A

4 -1

.c 440' c

4c.

toi C N0C I 6W

AI M Id 44f10

*NOW 41

C.. J

X 4

W 'A0P 4]a 0 A *%* 0

labP4.0

04(N.

'4 1aa *a% ý p ~*Aaký 9S.W P 4k'R N 'Q A

......... R A 00.0 f

%gRd IR W. A 1 .2.0000A

I 0. 4 . 0

~0JCO~~Srn

01 .,4,.

p44

4)'

44

4% 3 fd imr 4 oo -ft4D0 p.f

u4 4* 44

464

145 F-P P. F. NII~U

M. Q

P- e 0 0 N0a h0r A ' kau

ck~. p.: 0,

p.~~L 9A4O~

A4 A 4wo NP AA:.A0 4AG (

:.A:. w.w*

P.- 44 jQ 1 .!aI m Vlod 1; 4 Q -a

z~~1- .- *- e OAwb..o1 0 (A Aa;- !ýf 'a~0 *ip .h 4, aQ 6 L

MAI Q 's b 5U; ~ ~ F.5.~:KL 16 46aa. 4A 4:w pA 6

* ~~'-..Li

L~I AJ~a241

C '1

NN

ei I VLw

lo r- . .0 c3 i

Nd

0Q *f A1. A 4a Q N1 (p I,

- a ,w ]j t, tI

;A1 NR MP-TF

~~~~ ~ 4~4 ( , ý

~~*~16 4 NI~~~~b 1ý6..1'1-.3.

_ _A- " I.

II ~~ ~ FIS ll~ 2.

1.I ~QP;

"Ha a I - %. ..9.it~ 4 -l ~ ~ FF'~.0F

I j j L -t A

4~ 4 4 4

P.r J I -

a 4 G ii

4J 61 (J..,A4 n

C*w 4 mm0,w 4 F "64 ýtI-" i

4 N I

41 11

g. f

4 4j

4 U I ~ 4NAI, * .~4aI I 4~47

a a

a 4 m -N ~ InC-IWM A O CI- .41- 4'ý r-C

-4 -4.0 44 -1 10

.4 U, C 01 '44 t--aW IU

4I a: Mc. e4nra- a 30' a: 4

0.- IV43 ,' C04 0 04 Oe (n4N -t a 0 .- 1 C

t- u0m N - -1 ot w a 4l 0 3 ..v

* ~ c >- ~ cc-

.3033 0 0. 0 00 0 .0 3 .J-.0 . .

' ~ a o rr-..i4 ~ ... ,:0.iN0 ~ ~ a:* -

I IJ *-a0'... ~ ~ ~ ~ ~ ~ ~ ~ c a:0alaol4:0.-a'4.l0 -41-'Noaa.li-.-40 ~ '

'0j.J~ ~~~~~~ u .''''' a: N ~ ' 4 -

It1 N~ .- a: G, It- wm 4.> c 4 444 N4 N4444c 4444.4z It3 t- 33

Ua w aD0 1 ' 4X0 cS 4~ ~0 '0 4 00 Z r 4 Z

w~t a-<m o z 0 .l . .N .- D . .1 a. m:- z-.-jZ0 a0 40000'0'a C 0a 0 r-4 a4z-j :E~j.1-

>J.O 3ý cý 0. M. 0.- 10.1 2

03L000ý000000-4000000o 000000 c IdnZ ,

>1 I 0 -

0 m4* 44 * *o 4 0' a:<'<3j- 4'N0'N.ClN- 44 '-4 4-4 a-.A4-.4 ml m-i~ m~ N .4

04--0--44-44- 4 4-4 4 4-4-4-4 4-4-x

i ~ t3 I4 =LLU

Ij LU->Dl 0 UU. < 0m241'm(' rt03 w'4P.-0 nw a.4aN O.-.0 4: .4 I

z o 4.0n c 44 r-0'- ma,0%0 .~ 4 NC 0 C-4t w)' 0400 X-LL >* (n 'o CO 0 0, 0, 0,0, .000 0, 0'a 0a, ~ a 11)0 0 a

* 0000 -4.-4.-40-4 0 0 7 w'-.

>. in KOo 0030. 2US w uClNNn 0 NrI0N r-4n 0 NN'NN- o-a 0 m ~ * om .00 .4~4~O a0 In r-a 0'04 ~ 01I -a, Ln U- ')4

00 0 00 0 wor-.NN I-I- 411)ol oa- M.i cm 0 000 00 00000000q(q( C NNr, w a

M N 1 -aunor-e 0 O'~ 'o cI I- c 1'-4004 a N < zaa- D ., 0I- 0 00 4 NN olr 4N' - Lf:0 t-'0' l IIi ~ k11 1 Ta 01 0>0, .7 .7 Iao C 0

0 US 0 0000ý , H 0 0 0 30D0 00 I1IJ'J -d N' x0. InI

*~~~ ~ M.Li hI~P

x o6 0C, ,o'0 Q 0 l 1 o l -ol4,265 <

of- 1 d m

14 0600 0

Fi.

a. goa

> N0.0 43o1 14

<f 'I in

r,0in~ ~ ~ .N0mIN

I (q 0 Z N NN

00 01. 0 4 4

D01 I II

*~~~u 00 0 OA? ' q4 N

5- A ( mr- 1. 1 c

El~~ .... < e4 0: Ov0 ) ,4

0.I00f t- r4 V 0.4.

Ik 1 -1 0000*q r- 001000 wILL~~ ~ Iitcma,>__w1

w , 1 0- '

4y266

4' ' -10 4 .014 e

4 in4

I 0

I .0

o1 >

0-0 N-~~e t- IAN - 1 .w Mf - -' 4 a0

> 40 0:<c 1vi 00 00 00 0 0 0 0 00 0.A-

zu..

o, 4 -. a -o m00 - - -0

0 ~ ~ ~ ~ , mr- w.oc 1( ot N ,604 t

15, Z.0 -t <

0 0O000ý000C.000 000 00 .- o.w Z! t 0

t- 4 0 InU t' 0'4 4 9.f 1

> w m O 0 r-0 0 44. x-444 lo4 .4.4 .= WQ WZ olr 0 O N -

< 1.0*- .

000) N 04'O00 IA0O.00 00

4 0 00 cc 4I30

4j", ZN0 %0 H 04~0 ' .0 44 0 0Zm 0 r-4 04I 0rNC,00m .U

o O .O.O.O.4.0. 0.0 00 A.""W-.-.o 4" 0o 9w

00. 0

0~ I.-~P Ol- ~ X I A

N40O440ormNN0woNomo In0. w N 0

* 00-4-4-4'0-40000 coo 00.4 .- 00 a~% w -4

M 0, V .- M W%~ Q)~

.04 .-. 4N 4 if ,4 r-Q. MC - r- 4 0 0 c

4 ~ ~ M 0O000 O0~ 00000fw Z7 4 -m i I ' .

- 'i'I 4 4 0N N I0 4 ýN4 .>4 Z;-414~'tO .00IA@4 O~V4

** .I .I .' .* * u x 0 0 0 _

. .. .= 4 44 1 .1 44 4 0267m4

>4

A A Goa00-10

0 0

OA- @4(44*0 f

to '0 n ma

1 'AJ u~.44

4L 0 0

4i z

0 ~ ~ G ON"44 Arme

-j0 39 aOt~l. 3 04044

r..- *000,4 nm zILU- 400 - 0d(4.4f

10 4N r4%.Ifco (PI oN14 N

"r.0 G4N

a. r- w 10 n.

'4 14Nm tinf4'

al 0.4N Nq*0N

-0 u0

41 in .4 4p

-4 ff0 0 0-1 n

0.4 (4x me

NiL a, @05 0 c- t-0 P. A NA- @4 4 co a.

0.e- -C 3c nat 0 0 CA *

it 000004 CL 01 ' 1-. 0O 0

1'~~~~ * 0000 CL-4 0. *40

wo~Cl .A0 N in P.P w wP'- 40% 1 4 A -0

*~~ ~ -4 4 0 N* * 4 ( 4 . az( in 0T05 t"-i- -3 N'44

Li o. A1PrP o 0C£; t

03 Li4 5 4W268

A 0 A N a. P- r 0 do P o N a ,- M 00.. N N a f aMN a 4 100I n %' A in- 4% A A (14 1t 3

e0M-40 .In A N M or N 4 ý4? N4 00 A4 0 N- 400 Na

P ga,4 N 60 N6060 0666NIn06 0 .0 044mAM -Ct: .N N 00 N 004N 0CI 91 71 1P-i7P ýi1 !113 00 ~ pit 00 0 * *0 9 0. 9 m . ... 9 4* - 1

A 0 N 0.6N 0f- 0 AA0 rN 0 0.m4In0 (P -40Nr N0.# NN4444-4 .-4 .4 .4 -4.4 N N N N N M 040

01 zU

4****0~~~~ 0k00 4 .0 0 9 9 -

A M0 N .0.0.0.N0.0.0.0.0.0.0.0. 0.N0*1.0.0.A.0.0A0. -4 us.4 .44 A .4N 0 0,N .4 M AA 044.4.4.N.44 .4 0 IA.v

M.)U N4a0 -Na

3~ ~~~ 000 0 0 In.. . 000 ** 0- aI

4,4 4.-4.4.-4ý -4 -4 -4 .4 4 4 4 0in.

;:I- U.A C A m N0 4 - 4r-0No 0 0.64No-0, M406 Ný P N U.

000-400000000000N40000NN00000 us 09 ~ ~ ~ ~ ~ ~ Q LL 000..0006N04N44 - ~ .I

O O O - O 0*CC 0 0O O0M o I~ 00 0.-. O -CU.U x -

U. NA 90 '0014 AA AA MMN90 .94 0 NO w

of * 4NN to Lo0 NN1NNNNNNNNNNNNNNNNNNNNNN -0 (K 0ui C 0. 0 a0000000......... m4,41 ýG l ,(.0a .0 2a.*10. *1.*10400 m - 3-N~~~~~~~~~~ 4 .- 0 NN4 000.06 Noo4.0U. U.

-4N4 1.4 .4".4 - .4.4 Ia wz %CooNl- ON. 0

N N4 0 N N N 1 -ý

a *00000004ý0000000000000. 0W ZI. 40 00 00 00 .. 00 00 00Mo . .1--1 0

4ý00

.4 *NG.4 0 ..o 0!.-

3 00. 4ONN OOOO .. ONO 0.O~~a.O I.- >4 a.* .0.4N O606 0000 NN NO00 A w 0. NOj N

334z 0. VO0 0'A

.00..400.NN~o*NO.4 xt 0 0c

i *0 0iN0 . . 0N 0 ~ 4 6 N 0 of U.U.00 0 OO 4 4 W FL0 ONm0. a (.(4Q.0-0N N a, W *7 N-p- ý w z U.O

000000000009900000000000 WZ V91

0 -4* at 44 4A44>00 - .. tU

Ia ~ P * O NNN N 0 N N N N N ~ NN a 4 0 QANo;0~4 040 N. .o. 4N0 O P. I-a a CPO0 00 NO in Itn'A

W~~~. U. 44.0 I&. ~ a~~ 3c kt L - -00

4~ ~ 0.000. 4 1".0.8.

a in*. t- **9 9 9*-

o L 0- 0 N 40000 I

oa o

0 0 rq N 00i 3 pA n - 04m 0.0

w 04 0 t4

N f4 .- t o

00.0-, N 4 4Noý O''0-1

v)) 10 0 0Nf01 0 ?-- 4Otý4

> m 4.m 4 0 A 3 00 4. 4

-. 0 3 3f f-~g* 3L P- dim..0 N.4, r > n0d

:3 o0*. A.z .4 0..(m in * 4nt- n h.4 cc

at 0 M I.00 f-. 4 ý 4ý

o04-.d4.f0It 10 I44NNN do

.0~1 a0N 0 a

-c A

4~~1 Ilk.i~ N'4*44A

('000f. 0 4dee x a- ak*0

