Home

05Sep_Villescas - Calhoun: The NPS

1. EHE RUE bool 11 INSTRUMENTA TION essen OPI 13 A PROBES oeras aE ic 13 B L C SMITH PROBE ACTUATOR 4 202 13 C HPVEF SCANIVALVE SYSTEM eh 13 OMEGA PX 135 PRESSURE TRANSDUCERS ecce eere eene 14 E PMD 1608FS DATA ACQUISITION DEVICE eeu 15 F USB ERB 24 REMOTE RELAY CONTROLLER 16 G MATLAB DRIVER 17 CALIBRATION SSS ASLO areas 23 A PURPOSE SOR bp nM Hi Md 23 B 3 RTV 23 CALIBRATION PROCEDURE 24 D CALLIBRATION PROCEDURE WITH AUTOMATED CONTROLS AND INDIVIDUAL PRESSURE TRANSDUCERS 25 E CALIBRATION EQUATIONS use 28 28 2 Setup and 5555560 een 28 3 Calibration Equations ccce eek 30 E PREDICTION OF VELOCITY AND YAW ANGLE 31 F COMPARISON OF SCANIVALVE AND PX 138 DATA 32 TEST XRUNTROCEDUNDE dive ndi 35 A OVERVIEW sanieren 35 B IESE PROCEDURE aaa tans 35 RESULTS AND DISCUSSION sense as nn 39 ANALY a d 39 1 Upstream 39 2 Downstream OL nS Aree vd 41 3 DIHUSIOD Factoria its 46 B OBSERVATION ss 48 1 st 48 2 Magnitude of Diffusion Factor 48 3 Comparison of
2. 18 Matlab calibration code block 19 Matlab data collection graphical user interface pp 20 Matlab experimental data acquisition and feedback control code block O __4______ 21 THESE OLE probe numbered scan 24 Probe actuator and data acquisition wiring diagram 26 Three hole probe pitch and yaw angle definitions 28 Scanivalye AS Cabanon SUIC ua Dr 33 PX 138 Transducer As Calibration Surface 33 Scanivalve At Calibration Surface 34 PX 138 Transducer At Calibration 34 Inlet Mach number at 70 0 088 39 Inlet Mach number 2 80 nie a 40 Inlet Mach number 4 90 Speed aio 40 Inlet Mach number at 100 Speed 4 Exit Mach number 7050 Speed tendunt ta 42 Exit Mach number 80 Speed tian 42 Exil Machnumber 90050 SPECI SS SSS eS 43 Exit Mach number Dd 43 1 Figure 36 Figure 37 Figure 38 Figure 39 Figure 40 Figure 41 Figure 42 Figure 43 Figure 44 Figure 45 Figure 46 Figure 47 Figure 48 Figure 49 Exit Tow angle SDOGQ ee 44 Exit TOW angle 6070 peto idees aee 44 Exit How angle 9070 sida 45 Exit Tlow angle 100 une ua ee 45 Di
3. ear Stall 0 06 0 08 Probe Position m Figure 28 Inlet Mach number at 70 speed 39 T 2 3 Choke F eg E ff ear Stall 0 06 0 08 Probe Position m Figure 29 Inlet Mach number at 80 speed Mach Number Choke Peak E ff Stall 0 06 0 08 Probe Position m Figure 30 Inlet Mach number at 90 speed 40 qi 2 E LE a ME E hoke Peak Ned tal 0 06 0 08 Probe sition m Figure 31 Inlet Mach number at 100 speed 2 Downstream Data Exit surveys were conducted downstream of the rotor Results are shown in Figures 32 39 for rotor speeds of 70 80 90 and 100 with Mach number and flow angle plotted against the hub to tip radius ratio Graphs of the downstream stagnation pressure ratio for each speed are shown in Appendix L 4 de 2 cr 5 Hub to tip radius ratio Figure 32 Exit Mach number 70 speed Mach Mumber Hub to tip radius ratio Figure 33 Exit Mach number 80 speed 42 0 5 0 4 0 3 FE Hear Stall 0 5 0 85 0 95 1 di 2 5 c 0 2 Hub to tip radius ratio Figure 34 Exit Mach number 90 speed Mach Numb er Peak Efficiency bear Stall Hub to tip radius ratio Figure 35 Exit Mach number 100 speed 43 W d gt
4. 1 gt E LL Choke Peak Efficiency Hear Stall Hub to tip radius ratio Figure 36 Exit flow angle 70 speed Flow Angle degrees ear Stall 0 0 8 0 85 1 Hub to tip radius ratio Figure 37 Exit flow angle 80 speed 44 ut D d c c 2 Chake Efiiency 0 8 0 85 Hub to ip radius ratio Figure 38 Exit flow angle 90 speed Flow Angle degrees 0 85 Hub to tip radius ratio Figure 39 Exit flow angle 100 speed 45 3 Diffusion Factor The diffusion factor 15 a method of assessing blade loading in an axial compressor For a simple two dimensional geometry the diffusion factor was calculated using the equation 8 26 Vj Where and V2 were the inlet and exit velocities relative to the blade Vo was the turning of the flow assuming the flow 15 axial at the blade entrance and 15 the solidity of the blade corresponding to the flow It was believed that diffusion factor values above 0 6 could indicate blade stall and values of 0 45 may be a typical design choice Cumpsty 1989 The relative inlet velocity V was calculated using the velocity of the rotor and by taking the area average of the velocities found upstream and calculating the flow at the leading edge of the rotor assuming the decreasing area acts as a nozzle The relative exit velocity was ca
5. 118 5 5 05 Pitch xis Sheetl column l77 89 5 O Degree pitch A end 1 xlsreadcells 5 10 05 10deg Pitch forward xls Sheetl column 1 10 Degree pitch Mach 0 3 Nominal A end 1 xlsreadcells 5_10_05_40deg_Pitch xls Sheetl column 3 5 40 Degree pitch 1 xlsreadcells 5 9 05 30deg Pitch xls Sheetl column 43 5 30 Degree pitch 1 xlsreadcells 5 9 05 20deg Pitch xls Sheetl column 43 5 20 Degree pitch 1 xlsreadcells 5 9 05 10deg Pitch xls Sheetl column 45 5 10 Degree pitch 1 151 251 251 271 xlsreadcells 5 5 05 O0deg Pitch xls Sheetl column 1 13 5 0 Degree pitch A end 1 5221 xlsreadcells 5 10 05 10deg Pitch 15 1 column 1 13 10 Degree pitch lt Mach 0 5 Nominal 1 xlsreadcells 5_10_05_40deg_Pitch xls Sheet1 column 22 34 40 Degree pitch 1 xlsreadcells 5_9_05_30deg_Pitch x1ls Sheet1 column 60 72 5 30 Degree pitch 1 xlsreadcells 5_9 05 _20deg_Pitch xls Sheet1 column 60 72 5 20 Degree pitch 1 xlsreadcells 5 9 05 10deg Pitch xls Sheetli column 64 76 5 10 Degree pitch 1 xlsreadcells t5 5 05 Ddeg Pitch ls Sheetl
6. poj pl q pitch 5 Centre hole Initial estimate of Mach number ps m m m p3 m ps m mini pl m 2 m Po m ml j po m max pl m p2m p3 m po m max pl m p2 m p3_m m Mach_m sqrt 2 Gam 1 1 Gam gas Gam gas 1 Pitch m O ones size pl m 5 Initial estimate of pitch in degrees Yaw m size pl_m Initial estimates of pitch for i 1 length Mach m if max max Mach lt Mach m 1i min min Mach gt Mach 1 max max Cp3 lt Ops mii min min Cp3 gt Coa mbi max max lt min min gt Cp6 m i disp Warning data may be outside calibration range Yaw m i 0 0 5 if max max Mach lt max Mach_m end 100 temp find isnan Yaw 1 5 Data that is within calibration range Initial estimate of yaw temp griddata3 Mach Pitch Cp3 Yaw Mach m temp Pitch m temp Cp3 m temp 5 Variables on the tolerance tol 425 Patch m old Pitch m 2 tol Yaw_m_old Yaw_m 2 tol count 0 while max max abs Yaw m Yaw m old max abs Pitch_m Pitch m 1 7 gt tol count lt 10 5 Stored variables for the tolerance Pitch m old Pitch m m old Yaw m count counttl if length intersect m p2 m 0 5 is this is
7. it to an EXCEL Spreadsheet eval fred pause to ensure Excel is closed for a slow computer pause 1 m 1 Pitch_m Yaw_m Mach_m ps_m po_m hole probe output m P3 P2 P1 Poj 125 Change the value of freddy to next row number freddy freddy 1 Save the row number to a text file so that it will not be cleared save freddy txt freddy ascii function sampledata_run begpoint endpoint points Tp Tj numpos is the input number of positions to take a sample of 15 the distance to take samples on with zero beginning at the thelength begpoint endpoint Create a linear array thepos linspace 0 1 points output an array of positions based on cosine cospos abs 1 cos thepos pi 2 create the array of final sample positions sample pos cospos thelength Change the starting area if needed flipped tfliplr sample pos Yo Yo Yo Yo Yo Yo So So Vo So So Vo So So Vo So So So Vo Vo Vo Vo final_pos flipped endpoint plot thepos final pos break efor j 1 points handles guihandles gcbo generate handles struct set handles Stop_Points j String final pos j 126 for i 1 points Pos Input final pos i handles guihandles gcbo generate handles struct set handles Pos Input String Pos Input probe pos req Pos Input traverse probe pos req pause 1 OUTPUT_TO_EXCEL Tp end 127
8. i 7 Figure 13 Probes positioned on the mounting bracket for calibration procedure on the free jet test assembly D OMEGA PX 138 PRESSURE TRANSDUCERS The Omega PX 138 transducers used in the latter part of the experiment were miniature pressure transducers one of which is shown in Figure 14 The pressure transducer used silicon pressure sensors which were temperature compensated The PX138 015D5V model was capable of a differential pressure range of 15PSI to 15PSI with a linearity and hysteresis error of 0 1 typical 0 5 max The PX 138 was given a regulated 8VDC power supply and provided an output of 2 5VDC to 4 5VDC for the range of pressures applied in the experiment Omega 2005 14 weg rgo vM ga oe convence EX i o 5 Figure 14 138 pressure transducer The pressure transducers remained calibrated throughout the experiment Their pressure to voltage ratio was nearly linear and all five transducers would generally read atmospheric pressure within 10 Pascals a less than 0 196 error However the PX 138 transducer had very small pinouts and came with a CX 136 4 plastic mating connector Maintaining electrical connectivity was occasionally a problem due to the fragile nature of the pinouts and connectors E PMD 1608FS DATA ACQUISITION DEVICE The Personal Measu
9. 170 88 kg s m 35lbm s ft 246 1 28 1 2 Sl 28 2 22 1 3 0 2794m 11 nches A 47 compressor map A Transonic Compressor State Part 1 Gannon et al 2004 for the rotor only 15 shown in Figures 5 and 6 The stage design point 15 highlighted for the 100 speed line The CFD data on the compressor map was obtained using SWIFT a multi block computer code created for analysis of three dimensional viscous flows in turbomachinery Chima 1998 A complete stage compressor map including rotor and stator can be found in Appendix B 1 8 4 Ha Retor Rotor 1 1 6 Gua Speed Speed de 1 4 1 2 1 0 3 5 4 5 5 5 6 5 75 9 5 Mass flow kg s Total to total pressure ratio Figure 5 Total pressure ratio compressor map e Isentropic Efficiency a 8 0 75 Nass flow kgs Figure 6 Isentropic efficiency compressor map 8 A steel case wall was used to mount instrumentation for measurements The four holes labeled in Figure 7 were designed for mounting a probe actuator for flow measurements The gold band represents the location of the rotor with holes one and two located upstream of the rotor and holes three and four downstream Additional details of the case wall can be found in Appendix Figure 7 Compressor case wall Hole locations one and four were used in the current project Hole on
10. else Pitch m reshape Pitch m shape pl m Yaw m reshape Yaw m shape pl m Mach m reshape Mach m shape pl m ps m reshape ps m shape pl m po m m shape pl m end if point no 5 Cpl and Cp2 are plotted figure 1 close figure 1 plot Yaw Cp1 Yaw Cp2 temp axis axis 30 30 temp 3 temp 4 Cp3 is plotted figure 2 close figure 2 plot Yaw Cp3 temp axis axis 30 30 temp 3 temp 4 5 Cp4 and is plotted figure 3 close figure 3 plot PILE cpat BILO temp axis ax s 30 30 temp 3 temp 4 5 15 plotted figure 4 close figure 4 Plot Pitch temp axis axis 30 30 temp 3 temp 4 1 Surface plot of the data is plotted figure 5 close figure 5 TRI delaunay Yaw Pitch cCrisuri IRI Piccn Epa hold a tplor2iraw m Pibtoh 5 m OE end if 0 m function file to pull in certain data sets 102 function Load_constant column 5 Mach 0 06 Nominal A 1 X sreadcells 5_10_05_40deg_Pitch xls Sheet1 column 110 122 5 40 Degree pitch A end 1 x lsreadcells 5_9_05_30deg_Pitch xls Sheet1 column 1 5 30 Degree pitch 1 x lsreadcells 5_9_05_20deg_Pitch xls Sheet1 column 1 5 20 Degree pitch 1 xlsreadcells 5 9 05 10 Pitch xls Sheertl column l 5 10 Degree