-4 .4.4A In 0 'A.-.4N

w~. f'4. 00513 44.4

S-0.4-90. 4 10 w L 040 0.44 4 "( - maI- 0 : 044A.wk

,-) J4 "11 44 .4

0 .4 %4pN'1 0. C4 0 .4; on>L .t4O # 0 f.I.u I 0.46 A40P.

0d 'A Li 00001444'i4

30 000i 0400

U * 00000 0.-- 4 40000.0.4004 N a4a r-a

270464044

V1 nn.44

co .* .*. . . . .**.* . **.. . . . . . . ..LQN ('4 444 4 4 4 A -4 J wr-4 1A

o 00. A0N 0 mN N Aý -1-4 I)'0 t-m hA N 0.*. . * . * . . . . . . .* . * *

o. .4 AlN. 1-ý NNNAmmAa

0 0 101!000000 00000 00C.01ý1ýC!000 0 014

a,0 l ' .Oka .cým0 .P. Q** . 0.0. 0. 0, .K aa 1 -. o L

'0

o"Al0 N m r . 0 tm0,AWD" -1

1110't04 1,0ar-r ýt-0 - am 0 04(0 N 4% CA, N t0'00 .a Inxi 0. 4 .4 0 00%N4 c" m m 40 c 0 .Q.'0 InI%9,,au

u.1100-

4 - t 0 M N4I N04 0 40N Al -0 4 -0 0 ( ' Cw "r noo e -wa aj o 4 0 4 A 4' 0 -0 t 4WI

ri " I~E *** ** * ** * * *

x 3:r- .U <4

4 N- 4 M * 4 OILIL ýA N -Aa20 Q, In 0 NQ, 4 4 4* t* A * 4 * 0 a.n

LL 0- 4 >l

Aw I-. 1ý 0 `0 4 AAlol '000- 0. SI'-'0 *4 0N~ co)40 Ala(I Q .

CL .* * <3

0%1 C10 0 12j 0

00.0 - C-1w a, L( 0- kU.0t, 4 'Il'0,'m'0, (I'40 ,44 0 -4P L m 7.0.t 4.l IL- L

o0 ct0 000 0.1L) I1"0 .O-.4 ~ I-() 4? _01. ('

r4 00'O N N 0040(' M 0I'-4 0. -4 M4 0'i *w z 0

z 11 z0 '0'0,0 0 Q 0 0,14t-0'4'10~ A .1--0

A 04Q 4 ý NN44.1)' o. 0 4 (7 a,

v ~ ~~ x . 4 4 4 4 4. . 4 4 0 ml4J~ ý m Iý(1 cýa %(1wt nA hr 1 n_j >'- 44) *ojI,0,2 la Q " 0(.2 0 ,9 MW. 3r-IA2-4 44* I *j) -- l-

4 ~ ~~~ -t00 0 0 00 0 0 0 0 0 w o zo

T) 04 44 04 en U, '04)(nM 0 I% I 0 tJ M I

*~~~~ ~~~~~ .- .0N ~ (% 41 LI'''t 0 0 Q . . .. ..... 0.

0. 0 S NNmmMcqN0a t0P . l4L -.- ' 4h C. 'X0

rýc 0 ac c *(K I C Dcor - -1271' 0m0 u . r

S.~ .... ~ ' 4 W ~ 4 4 V~ 4......... . ...

414

m0040 (

.44 4 *..

go (4 1

z .. N .410P f

.4 Ot.4.N4Y~NO.04-

o 0 0 4 '('4 3 fu%4tjB 0 P

Aw cc 4N ce . 1 LL4 PA Q.M A 61 'oN. A C

40 444

W N NA - Aa A CP

-a .4 N N

LL -t0 0,A z 0 q 4 0. a

40'f"r a p Q

044- N t-, P-~o i 0i 4 1A 0 44 0 4

de 0'4,00-IL. f" P.1 n n**.NP.a 0 4A4lN N

u~~a 4- . 4 4

A 0In m l 4 M n-.4

o '4 N N N 44.0 u 464-4x x ~ 41~' 0 CF P.4-

444 400. - 0f.- 90 W~0 IA JI

N~ ~ 44l a -w 4

J,4 j 4E44'0- 444s'o4 49 IA n ID~) a- 041

m4 *0-,-- " A04 0- U1 0k

04 (30C4 0 0AJ4 43-4 OR P.- Ok

4 0 04&ZI272.

*40OAA4-4 0~P.M-ONO O"OreO00

is 4. ab 4 ONp. 44PObk -e-4*t "0 e4 4 .4 :1

P-4a 4 ft-0^.4IMa ý0a 7 anq# ".40

NOO-4 op. M AP- q sot 6- ON4P 0 p1

0^ .a a4@ ~ * A 4N4.4@ a OaNý1ýr-a 64 4

at 0.0.us 6

*a N W 0 e i

0. Wj *O ,* *o f" f 00

* "11 0 *9 9 @ .. 0. * 9 9 .. ON

w ui le Na s f% ý.606-1em

4~~-44aaa~o4...eNNmi-e 00 0

w4 4 6 6---pP4444 w. b-

0 04 00N4 G*4f0 w Ra Nok406 44 .4.4 ON 40lo a 00 a: J~-I O -A.; 10 : ll1 1: 1; a* el;4; "*W a.

00 E 0.300 0 34 b0D 00a0a

041, 99 w 0

2ta 4 4N06Ae~b-44'44N640460000.

ft43s 0 0 *.. 444% In-4 11000. WAN4pp-..e .4 I118.

. .. .4 . .. .. ..

6616 p 400g 2-aAAA0p 0~ OOWP *.ew....eee04e.eele.*0 0

dl0AAd 04 A- 4 * 4 O 6 'a ftI- P 0 t r&00 . ft 0 . .906

5 4 .46100" 000A

ON"* 0AP " 4014 - 00- 06. 4 a3 .*Op 01 ~ w 0*0*004.wA A A4

.4~~~~~~~~ 0 O 00 0 0.4 0 0 00 0 0 0 4 9110~

'. ~9* 0- O73

go, 904

4--

P4 P A

P44

P9 4 F^ W A0

6ý~~ 994994 * a.* 44C2 4

dwon .44.

o~10 494 .4

19 4 at d4 4 _64

06 449999.

2. *

P4 09 N'4 4.4 4t f" N O

.09 IP 44 P. 0 N 90.a4-

0 ,0 M. A. N. - ft -A P.

0 l 44,P .P 409 4 el 4"

ONNIn

.4N4 -'4 .91 P.4 o.

-

P. 4444 0*444*4 AL F-"a I* In.

0P.NN4P04NP..'40 oUd% 0 P ft P. a 04

'A 0.- 44Pt4.0'9

0-..00"oC0(30.P.440.P.Q

.4P4 ok ~p 0% .4..). P#

a~~~~ ~ P#awAwQ eai~h

4 - ~~4e* * ** Q.

* 04

f. . .O.0.0-C P4.400t.0P fieAir4 4 *c r49 *A -, 'a'a 3-, SANP~4 Ocf

a "ae

-' - ~ ~f'Alt

'A *094P. w

.4 ~ ~ ~ 2 Voa4.Nf.0 a4 .4P. C '490 P.4P.4 0 ~ 0.

S C.AV-..S. 77

0

4405

44 40 i4P

vil^4 ft

.00000

A AA.4

p.oo 0.0.4P4

00

44

N~~ * 5 4

a t

-0 - o-0

41-1 *.* 0 0 a 1ý 0 -

*4P9 P..4.4P 4 40AO4NP, 4044;1:44;

a P. no'P.. -P.4:900"P A9

UP

*P. *0% $ -t . #a 00 .Pd

4'00 ft2" 0 P.4a4 P9 04 Ok9

3 4Z4 9;444 91#40 * OW

2f a3. 99949PPPP0P 39 Ii.

Ms4. 4P 04 S 4

Cbi a*9 *":99 000.

9090

.6w.

pa 04. 1 4L

w 0 O u 0 0:0 0,O Im Ok aA

V4d A ib44 t &*.a6440 MA

w~~ 4 a* a4 0(.0

LjN

ob00. Ot. e 4ef0

a Cciee cceacc

04

o- 0 P. -a

'.8~~ 00 0. 6 80 0

44 ; 00. 000-40c

ve " **a0 1044c.i..

UsV cIca-. *4 Oft ft.

v4 wo

a& .ft 10ot 0adwP-0 0 p.4 fcpi

a a 644d fib 0c

.0.

~4 w t dh *4. .40A. 0 . .0I

- ~ ~ 4 46t "t P

cc......f 00,to. Vie.. I- MaeI4 . -A fd c

A4

:woo&o~

on 06 .4.4ft 400 'i

0 "a-ft 1 2 11

,4 0.0444A0inan ^N-40.

&.N Sn44 Q#4)OM 0440.40

mm M4'0.4 t00 a M0,4054n 0.41.0.04M400.4o0

0N 4,410 of- 04 in04ai.40'00 444 Nt 00. k n

44 N .44 ý4 -4N N ý4

't4 -us .NNNNNNN N('JNt- v

M4.4.4au\4%\ 4n A 4 mm ILinA

a- 4- I.o

400.A03-a% 000 A Mn4 MOx t-Amln ~rr- 4 omo0 4c4z

us * *% 0* US

>- Snn-,O N A0(J0 U~m W% 4 m t-N "0 4 I.-.5C

W N M4 Md\'A04 4 MN 4 9% .-nW 0L

* CL a a oIt,00.0.0 I-naN o Aa\-.I.- A UN 00 r 0 .- V 04.

E-4~W. 0 4 .

A -NO W 0 0 0,0 0 wzx-

OD 000C! ccSnworo

4 N.-4,4t nr % - . - o

0 KN 0 000 4 in ( W Z N 5... wi-i

4 nn100Qlon- ~ 0 ~ O N t4m w a w 1%

NJ 0 ZZ$31-N0 00000. 404N00N000 4 M0 0 U. U

0 n N 4 I0.4rg '140 M 0 1 mMý"#9 lZ

X h1 4 N N N `-.4 00 *lW0.tU Sn* * * * * * * * * U'0 X . . - z ' 0

O ~ ~ C 0.0 S C 4.0a V v. I.-1

F-- 0 'O.$.- U- w zwI.-J -1 - - cr 44 0 ir 4 .W

U 04%4 0 0 a, j, 4( Q40ON 47-4 1.- '0 In e 0

*3 00004- 000.'-00.-"--4 wIs4S cc. 43 d) 60I u4

uIj.&l 2! 0- LL W A' 000O0O00000. 4400 ii W~% t-U WV

W L In 00 n.N`40 w.ý0. 00 W%.d 4 m 0 0w~v o0.44N r N00N4 OOO 00 4-S.lO W\ >v

NI 0 N ** m* ** % * ** N *. I w m

A A OO O oOO O O W00 0- 00

a 4 CL4s.

~ 0 40 IJ.400'4 440 WW279 40

0. O Ook fnI 4 - S~

w a o4 000004

in 0%4 -r t -i

4n-N 0 0Qým39 o aa4.0

0*4r-0 40

ca fm m N ýý4-to .0% 4"m4"

ha 000 Rt04 Xa It0000

co t- --b I

In00 i 0% tq 0 '004300

*U.0WE v) 434

00 4,4

m a 0 a c".a.1.-0., In N'.* 4 '0fIt

40 r-f 0m0.*~ ~ 0 o4 45=005

CL. Q0** 4 '60#3n 0.0.'(3 3 > 4.4

0i Q - fmA r r4f

4- 0 oo- s.- 43'440N 1

L. '0 e 1. 44 e-4a 40

o .4W4.4 0

>4 10'06 4 * .4 emtq100MR ~ 4004 me > I- f4 a

0.5 0 OINOus k . a r-

0 0."~ 0040.4 u w4( -C w Z,a G401- 5ux:- 00 O4

in ft 04 #4 4 -

3AZ So,40.4-0,

15.001-300 ~ 0400000-w0 In r- Co. In4 f0 0 0(4

~W **** W1- .280

3; 'A4 a~ 14~ N~ .4 t- N4.40

At 0 ( WI N 0

ITA I % 4 ( ft 00 4 .44

P- 0.4.4% O Q - inor-

10N N m~ 44 m N 0400.4

40

&ý*4,4 0 ?INOn a r

. . .. .*. .. . *..A 1 coN N I* v-t- t- r- 0. NL

~o NN1

*N L W

00z

('4 n.A UWn4 4(1 mmLO

44, " 0.0 ý 4,. 0.('l.4.a

o * m 00.0 Ao.m&04 &A qN4 de.-zQ: x - (" a*..** ...o.-N JU L OC

w. 0 .0 -4 0 0 0ON 0% 0 0 0 0 0N - -3 - . 0 1; m *

3~ ~ .W4.M' (n1 M' N go--4go0004 04 cz4 4 r- 00 M4 (' 00 at 0 of- - 0

> M 10('LNN9 1 Q 0 n InQI -.r- ('1 4)wI-4

La m. N4N N" 1- N N 0 0411

44 4 @0- -. in

C 0%.. 0Z-fU. *.

4~ m0

00004 0 000 0. -j> ~ N fm A 6

N CD - 4.4.4 0 400 M 40 "OR 00--& 4 - -

> -. I *.l.4-ll .S4U 0 Ml.4. in400 oni 06- *sWo9

"..4.ý4.4.4--4 4.4.ý4.44 Cf - v 'A .I -

N0f-4gN 4-4NItNOON 0C I-t

0~~~ o!N44 t NM C W 0 4 I 2* ~ ~ ~ - . *.N N N- . O G 6a~O Z i. o

**M *L* M4* . Oý 0 -C*~~~~~~~~ 0.-.-.I44444 0W. W

00n a-gin

004 2- "m~O !A Of C 4w 0 0000eN 40C. . . NN 6 llD 4U 0 ~ u 0

* S00000OU00000 w4 '4wo0 *CaIUU CCCzoe SCL

It a. LL (Y-

X 0M ý 01- i A0 0V pe28C1(

1ýNP . f ,ý 9U

o04 G0.4 0 ph0 aa-0 I

.4ýq400.40-M % 0 000 .4

oA 0

-~ .0 0O a

Q-

-. f-. **** 0 nr4J MQ.4 it4 a-5 0 000

0 . 0-4I'aP 04 A vr-

N 0043. 0

(Y ~~~~4 .4- 0 .4..o4(4L700 ý 2 . 4 Me s .-ANr444 cc * 0 9

li Nr

0 o01 0.. 0 ty4 104.N 't h3 (Pk Is a-

% .14~* 3 0 PN0a

N tN %. * 4 P4i*Nq

0

*UU 0. F* 11o A 4 0

fm 0 fm 44 4%~~zo~t &- l04e1 Q-o0. u&. mI¶J N.at0-

P. a cl 0 - 6

x1. T- ** O 4-0 01 --W W

04W4 2L ok.0

OR v 0 tN 34 b. P t6 0FdVI .O.410 -o- * Nf a94 .

*~OI- d- N 9% 0

54

0 00.400 0 00 a0 0.

04 00

0 0.4 4ý4 w 0000

-a 0 4D 04 0

US , 00.r44 NS 00000

04.4 t. in o,t4 '.

rN-I*.4r-*A st#0%

A0l at

10 N 0' A t-04.0-40

VR 0 n%

us w

lp . fm p.*4NN

ýs *6S 44 A

1? 44 1.

W4 0 %.4

9-0 * e.i -*A"a

~W 6* 4 3 ý0 P

~~ 009-U~~fd .4K0.9 U.A

US~~~1 *oft9-9 1* I- US f

US ~ .4b a ~ ai~U

293

.0 Ný 0' m. M' '0 mt o, (144 0'0 **..**Mcc** A * 0

N 440cc 0.0. caaQOQN A4

(%4AA A' A 4% 4 N,0 F-

-h co0 mA ItNo A4 N 40 I'..0

.4. ' (4 ,4% .. 0 N O 4 r- 0

07. 4..04 4 N N4 A47 44~~~~ >. * .*.** .

0 L

, ~ ~ ~ ~ t . * . .* * *

.4 0 0 NA -4 4 NNaOOAO0'A a,(. m-... -4n N Qr N (4 A4J

Dt 0'.0

m 2 f-t-r ýI o "mo oo.octu0'0''0' '0M'0'. 0 to.

4~4.4.444 (4 ID

0 A 4'omtAo.4 4A0(%PI v1uI

0. IL. o

0 4~~ ~~ 0' 0O 44 0' 7

*. N ~> I ** * *** * * JJ'

a 0~ m:ii.---

ý4 ID f-.-.A 7000-7 0~~~o 0'.00 o.t ' 0i I-

Id (4 r- NN r-M r- .- t PN -4 (14 7.'0:OW

I-~f 40..4L

4 t40 ,0Q% .094- LXP

hi Q. N7N1p 71- 7o.oN ,1-. NNz0 I- GI InA.'7N--9n'AA 4.

C: 4-4. 11. T ii% a zw a roo c.o4o 0 0- 0004:

40 12.-. * C

4 .- rO'AN10(4% 4m3 a 0 C.I-~~~~~~~~ 3E 'a'0A'.07.4 0%O4-.

-. ~ ~ ~ 0 N 7.00Q000007 440. 1 *

N **4 ** * .4. ý .4 at 30 U O

'A I. >cAI

O4 A N 4 0 os4 4 A AN t- -U orA 'S 0 - N V

1--0 hp 0 ' 4 4 A 0" 4 4 -M 0 .44olocp A40,30077a.A4 P i .I- W .C,~~~~ ~ ~ ~ "N AA of'M .1

We! '0 4'Ai47 .4 r.A 0 b U W 9.-X

4 ~ zoo U)

14 O4 a 0 c. P ~0 0. (P y, a 44000 W a O. .

40(A10 I000" 004 b-440 * 0

,.I~u b-ooo o o couoE1

gW U O 91 1 0.4e 64 A7 A&4O404 w J"t* 0. ýN4.44-UO .I j p.W4 e .44

~~ *44*44.** 440 4 fU31 4

284

'q, 70@.?

4t 40ONG0

0't

*.0- a0.(P

w Mc N* ý44.Q

w Go 04.4.I t4 (ft 00 00

..- ,4,4 4 00 g 0 t" 0

Nr D i r , D r o.* 39 1 0. 4 4O " t43c -4.0 A 0 i

v f G -c O p. 04 i n1 -

0z

2 a( UNQIAN 0 LLA0 co In

4~~~f 0.-I.0- (a 2D mO4

00 Ll N.

.4n .a 0.0Y4 04 0N N N''-Q% 0 N.t4 A

kn0 40 IA

LA 104 0 UwJ U4(i *i r-N ' r c Q 4g

3 0..4Ai .~U OA 0cc IV-Cm Q. WWWinlý

0 1 a N I& It WS InWL4 %n 0 .IA %n21 CL 00- 4 0q *

014P 44 24wORNa 0.04 U%-~~~1 4044 0 1%S

* 0 w300 4 wQ 00000* I -AK j u1

LS Sn2000 0000A

9 9o i to

2 n85- ftaa6 >4 4a

ii

w4 ý4 in c" t

0 0 f

39 04e10 .4 Q

-n~y

0x 0* S & 0**.50 1:.4E 0; fm.4 CD .4-

'4 0 4 0 p , 4C

04'~~ w N~NC4 S Oca U a nN 0%

za~~ in ' 0 1-

Na M f.4 fmiJb .

ce 0ý4 -W

-, 0.- 0A a~ ý4 Y0> ~ ~ ~ i .405M kEKW 0 a44m>- a. 10 n oP- 0

M- Ul 40~ 404 f.tnV G! EX sS. #* Na%

-~ W-.. -la .0 0 I00 a. 0001.07 0004w 0N0 #0000 -4 09o0q0 00

intf 4 A-to 6-6 A6 .m 1 1

286 F - -F.I

0

t" in 0 in i .4 N C4

m'~A t" O tý 4N at* 'ý0

31 AmA A4A4P P

OW ZC me C 00 00C6 6006

**P. * 0%** 0. 0kc - 02 ~ 0 4'.O4400. Ok4,a.0 U

M to ( a 10,9, m 4 a ca.

.0.4 rq>

> 0 0 A 4.44 N4.4 "

m (PP-- *

'(~~ OA0N OQ 00a.AO IL0004N0

*...NI ***. O_

4 P:

oA4 ~ ~ ~ ~ ~ 4 0.NA AA N 4 O 0

I-.~~~ 0.-.- -

ui~~9. m 0A A--J 0 0 Om0,0 D>0 00 a% m I 6

0. " 4 ;-I-* *,*4- -

In en * (P*. *t**0 4 0'4 0 0114 00,40000 'd A

-- W

A. A 4

0 0 4CiAO 3oo 0,, C) 0 ' v z

287,

U~~~~ m 0.m4a"1

~91

aI * * 044a 4 op.

.414 .4

V.o .3 .

*0* Po- * Ael *44N0@o

0044s0o4#*000

04..4ft0 0 4

400 dk 0.24P P.

0 ~ ~ 1 *0.9. 4*9

10 0 P9N 4 Is. 90094

J 0 so P. .00%0 0

Is Oft 0

.4j0 .4 PP! 040g

4-0 I' 4 0%

.4a 0sE .4P. 55.

0000

LS

.4.1 666. h& ft404

- *000,06% . 00.9

0 oft0440 0*al 40 .4; 4

eg oP.6OP b-.W.S E 65.

*E .4 -. 0-0 AP. g *p 4

P- 00as* ow *.1-*Y.

A0 p.k AAU OP4 V. a-4t9

@NP.a* * PO4 It00.404 ft"-0-.4 ~.0" WOOA

AO A%% 0044 06440% p.o

43.~P U

4 9 40 P, Ot 044 9.

Aa a A A 0 F. I- 19 AA Go

* 9 * *.* . .CL IL 0

44,. ot 9

4 0. 0) "ft

up X 9

049Ap-r0 4 * o- c

Ow W9~ a....0*0 . 0.. 9

Id 0 -a

9...A*bo 4 9449~~~~~~~~W W .fg 6 4 9 4 1 4 EP-

4~*: Q,4 99..eeeppa.~~~.~g 94 do"9 9~ 9 9 9

al

A~ 4%

000 A a

*400

a, a

*4 0#04

0~~ 0*@

4(2 4P44 ao~ Oe4... Q40

0 lp44. * *** a 0 (

0, 0

g. I l 4b"A$4p4

0Ok~

on to

66 A U

A vU- -Op~p- 44*40

S....S...

-4.,E 04 * 45 *411-

0 I

d ~ ~ 4 V4 ul4~4P

4N 'n oft 0 'a 0-

i0 .0,0 ONP ~ A. d

.4,- ~ ~ 4p ?- ~ Nd4 'o

A 5 . *. . *. . . * . -1D o o4 04,U

a

LO-I 004 PSE .4

* 4. . N 04 P4 IS(L

PA I"4

A* 0

. 4" .44 4 & t4 ".&. * . I 0

Q9 ~ ~ ~ -p 6 .

fK

* 4 OU0N P.I.S4I ukLiU() NI~r 441'

' ~ ~ ~ ~ A 4 1AN*'4O'

I

400 •4 *0 0*4 •

I4 -

w 0 .° "°o

... ~~to 0046

.aSk e * : "a '

te e -do. .. .. . 4 .& ..

0 1; ft a

e • Ok C 0a• "

0%.0 4

., 0 0.