11. 7090 Speed Mass 63 Bn Speed B 6 5 Mazz flaw Kgs 90 Speed stas e in Point 100 Speed Ple zi an 5 THIS PAGE INTENTIONALLY LEFT BLANK 64 COMPRESSOR INTAKE SYSTEM APPENDIX Laid SE HI NOLLdu cHd 1 BALE ES gH y kascaxaorowa z Alae lode e MLS E x go ee EI ON 2v AW LO 11 e dag E 3 00193 9 x 01 zt 006 OMY 030405702 HO 09 3742159 Sl 07140 5234 til Tl IE 65 THIS PAGE INTENTIONALLY LEFT BLANK 66 APPENDIX D SCREEN SHOTS OF THE HPVEE PROGRAM 67 o m E EOM u E zd i E i u N ps E 68 69 70 71 THIS PAGE INTENTIONALLY LEFT BLANK 72 APPENDIX E CALIBRATION EQUATIONS AND SAMPLE CALCULATION In the calibration 15 the yaw angle taken from the angle of the probe actuator which was set to 20 with a 1 tolerance 19 928 The pressures P P3 were tak
12. F eak E ficiency er Stall E r3 c ut ut 2 rm e 3 e 0 8 055 Hub lt radius ratio 80 Speed 2 hok stagnation Pressure 1 1 Peak E ficiency ear St all 1 0 65 0 5 0 55 Hub to tip radius ratio 111 90 Speed AAA MAA meres hoke E m Peak Efficiency 1 Stall 0 65 0 7 1 75 0 8 0 85 0 9 0 95 1 Hub to tip radius ratio T 7 m o 2 100 Speed Stagnation Pressure Ratio Near Stall Choke 0 75 0 05 Hub to tip radius ratio 112 APPENDIX CFD VS EXPERIMENTAL DATA 90 ROTOR SPEED Near Stall Rotor Exit Flow Angle Near Stall Rotor Exit Total Pressure Ratio Peak Efficiency Rotor Exit Flow Angle 114 Peak Efficiency Rotor Exit Mach Number Hiub In Amu Peak Efficiency Rotor Exit Total Pressure Ratio Ando bo Tip Aei 115 Choke Rotor Exit Flow Angle 0 8 0 Hub Ene Ti ps Amim Retin Choke Rotor Exit Mach Number Choke Rotor Exit Total Pressure Ratio 5 85 0 55 m Ratin 117 THIS PAGE INTENTIONALLY LEFT BLANK 118 APPENDIX N MATLAB EXPERIMENTAL DATA COLLECTION FILES function OUTPUT_TO_EXCEL Tp Tj this function gets the data from the devices using PMD_1608FS_2 and PMD_1608FS_3 and sends the data to ExcelWrite to be written to an Excel Spr
13. NAVAL POSTGRADUATE SCHOOL MONTEREY CALIFORNIA THESIS FLOW FIELD SURVEYS IN A TRANSONIC COMPRESSOR PRIOR TO INLET STEAM INGESTION TESTS by Ivan Villescas September 2005 Thesis Advisor Garth Hobson Second Reader Anthony Gannon Approved for public release distribution is unlimited THIS PAGE INTENTIONALLY LEFT BLANK REPORT DOCUMENTATION PAGE Public reporting burden for this collection of information 15 estimated to average 1 hour per response including the time for reviewing instruction searching existing data sources gathering and maintaining the data needed and completing and reviewing the collection of information Send comments regarding this burden estimate or any other aspect of this collection of information including suggestions for reducing this burden to Washington headquarters Services Directorate for Information Operations and Reports 1215 Jefferson Davis Highway Suite 1204 Arlington VA 22202 4302 and to the Office of Management and Budget Paperwork Reduction Project 0704 0188 Washington DC 20503 1 AGENCY USE ONLY Leave blank 2 REPORT DATE 3 REPORT TYPE AND DATES COVERED September 2005 Master s Thesis 4 TITLE AND SUBTITLE Title Mix case letters 5 FUNDING NUMBERS low Field Surveys in a Transonic Compressor Prior to Inlet Steam Ingestion Tests 6 AUTHOR S 7 PERFORMING ORGANIZATION NAME S AND ADDRESS ES 8 PERFORMING Naval Postgraduate School ORGANIZATION REPOR
14. by the CFD Swift code 11925m Entrance Flow Velocity Triangle U 1s the velocity at the leading edge of the blade at the point that 15 measured The points that are used for U are the area averaged points corresponding to the area 135 averaged points sampled It is a product of the rotational speed and radial distance from the center U m U 338 233 8 V is inlet velocity relative to the blade V m Vy 372 235 S Vo was assumed to be zero as the flow was found to be axial upstream of compressor The flow variables were then calculated downstream of the blade row The flow was sampled at 15 positions using the pressure probe gt is the Mach number measured of the flow measured with the three hole pressure probe at point four downstream in the case wall The position was taken near the midpoint of the flow M 5 555 136 The Mach number allows us to calculate the static temperature assuming the Stagnation Temperature has remained constant through the compressor 1 142 1 DE 274 657 The Static Temperature allows us to calculate the Static Velocity of the flow downstream of the compressor rotor m 185 029 S The solidity o was matched to the area integral of flow of the trailing edge of the blade The solidity was calculated using flow lines taken at 49 points from
15. Angle Degrees 94 15 Mach 22 Angle Degrees 95 ce THIS PAGE INTENTIONALLY LEFT BLANK 96 APPENDIX I MATLAB INTERPOLATION FILES 5 m file to call the 3 hole probe output file clear all Type of file to get data from 5 Call from excel file type all from text file file type m E E 5 Measured values Can be changed to a function later 93692 44746 93648 41833 93540 03893 93419 38217 9325004009099 93144 62346 25159 17099 93047 25134 93104 40454 93123 95925 95103 993118 223125999515 2219033490213 92046 01196 91071 86744 1 paz m 95925 96792 25915451443 95922 96894 95927 52448 95922 008001 959005290436 25919 04 1255 95897 70643 95905 76022 9006024 95831 44412 95680 69734 04890 93334 22410052052 91121 43572 1 pam 93648 22584 935592549472 93429 69185 95291 13094 93106 09345 95066 76159 93009 85444 SLOVO UL 23164 0989 93188 46395 97 three three based the same into the same directory 19 2209 92044 91100 22687 poj_m 950930 2396097 235024 22520 2020 520021 por 99905 227 920215 95846 95884 22 47 92201 265959 94203 62365 95852260365 08642 ILLOS 71878 11268 9255 97129 8053 47151 03452 40784 73012 06898 102794 6236 10720639565 1051 00 2270 107363 8443
16. flow field and rotating 1t through a spectrum of specific yaw and pitch angles The data from each position angle and velocity was converted into a calibration database utilizing a set of non dimensional coefficients Accurate predictions of the velocity vector in an unknown flow field could then be determined by interpolating the non dimensional coefficients Johansen Rediniotis Jones 2001 B OVERVIEW The calibration experiment was conducted twice In each case the three hole cobra probe was calibrated using the free jet test rig pitot static probe a stagnation probe and a temperature probe were used to collect the calibration data The probes were mounted on a bracket and situated around the free jet The three hole probe was mounted in the L C Smith actuator capable of both axial and rotational movement Pressure data was initially acquired using the Scanivalve system The Scanivalve system required an excessive amount of time as 1 contained only one pressure transducer and had to step through each pressure and then wait for tube pressure to stabilize before taking a reading Once an initial calibration run was completed the setup was converted to a computer logic controlled system using USB devices accessed with Matlab Data was acquired through these devices and the run was completed The response time of the computer feedback control system greatly reduced the required time for the three hole probe calibration In addi
17. parameters It was believed this erroneous data 1s caused by the probe tip s proximity to the probe s insertion hole in the case wall and possible swirling of flow in that region Each test run was conducted at a constant speed with mass flow rate adjusted for each set of data points Runs were conducted at 70 80 90 and 100 percent speed 100 percent speed corresponded to the design speed of 27 085 revolutions per minute 36 On each test run the first set of data was conducted with the compressor throttle fully open This setting was labeled as the Choke setting For the second run the mass flow rate was reduced to the predetermined Peak Efficiency setting for the each speed in accordance with the Mass flow Isentropic efficiency graph contained in Appendix The final run of each series was with a reduced flow corresponding to the near stall mass flow rate Table four identifies the rotor speeds and mass flow rates used for calculations A 4 Rotor speed and mass flow rates Speed Mass Flow Rates Kg sec Speed Speed Choke Peak Efficiency Near Stall RPM 37 THIS PAGE INTENTIONALLY LEFT BLANK 38 VI RESULTS AND DISCUSSION A ANALYSIS 1 Upstream Data Inlet surveys were conducted at 70 80 90 and 100 percent speed The radial Mach number distributions are shown in Figures 28 31 with yaw angle and stagnation pressure in Appendices J and K 3 c Choke P
18. ready with answers to all my questions John Gibson and Rick Still thanks for redirecting me with common sense when necessary The atmosphere at the TPL was perfect for learning and working and many thanks to all students and faculty who work there for making my thesis a pleasant and rewarding opportunity xiil THIS PAGE INTENTIONALLY LEFT BLANK X1V L INTRODUCTION The F 35C 15 the Navy variant of the Joint Strike Fighter JSF and will utilize the F 135 engine which uses a transonic compressor Experiments conducted with catapult launches of an F 18 at Lakehurst Naval Air Engineering Station shown in Figures 1 and 2 demonstrated that current jet engines are susceptible to catastrophic pop stall when exposed to excessive amounts of steam The Joint Strike Fighter will be the first single engine jet fighter in recent years to be assigned to aircraft carrier based operations The single engine nature of the JSF increases the severity of a potentially catastrophic pop stall particularly if it were to be exposed to large amounts of steam leakage during a catapult launch aboard an aircraft carrier at sea Donelson 2003 Figure 1 Lakehurst steam ingestion experiment F 18 approaching steam leak during catapult launch The present goal of the Turbopropulsion Laboratory TPL at the Naval Post Graduate School NPS 1s to evaluate a small scale transonic compressor during stall and surge when exposed to steam
19. the same equation 15 then used to solve for the Mach number at the entrance to compressor blades The sixth order equation was solved using Matlab A 1 5 1 Ma gt 5X 1 star x41 Of the six solutions only two were non imaginary The first could not be used as it was greater than 1 and the second satisfied the requirements of being less than one Ma 0 461935 133 The Mach number was then designated the mach number of the flow directly before the leading edge of the compressor blades M 461935 The density of the entrance flow could then be calculated 0 2 Pe p 1 0 k 0 937 3 m The Static Temperature at the entrance Tse could then be calculated assuming isentropic conditions as the Stagnation Temperature remains constant in a nozzle 279 762K 134 With Mach number and Static Temperature the Static Velocity could then be calculated C 1 l se is then the velocity of the flow as it enters the blade row m C 1 155 427 a S The radial distance at the leading edge of the blade was used for calculating the relative velocity of the blade at the flow entrance The distance was calculated matching the flow area where the data point was taken downstream to the flow area at the trailing edge of the blade This flow area was then matched to streamlines calculated
20. 2 5 m 2 po m 2 flow angle Set the value of the Mach No textbox to the current Machno set handles probemach String Mach m 2 Yo Yo Yo YoYo Yo Yo YoYo YoYo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Y Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Vo NULL YAW LOOP Added to rotate the probe directly into the flow for increased accuracy when evaluating the flow behind the rotor flowangle is the angle relative to the probe flowangle Yaw_m_2 124 if abs flowangle lt 5 abs flowangle gt 60 break elseif flowangle gt 0 dur abs flowangle 2 9 ERB24_RIGHT dur pause 5 current_angle elseif flowangle lt 0 dur abs flowangle 3 ERB24 LEFT dur pause 5 current angle end end Ao Yo YoYo YoYo Yo Yo YoYo Yo Ao Yo AoAo Yo AoAo Yo AoAo Yo AoA AoAo A Ao Ao Ao A A A AoA A AoA ooo oo Freddy txt is notepad value used to hold to data outside of the program to ensure headers will be created load freddy txt restart txt is a notepad value used to hold on to data outside of the program to ensure data will not be overwritten if the run is restarted load restart txt a freddy 1 restart create a variable that can be evaluated so that the values of lines be changed fred ExcelWrite idi CADocuments and Settings user Desktop TransonicNesting1 xls 0 T num2str a num2str 2 header evaluate the variable and send the info to ExcelWrite to evaluate it and
21. September 1997 Hobson G V Gannon and Shreeve R P 2004 A transonic Compressor Stage Part 2 CFD Simulation ASME GT2004 53927 Turbo Expo Vienna Austria Johansen E S Rediniotis O K Jones G The Compressible Calibration of Miniature Multi Hole Probes Transactions of the ASME March 2001 Vol 123 pp 128 138 2d 10 11 13 14 15 10 17 18 19 20 Measurement Computing PMD 1608FS USB Based Analog and Digital Module User s Guide May 6 2005 Measurement Computing USB ERB24 USB Based 24 Relay Module User s Guide May 4 2005 O Brien J M Transonic Compressor Test Rig Rebuild and Initial Results with the Sanger Stage Master s Thesis Naval Postgraduate School Monterey California June 2000 Omega com Low Cost Silicon Pressure Sensor PX 138 Series 2005 Stamford Connecticut Papamarkos I Inlet Distortion Generation for Transonic Compressor Master s Thesis Naval Postgraduate School Monterey California September 2004 Reid William D Transonic Axial Compressor Design Case Study and Preparations for Testing Master s Thesis Naval Postgraduate School Monterey California September 1995 Roberts M W Unsteady Pressure Measurements on the Case Wall of a Transonic Compressor Master s Thesis Naval Postgraduate School Monterey California June 2003 Sanger N L Design of a Low Aspect Ratio Transonic Compressor Stage
22. USB port with Matlab programming A detailed block diagram from the user s manual 1s shown in Figure 16 16 500 m USB 2 0 USB compliant Microcontroller interface 3 ic driver Relayi2 Switah 1 Switoh 52 imeinan imv pull upidown Figure 16 USB ERB24 Relay Module block diagram from Measurement Computing USB ERB24 2005 G MATLAB DRIVER Matlab Release 13 Version 6 5 was used for programming and control of data acquisition devices Matlab was able to address the relays and guide the probe actuator through feedback controlled rotations In addition Matlab was able to collect and process pressure data from the transducers prior to exporting 1t to an Excel spreadsheet A single Graphical User Interface GUI was created to perform the required functions The GUI shown in Figure 17 was capable of probe actuator control and data collection in the calibration procedure 17 DURATION LEFT ouT RIGHT Figure 17 calibration graphical user interface The calibration control GUI required the manual input of the stagnation temperature of the jet read from the HPVEE system and the starting angle ending angle and number of calibration data points or angles as input The function calibration run m was called from the Excel control box and outputted the required angles to use for calibration a
23. Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo 9099090909999 Right Hole Pressure P3 Pressures 1 Center Hole Pressure P1 Pressures 2 Left Hole Pressure P2 Pressures 3 Total Pressure of jet Poj Pressures 4 sStatic Pressure of jet Psj Pressures 5 YoYo Yo YoYo YoYo YoYo YoYo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Temperature in pipe Tp is entered manually in the GUI Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Vo Vo Vo Vo Y YoYo YoYo YoYo AAC ALCU LA SY Y gt YoYo YoYo Vo Vo Vo Vo Vo Vo Yo q_probe P1 P2 P3 2 V_jet i sqrt 1 Psj Poj gam_air 1 gam_air sqrt 2 Cp_air Tp Cp1_Probe i P1 P2 q_probe Cp2_Probe i P1 P3 q_probe Cp3_Probe i P2 P3 q_probe 122 At_Probe i P1 Poj q_probe As Probe i 2q probe Poj Psj Mach Number Mach No i 2sqrt Poj Psj gam air 1 gam air 1 2 gam air 1 end YoYo YoYo YoYo Yo YoYo Yo YoYo YoYo AoAo YoYo YoYo YoYo AA YAYA AAA AAAA A AAAA AAA the values YoYo Yo Yo YoYo Yo YoYo Yo YoYo YoYo AoAo YoYo AAYA YAAA Vo Vo Vo Vo Vo Vo o probe pos probe pos 1 probe pos 2 probe pos 3 3 cur_angle cur_angle 1 cur_angle 2 cur_angle 3 3 Cp1_Probe Cp1_Probe 1 Cp1_Probe 2 Cp1_Probe 3 3 Cp2 Probe Cp2 Probe 1 Cp2 Probe 2 Cp2 Probe 3 8 Cp3_Probe Cp3_Probe 1 Cp3_Probe 2 Cp3_Probe 3 3 V_jet V_jet 1 V_jet 2 V_jet 3 3 At_Pr
24. a CFD grid along the actual blade dimensions following the flow line The solidity data was graphed and 1s shown below Solidity calculated matched the predicted values by Sanger Sanger 1996 157 Solidity of Rotor gt 2 15 0 4 0 6 Percent Area Hub Tip The solidity was then interpolated for the points required for each experimental data point based on the ratio of the area of flow For this data point solidity was found to be o 1 642 The value for the yaw angle of the flow was provided by three hole probe from the values taken downstream of the rotor The angle was used to calculate the velocity relative to the blade The velocity triangle 1s shown below in Figure 45 25 27 deg 138 Exit Flow Velocity Triangle C 2 U The variable Co is axial component of the static flow velocity and is calculated using the static velocity and the angle of the flow Cg Co sin B m 78 986 S Cx is the radial component of the static flow velocity and is also calculated with the static velocity and angle of the flow found by the probe 139 m 167 322 S U was the velocity at the trailing edge of the blade at the area averaged point of measurement Like it was a product of the rotational speed and the distance from the center of the measurement point however in t
25. closer to the case wall improved the resolution of the boundary layer Positions of the cosine z 2 function are shown in Table 3 For the downstream position 15 data points were also taken For this position a cosine z function was used to sample data more frequently closer to the hub and case wall A depth micrometer was used to identify the exact distance from the hub to the case wall The probe would begin at an approximate 1 5mm offset from the hub and 35 traverse towards the case wall The cosine 1 function was projected on to the distance between hub and case wall Positions of the cosine z function are also shown in Table 3 below Table 3 Probe sampling positions Null yawing was used in the downstream position The angle of flow varied from the hub to the case wall and with speed and mass flow rate To ensure the accuracy of the measurement an additional loop in the Matlab programming was created Logic was added so that 1f the measured angle of the flow were greater than 5 from the angle of the probe the probe actuator would turn the probe into the flow to null yaw the probe The calibration of the probe was assumed to be very accurate within the near linear area between 15 and satisfactory results were obtained However the final data point closest to the case wall occasionally read a large angle close to or 90 This caused the Matlab program to fail to output the final position as the angle was out of normal
26. conditions Compressor speed and mass flow rates were repeated giving close approximate reproductions of flow conditions To ensure less error the use of two probes giving simultaneous data upstream and downstream 15 recommended Data downstream of the rotor was collected with a three hole probe that was not sensitive to pitch models have shown that the flow at the exit of the rotor contains significant radial components as well as axial and tangential velocity components five hole probe would be better able to determine both pitch and yaw angles and give a better representation of the downstream flow The temperature data used in this experiment was provided by thermocouples connected to the HPVEE system run from a separate PC More accurate data collection would be possible with a thermocouple connected to a USB device controlled with Matlab Temperature measurements would then be collected simultaneously with the pressure data The diffusion factor calculations require the relative entrance velocity and relative exit velocity of the flow referenced to the blade The diffusion factor was calculated from holes more than four inches upstream and two inches downstream of the rotor Gathering data closer to the rotor would increase the accuracy of the diffusion factor calculations The Matlab functions used to control the probe and sample data could be optimized A better logic loop for rotating to an angle or traversing to
27. of lines be changed fred ExcelWrite CADocuments and Settings user Desktop ERB24Ntesting1 xls 1 num2str 2 m evaluate the variable and send the info to ExcelWrite to evaluate it and send it to an EXCEL Spreadsheet eval fred Change the value of freddy to the next row number freddy freddy 1 Save the row number to a text file so that it will not be cleared save freddy txt freddy ascii the run is not the first or a reset just output dats elseif freddy 1 a freddy restart Call the calculation function calculate Tp Tj a variable that be evaluated so that the values of lines be changed fred ExcelWrite CADocuments and Settings usen Desktop ERB24Nesting1 xls 1 num2str a num2str 2 evaluate the variable and send the info to ExcelWrite to evaluate it and it to an EXCEL Spreadsheet eval fred Change the value of freddy to next row number freddy freddy 1 120 Save the row number to a text file so that it will not be cleared save treddy txt freddy ascii end Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So So Vo Yo So So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So So So So So So So So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo SUBFUNCT
28. of the direction of the flow relative to the probe 1 _1 probe 4 3 LL uw 2 q probe 4 4 AN 5 probe 45 30 The total and static pressure coefficients provided the measurable data to be used during flow field surveys Piro _ 1 t Probe probe A6 b N q probe s Probe P oj p The Mach number was also calculated from the static and stagnation probe readings MachNo 4 8 The cal bration constants for each speed and angle were tabulated n an Excel spreadsheet When the table was complete the data was stored and placed in a binary Matlab format MAT file for use in subsequent flow field surveys E PREDICTION OF VELOCITY AND YAW ANGLE To determine the unknown velocities and yaw angles the Matlab function GRIDDATA3 was used The calibration data was accessed through a saved Matlab workspace text MAT file and loaded each time the flow field survey was conducted GRIDDATA3 interpolated hyper surfaces using tessellation based linear interpolation based on a Delaunay triangulation of the data The Mathworks Inc 2002 Interpolation constants and Matlab programming are contained in Appendix I 3l COMPARISON OF SCANIVALVE AND 138 DATA The Scanivalve and PX 138 systems created similar calibration databases However the PX 138 system was much quicker and produced smoother calibration curves for the As and At calibration constants especially in the lo