~~~~~ *. "k Pt4. lI•4 4

4 . .4.4. 4..

CA 04 0*C*f

* C.4#4,* * , * ° - . ° .

MOM,% 4

C i0 ,

ou a 04 9*0

4 06 Ok0 d

* A t m .6 6 6 6hi Oi 49

2 Cte ~ 6

wook in0.4.6Oin

M-. 0 40 rb t-

o 0.dlit0N 0 00 %

o 0 nwoincar

0 4000,

-Wi4

cr 4z

0.0.00. 9% r- 0%

*00.LII 01(n4.4 d% 00000 6 )

C4I I kOk0-

In a in~im0 N.#A

0~0 0 04 *04Id 0 a n0'

N AM t-rt- 3 00~

an~p 00'in

1-0~~ *4~nA 0

3 43ý N-%'n41.P- 3c 4% t"It, ..4 4

V. M vO .4 N 0% t 00. (P. 4LI 0K

0

0 o M *. 9% 4.00 J-3 1 91 ;13 ~p-i -.o . 44 NI. N -t

A.*I4n a-yt W ~ o- > 4 0N N

Cli- 1 I-A r .4 Io ~ ~ ~ ~ ~ ~ ~ LL1 aI t 00 '.-vQ 4a

IL0 10 N r-4

o 10 p- Ch fn w 0 a... 0~A ý a 4-1-aý4 Ift .- 40. mnK0.

uiO~ I-a.*0.w- 0... 0vU I -NN4IGmw Iin 7:1- 0 004 -00

CL4 t- 1- iI4~ 4 3w .0 >4 4 9% m '-IM

r * *40. 0.-0- -C

X. It N P t"A-.-I-W t-XC

. o. at 0. ChQ 0 0 w- ~ 0* 0 * 0

.4 ~ ~ .I-.-~ .

3J WWV) 0 0

N N41in In.41 44 WI. w 004p-4t- a10 a.N a n - mWW.IO It - 9%O~0.%o . ... a> ~>mwc! 0- . o-

0 M AAP410- us 44wbIIA -0 d" 1*o0.

in~ ~ ~ nu% ona a4, - Da1

J- 11 44m lm 4 4 01.4 4 0-PNO0 z In 40 00

ol1;* 0 00 0 00. OQ -0 000

00000 80000* 4 lm44f-0. 4ý4býf -

293

0m@m" M

.. 0 0.. W . 'S

a. ', - 10 'lw 14

-0 -, , ,4 > in o'o 0oo~.4fl04 fit44no to

' to I N -24 t- a4

4p ,f' .40,Sto r .•. 1

3 P 1 f - O

in- o,4 .

6" 0 .J in m 44.40 EO

4I14 0.3- -in o% n • 0

w I NNNNNf n f 4444 C41

V04 /I. Int 91 C4 10 . in."4

0..

44a-Ot-.0 .4

. in ont- 3 .

:•. nl M jn I n Op wok

O.ý4-4 N44

N-j

0 4 be4v%OW u 443I 0 044.4

- ~~4*4ON"- ft0I *0@4 fmt . e 0 .4ft MN A

I - us,ul- J a-.ý40P 0 (o

44mfna in N . 04 N-4 1-EN

to4 0 .w0. 9 * 0a I-d4a m -t N M1 tA

0 ix 0t"4 40' 0Ofl- ÎnOK > '' a a .4 N

4 a. 444 cea 'dDftILu4 ) O.4 0* f

ee CL0. , in3-- t n0P~~LW~ *0~N 0 ~ .00 w > 4KO. IL I-4 P 43 4N4

v~~~~ v * 444440.00 w3C4

oo4 a- CL6 4 1.-J 6- a- "I~4

hi 4 at 41h4 aU4P - na*dww00.- 00 ow a1 It. 4

NOOý4 1 I a- &WL '. 4f FI- 0 0P'IN.944M4' us doW51 'qe - 04 in

a- of.5 *tuEP K zw ORot0P

0 0o: 0 000

294

.0 ý4 .4 0? N0 .40 N 0

IA4NIA *'AIA fA *o-~ IA

01-***.*000.t- r-p

IV, ~~4 0 0 0 0 inyi 3 rý l. t in0 4A e4 4 -o40 0f

3 .NUOIA4~~P-.r1.