29. 978 1 3665 1 6335 1 9022 2 1701 2 4381 2 7086 2 9846 3 2541 3 5261 3 8 create a square polynomial to fit the data to p polyfit y x 2 evaluate the polynomial at the analog position and output the position in degrees current_posit polyval p analog_posit 79 function analog pressure PMD 1608 5 3 This function is called to take an average of channel one s output will be used for feedback on the pressure of the three hole probe clear all global analog pressure for i 1 length openDAQ stop openDAQ i end stop al delete ai clear all analoginput mcc 0 create channel 3 7 called 2 6 addchannel ai 2 6 set ai SampleRate 500 ActualAlRate ai SampleRate set ai SamplesPerTrigger 30 the t is left in for plotting and troubleshooting start ai a t 2 getdata ai start ai b t 2 getdata ai start ai c t getdata al start ai d t 2 getdata ai plot a t take an average of channel each output analog_pressure sum a 30 sum b 30 sum C 30 sum d 30 4 stop and delete the stop ai delete ai 80 function Pressures pressure calibration insert data into x and y to get the calibration curve for rotation the x data used for the angle was taken from digital inclinometer the y data was read from the digital pmd1608fs 2 First calibration 5 3 05 in inches of mercury converted to
30. Exit Plane CFD Data to Experimental Data 49 4 Comparison of CFD Diffusion Factor Data to Experimental Res uan 51 CONCLUSIONS AND RECOMMENDATIONS sssssssssssssssssssssssssssssssssssssssseee 55 A A AAA O 55 B RECOMMENDATIONS 52 euere 56 REFERENCES APPO O 57 APPENDIX A CASE WALL ENGINEERING DRAWING ssssssssssssssssssssssssssssssssssse 61 APPENDIX B COMPLETE STAGE COMPRESSOR 63 APPENDIX C COMPRESSOR INTAKE SYSTEM 65 APPENDIX D SCREEN SHOTS OF THE HPVEE PROGRAM eee 67 APPENDIX E CALIBRATION EQUATIONS AND SAMPLE CALCULATION 73 APPENDIX MATLAB CALIBRATION FILES 77 APPENDIX SCANIVALVE CALIBRATION 89 APPENDIX PX 138 CALIBRATION GRAPHS ee 93 APPENDIX I MATLAB INTERPOLATION 5 97 APPENDIX J UPSTREAM PROBE DATA STAGNATION PRESSURE 107 APPENDIX UPSTREAM PROBE DATA YAW ANGLE DISTRIBUTION 109 APPENDIX L DOWNSTREAM PROBE DATA STAGNATION PRESSURE RATIO EB OO 111 APPENDIX CFD VS EXPERIMENTAL DATA 90 ROTOR SPEED 113 APPENDIX MATLAB EXPERIMENTAL DATA COLLECTION FILES 119 APPENDIX O DIFFUSION FACTOR CALCULATIONS 129 ENITIAL DISTRIBUTION TEST 143 viil Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 Figure 9 Figure 10 Figure 11 Figure 12 Figure 13 Figure 14 Fi
31. IONS Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So Vo Vo Vo YoYo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So So Vo So Vo Vo So So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo function header calculate Tp Tj this subfunction will get all required data three times and average the three sets of values before outputting the final answer 9 o9 69 69 09 09 69 69 c9 69 69 699C ONS T ANT SY YYYY AAAY A A AAA of air Cp_air 1005 gamma of air gam_air 1 41 gamma of air gravitational constant m sec 2 g 9 81 gravitational constant m sec 2 density of Mercury kg m 3 calibration pressure in inches of mercury rho Hg213550 for 1 2 loop for data for i 1 3 Get the data from the PMD 1608FS s Yo YoYo Yo Yo Yo Yo Yo YoYo Yo YoYo SoS SoS So SoS YA Ao So So So So Vo Vo Vo Vo Vo So Vo So So Vo Vo Vo Vo Vo Yo Vo Vo Vo Vo oVo 5 Get the current angle h PMD_1608FS_2 posit PMD_1608FS_1 Convert the angle to degrees cur_angle i angle_calibration h probe pos i traverse_calibration posit get the pressure variable info Pressures pressure_calibration Pressures PMD 1608FS 3 121 Yo Yo Yo YoYo Yo YoYo YoYo Vo Y Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo YoYo YoYo Vo Cobra Probe pressures Yo Yo Yo YoYo Yo Yo YoYo YoYo Vo Vo Vo Vo
32. Pascals global Pressures global analog_pressure analog pressure PMD 1608FS 3 x 0 2 4 6 8 10 12 14 16 3386 38 29 795 3386 38 Values for upper pressure connection y 1 3 50302383 3 339436942 3 172269271 3 010462603 2 844668243 2 679341828 2 519305206 2 354955367 2 191195544 y 2 3 506919966 3 34128837 3 171496147 3 007471834 2 83968363 2 671834389 2 509325806 2 342646423 2 1769131 y 3 3 504193688 3 342549783 3 1 76033163 3 015854124 2 850883752 2 686930648 2 52751456 2 364995804 2 202263422 y 4 3 512240279 3 346832481 81 3 17700974 3 013300781 2 845888965 2 678171969 2 515765113 2 349106076 2 183474479 Values for Upper pressure connection 5 5 05 y 1 3 5036 3 3389 3 1709 3 0046 2 8464 2 6802 2 5208 2 3544 2 1891 2 3 5082 3 341 3 1703 3 0016 2 8412 2 6727 2 5107 2 3419 2 1742 3 3 5048 3 3418 3 1744 3 0098 2 8527 2 688 2 5292 2 3641 2 2001 y 4 3 5 3 3348 3 1685 3 0069 2 8396 2 6759 2 5129 2 3494 2 1863 5 3 5129 3 3457 3 1751 82 3 0065 2 8469 2 6785 2 5167 2 3473 2 1803 for li 1 5 create a polynomial to fit the data to Data was nearly linear p ii polyfit y 11 1 evaluate the polynomial at the analog position and output the position in degrees Pressures il polyval p ii analog_pressure ii en
33. T Monterey 93943 5000 NUMBER 9 SPONSORING MONITORING AGENCY NAME S AND ADDRESS ES 10 SPONSORING MONITORING Naval Air Systems Command Air 4 4 Naval Air Warfare Center Aircraft Div AGENCY REPORT NUMBER Propulsion and Power Engineering Patuxent River Maryland 11 SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the official policy or position of the Department of Defense or the U S Government 12a DISTRIBUTION AVAILABILITY STATEMENT Approved for public release distribution is unlimited 13 ABSTRACT maximum 200 words Investigating the effect of steam ingestion into an aircraft jet engine 1s necessary to improve understanding of stall and surge in transonic axial compressors Specifically to understand the pop stall phenomenon experienced by naval fighter jet aircraft during steam catapult launches Steam leakage from an aircraft carrier catapult system can be ingested into the intake and cause stall or surge in a jet engine upon takeoff It is important to understand the conditions under which this occurs as the Navy prepares for the fielding of the single engine F 35C the aircraft carrier variant of the Joint Strike Fighter This project prepares the structure and instrumentation to investigate the inlet distortion and effects of steam ingestion on a transonic axial compressor compressor test facility including mechanical equipment data acquisition system and remote dig
34. THIS PAGE INTENTIONALLY LEFT BLANK 128 APPENDIX O DIFFUSION FACTOR CALCULATIONS The diffusion factor 15 a method of assessing blade loading in an axial compressor For a simple two dimensional geometry the diffusion factor 15 calculated using the equation where V and V gt are the inlet and exit velocities relative to the blade Vo 15 the turning of the flow assuming the flow 15 axial at the blade entrance and o 15 solidity of the blade corresponding to the flow It 1s believed that values above 0 6 can indicate blade stall and values of 0 45 may be a typical design choice Cumpsty 1989 The diffusion factor was calculated for each set of speed data and results were then compared to data created with SWIFT Code sample set of calculations for 100 Speed Peak Efficiency at a point near the midpoint of the blade 1s show in the following D is the design speed in RPM D 27085 o is the angular velocity in rad sec T 60 sec o D 2 S 29b 129 mdot 15 the mass flow rate in Kg sec kg mdot 7 687 sec To 15 the Stagnation Temperature measured in the upstream flow 292 15 the Mach number measured by the three hole pressure probe in upstream flow M has been area averaged over the 15 data points taken upstream 3344 A 18 the Specific Heat Ratio The value of 1 41 15 used throughout the calculations for air and was used dur
35. The present study documents the equipment preparation and instrumentation design to conduct the evaluation as well as baseline data obtained Figure 2 Lakehurst steam ingestion experiment F 18 Pop Stall caused by steam leak during catapult launch To evaluate the pop stall phenomenon a rotor similar to that used in military fighter jets was necessary The rotor of the first stage of a transonic compressor was evaluated in the present study The Sanger Rotor is a low aspect ratio transonic compressor Nelson Sanger designed the compressor stage in 1996 at the NASA Lewis Research Center The rotor was designed using CFD techniques and provided to the Naval Post Graduate School for experimental testing Sanger 1996 Pressures both upstream and downstream of the single rotor were evaluated by traversing a three hole pressure probe at various locations Prior to conducting stall experiments a three hole pressure probe was calibrated and a full baseline survey was completed Data was obtained at 70 80 90 and 100 percent speed with mass flow rate varied from choke peak efficiency and to stall This project 1s the beginning of an in depth study of stall and surge in transonic compressors The current project prepares the instrumentation and baselines the performance of a transonic rotor Future projects will include the injection of steam upstream of the rotor With the baseline completed the calculation of stall margin and
36. Using CFD Techniques ASME Journal of Turbomachinery July 1996 Vol 118 pp 479 491 Sanger N L Design Methodology for the NPS Transonic Compressor TPL Technical Note 99 01 August 1999 Sussman Automatic Corporation SVS6 Electric Steam Boilers Specifications 2004 Long Island City New York The MathWorks Inc Using Matlab Graphics Release 13 Version 6 5 September 2002 56 21 White F M Fluid Mechanics 5th Edition 2003 McGraw Hill New York New York 59 THIS PAGE INTENTIONALLY LEFT BLANK 60 CASE WALL ENGINEERING DRAWING APPENDIX A seyadio 111 i S QD 010195181 N ICE E y seusuiaBd LOLI AN SS gt N 2 EN MULES S6 J Y UMUZB PE 1 5243 BR Z 1 seu3ulg BOE J z e 61 Probe Hole Distances from Trailing and Leading Edges of Rotor 111 5mm 4 39 Upstream of Rotor Leading Edge 8 64mm 34 Downstream of Rotor Trailing Edge 59 4mm 2 34 Downstream of Rotor Trailing Edge 9 91 mm 39 Upstream of Rotor Leading Edge 62 APPENDIX B COMPLETE STAGE COMPRESSOR MAP Total to total pressure ratio CFD Stage Rotor Stage Rotor Only Surge Lines Isentropic efficiency 0 F5 4 4 5
37. a position could reduce the time required for each experiment In addition the calling of calibration constants required a large amount of computer memory to load the constants in the MAT file and interpolate the angle and Mach number and may require even more memory for a five hole probe S6 LIST OF REFERENCES Chima R V 1998 Calculation of Multistage Turbomachinery Using Steady Characteristics Boundary Layer Conditions AIAA 98 0968 or NASA TM 1998 206613 Cumpsty N A Compressor Aerodynamics 28 Edition 1989 John Wiley amp Sons New York New York Donelson S Briggs T JSF Team Probes Steam Catapult Environment JSF Integrated Test Force dcmilitary com February 6 2003 Gannon A J Hobson G V and Shreeve R P 2004 A transonic Compressor Stage Part 1 Experimental Results ASME GT2004 53923 Turbo Expo Vienna Austria Gannon A J Hobson G V Shreeve R P Measurement of the Unsteady Casewall Pressures Over the Rotor of a Transonic Fan and Comparison with Numerical Predictions ISABE 2005 17 International Symposium on Airbreathing Engines Munich September 2005 Greco P A Turbine Performance Mapping of the Space Shuttle Main Engine High Pressure Fuel Turbopump Master s Thesis Naval Postgraduate School Monterey California September 1995 Grossman B L Testing and Analysis of a Transonic Axial Compressor Master s Thesis Naval Postgraduate School Monterey California