416 (4 a N A444.. a f AIA*a a a rý - 1 A

(0N N NN N N IA

0 A N 0 0 0 I *a N-A 0 a W

IA @0 ~ ~ -.ICII I-t

N A I 4 0 MN0 04 0 M 0IAeA N l

~~~ 0.0000000mo0.

4NO4,4.4PW 4dM I

CL N 4P.F NNp- cm f- F- (A 4 44 4mI-

0 ?.04 0 0.4IA04t0r-00 - *1 0 . 0 qmo w A a* 0

0 *0 NoO O O ~ o A P--0a44"-4~ f o ý I

4A t.- .J AN4A N4N0eNt-.4*.4*44. . 0 ew40 - 10

4 a 00p- 4 0. P- A 4%'s O 1 A

w 0.

N4040 0IANP-a, O. 04ga 0N 4 **o F 4 0 oin4-0

to 00-000.00 00 00 ..

IWO

o ~ ~ ~ ~ ~ ~ ~ r bre OA~.A4 . - *.- Al

op~~ ~ ~ 0 004 M O ,To P 40^4?- 0:0444.4A4 : 44.AI , p.

7 101:t 0-PIIIIIIP. .4 IA. .*****544*.*..* xi i

* CL.. -. 44 0 2W4

2954o p.

0#100 000 0 0 In

0 0

f. alftf 1ý 0 R. 4

of. I. N.ON

ul .44...4S 00000

K z

0 ab ~ ~ 0 Ind.0 Nin 'a M % 0 0.4F- f

MS on4 n -4.6

4~f NNC 11100 1& N .4 a f

00

t04 P.0 4 l f-o~- w 1*".4 6 4.4'4O4

ono

U- QPI #1 t-4

ljz 19 * 4 4J of .9 wl N tI

J0o 4 =AJwz N. 044a

4Lw ft01 0nItI#$O-44 W~ N ,0" t10

00 *.4OE4'l04 0%

4 * 43

u&s ka 'A 0

~ 44444' EOth 40 l w U p444- & IS~. Q * 00044

SM-

A~~~ TOO~ *~h

00000 *@0

10, 000'SIX

44

ot 014 4F4 44ft ft N M0

o04

0.4 I0% ^4~oe a tr 4 -

IA4. ?4U

14 f0în004 M4 t 4F OW 444 000C4Fftl4W#4C:t4 U~

z ^4bo lp44t0f444 -

00000000000ft 0 ".

e4Nftft#4Nflce40 .-

144

ow wti *** *00 090 W" 0..

*we 00C as w1:0 A 4 'A

am "" 44 a*-,4e4

0 9990 00 00**0 .00 WE

oQ 06-.. .44 4 4444 0 b. 6. S

6i A. '4kC ' Or

71 4 d* 0 0 Aft-6 a0 & S

-**p4*.* a* &-*4 *4-4010MIM P0 U0

30 *0-MaoOO 06J.444

4ooook 0ew

tv. 0O a 400Noi0

0 000 0 .oI

e-,OI~.0 0E.4 f"

000.4

a LL, .4 .4 us 0 0

4'4'aa'

0'n04I ' N-r - - o

us P4.4 '

0 dff ?' 0 ýo 0 cO@

W CNI r444 4f c

ac 44 .. 33

N ef AC4aa0

4~5 00'04 0l4~000-t a % mP nP

M 3 -CPON0 V"7 00#v; 0 UA

*I PCP t N a333

P40

4~e-a0- 0-4 0

t 3 ti4 0 4 "w - al%3 * 10CInA fta

A01 4 Ito ft?.0a3~dt~44~a*

P4 "'a4

4% 41 0-30,

- ~ ~ ~ ~ ~ ~ ~ 6 0 4hJ*ý - ~ 3*aP444. X~* .4

1 9i A C

-~d P4O . -0 P

o 0 k @4.0 1 0 -Aj 0- 4 toa0

a Q C

-'' -, , . . .

I, I,4 t l,,-t

4 080,0.., •,

0'2 00

4~~f C4.#o C4 N .f

,,.4 v, 0 in

0 *OO,;#

0 nhwf 4p4e 04 a t-

A 1000-4.4I •

U_� . 0 " %. ....

S0.0 0 " elNr :! ftN I,

4 0i000 0 .. ..

00000 0 04

0- %54 45 , 00

GL in 0 0 N

old, 004%4w *0.ot .J fit00

Cr # 41000 i Mr4 9 4 C4#4t aA0i

.4 ON ft ? 1 4n .t-f"qIn91.0 o tA4

6000 ~ ~4480 00Oo

Ifoo ki 45f

4n % 0 wEO 2 pO

.4 in CO~ *. t L$".64D *0-u "1, 1 04440- *009 .t ~I. u- I4.~

'T , - w. $lm o. 8 w -0--4 -a J 0*. 4,man* 0* 00 d%

U0 on 3 d% X'S 000100'4 W 4 44 .00

04 0-p4* 00 0600 9 3 £ 0Ow

VI w a 9.0 .b

299~

43r" U* .

N -AC N N 0% m N M40

N N n 4 N N* 40 N N4 N N N N

N~~~ 400 40 0 Z N w

M M ?ý .4 * V

10A M00 0 .4 (oco A a " A0 0

'C~n >. . .4 . . . .iN .4 0 A -aa

U 0000 000*00000 .

on .400e"N.4'40MN40nl

4 No 4nNO.4mNO.NO 4p NOc

4 0. AM.00 Co o.M mt 4LI

-A 0 054~~~~~~~ :4.40.*~4N0~)O4

z *Noah00N440.44 Qb 01.> 00 .4421 M0 . 0.N4- 0*b 04 W~~~~C 0OOO..O...O00 V4a

a00000"400000000 0 -. WE w04 e 0.

V. a4 ILzI -*N r4 LiL 4 0*40,t og 6N .

O *0*O~06PN W a w -

.4..4'4.44.4,4,4 ""4.4 " 0 4 A I.-32.A .4 2I

N 4 4 0 0FFNN 40 04 0 LiLi. 4 N01

0~~~~ W40O0r'ACN IAJ Z

-6;z)4 4c -44 0 a 00, 0N -P1.0 w o 4

us I- & -ýb~i0 A0A "R4 0. M 0 00 44K

Li 4 n~~~e~Ee N 414 cl N

A 4 A. p~t " 00004'...4 P a . U4'49 V I&- VA

444N.400N404.4A Q 0 0 L. Li.h

300

z 011 1",.

0 k 00480% 4.0 00i I

0 0040

-id4p in~~4

oh44tof- 4ir .

w O"4-ý4 w 60000

."" I

lo00 P f-1. 03 n1.4

#AJ 0.10. 40 -inf 3 I00.0M 1- A 4" 4Nin N ft c in

in4.

0% 0

*f z I a mO*0f # *Ou0 - p - .c

4 9-

am3 1 -j

0 inK

04 ID0 4N0

in 0,0)4%( -0

14 4 [email protected]. 4 '1 4e

ae 39C- In,9 t" 01- 4 a

w 4 v f,-4 0.D f0 *... 0" ftgw -- 9% 1-

w- dO n -0

*Jt 0.40n- d%001-S

'A&

301-a

o 0.0

on..'.-~r 04**.

ON* O.4 4 1

0 20 P0*

N ~ ~ t ON460 .4

4.4 0~~~ 0%4.* T 0.4o(L K 3 ;o - n -

TI "00.44.44 * vO4p.

'00

4

43

01 Aqw aQ44 4f%

'A 0

441s

wo V 0.-0 fit 3 90 44

%A f-al 00,n W ft#;:00 d'4 4

0 44 3

600640 a4 a A 44f

03 P4 Ob *od%

-w~% so.. %14 a

.4....a Q41 04 .1 40%4^P- 01 dl -

30204

0- 0 00i-

ob4 a I- 00 a in 4

.43

A. PiP. 0 0 PIP P0 I- omi

S0 A 4 .4 P4 ft 0 .

*0 4 N PI04P-004N 45.x

.4.P N. ON. 40 P. AP

b.

... . .**.*..A toi

04' P .4 fl- a Q* aP S

444 4 4444 ft d

000002000 '2

xuO~oPP 44 44

#4 b* VP0II4P P.. P-E if

@9999S 0 a 00408

.9 @we9 .. 00000

.P4

0.49.91a

00

ago .4m49

a . 6590 4046.4,

Me o a..ok~

ut55 j 0-64 A

0- T Of A -dvosStg .o..

a ~60-442 2 00

* 191 ~ 56945 4 ~ 99154

06 0 0 9101T%I" 11. 04W 6. .1404 06

aNA

W~~ 44 6 4.5

304*34

O 04

Ok P.

P.P p. P0 0.

In.1 0 P. 40R.*.

* Q 0 0 do%~

MA 00"0+

** P-~0 0

0*4 WOOF* 0

04

Vt .*4 .:4 ~ 44...WOO"*

'A4 LI%1 9. .a

:! if*

ca 14.44 a *

49 ; 3 *P'. - - 0 R'

a ~ ~ ad .al

44 C'k4

II ftJ&1 10

op *@* am 440 "a*

.0 .00"Moll.4

OUR

N4P4#4 i4 N #4t ft "

&*S

'C~~~~~~ 0 .Ot.E *44

69-i

.04. ... .** * 4. 4* . .. 0 Q i

moI

4~ ~ 4 *Q....

~~~~S 4 "A* P*~L

S .40

*.e.aa*a.S~ 4w**-

b-,4IA44bA8 .ý.0E I

*444646 6046b44 44 4MA. *~6.

ok .

40 _ _ _ _ _ N__ _ 0 __m

-4 0PN0 ".4 40 01

At f" 14. 044.0 -4 N# f% A

0.4~ ~ ~ 0 0) ft w-0 t

V) 1 U- 14 4f-N4 4 .4.4 a. IL.

in iWO 0 a L 0 Nt P4 4 0 a A '0 0- 0

Q. 4.4 444 04 m

W% Ia.

(K N 4ft ft O C 4OC4It (4N co0 9tol -C * ^ * w Mt 0^ A A ft.0I t . r(AN4)-ooftoft 0 a -.

xu -- Ij4e Ic a

4j .4. 08-AffO

0. 0000.0ý000000f4 NLhj.0 >0 I- a"I

I-I I- *a Iw 00 *.404f(-ft4

[1 O.W A4 ..i ~ ~ * .4t4 (a N 4 6. (A f

0- 0A 0 N4f040 N N a4 a,*i 1 #o v p

0- Am* 44

Inpl 44 0

w- 00 0fII0I0000060 4-"". 0% 0 * 3 0 " 6- A9

V.4P4

%s2 .a ~4d

0 1i 4944 6 I

Z K ft 46(0A-N. al *4 60*00 I- O -L 6( (~m, a

3070004000

P-fl m4.4.00n* Won 4~ A

is . A 4 A A Q 4

t. 010r O64oa.24u -4 .4 Q.91

030

043~~ N6AfN40of- 0

A"P N ".4.4.4 .4 P.4 Ni 4

op 3

4~ fui

10

a IL

r-~. m. Coor o

0l NO U 44t 4Ao

V) % A *. *4 F.II -- P. P. 4in 0.4w 14..4-4.4 CL [email protected]

6- b.-

Or-- I- l

0.W %:4N 0 0 u. 'I 0 NPP

L b1 * 0 * a * . o ft * *N

144

0.54 PO 0. No - 0 d% -P. a

4~( r4 4*, *..* . ..4vi44 14 1412

e5 UA .6 *Ob

C6 * 'W jN*400a406 440

-~~ wmW Z211 *S*b94lf4 *

%I a. * e.a *a .0 a aIP... P.a

Z. li .."I. w3 I o I

a.~f 0.. a a . * 4ý a0 tE4.04~~f4N0M 4 .4 *,A

44~~ ~ ow a w op : 4

3~ W

4..4P4 .4 4 Pe O 4 o amonO

3 , n P44.4 N P- M W 4

64 0 m4 m m4 9 . SI 4N

In. 0.6N~0. 4410.100400000*00.000. ..... w ...*0 .... 4.911 b

0 0. 0 a 0CD0 0 06 0 0

-40

(3 004 P.O00.04% N 000044Nf fp.00

0 00000 0* 00*

.4~~~C O0 0.fl l 0 -

* ~ ~ 0Ad f ... 4 4% .04..4 4 .

.4 C U.

w 41 04. M4NP CS 40 04C

a -4444400 O P.A N0 4.4UO .44.4 .0.4" f

li w-) N

-1 C) 0004000"000000

4 O4Q411!" 0.-4-4.4

-tf Qf 0 IM.0.0P0.N P.

o1 00P- 4 040 w ol-00A4644

-WII"^ 4 4

Am.P 40

0 4~4 m 4 4, G*a0*E. 41 0 43ft

4 ~ ~ ~ ~ 4 4044*~40*.

0~~~~o 44 A44 00344.0 0

3090

C4 0 .0-y0 4r

40

P4.P4 0 t-.44 t. In000

.4"AU -4"4 A to.

n4 P: 040 P* ;.44a914..4mom.44..4.

0 AIZ 9 . I- 4&L* .... .. *I.

'4-~~r I-*4@ 4 e 9 I*

94 404.4 4 4P 900

O94in 0 40, 1400%9-

!941 10000000M"1If0040 U;us P.-4.. IF4..4I-

.4 ft44*~e999 61 Q h.

461** 0-

.444. 4

4 .600494** Wgo a.. go ^-, 0 4009&94a44.

6 00.00000 4-00

CA ft Mot WUa 4 4

* - a. .J 94.40944r,440.

0 of-A Oft 0- 494*49 fta -1 4

94 .. A*S9495~4090Ed 9

310*. * *..

40 .4-0 - 044am00.40A

A.40 o 4Nfp.MORO.#o*

-4eo A4 ;10 0%t. A so 4AE41 P.4 1- V P 4 -4S SO

* NA A4 P00.0000.4.4

1. %2 CC 4 00 NNoa 4L 0

Uh w *..IL...b..I.-O OftA 4 aC 4o AN 04 AN

a&,*I-me mol A.40 *0 .1010.1

Ch

NP. AA is 40 aNN~.,~ ~ ~ g iR w44 4. 4

AA446OA.A44p.A NP, 0a 0 obA 40SP

to m. A

* ~4P

40 4444 4 44 "aOWAAG&0***6O ** O 9i 4. -F

4 *m.*mV.*let&

-0. 0 -.

:.44-~ 40 d*v v-

00WOOOP-aft"".

o t 0ft P 40Vf

0 02..* 0*g 4EsAA444Aft4.s***** *~ *p. s,

to

s, 4:.4.** * * ' A

oft-61u

4~~~~~ 4 1ttff~tft~

vu 0.44q140 0000 -6

'4( CA 000 000 4 0.

IV , OIiii 4.4 2 e

Laa.~~~~~~~ sa4.444... .4. . I

w~ Lt

WI -AIft

OR Nm 4 0.4 4 A404 N .

* ~~~C -ns4 N 0-44.40P.04 4 MAVs.4 Ono 4,4.0An

p-u...0000* P. .

4V A 041

d" 00000000000000 0

44 P. p4 P. .p.4 p.4 NA4 P.W

... 4. a 0 1- .4 0 1- 0 RN -

49.

W N* * * *. . . _44 ~ ~ ~ ~ ~ ~ ~ - u P........PP9PPP4400000000000000 4 I00

itt

*~~~ * 04 0":F. s00000 00 4 4NpU04

do 2 . .. o .. . .0-W ~ O4N~pp4.4..4.Zola.

FA 0 fP0.o o .6""N -44P.UWAA..P..4P Lo

A0 Z ZO: j 4W

00 4F 4. Pd

4 K 4 404NP 00A .e #A I

000.00-00000000 0, 0

4.N4f R. 0.

00 P, 00 NN 0 p 4. 400; 4p, .0O- in

-s-6(s~P. . ...

4~~t ot 0.4A A * 4 . 0 S 6

30s~ko ""00-* O ** * A!0 OOOOO GOOOO~o N ý 6.I

A -8 C 4cb4 * P4% 0 : 0 z 0p.04.a '40 it

0~~~~~~ N4 .N . A44.A P.11 P..4 ~ ~ ~ ~ ~ ~ ~ i 0 N4 A0U.4.N 4

*OAAA O.AWP4. 29 lb 0 '0.4.4.~.4.440:8.4...... 00 -OC

46.6 ~ 2. a-a4 A ;I

*44N*A4..0*Pe. A mi A( 464

313 NN0~t~4O. 4

0o 1-* A00040440ORO

.4*.40 0 .1 P

t4ft*.n 4*40v4".400.

.444 *f* **06 NRC o

4r

9. .I

3~~4 0d -4 ---~4.0u

S2b ftdftfta ttftfttrAPft AU

*j u

2 ft000000000000 W-40p ' 00)L .A

41 2. b

ac 0":::: o* 17 ft .4h 0 N 4 0 1?U t I.1

-j -3

0~ 000 g S I4 49 #3

of a- m tA t.0

I -)0.f0V .4 A t. 40 w 0~: s-40 U

N ýO fý *0 * 00 a44 a P IO 0

%j 00.4000000000 k* k i

06 ... 19tP0.6A

-J~a P. d AN4~M

.9 oft cl 040 wasko*ft ft 4 4 Pah a 04% 6309.43 0 .00 &. i #I .

O.~ O 4-0060.4-

4 .004

i~~~..J00 A *'14.* .4 ~ 444NM ~ ~ ~ ~ 3 If.444.. If a

UJ4 .1 00314

STATIC PRESSURE DISTRIBUTIONS

This section consists of static pressures along the entire flow path of the rig forBuilds No. 3 and 6. Figures 220 through 229 consist of static pressure profilesalong the inlet flow path; Figures 230 through 239 of static pressure variationsalong the impeller shroud; and Figures 240 and 241 of diffuser static pressureprofiles.

Figures 220 through 225 are Build No. 3 data, which had a straight axial Itnlet,shroud, and hub static pressure taps from the inlet guide vane leading edge to theimpeller. Figures 226 through 229 consist of Build No. 6 static pressures withthe converging inlet, which hd only shroud static taps. These taps range fromthe inlet guide vane trailing edge through the inducer. Figures 230 through 235show Build No. 3 shroud static pressures along the impeller shroud and Fig-ures 236 through 239 show the pressures for Build No. 6. Figure 240 is a BuildNo. 3 diffuser static pressure profile for 95% speed and inlet guide vanes of 0,10, and 20 deg. Figure 241 is a Build No. 3 diffuser static pressure profile fora 10-deg inlet guide vane at speeds of 30, 70, 95, and 100%.

J31-

0 1!

311

0 .

Ir~

---------------

----

[~ Z~ii>

- -- - - - -

50 u ----------

V El

I r

-W4A

4a Yt

ICI

-If

---- --- ---- ----

00

4.Aj

D~lowNOW

to 00---------

Iz

ot

at 0Wggi~d~LCl

4.

I --

I "p 4

/ a

VE te/ S '7to/ iI&

P' ft~

0 Sitowu swiac Pimea Tap

* Kim Stati

Snl

1. 40 [nntSplitter

LeEn

30 Edge

0'ý0 -0.2 0.4 U. old 1.0 1.2 1.4 1.6 I's 2.0 2.2

DISTA?4cH ALONG IIIILLER SHROUD tm

Figure 230. Static Pressure Variation Along Impeller S] -oud, Build No. 3,100% Speed, 1O-deg IGV, Near Stall.

-0 Shrind Static Pmrvsur Tap

Etta Swaic

Av~~ of Five TopIt40 Sma

30

Exit

Figure 231.. Static Pressure Variation Along Impeller Shroud, Build No. 3,95% Speed, 20-dog IGy, Near Stall.

321

0 SMioi. ShIM Paww.tw

so

1 40 k~u

II

L41

420

UniU 0 0 A UA U 1.0 U. 1.4 116 1.8 %a 2.2

OWSANCI ALONdG WU.MUA 51404J -u.

Figure 232. Static Pressure Variation Along Impeller Shroud, Build No. 3,95% Speed, lO-deg IGV, Wide Open Discharge.

MUi

,91 40~S. , U 1

Figure 233. Static Pressure Variation Along Impeller Shroud, Build No. 3,9ki% Speed, lO-deg IGV, Near Stall.

322

$"(0 A* I $1141W TAP.

so AWO

Figure 234. Static Pressure Vanriation Along Impeller Shroud, Build No. 3,95%i Speed, 0-dog IGV, Near Stall.

tq 'I 'A"

vi0

707 Speed 10dgIV erSal

323

~~Su~~ -27 nw.7ý 2 ~~c

~~~~~~~ I01j Sped -4d IGV Tieopeoichre

S. _

Ki '4

0 -4-d Shgle IWV Ne Too

30

-aG ~ . CA OA 0. 1 .0 12 1.4 L6 IA ?.A L2.

USWAMc ALWUMUIGWNL SNIOUO -

Figure 236. Static Pressure V~ariation Along impeller Shrvoud, Build No. 6,1011/( Speed, -4-dog IGV, Wide Open Discharge.

0 Ake"*SLJ U" MI 14.To RIM SIwAk

Aaupotf A" Tp

LEU

Figure 237. Static Pressure Variation Along Impeller Shroud, Build No. 6,10le Speed, -4-deg IGV, Near Stall.

324

I. .- _

.'I

Va I i

i...

0. L .

LOAS A WftwIJ MUM- 6

" ~~Figure 238. Static P~ressure Variation Along Impeller Shro•ud, Build No. 6,•,1O95' Speed, 10-deg IGV, Near• Stall.

U4 4A I

W-, 1101 0cegIV.N SalIaktg

4325

.4 .I 4 .O t .L & .& I . ...

0 N

0 0u ca f

-ado

UG 02

000

3262

lIG;I--FAMTE NCY RESPONSE INSTRUME.NTATION DATA PLOTS

11igh-response (Kulite) data were obtained during Build No. 6 testing at a near-stall, steady-state point and during a stall transient for both 70 and 78% designspeeds. Data were recorded from one total pressure probe at the impeller exit(header PTIDKI), two static pressure probes in the diffuser ' aneless space(PSDVKI and PSDVK2), and one static pressure probe at the diffuser exit (PSDEK1).

Plots produced from the data are shown in Figures 242 through 296. Each figureincludes a plot of pressure vs time, peak-to-peak pressure vs frequency, and across correlation. The cross correlation was generated to determine the com-monality of the signals. It was accomplished as follows: Time samples of tie twowaveforms to be correlated were divided into a number of segments. The cor-responding seg.ments of the two waveforms were multiplied together and the re-sulting products were summed to obtain a value that was representative of thesimilarity of the two waveforms at time delay zero. The segments of one of thewaveforms were shifted in time by a chosen amount. The corresponding segmentsof the first waveform and the time shifted segmnents of the second waveform weremultiplied together and their products summed to obtain a value that was representa-tive of the similarities of the first waveform and a time shifted version of thesecond waveform. The shift in time, multiplication, and summation were repeatedfor all time delays of interest. The cross-correlation plot is a plot of thesesimilarities vs their corresponding time delays. Plots produced at a near-stall,steady-state point at 70, speed for each parameter are shown in Figures 242through 245. Figures 246 through 251 show the plots produced from the total pres-sure Kulite data at successive time increments proceeding into stall at 70V speed.The time statement on each plot refers to the day of the year, hour of day, minute,second, and fraction of second. The fraction of second code is included in thetitle of each plot to aide in identification. Transient plots at the same successivetime increments for PSI)VK1, PSDVK2, and PSDEKI are shown in Figures 252through 257, Figures 258 through 263, and Figures 264 through 269, respectively.Near-stall, steady-state plots at 78( speed for each parameter are shown inFigures 270 through 273. The transient plots into stall for each parameter at 78%speed are as follows: PTIDKI, Figures 274 through 279; PSDVK1, Figures 280through 285; PSDVK2, Figures 286 through 291; and PSUEKI, Figures 292through 297.

V2

AA

'4+

. ~ ~ ~ ~ ~ ~ I . ...I tI44411, 1t

+ i-V

tA 4-4~4

tIJ~2 4 4 4-

4 4 4 ~ -*4~ 1;4 tt

2~~ I. :22 4 ,.444, ,4t4.

* ' t4-t -~ ~ '-~- 414 4-.4 - 4 -.. 441 4-4 4-44 -F'-U

tt 44 4- 4 - "

-.- ,-4---4 t-* +44- 4 . * 4-f4

4-ii*-44I- ,- 4-4 44-

~4 ~4 I -.4-ri t4 -ii 4-

tf'4t +fý4 ~ + ,~--

I + 44-1-4 4-

4-.4-f, * ''-4- .4- * ~ t4- 4+

.-.- ,-, ,,,, + 4, .,,' ,'+

.~f~~4t 4'.4-f -4-+

.. . . . .. . . . . .. . . -- Sri.t4.. -4t'---4'''-4

32 S

ci~~~ ~~~ 4-T~' H W' TF1~

.1 01

'444

p 9

..11 1 . . A

4-)

-it,

32

~ i i

1~1I I i**

l~It I

t I~

I Tt

~ nt 44[I ii I~ :0.,

Ititt

ttF

0t tI

I T 1 11 f

¼~~ itftt

II 1

y I Týr44

31

-VA)

IA

4l

Tf~i 7rr11 .0

ri-ri, 1 tt -

T' F 7W i I T Q5S l

Tr r

t4-'i~d Is''~P r

T3?

1 -7 r.r- .~t--r--'---r--- - -

3t

Jit,

ni~~~fi LuyJ.

~ irti

-~1 3

TI r '~V~ ~ .I7I rrC- r

L4*3,, 4 i .t 3-

7`7~ ' ~'i

7 * : 7V

* 4-Y I 1 4 3 4

[T[4 tflfF{I~f It I t J I 4f 4; 4

L44 44; I 144 ~I' ~~[1Alit*f4f Ijf I4t IT-- itt'4 ;j~

444 4 .~ tfr rillf ___

I.11 , fc l I l : t

WI I.U

lt I -

T a'

1fl I{4 r Li tttvr 1j 4 i +4

rt j-,

*1~~~~~ ;t"tt411t1j4t

,~V7,

-A4

.FH'

i-A

0 ý - I - - - --II~ ~ ~ ~~~~- 1,t 1---'4*,'II4-- i-I- ti I--& Fl F4,--

a-11T- -- '

4r 44k

~H#-1. j~~ - -~ vg't-.co

. ~ g .~ ....