38. ata to an Excel Spreadsheet The transducers were calibrated by creating a table of voltage outputs for known pressures The transducers had two ports to measure a differential pressure One port of each transducer was connected to a pressure standard To verify accuracy the calibration pressure was then connected to the other pressure port on each transducer and the outputs were once again tabulated Once complete the pressure transducers were disconnected from the calibration pressure and connected to the pressure probes The response of the transducer was found to be nearly linear for both sides positive and negative pressures A linear interpolation method in Matlab was used to find the calibration for pressures When the PX 138 was used for calibration it was commanded to take 30 samples at 500hz average the results and repeat this process four times The four results were again averaged to create the output of each pressure transducer Calibration data and Matlab programming for the PX 138 transducers can be found in Appendix F The calibration experiment was repeated using the control system and PX 138 pressure transducers Using a feedback loop for angle position and automatically rotating through the 30 to 30 the time required for the experiment was greatly reduced Data obtained for calibration was comparable to that obtained by the Scanivalve system 21 CALIBRATION EQUATIONS 1 Purpose The purpose of the calibration ca
39. column 21633 2 5 O Degree pitch A end 1 xlsreadcells 5 10 05 10deg Pitch forward xls Sheetl column 1 10 Degree pitch Mach 0 7 Nominal 1 xlsreadcells 5 10 05 40deg Pitch xlis Sheetli column 39 51 40 Degree pitch 1 xlsreadcells 5 9 05 3 deg FPitch xls Sheerli column 77389 30 Degree pitch 17 29 1 xlsreadcells 5_9 05 _ 20deg_Pitch xls Sheet1 column 77 89 5 20 Degree pitch 1 xlsreadcells 5 9 05 104 _Pitch xls Sheetl column 82 94 5 10 Degree pitch A end 1 xlsreadcells 5_5_05_Odeg_Pitch xls Sheet1 column 40 52 5 0 Degree pitch 1 xlsreadcells 5 10 05 10deg Pitch forward xls Sheetl column 34 46 1 10 Degree pitch Mach 0 85 Nominal 1 xlsreadcells 5_10_05_40deg_Pitch xls Sheet1 column 59 71 5 40 Degree pitch 1 Xlsreadcells 5_9_05_3Udeg_Pitch xls tSheetl column 93 105 5 30 Degree pitch A end 1 xlsreadcells 5 9 05 20deq_Pitch xls Sheet1l column 94 106 5 20 Degree pitch 1 xlsreadcells 5 9 05 10deg Pitch xls Sheetl column 101 113 5 10 Degree pitch 1 xlsreadcells 5 5 05 Pitch xls Sheetl column 59 71 5 O Degree pitch 1 xlsreadcells 5 10 05 10deg Pitch for
40. d function OUTPUT EXCEL Tp Tj this function gets the data from the devices using 1608 5 2 and PMD_1608FS_3 and sends the data to ExcelWrite to be written to an Excel Spreadsheet global Pressures Freddy txt is notepad value used to hold on to data outside of the program to ensure headers will be created load freddy txt restart txt is a notepad value used to hold on to data outside of the program to ensure data will not be overwritten if the run is restarted load restart txt it is the first run or the run has been restarted this loop will the header in and then write the first line of data if freddy 1 first line put the header in if restart add the number of lines so that data will not be overwritten 83 a freddy restart header m f angle P1 P2 P3 4 probe V_jet Cp1_Probe Cp2_Probe Cp3_Probe At_Probe As_Probe Mach create a variable that can be evaluated so that the values of lines be changed fred ExcelWrite CADocuments and Settings user Desktop ERB24Ntesting 1 xls F num2str a num2str 2 evaluate the variable and send the info to ExcelWrite to evaluate it and it to an EXCEL Spreadsheet eval fred change the row number it will be the next available the calculation function calculate Tp Tj create a variable that can b
41. des angle is input from the For Loop to get to the proper angle for i21 15 Get the current analog angle and call it h h PMD 1608 5 2 Convert the analog angle into a real angle cur angle angle calibration h handles guihandles gcbo generate handles struct 77 Set the value of the Cur Angle textbox to the current angle set handles Cur Angle String cur angle Assign the cur angle to be used in the base workspace assignin base cur angle cur angle Find the difference between the current angle and the required angle diffzcur angle des angle if abs diff lt 1 break end if cur_angle gt des_angle amp amp abs diff gt 20 ERB24 LEFT 6 pause 1 elseif cur_angle gt des_angle amp amp abs ditf lt 19 99999 amp amp abs diff gt 10 ERB24 LEFT 3 pause 1 elseif cur_angle gt des_angleg8Sabs diff lt 9 9999988abs diff gt 5 ERB24 LEFT 1 5 elseif cur_angle gt des_angle amp amp abs ditf lt 4 99999 amp amp abs diff gt 3 ERB24 LEFT 1 elseif cur angle 2des angle amp amp abs diff 22 99999 amp amp abs diff 1 ERB24 LEFT 1 elseif cur angle des angle amp amp abs diff 2 999999 amp amp abs diff 2 2 ERB24 LEFT 02 elseif cur_angle gt des_angle amp amp abs diff lt 199999 amp amp abs diff gt 15 ERB24 LEFT 01 elseif cur angle gt des angle amp amp abs diff 14999998 amp amp abs diff 1 ERB24 LEFT 003 end if cur _angle lt des_a
42. e evaluated so that the values of lines be changed fred ExcelWrite CADocuments and Settings usen Desktop ERB24Nesting1 xls num2str 2 m evaluate the variable and send the info to ExcelWrite to evaluate it and send it to an EXCEL Spreadsheet Change the value of freddy to the next row number freddy freddy 1 Save the row number to a text file so that it will not be cleared save freddy txt freddy ascii the run is not the first or a reset just output dats elseif freddy 1 a freddy restart Call the calculation function calculate Tp Tj 94 create a variable that can be evaluated so that the values of lines be changed fred ExcelWrite CADocuments and Settings usen Desktop ERB24Nesting1 xls e 07 T num2str a num2str 2 m evaluate the variable and send the info to ExcelWrite to evaluate it and it to an EXCEL Spreadsheet eval fred Change the value of freddy to next row number freddy freddy 1 Save the row number to a text file so that it will not be cleared save freddy txt freddy ascii end Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So Vo Vo Vo YoYo YoYo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So Vo So Yo YoYo So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo V
43. e was located approximately 11 15cm 4 39 inches upstream of the leading edge of the rotor while hole four was located approximately 5 94cm 2 34 inches downstream of the trailing edge of the rotor Hole one was also located axially forward of the nose cone giving access to a complete radius of inlet flow sampling from center to case wall 9 In the TCR air was drawn from the atmosphere through inlet throttle valve and into a chamber labeled 1 in Figure 8 A 46cm diameter five meter long inlet pipe connected the chamber to the test compressor The inlet pipe also contained a nozzle labeled 2 in Figure 8 for flow rate measurement and a transition duct to the 27 94cm diameter case wall Papamarkos 2004 Figure 8 Compressor test rig air flow B STEAM INLET DISTORTION GENERATION A Sussman model SVS600 steam generator was installed in the test cell adjacent to the TCR to provide steam for the compressor inlet The SVS600 contained a 612kW boiler and could provide steam at a Boiler Horsepower BHP rating of 62 4 This corresponded to supply steam of 82 1 kgs 1806 165 per hour at 100C 212F The design pressure of the boiler 1s 1 034MPa 150psig Sussman 2004 The steam supply system 15 shown Figure 9 Free Jet Test Cell TCR Test Cell TCR Inlet Ducting Figure 9 TCR Steam generation system 10 The steam generated by the SVS600 will be injected upstream of the test compressor Steam fr
44. eadsheet global Pressures Freddy txt is notepad value used to hold on to data outside of the program to ensure headers will be created load freddy txt restart txtis a notepad value used to hold on to data outside of the program to ensure data will not be overwritten if the run is restarted load restart txt it is the first run or the run has been restarted this loop will the header in and then write the first line of data if freddy 1 first line put the header in if restart add the number of lines so that data will not be overwritten a freddy restart header header probe pos angle P1 P2 q probe V jet Cp1_Probe Cp2 Cp3_Probe At Probe Probe Mach No Pitch m Yaw m ps m po Pitch m 2 Yaw m 2 m 2 m 2 po m 2 create a variable that can be evaluated so that the values of lines be changed fred ExcelWrite CADocuments and Settings user Desktop TransonicNesting1 xls 0 T num2str a num2str 2 header 119 wi p 07 evaluate the variable and send the info to ExcelWrite to evaluate it and it to an EXCEL Spreadsheet eval fred change the row number it will be the next available the calculation function calculate Tp Tj create a variable that can be evaluated so that the values
45. en from the three hole probe 1 117780 4Pa P gt 117318 3Pa P3 101240 6Pa and Ps were taken from the Stagnation and Static Pressure Probes 120177Pa Ps 100783Pa 73 1 55297 OK p 1005 p kg K y 1 41 The pseudo dynamic pressure probe 15 given as P2 P3 probe 1 2 probe 8 501 x 10 m V jet 172 72 S The angle flow discriminators give the direction of the flow 74 P1 Po C probe 1 0 054 Ni p 2 b q probe Cy 2 1 946 P5 3 b probe Cp 3 1 891 The total and static pressure coefficients provide the measurable data for discerning unknown flows 75 P1 t Probe q pr ob e At Probe 70 282 _ probe As Probe 5 _ As Probe 9 438 MachNo MachNo 0 506 76 GUI APPENDIX F MATLAB CALIBRATION FILES function calibration run angle1 angle2 points Tj positions linspace angle1 angle2 points for i21 points Input Angle positions i handles guihandles gcbo generate handles struct set handles Input Angle String Input Angle des angle Input Angle rotate des angle pause 2 OUTPUT TO EXCEL Tp Tj end function cur angle rotate des angle this function calls the data from the GUI PMD_1608FS and tells the _24 to rotate the probe actuator The desired angle
46. es have been conducted on the stage Roberts 2003 Gannon Hobson Shreeve 2004 The rotor profile 15 shown on the TCR in Figure 4 Figure 4 View of the rotor only configuration tested The rotor was manufactured from a high strength aluminum alloy 7075 T6 and consisted of 22 blades In this experiment the rotor was tested with a parabolic spinner which replaced the conical spinner used by O Brien O Brien 2000 Gannon et al 2004 In the majority of previous studies the entire stage was evaluated to verify the accuracy and validity of CFD design For the current experiment the rotor was evaluated without the stator and data was obtained for the rotor only which was simpler to test mechanically and the future steam ingestion tests would quantify the effect on the rotor only In addition the honeycomb used to straighten the flow downstream of the stage 6 was removed Design parameters for the rotor are reproduced from Sanger s Design Methodology in Table 1 below Sanger 1996 Table 1 Sanger rotor design parameters Parameter Pressure Ratio Tip Speed Design Speed Design Mass Flow Specific Mass Flow Specific Head Rise Tip Inlet Relative Mach Number Aspect Ratio Hub Tip Radius Ratio Rotor Inlet Ramp Angle Number of Rotor Blades Tip Solidity Rotor Outside Diameter Rotor Diffusion Factor Tip Rotor Diffusion Factor Hub 1 61 33 02 m s 1300ft s 27085 rpm 7 75 kg s 17 051b s
47. fied angle using the manual controller was laborious and time consuming The final system of instrumentation created to evaluate flow upstream and downstream of the compressor reduced the amount of work and time required The Matlab control system using USB devices worked rapidly decreasing the time required for probe movements The pressure transducers provided stable accurate data to a system capable of processing the data as 1t was taken and were capable of being calibrated to a lower Mach number The USB A D converter allowed rapid communication with Matlab and allowed the fast processing of data during the experiment In addition the USB A D converter provided feedback for the probe actuator as well as the processing of pressure data to allow feedback for to null yaw the probe However the Matlab programming could be optimized as the system slowed occasionally and failed when the probe was placed in a highly swirling flow The nature of the diffusion factor calculated was found to be close to what was predicted by CFD code and the magnitude was within the expected range In addition the shape of the diffusion factor data shows the higher loading at the tip and validates the prediction of a stall 55 RECOMMENDATIONS All data was collected with a single three hole probe To calculate the diffusion factor data from both upstream and downstream was necessary The data was collected on separate days with different atmospheric
48. gure 15 Figure 16 Figure 17 Figure 18 Figure 19 Figure 20 Figure 21 Figure 22 Figure 23 Figure 24 Figure 25 Figure 26 Figure 27 Figure 28 Figure 29 Figure 30 Figure 31 Figure 32 Figure 33 Figure 34 Figure 35 LIST OF FIGURES Lakehurst steam ingestion experiment F 18 approaching steam leak catapult line A een 1 Lakehurst steam ingestion experiment F 18 Pop Stall caused by steam leak during catapult launch oooooonnnnnnnncnnnnnnnnnnnnnnnnnononononononcncnnnnnnnnnnnnnnnnnos 2 Transonic compressor rig in test cell with inlet piping 5 View of the rotor only configuration tested 6 Total pressure ratio compressor map pp 8 Isentropie eiHieieiicy COIIDIOSSOF MAP 8 Compressor caso 9 Compressortest BE ke 10 TER Steam y stent 10 STC ANTS CSPOT Marcas 11 Three hole probe within the probe actuator free jet assembly 12 Threeshole cobra probe 13 Probes positioned on the mounting bracket for calibration procedure on ai jel test as MD ae 14 PX pressure 15 PMD 1608FS A D Module block diagram from Measurement Computing 60865 2005 d een 16 USB ERB24 Relay Module block diagram from Measurement Computing USB ERB24 2003 Silanes Matlab calibration graphical user
49. his case the point was projected onto the trailing edge of the rotor blades The radial velocity 1s found by measuring the distance the point was from the case wall subtracting it from the full radius and multiplying the result by the angular velocity U 0 1397 0 0228223 m o m 02 331 504 S Vo was the angular relative velocity and was the difference between the velocity of the trailing edge and the axial component of flow 02 U2 Co m V go 252 518 1 S was the velocity of the flow relative to the blade With the angular and axial flow components found V was calculated 140 2 2 C 99 m 302 923 S With all components of the equation the Diffusion Factor can then be determined with the diffusion factor equation Va Ve DF 20V DF 0 3931 141 THIS PAGE INTENTIONALLY LEFT BLANK 142 INITIAL DISTRIBUTION LIST Defense Technical Information Center Ft Belvoir Virginia Dudley Knox Library Naval Postgraduate School Monterey California Distinguished Professor and Chairman Anthony Healey Department of Mechanical and Aeronautical Engineering Naval Postgraduate School Monterey California Professor Ray Shreeve Department of Mechanical and Aeronautical Engineering Naval Postgraduate School Monterey California Professor Garth Hobson Department of Mechanical and Aeronautical Engineering Naval Postgraduate Sch
50. ing the calibration as well 2 x9 LAI The Static Temperature T is calculated from the Mach number and Stagnation temperature T 285 456 130 P 15 Stagnation Pressure measured with a pitot static tube upstream of the flow P 69153Pa 15 the Ideal Gas Constant specific for J Ra 287 K kg The density p 1s calculated upstream in the flow prior to decrease in area caused by contraction in area prior to entering the turbine blades The diameter of the inlet d is 0 2794 meters 1 linches 4 2794 131 A 18 the total area of the inlet 42 Ay 4 A 0 061312m Knowing the mass flowrate density and area allows us to calculate V upstream the velocity of the fluid prior to reaching the spinner V mdot stream up m upstream 148 534 The area of the inlet 1s reduced as the air is directed through the compressor blades The spinner and reduction in area acts as a nozzle and a new velocity must be calculated to better approximate the velocity at the leading edge ofthe blades The restriction in area 1s treated as a simple nozzle in the following equations 1 2 073025 m Aentrance At 4 2 0 044559 entrance 132 The critical area Astar can be calculated assuming isentropic flow and that there are no shock waves in the duct White 2003 Astar 0 022231 m Once the critical area 15 know
51. ircraft jet engine 1s necessary to improve understanding of stall and surge in transonic axial compressors Specifically to understand the pop stall phenomenon experienced by naval fighter jet aircraft during steam catapult launches Steam leakage from an aircraft carrier catapult system can be ingested into the intake and cause stall or surge in a jet engine upon takeoff It 15 important to understand the conditions under which this occurs as the Navy prepares for the fielding of the single engine F 35C the aircraft carrier variant of the Joint Strike Fighter This project prepares the structure and instrumentation to investigate the inlet distortion and effects of steam ingestion on a transonic axial compressor A compressor test facility including mechanical equipment data acquisition system and remote digital control system was configured to test a transonic compressor rotor similar to what will be used in the Joint Strike Fighter Rotor inlet and exit velocity profiles were measured with a three hole probe to obtain a set of baseline data before future experiments THIS PAGE INTENTIONALLY LEFT BLANK vi TABLE OF CONTENTS INTRODUCTION lt a 1 EXPERIMENTAL FACILITY AND EQUIPMENT e eee eee eee eee eee eene 5 A TRANSONIC COMPRESSOR 5 B STEAM INLET DISTORTION GENERATION eee eee e eee eee eeee 10 C TEST
52. ital control system was configured to test a transonic compressor rotor similar to what will be used in the Joint Strike Fighter Rotor inlet and exit velocity profiles were measured with a three hole probe to obtain a set of baseline data before future experiments 14 SUBJECT TERMS Compressor Transonic Steam Ingestion Inlet Distortion Turbomachinery 15 NUMBER OF Pop Stall Rotor PAGES 159 16 PRICE CODE 17 SECURITY 18 SECURITY 19 SECURITY 20 LIMITATION CLASSIFICATION OF CLASSIFICATION OF THIS CLASSIFICATION OF ABSTRACT REPORT PAGE ABSTRACT Unclassified Unclassified Unclassified UL NSN 7540 01 280 5500 Standard Form 298 Rev 2 89 Prescribed by ANSI Std 239 18 THIS PAGE INTENTIONALLY LEFT BLANK 11 Approved for public release distribution is unlimited FLOW FIELD SURVEYS IN A TRANSONIC COMPRESSOR PRIOR TO INLET STEAM INGESTION TESTS Ivan J Villescas Lieutenant United States Navy B S Rennselaer Polytechnic University 1996 Submitted in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN MECHANICAL ENGINEERING from the NAVAL POSTGRADUATE SCHOOL September 2005 Author Ivan Villescas Approved by Garth Hobson Thesis Advisor Anthony Gannon Co Advisor Anthony Healey Chairman Department of Mechanical and Astronautical Engineering 111 THIS PAGE INTENTIONALLY LEFT BLANK 1V ABSTRACT Investigating the effect of steam ingestion into an a
53. its change as steam 1s added to the system can be conducted THIS PAGE INTENTIONALLY LEFT BLANK EXPERIMENTAL FACILTTY AND EQUIPMENT The current research was conducted at the Turbopropulsion Laboratory within the Department of Mechanical and Astronautical Engineering of the Naval Postgraduate School A TRANSONIC COMPRESSOR RIG The Transonic Compressor Rig TCR was originally built to test a prototype transonic compressor stage and was designed by Professor Michael Vavra The configuration tested 1s shown in Figure 3 and was detailed extensively by O Brien and Papamarkos O Brien 2000 Papamarkos 2004 The TCR was set up to evaluate the Sanger transonic rotor The test compressor was driven by two opposed air turbine stages which were supplied with air by an Allison Chalmers axial compressor and was capable of testing axial compressors up to 30 000 RPM Be A ERA 1 d D M a8 Figure 3 Transonic compressor rig in test cell with inlet piping removed Sanger at the NASA Lewis Research Center designed compressor rotor as part of a complete fan stage It was a completely Computational Fluid Dynamic CFD designed stage Details of the design and design considerations are contained in TPL Technical Note 99 01 Sanger 1999 The stage was manufactured by numerical machining specifically for testing and evaluation at the NPS Turbopropulsion Laboratory Several experimental studi
54. lculated using velocity of the rotor from the downstream probe angle and velocity measurement and calculating the radial and axial components of flow From these components the relative exit velocity was determined The solidity was calculated using data from the rotor blade that had been used in the SWIFT calculations A graph of the solidity can be found in Appendix O The diffusion factor was calculated for each set of speed data A sample set of calculations for 100 Speed Peak Efficiency near the midpoint of the blade 15 shown in Appendix O Three dimensional representations of the diffusion factor are shown for Choke Peak Efficiency and Stall in Figures 40 42 46 Diffusion Factor Diffusion Factor won s msional ion dime Figure 40 Diffusion Factor at Choke Figure 41 l position Non dimension Diffusion Factor at Peak Efficiency 47 Dittusion Factor 7 a 1 i Qi ye n i J L lt i Figure 42 Diffusion Factor at Near Stall B OBSERVATION 1 Tip Stall The experimental data predicted a tip stall as the mass flow rate is decreased In Figures 40 42 the diffusion factor peaks near the tip Some research has been conducted on the shock structure and location of the tip leakage vortex originating at the location of the maximum pressure differential at the tip of the blade Gannon et al 2004 2 Magnitude of Diff
55. lculation was to determine the non dimensional variables for each angle The total pressure coefficient A probe and the static pressure coefficient As probe had to be determined for each angle in a flow field of a known velocity to be later used to determine the flow angle in an unknown field A sample calculation of the calibration constants 15 shown in Appendix E to demonstrate the calibration process The calibration calculations were adapted from five hole probe calibration calculations Johansen et al 2001 2 Setup and Nomenclature The center hole was designated 1 in accordance with standard industry designation When facing into the flow the left side was designated 2 and the right side was designated 3 Pitch and yaw angle definitions are shown in Figure 23 Yaw Pitch Figure 23 Three hole probe pitch and yaw angle definitions 28 Table 2 Calibration nomenclature rT CON EEN Pat e 2 00 MachNo Mach Number Calculated with Jet Static and Stagnation Pressure Probes 29 3 Calibration Equations The pseudo dynamic pressure 4 probe provided a single pressure value for each data point for the probe and was given as 3 probe P 2 4 1 The jet velocity was calculated with the static and stagnation probes and 1s given 95 4 2 The angle flow discriminators Cp 1 Cp gt and provided a sense of magnitude