..........~'1 A j.ALL i~±

r4 - - -

tl ~ ~

I - .1 - - - -

335

I -

44f -

7 1 ) -, [L - r'

i~~~~i~11 41 ' j 1 { Li V

'A t j Af i4 T i - I ifj l

IT~

a.--.--.- itjj .jj... ......

lN3 (d) .:?&nýS3&d

336

4114--> --------

4 4-l4-

... .......

tHH

4_- 4-*4

133

-Fig2

r0

-T-

"4---q -4-4

..~ .. ..t .. .m{ .K U .. --

(I~d4 +1/?Y~ C

733!

I

Is-

IM:I

XAs- I

. w.

------- ------

to~

((rd)-----------

---------9

1ii-

I

Cl)

-19-

340

-T TIT -4

- -- -- --- ---~ ~

* F

3414

2iý

------ - -~ -4 -

a)00LI4

- U --------- -- -.

It 7.

4R4

VV

'41

1411:

-1 '.fiITFI At

Icn

TI 4--

ii' ii~f ~j~Ij 1tf{~ ~ h~[1~ ~iT'h

iiMi

CL

cUQ

t1 44 tf-Ml

+4 1-4 ~ I~

it M

t-1~

344

- -

t tt I

1If~ t

r T111t-t-* t;

t f i n

345

JI I +

tt I T -2

1 i

fl 0

H:;''~~ I +I it4t1fTr~iiii~~ ff hi f,12ilr4~~~~ ijtil Fi; L;* 4 "'

T 11 i ~+t

if 4 1-

ftV~if 4 t-r

I T-1

is +

*~~~~ I ~ ~ ~ ; t ~ ~ t rt I t ,.

41aiCue)~~ KMlt3r

34 CI

1t 4i~a

1T;1rtf 1 1 f P:in b5t' -

..... ------~t tt t

I i.-.444#f4 II-; ti 11 ' I'

l.?ti

J ij I4 T + iiiL

14 tt4 4

tt ~

A~itt i4t-- it '~ '$

r,-wjfi lit tt 4"t i 1 - f-4 . 0t . .I , l it -t --t i t

Ii 7it -1*+1 4'rttt Lu it - ---'-tH it- -~q

'd ýi t-sI, t I, ýi-TMitU it ii -

44f-1- 4't -~ i -it-t

VC{4I*I ~ ~ It I4-4Ti Tz4 ' Tr~

ii 44V4I . ~ - , -r~,r' t~'PIuVA~~rp ctt i±t 44{,itt4

owmr t1 *

tA I

it,~4 Itj AT ~ ~

f*41 4J -1- ý__ IT

"C' itt 'C

*- - -c- -----, -

)+

1-14

II E ^-Ti-i t14~iw 4hf

II I

A' 'I'll

0D

'349

414

LIA. I I

i-4- -t

A 41y 1i f~41 Ft

-4-i.-*1 1.,1 - 1, ! -+

L-Li

3500

-''7 'ý7 77-.1"IT4

---------

J-12

tt

K7 ' -

-

- ---------

-- ------

fill

j t

4 tN

! , * + T lv _ýIt ! ,- -t~~4Jf-

(Ii~) 34-

-~ fjLi t~ Ai

Ma UT

tt{

T0.

353I [if

t Ii

itt5]t

-it

a a a .... ~...... ..... i: :; :

"I"P

ff1 IL a a t

[

ttt6j* I

354ak,

.k4

'Ii

*~~ .. , .i . .' . . .

.1 :1

f 'ItPo-- - M' 2A~IN

AH

0~~

I ~- .i.f

..

717

~ L t t fv Z 41~T 4

4;

.tt . .n......

St

--4 - *-t-t

Tn Lt I I Ttiý'

• • + :,-d --- ". f + ,

i CA i+ i [ , .. ........ .i

is..... .. t'r*'.-,-•'..... -4, - t+..'''tf4-

44 I. 4-

I . . . I T

-4

-ri -,..rr -=T!. - .

+: ... + + ; :++++ : :a'''4.+ : -++ + " ? ; : + +',.I I- . .. . . .. .. 4;'I *:''t,-'-4- '-+j V

4 1 I

S +"4

ti~~ - rr4a r

fill 1 . 1ý ilii

,~ I1

I T -

if 1 TM r

r s- 1

I~~~~~~~ r~It -i-w.n~n -

144

i M 1 17.

..........P P.

0-

1 VA Lo

ITI4. i-

i1 1 4 -TV

~ ±t 7lh~~ti;~u 9T~4~ t is UU:' F

I T Z__T 7t

zt z 4 I ~ ~

ft kit I' -i~

1- J

t! ~ j Ui ijiEftHlT fjI L--

4I -'rI~

14 j 4. 44 jT j4-

41T J.244Hj C

~ .'T -~i ~':~t * --- *'* I -i 0

* i ;T 41 1 1 ,-j T...TT- 1 p

~ iT-I I T 4

314'

- -' --------- .- r ., -- i.ir ~ i ~ wr ~ -fti * . ~ q q n su p u

iA -1 A

I,~ i At;

4,~ 7 1 1

1.1 lL- it 4 -,-~ ft

-4144 17J A

4t4t

--4 --- 4 4 t 'o- -

+i J

-4; - -- It I 4Ft 1 '~i itJ }-UI$;II-

-- ~~~ U-tP1 pIo*PTO'~ &.

i-I 2 ,4t it~ 1 l

T- A +;

11 TI- His 114 [1Eli1 01b

4 412

I I-o1. 00I

~ 4 .~wv4;H Ii-'r' "I'-'--L"i- 1 tA Iv-

"ISI

-li

li

-4- iHAV -..

4--

44

44L

411

4.ý

4k

-11IJ---

4 ----

I)f

tf

V. _ 1½

Wii H5 ±S3.i

TVo V

bb

146

''77

j [i~~ ~~~~

1_____________

*~~~- I- .t'%-tfWbflrt1,

II

I T

.. .. .. ..

'fl 4 'ji i

WlWhAWNfliSiWiM

l'0 cc

it.!t ..

- .111 ~ 36i

T 7 -

-1 L

it ~ ~ ~ f II . oy l 1 1 ! !; -V

F T

': ~ ~ i t~~ itiff*, ,- --

___

- 7~tt 1 1ii

1 T I, I . ,

S.. ... .. . . .. ,.,. . .... .......

TT *

• , ,_, ,.. . ., : . . ., I - .• + i.. .... tt ...

"7 7 7 7 .ii = i 7 I 7 7F' ' .. . . i * fi, • t • ' • ' i i : ! • . . . t + I. . . . . . , ,In :+" ' , • L!: :- ' I ,:•L _•I_,•, ' .. ... . .. . .. . . •.. -•

LLoI-* ,-

' I t'

F• • : : • - • ' t ' ' ' + ~ •: '".. .,. . : •.- ' •

t-: !K - : + , ' lj:: : : ::: ::., .p - -0.to:

*: _ I

. . . . .. . . . . . . ., ,5 -++. . .....[•4 . .. . .. ... . . ... . .. ... .. ... |, • + :+vt .. . +•- . : • 1 t .. . .. . . . . .. . . . . + , + . . . ..

-j-f-- l•=+• - - it:€-,• . , _i

,.meell , • , - . . - +: ++ : .•. : -: ,+• +...a'+

" ; .i • . '~ ::+ : P + :; "5 + : ,+ :+ , ,

,sflrtT

I ti

-. .. - -,l ,, tr ... w -' -jt, -ii , -• ; ra .- n -

. ii . ..... ...

S........;4 .. ........ .. " !

, . , ,:: !1rr , .,. . .. • ., . . .it j ~ jt ' ........

I . . . . . ... . . ..'.. . .

I .' . ..4: -ti , .. .. 4 "H

a• )

1; 's

'itf~ 77 I

IN

~7

Nix

oii

370.

... .. ..... ..

444

t .... - .. t

Li "t 4 ...- 17t • ,

44K I . l•it I :I

', i I .i i -I i i;I I i, ':3:i !. . ... .

1k I .cjm

I i

* F

2 1 _ . • '",

4•.•" _ ..

'•••,•',,

rfltt f1

hR i

fitF F 1 it t t

111 4 1-I

T T

iT r

'Ftj

va 7t.

I IT-

tit 1-

44

*- ij, 4f w4, i .

4i

7TT TT t -tit z i

+Hlr* -4 ý

44~ ~ ~ ~ t tI' Ft~ {~t

1_4 -t Ii

t' t

NO T'iý T4-

-1 Jt1 1 4--

s-7t T 1'

4 4

4-

-T

45.looff"r r 4* I

-- - -.--.------- �-'--'-�, � ________

'�?j �* * - .*,�ft �'l�t � I'r}:.: .1

I,

TftIttt * S1-� -. -

-4t -

�tJ,1I, *SSII -�S.,, S.,,, VI . i�J

S� 5,

4-.

- j'':' -�

-4

� .S, i:LJ -

g"5" 1;. '-.5

0

- � �#SI*

)555 �5 t

- k,� '�S�'

-. .. - 2 �

1 �S: 4,.

r'-� �.. -t -

_____ I f I �C .�xr��.-=�v--.- -fl� �

-I

* rIrILJ

� 1

"�

74

_j 4

'T1

..~~t,.W 4tr iti 4[;I ''.-I1-4-

Zn 1-4- 4--TA+Th4

.1Ir 4. Pt-J 1~ i tj

-, *,L' I j.Jt~

4 :4A- a-

444~ ~~~~~ 4'4' 4I---

Li.iL4j4 I4.a

4~~~~ t4il1 I .LE .± 4t~l-I L'4 S C)

H t~ 4 tli T~A-A44zi 2 1 . A~ i /

-Lo

27t¾Ji c)

10 w

3751 ~

4-~

00

4 T.- j 1d0)

~~t4

.4I V.1 I lw.j

H4

376

HjH

i-44)

4r-14- - T .-i .4.

f 44 4ý

TT

17L4

I 1 111 I- -4- I

L4 377

II~~~~~~~ ......... i~jiPt~ilHK'

I-'V f 4 -t hev I 'li li l iitil

11 tflji

T14 4-- F.

t'

-I -A

T, 4- 11(

f4- 4-4. ti [4 -1

-H':t-i ~ 117TI.. > ,.

FIR.-. t I. {L.2 w i

T. IT

i t-

378

~~7 I M_____

JJ

, tt

lf. t+

t+i l

Il~jr )

Ii '00-7177,41-~ -

3749

----------------------------------

:7.

*-1

fIIIt 4- t - 7 hi

T -i -7 1, I *

lit

9.T

AA,

3,:11

, , . r , . .. .'t ' . .. . . . . . . .

*1t

77i4

S..... . .... ..

IIS...... .- - r, - ,-r. . . .... ..77

S~ ~ ~~~~......... ......... ,- l . ,,;............

S; . ... .. ,.'

-•i *----,r ° -

i:I

l - -

, " . . . . .r, 2 rj't r -

-41

1 4

LIT

-14

- - 7 .... ...... i' . . . t

, , . . .. .4,,- Tl'•

""t 'L,, ,�i t• 4 A l

*~~ * t iT

'Vt4

V.. . . .. . . .4 42U fV, i•- . .0

~~ ... . .t'' ... . ..

$ThI -) • ' . .. , - , . i- 4 • - • • -H n, - f l -i• - ' " . - , ,

k -4-4 r44.. .... ' .. . .- :..

nw-

-- ,,--W'*t-~ S~rf -

IMPELLER M.IT 17(' STALL TRANSIENT DATA

This section contitns stall transient data obtained with the impeller exit (Station 2)cobra probe sensor positioned 17% of the flow path width front the ahroud wall.The cobra probe and recording equipment were identicl to that used in regulartraverses. Frequency response of this system should be considered poor fortransiont ch,lracteristics. Impeller exiL air angle and total pressure and collectorstatic pressure are presented vs time. The pressurcs were ratioed to standardday conditions. The key denotes the proper symbol to the scales from right toleft in ascending order. Each condition was recorded at two scanning rates, onescan/sec and maximum (approximately 6 scans/see). Conditions for Figures 298and 299 were 101% speed, 5-.dog IGV- Figures 300 and 301 were 1017, speed,0-.dog tGV; Figures 302 and 3u3 were 101% speed, -4-dog IGV( Figures 304 and305 wore 94.5% speed, 15-dog IUV; and Figures 306 and 307 were 95% speed and10-dog IGV.

3 4

44

~~~~~~- -7---v--Ti

* - 4 I I 7

-iui 1011 *P d, 3-@9 I V

IS Rat-e. ~i

- 3-- 93-

44

.............

0A

Fiut 4M S-1TrnettMa,11 po,3dtLV

M-xi* Ra 0.

94 jŽ

-16

-i -

rLwrq $tal - vt~id Nt,11L p.4(-4,g

3% T

* * f * - *I * r • . - r : , .#

SCA

""mop - I

..... •-

...

ij~ii

I-, $ -,

0-&

L 4A ! : " i .~ -t

.!.I '

T '] l • "I , -r :i ,

K / /-", !.: ----.-- I ' 1 . <- t--4 i :t w' ." -

• • . "- -,4 '_2 •._-

: ;-• - -- - L l- -. " i ..- ; . .. ,.4, " .rr a-. r : 42,:.:.tr _2! .tzrr .4,, --4IS .:.-• ,,.:'m m . "i/ : t t " It ; f , " • - • .• -.- • .AF."--'£ "-.- | --. -- "' • -"" " -t" • : ;4=I il .> .! " "- " 2. k- ---g1

ttfii

:41-Ž-_ .

-FIT77TjV77.

44

-11T

to +_v .... 3

J,.-

Ir -4 ZI-7L

7 Ip

Fiur 309 tlirnin aa 0%Sed 4dgIGV

ISP2 Ra1e

I 389

-17

torip

-A-

4t 1- - 7

vi- 4 -47 --- j -- t - -i - --

Ju I

-:: * .;4L-- "-4 4T 4

_

F i u e 0 . S t l r a s e t a a 1 1cS ed_- - e I G V

Maiu Rate

390t

6 i .~

.3:

7 7.

~4 4 1t-U

iI~~--i -- -- ----':1 -: A,

'M11

I~ 1 I 4. 1-I

4LL

Ir. y

"r7! 77.... .IaK; 4.- iIi__

Fiur 304 Stall Trnin Daa_45 ped 5dg1VT + L~iSP Rate. 4 1

-39

nw-c Ar

4mA q ý11--

-4 J t.7-77

* IIT

4. .......4.. r 44471

I .. -: - -

-h It ..... ..... 71.

42 EFiue35Ttl rnin at,9.%Sed 5dgI~

Maxmu Rate.

392AI.,. *'~ ~4:

77ioil. _ _

'I-

T T

Iu - --Ju 11-01I

HI

'7 i- I- i I

a L ~1Figur 306. Stl Trnin Mts9%Spe*1--gl~

~~SP Rate.---I

x ~ '' j393

777 I"

-~U I c Ii

04

0 VG3l

i 1

34 JJ o

Fiup37 tl rasotDt,9%SpoW 0dgIV

M44u Rate

~.......

DISTRIBUTION

Director of Defense Research & En~gineering 2Assistant Secretary of the Army (R&D) IDeputy Chief of Staff for Logistics, DA ICh1ief of R&D, DA IArtiy Materiel Command 3Army Aviation Systems Command 2Hq, Armiy Air Mobility R&D Laboratory 2Systems Res~archi Integration Office, AMRDLIAmtes Directorate, Army Air Mobility R&D) Laboratory IEustis Directorate, Arniiy Air Mobility R&9!I) Laboratory 20Langley Directorate, Army Air Mobility R&D Laboratory 2Lewis Directorate, Army Air Mobility R&D Laboratory 2Army Aviation Systems Test Activity 2Army R&D) Group (Europe) 2Armiy STINFO Team (EuLrope)IArmy Aeromedical Researc:h LaboratoryIArmy Coating & Chemical Laboratory IHarry lDiarnond LaboratoriesIArmny Ballistic Research LaboratoriesIArmiy Research OfficeIArmiy Materials & Mechanics Research Center 5Army Test & Evaluation CommandIArmy Materiel Systems Analysis AgencyIArtiy Missile CommandIUSACDC Aviation Agency 3Armiy Transportation SchoolIArmy Arctic Test CenterIArtiy Field Office, AFSCIAir Force Acro Propulsion Laboratory 2Air F'orce Flight Dynamics Laboratory 2Aeronautical Systvins Division, AFSC2Naval Air Systems Command 90Wbi of Naval Research 2Naval Air Develpmn IetNaval Air Promlsion, Test (-enter 2Naval Ship R&D Center 3Miarine Corps Liaisont Officer. Artiy Transportatioo Sm~toolIAmes Reseamch Cenitcr. NASA 4Langley Research ('enter, NASALewis Roevsec ('enter, NASA I

Goddad Spce -iht Center. NASASI INFO Facility. NASANational Aviation Facilities lFxp~efitental Cwi~tcr. FAA IlMcpartulent of 'Transportation Library IEastern Regiun Library. FAAIFedefal Aviation Admnini~trtion. WWsii~wGave'ranent Printing Office

UnclassifiedSecurityw classification

DOCUMENT CONTROL DATA. R 0(Sifeity lasstleofeff f W4 eSIME of filistfo lan fonift oneilaion *woo 6e Whie n *0w *P cerot$ report to Et18081"04,

Cal sl'on aamw ,eacu"mVe Ct.466$PICAfION

i~mieUnclassifiedUnited Aircraft Corporation Ucasfe

[Florida Research and Development CenterWet P FmRnnwh Florida

10:1 PRESSURE RATIO SINGLE-STAGE CENTRIFUGAL COMPRESSOR PROGRAM

A. 0&ýcft1VT1wa NOTES1 (Type of ftepar 4ad leelvalve get*#)

Final Reporta. AU TH0111ier (Yuf fe.w*. Mwo MWXi less Nd00)

William J. McAnaily, Ill

a. R&PONT OAE I&, TA.NO .o r "Gas s.NO. 09 maps

April 197442B.CONTR&C T OR GRAPL? NO. aa. 00041reATQRI EOR NUERS

DAAJ02-7&C-0006111111ojacy No. USAAMRDL Technical Report 74-15

Task IG162203DI4413 Ms. OT1111P1 Nf' OM (AWa 0*. AM imob" 16SA~hW so&&b

FR-6086to. OwSRIOUuIO. STATEMEAVNT

Distribution limited to U. S. Government agencies only; test and evaluation; April 1974. Other requests forthis document must be referred to the Eustis Directorate, U. S. Army Air Mobility Research and DevelopmentLaboratory, Fort Eustis. Virginia 23864.

ii. w.~eE~T~v ~oi. I Eustis Directorate, U. S. Army Air MobilityIResearch and Development La4boratory

Fort Eustis, Virginia

The oblactive of this pr(Wafn was to daelion, tebric.'te anti test a 3.1- bI- aita flow sifigi4talgb centrifugal cowlratlor that could be IncorporateiM a future ArMy UPacrdied teChIM1[4VV- WAI, euJiqangine. The dearn n pa tornwiace anal, were to excead 75% eIfficilinfcy 41 10:1 Pressure ratio.Since Wa turbine engines fot ArmyV aircraft appligiitIon, operate undit oart-owar coditions a majwority of the tiffe, an oft'dowits per gftato"0FXgoal of §0% ifffiglisft at 8:1 prtW41610 ratio wasestabflishted.

In the oanion ot the corepruao.0. paraffwtrfra studies were Conducted to select ant overall design conaistenlt with 00tltiffi~mm r~o pvrf0X966ante at both prftoewiance "Ias. Theta studies defined the compriessor inilet corrected flovw rate, iftspeller ialet hub a" tip rasdii. Correctedimpieller rotational speed, PAd intlet piedhil. Airflow selection and the selection of the hubs radifu mere "mlta-e toy thte 0WitOn to dagaign a

cnpaauthe' co~uk be uaed in a affiall turbotlieft etigin with a crtncentuic sihath front drive.

The tip radius *a# selected afteir determining the affete oa aelala Math Rlnusff . inducer t.P ralatiVe LaCK nlumbier', and isleti Choha flow maursin.The affect of inlet guide vane lossea. hilet shock lowts, dittusee toasts. and shroud friction heatinig ware parwfliatnic:411V evaluate beorsteecting# an IGV previhii wn4 rvto. spied to provide optimum overall compressor pertaforisace. A remote inducer designl was selected Owfe anintegral ndcrmeel Configuration, so tkat the inductr colteaesduigtasneailwcmrsor Meho ogy. Ta wort so~bawaen the inducer and imfpeller vas setaledne so the! t diolatiye Mach nurfber into the imfpeller wor4ld tw sutcison. A Ppeg diffusrwaselecteol over veASe islad and Cascade dift~%sre 40c.-use it has the lovwst denftonatretd lO1isesl tOW the ftrgV§9 Wa',p Of Malk--A mu4e ati -

ereP&WAts Wa sustanil~tial eegeastc 1A dAWAusg a9d falrifatirie the type of diffusar.

00OPArnontrttd toWtal-111stiee Partorirsane Was as hinjs as n~.6% ettkwwn'v at 019?-I pressure ratio an4l ?3a% Effir-eAcw at t0-03:1 presureratio. erformance adjustedi for inicreased losses tfrom a iiamaged drjt~uae (totS)' pfalsufe ratio adteft~ie~inCYl fndicateh thst El~t basie

"omrSo devigh VOULd 6urpAss the 101 PresSure ratio pr-ogramB goa. Evaluezoi; at coeofiaeFRe Wfla~ aca datreaedittac$svlisses occurred in the induCerf akaoe W% fit 444Pg Wpe nititarraag f*s wpnnettdp0e neiueeo efesie

asSQ Spresrorewa

umeata toa~aaa

UnclassifiedI.LINK A 6100"U IN C

not.& WT MONKS ? ROLC W?

CompressorsFabricationPressure RatioEnginesTurbinesPerformance (Engir'aering)

Ufwlavf~d Y'

4 A

Documents

10:1 Pressure Ratio Single-Stage Centrifugal Compressor Program