56. n the HPVEE system The angle at 24 which the pressures were equal was designated the zero angle and used as a reference point for calibration A post processing program was created using the HPVEE system to determine the calibration constants for the three hole probe Calibration data was obtained using the HPVEE Scanivalve system described by Grossman Grossman 1997 Screen captures of the HPVEE system are shown in Appendix D In the free jet test assembly the pitot static and stagnation probes were used to calculate the Mach number The Mach number was increased and decreased by adjusting the mass flow rate through the nozzle This was conducted by throttling the dump valve on the Allison Chalmers compressor for Mach numbers above 0 3 and by throttling the control valve 1 the cell for below Mach 0 3 The calibration was conducted by rotating the probe from 30 to 30 with a manual electric control switch Pressure and temperature readings were taken at five degree increments and exported to an Excel database Each complete rotation was conducted for Mach numbers 0 85 0 7 0 5 0 3 1 and 0 05 with the probe perpendicular to the flow D CALLIBRATION PROCEDURE WITH AUTOMATED CONTROLS AND INDIVIDUAL PRESSURE TRANSDUCERS The probe actuator was adapted to a feedback control system using the USB controlled relays and a USB A D converter The feedback was provided through the Measurement Computing MC Analog Inpu
57. ncreased The results are shown in Figures 46 49 Exp Choke Exp Peak Exp Neer Stall CFO Choke Hear Stall CFD Peak o 6 LL D a 0 8 Hub to Tip Radius Ratio Figure 46 70 speed diffusion factor comparison 51 o o LL 2 un 5 a Exp Mez Stall CFO Choke CFO Hear Stall CFO Peak Eff Hub to Tip Radius Ratio Figure 47 80 speed diffusion factor comparison ooi A ___________ Exp Choke a Exp Peak Exp Near Stall Diffusion Factor Nee Stal CFO Peak 0 6 0 7 0 3 Hub to Tip Radius Figure 48 90 speed diffusion factor comparison 22 u Exp Peak Exp Hex Stall CFO Choke Hear Stall o hm o m L u Yen y c 0 9 Hub to Tip Radius Ratio Figure 49 100 speed diffusion factor comparison 53 THIS PAGE INTENTIONALLY LEFT BLANK 54 CONCLUSIONS AND RECOMMENDATIONS A CONCLUSIONS The present study set up the instrumentation and established a baseline for future investigation into stall and surge in a transonic axial compressor The pop stall phenomenon experienced by jet aircraft ingesting hot gas was the primary purpose of the present work The initial probe calibration was conducted by manually rotating the probe with an electric switch controller Rotating the probe to a speci
58. nd used the feedback and control function rotate m which accessed the both the USB relay and USB data acquisition DAQ device to position the probe actuator for each angle At each calibration angle the probe would stop and the output to excel m function called the pressure calibration m function to take pressure readings and calibrate them The pressure data was combined with the stagnation temperature of the jet from the GUI input to calculate the calibration constants The raw data along with the processed data was exported to an Excel spreadsheet detailed block diagram 18 showing both probe the actuator control and the data acquisition process for the probe calibration procedure for Matlab programming 1s shown in Figure 18 T1 P2 P3 Po Pe Current Angle de As Mach Figure 18 Matlab calibration code block diagram The Matlab data acquisition GUI was adapted from the calibration GUI The GUI was changed to automatically control both the angle and the traverse of the probe actuator The Excel control box was changed to require input for the starting point and ending point for probe traversing along with the number of points required for sampling Once the data was entered the SAMPLE DATA pushbutton sent the probe into an automatic cycle of probe traverse rotation and data collection screenshot of the Matlab data collection GUI is shown in Figure 19 19 Mach Number TubeMach macn Mach N
59. ngle amp amp abs diff gt 20 ERB24_RIGHT 6 pause 1 elseif cur lt angle amp amp abs diff lt 19 999998 amp amp abs diff gt 10 78 ERB24 RIGHT 3 pause 1 elseif cur_angle lt des_angle amp amp abs diff lt 9 99999 amp amp abs diff gt 5 ERB24 RIGHT 1 5 elseif cur _angle lt des_angle amp amp abs diff lt 4 99999 amp amp abs diff gt 3 ERB24_RIGHT 1 elseif cur _angle lt des angle amp amp abs diff lt 2 99999 amp amp abs diff gt 1 ERB24_RIGHT 1 elseif cur_angle lt des_angle amp amp abs diff lt 999999 amp amp abs diff gt 2 ERB24 RIGHT 02 elseif cur_angle lt des_angle amp amp abs diff lt 199999 amp amp abs diff gt 15 ERB24 RIGHT 01 elseif cur_angle lt des_angle amp amp abs diff lt 149999 amp amp abs diff gt 1 ERB24 RIGHT 003 end end function current_posit angle_calibration analog_posit insert data into x and to get the calibration curve for rotation the x data used for the angle was taken from digital inclinometer the y data was read from the digital pmd1608fs 2 First calibration prior to connecting all info through Amphenol cord 4 1 1 05 ox 50 3 40 3 29 9 19 8 10 2 0 10 20 3 30 1 40 2 50 1 y 1 093 1 350 1 641 1 906 2 160 2 432 2 712 3 3 263 3 530 3 79 Second Calibration Data after connecting through Amphenol cord 4 1 3 05 x 49 8 40 29 9 19 95 9 8 2 10 4 20 30 2 40 3 50 41 X X 4 y 1 0
60. o 3 3 Tj Tj Create the data array to be written angle Pressures 3 Pressures 2 Pressures 1 Pressures 5 Pressures 4 Tj probe V jet Cp1 Probe Cp2 Probe Cp3 Probe At Probe As Probe Mach Freddy txt is notepad value used to hold on to data outside of the program to ensure headers will be created load freddy txt restart txt is a notepad value used to hold on to data outside of the program to ensure data will not be overwritten if the run is restarted load restart txt a freddy 1 restart create a variable that can be evaluated so that the values of lines be changed fred ExcelWrite CADocuments and Settings usen Desktop ERB24Nesting1 xls num2str a num2str 2 evaluate the variable and send the info to ExcelWrite to evaluate it and it to an EXCEL Spreadsheet 87 Change the value of freddy to next row number freddy freddy 1 Save the row number to a text file so that it will not be cleared save freddy txt freddy ascii 99 APPENDIX G SCANIVALVE CALIBRATION GRAPHS Angle Degrees AngleiD egrees AnglefDegrees THIS PAGE INTENTIONALLY LEFT BLANK 92 APPENDIX H PX 138 CALIBRATION GRAPHS 35 41 22 SS 04 N Ml 0 35 E Mach 3 0 a As Mach 85 35 25 15 5 5 15 25 35 Angle degrees 93 Angle Degrees 8
61. o Vo Vo Vo Vo Vo SUBFUNCTIONS Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So So So So YoYo So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo function m calculate Tp Tj this subfunction will get required data three times and average the three sets of values before outputting the final answer 5 5 of air Cp_air 1005 gamma of air gam_air 1 41 gamma of air gravitational constant m sec 2 g 9 81 gravitational constant m sec 2 density of Mercury kg m 3 calibration pressure in inches of mercury rho Hg213550 loop for data for i 1 3 Get the data from the 1608FS Yo Yo YoYo Yo Yo Yo YoYo Ao AoAo Vo Vo Vo Yo AoA Y Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo A AY A AAAA Get the current angle h PMD_1608FS_2 85 Convert the angle to degrees cur_angle i angle_calibration h get the pressure variable info Pressures pressure_calibration Pressures PMD 1608 5 3 Yo Yo Yo YoYo Yo Yo Yo YoYo AoAo AoAo YoY AAAY YYA AAA Vo Vo Vo Vo A A Vo Vo Vo Vo YoYo YoYo AoAo Cobra Probe pressures Yo Yo Yo YoYo Yo Yo Ao Ao AoAo Ao AoA AoAo 9 0 AoA AoAo AAAA AAA A AAA A AoA AoAo 9 5 5 5 Left Hole Pressure P2 Press
62. obe At_Probe 1 At_Probe 2 At_Probe 3 3 As Probe As Probe 1 As Probe 2 As_Probe 3 3 Mach_No Mach_No 1 Mach_No 2 Mach_No 3 3 m 1 Output the data to the GUI Yo Yo Yo Vo Yo Yo Yo YoYo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Yo Yo Yo YoYo Yo Yo Vo Vo Vo Vo Vo Vo Vo Vo Vo handles guihandles gcbo generate handles struct Set the pressure values to the gui textbox set handles P Left String P3 set handles P Center String P2 set handles P Right String P1 123 Yo Yo YoYo Yo Yo Yo Yo YoYo Yo YoYo Yo So So So Yo Vo Vo So So Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo YoYo YoYo Yo o Yo Output and calculate the data from the Three hole probe Pitch_m Yaw_m Mach_m ps_m po_m three_hole probe output m P3 P2 P1 Poj Output a second time prelacing with P2 to correct for crazy numbers near the casewall Pitch m 2 Yaw m 2 Mach m 2 ps m 2 po m 2 three hole probe output m P3 P2 P1 P2 angle is the true angle flow angle2cur angle Yaw m 2 9 the data array to be written header probe pos cur angle Pressures 3 Pressures 2 Pressures 1 Pressures 4 Pressures 5 Tp Tj q probe V jet Cp1 Probe Cp2 Probe Cp3 Probe At Probe As Probe Mach No Pitch m Yaw m Mach m Pitch m 2 Yaw m 2 Mach m
63. om the boiler will be piped through a manual shutoff valve and an orifice plate before entering the TCR test cell Inside the test cell steam will pass through a motor operated throttle control valve before mixing with atmospheric air also controlled with a motor operated throttle control valve A fast acting solenoid valve provides the capability of intermittently injecting steam into compressor inlet Figure 10 shows the interface of steam ingestion system with the TCR air inlet The inlet distortion duct drawing in Appendix C outlines the steam injection piping connection to the compressor intake system pa Flow _ nozzle Hydrauhc Y Atmos Air a LA turbines Atmas Steam 2 lt dimos Air DE Air control valve LJ Fast acting Boiler EN lt Sisam control Vale Figure 10 Steam ingestion system FREE JET TEST FACILITY The free jet test facility was used to calibrate the three hole probe A mounting bracket provided positions for securing probes and also contained a sliding bracket capable of pitch movement The free jet facility used air from the Allison Chalmers compressor to create a steady flow through a 10 8cm 4 25inch nozzle For Mach 11 numbers above 0 3 a dump valve on the compressor was throttled to adjust the flow For Mach numbers below 0 3 a manual shutoff valve located in the f
64. ool Monterey California Dr Anthony Gannon Department of Mechanical and Aeronautical Engineering Naval Postgraduate School Monterey California Naval Air Warfare Center Propulsion and Power Engineering ATTN Mark Klein Patuxent River Maryland LT Ivan Villescas Monterey California 143
65. p2 m m poj m tic s Pitch m Yaw m Mach m ps m po m hole_probe_output file type p1l_m p2_m p3_m poJ3_m Pitch_m_2 Yaw_m_2 Mach_m_2 ps_m_2 po_m_2 hole probe output file _type p1_m p2_m p3_m p2_m LOC M function file to obtain the flow direction on the input data from a 3 hole probe and mach number To assume ZERO PITCH just pass p2 m to m so that they are This will stop the program searching along this dimension To load data from the excel spreadsheets they must moved the constants file will then also be updated file type e excek or m matlab tuncrion Pitch m Yaw m Mach ps m po m three hole probe output file type pl m p2 m p3 m poj m 98 If only one input point make griddata3 work if length pl lt 2 point no 1 pl cm pl m pl ml pz m p2 m p3_m _ poj m m else point no 0 end is needed two points are required to p2_m p3_m p3_m Input matricies or vectors are reshaped to column vectors shape pl m size pl m pl m reshape pl_m 1 shape_pl_m 1 shape_pl_m 2 shape_p2_m size p2 m p2_m reshape p2_m 1 shape_p2_m 1 shape_p2_m 2 shape_p3_m size p3 m p3_m reshape p3_m 1 shape_p3_m 1 shape_p3_m 2 shape m size poj m poj m reshape m l1 shape m 1 shape m 2 Physical constants Gam gas 1 41 Mea
66. pitch A end 1 xlsreadcells 5 5 05 O0deg Pitch xls Sheetl column 111 123 O Degree pitch 1 xlsreadcells 5 10 05 10deg Pitch forward xls Sheetl column 16 10 Degree pitch Mach 0 15 Nominal A end 1 121 19129 131 UAL xlsreadcells 5_10_05_40deg_Pitch xls Sheet1 column 92 104 5 40 Degree pitch 1 x lsreadcells 5_9_05_30deg_Pitch xls Sheet1 column 15 27 5 30 Degree pitch A end 1 xlsreadcells 5_9_05_20deg_Pitch xls Sheet1 column 15 27 20 Degree pitch 1 xlsreadcells 5 9 05 10deg Pitch xls Sheeti column 16 28 5 10 Degree pitch 1 xlsreadcells 5_5_05_Odeg_Pitch xls Sheet1 column 94 106 O Degree pitch A end 1 xlsreadcells 5 10 05 10deg Pitch forward xls Sheetl column 1 10 Degree pitch 5 Mach 0 22 Nominal 1 xlsreadcells 5_10_05_40deg_Pitch xls Sheet1 column 76 88 40 Degree pitch 1 xlsreadcells 5_9_05_30deg_Pitch x1ls Sheet1 column 29 411 5 30 Degree pitch 1 xlsreadcells 5_9_05_20deg_Pitch x1ls Sheet1 column 29 41 5 20 Degree pitch 1 xlsreadcells 5 9 05 l deg Pitch xls Sheetl column 30 42 5 10 Degree pitch 103 SET 1
67. r At Calibration Surface 34 V TEST RUN PROCEDURE A OVERVIEW The L C Smith probe actuator containing the three hole probe was attached to the steel case wall of the Transonic Compressor Test runs were conducted on both upstream and downstream positions The probe was inserted into the unknown flow at various depths and data was collected at each position The process was repeated for different speeds and mass flow rates The data collected is the baseline data to determine the normal operational parameters of the rotor at various speeds and mass flow rates From this data experiments following can calculate the effects of steam ingestion on stall margins and rotor performance B TYPICAL TEST PROCEDURE The upstream position utilized hole one while the downstream position used hole four The L C Smith actuator was firmly attached to the mounting bracket on the steel case wall and the probe was controlled through Matlab programming with a remote PC as described in Chapter III For the upstream position 15 data points were taken The Matlab program used a 2 function to identify the points to sample data The cosine 1 2 function was projected on to the distance between center and case wall so that the frequency of data points increased as the probe approached the case wall The probe would begin at the center of the flow cylinder and incrementally traverse towards the case wall The increased number of data points
68. ree jet cell was activated The free jet 15 shown with the probe actuator mounted on the mounting bracket in Figure 11 Figure 11 Three hole probe within the probe actuator in the free jet assembly 12 INSTRUMENTATION PROBES The primary probe used in this experiment was the three hole cobra probe shown below in Figure 12 In addition two additional probes were used in both the calibration and experimentation phases of this project A kiel probe was used to detect the stagnation pressure and a pitot static probe was used to detect the static pressure within the free jet The probes positioned for calibration in the free jet test facility are show in Figure 13 Figure 12 Three hole cobra probe B L C SMITH PROBE ACTUATOR The L C Smith probe actuator shown in Figure 11 was used to traverse and rotate the probe for calibration in the free jet and testing in the TCR The probe actuator was initially set up in the same manner as described by Greco however the actuator was configured to an automatic PC controlled system Greco 1995 HPVEE SCANIVALVE SYSTEM The first calibration process used the Hewlett Packard Visual Engineering Environment HPVEE and Scanivalve system to obtain pressures in the cobra probe The process used a setup identical to that outlined by Grossman and screenshots of the post processing calculation program used in the current project are shown in Appendix D Grossman 1997 13
69. rement Device PMD 1608FS Data Acquisition Device was a Universal Serial Bus USB 2 0 full speed compatible analog to digital A D converter The PMD 1608FS had a dedicated A D converter for each channel and was capable of the simultaneous sampling of up to eight channels Each channel was capable of sampling at 50 kS s with a maximum of 200 kS s for all channels The PMD 1608FS was powered through the 5 volt USB port and it was controlled with a PC through the USB port with Matlab programming A detailed block diagram from the user s manual is shown in Figure 15 15 USB Full speed USB 2 0 Compliant Interface USB Microcontroller Scr vw terminal IFO connector Screw terminal VO connector 32 bit Event Counter 1 chanmel Figure 15 PMD 1608FS A D Module block diagram from Measurement Computing PMD 1608FS 2005 F USB ERB 24 REMOTE RELAY CONTROLLER A USB ERB 24 Remote Relay Controller was used to control the traverse of the L C Smith gear It was capable of being addressed by the PC via the USB 2 0 port The USB ERB 24 was an electromechanical relay module with 24 electromechanical single pole double throw relays The module was configured in two banks of eight and two banks of four relays Each relay was capable of switching 6 amps at 28VDC with a minimum close time of 10 milliseconds and a minimum open time of 5 milliseconds The module used a 9VDC power supply and was also controlled through the
70. sured constants q yaw m pl m p2 MEDS m 2 q pitch m me 1 2 mtpl m 3 Cpl m pl m p2 m q m Cp2 m pl m p3 m q yaw m Cp3 m p3 m p2 m q yaw m Side holes Cp4 m m p2 m q pitch m 5 m m p3 m q pitch m _ poj m pl m q pitch m 5 Centre hole if file type e Coefficients of pressure are read in but are calculated later from the raw data Cpl Load constanti lz 2 Load constant 13 Cp3 Load constant 14 Yaw Load constant 2 Pressure are read in pl Load Constant 4 Load constant 3 Load _constant 6 At As and Mach number are loaded in At Load constant 15 As Load_constant 16 Mach Load_constant 17 99 5 Pitches are constructed Pitch Load_constant 18 save 3 hole constants Yaw pl p2 p3 At As Mach Pitch else load 3 hole constants Yaw pl p2 p3 poj At As Mach Pitch end 5 if file type excel q coefficients _ Pla 720 Same one as is used in the calibration patch B67 1 p24Pp 31 1 73955 SAt pl yaw 5 There is some difference between the spreadsheet and these values Pitch presssure coefficients are now defined for a 4 hole probe Cpl pl p2 q yaw Cp2 pl p3 q yaw Cp3 p3 p2 q Side holes Cp4 poj p2 q pitch poj p3 q pitch
71. t device PMD 1608FS which measured the voltage output of the potentiometers on the probe actuator and was accessed by Matlab via a USB cable Control of the actuator was also provided via USB cable with Matlab accessing relays on the MC ERB 24 to rotate and traverse the probe Matlab control programming can be found in Appendix G The feedback control system and data acquisition wiring diagram 15 shown in Figure 22 25 Probe actuator 1 feedback IN OUT GREEN Y control system gt 158 2 gt Pressure data acquisition system wii T car RUIN Figure 22 Probe actuator and data acquisition wiring diagram 26 Omega Engineering PX 138 Pressure transducers were integrated into pressure lines to replace the Scanivalve system Five pressure transducers were used to provide individual pressure outputs for the three hole static and stagnation pressure probes Each pressure transducer produced an analog signal voltage which was digitized by the PMD 1608FS and converted to engineering units Pascals using a Matlab calibration program The Matlab program also calculated the probe calibration constants and exported all of the d
72. tion the Scanivalve system was replaced with five analog output pressure transducers Each pressure transducer was connected to the USB data acquisition device The data from the pressure transducers was verified to be accurate when compared to the previous data obtained with the Scanivalve In both cases the resulting calibration 23 coefficients for the pitch angles were not sufficiently discriminatory and pitch measurements were found to be unreliable hence pitch was not used in the experimental data acquisition for the three hole probe CALIBRATION PROCEDURE The probe actuator containing the three hole probe was attached to the free jet test assembly previously shown in Figure 12 The jet nozzle had a diameter of 10 8cm 4 25inches and the pitot static and stagnation probes were both placed at the same radial distance from the center of the flow 2 86cm 1 125inches The three hole probe was placed into a Mach 0 5 flow and rotated until a manometer showed equal pressures on ports 2 and 3 shown in Figure 21 The center hole was designated 1 in accordance with normal industry standards When facing into the flow the left side was designated 2 and the right side was designated 3 Figure 21 Three hole probe numbered Once the manometer showed equal pressures the pressure tubes were removed from the manometer and connected to the Scanivalve ports 5 and 7 The Scanivalve system verified ports 2 and 3 read equal pressures o
73. true then 1t is assumed that the pitch 15 zero 5 Pitch 15 calculated using the centre hole but in case this 15 outside the range the other two are used Pitch_m temp griddata3 Yaw Cp6 Mach Pitch Yaw_m temp Cp6_m temp Mach_m temp if isnan Pitch_m temp Pitch_m temp griddata3 Yaw Cpb5 Mach Pitch Yaw m temp Cp5 m temp Mach m temp end if isnan Pitch m temp Pitch m temp griddata3 Yaw Cp4 Mach Pitch Yaw m temp Cp4 m temp Mach m temp end end Yaw is corrected to the right pitch Yaw_m temp griddata3 Mach Pitch Cp3 Mach_m temp Pitch_m temp Cp3_m temp The static and stagnation pressure constants are found At_m temp griddata3 Mach Pitch Yaw At Mach_m temp Pitch_m temp Yaw_m temp As_m temp griddata3 Mach Pitch Yaw As Mach_m temp Pitch_m temp Yaw_m temp The total pressure from the constants is calculated even though for the 3 hole probe it is directly measured po m temp p2_m temp At m temp q yaw m temp The static pressure is calculated ps m temp po m temp q yaw m temp As m temp The mach number is updated 101 Mach_m temp sqrt 2 Gam gas 1 ps m temp po m temp 1 Gam gas Gam gas 1 while abs Pitch m Pitch m old tol Points are reshaped to the original input matrix if point no 1 Pitch m Pitch m 1 Yaw m Yaw m 1 Mach m Mach m 1 Poom ps_m 1 po_m pom l
74. ttusion Factor ab C ae 47 Dittusion Factor at Peak PEE EVER EU 47 Diffusion Factor Near Stall ue ei ea b ee u 48 CFD vs experimental data 90 speed peak efficiency total pressure 49 CFD vs experimental data 90 speed peak efficiency Mach number 50 CFD vs experimental data 90 speed peak efficiency flow 50 70 speed diffusion factor comparlSOnL 51 80 speed diffusion factor 22 90 speed diffusion factor comparlSOn seeesssssseeeeesssssssererssssssssreressssss 32 100 speed diffusion factor comparison sssseeeeesssssssseeennsssssseeresssssssseereee 53 LIST OF TABLES Table 1 Sanger rotor desten Parameter T Table 2 Calibration nomerclatte en ai 29 Table gt Probe Samp lino ein 36 Table 4 Rotor speed and mass flow Tales ana 37 THIS PAGE INTENTIONALLY LEFT BLANK ACKNOWLEDGMENTS Producing a thesis at the Naval Postgraduate School Turbopropulsion Laboratory was a unique experience The TPL 1s a rare nexus of industry academia and the military Garth Hobson and Ray Shreeve thank you for the guidance looking over my shoulder and keeping me on track Anthony Gannon thanks for the great ideas and occupying the chair on my left
75. umber P LEFT gt CENTER P RIGHT mcs P P Righ Figure 19 Matlab data collection graphical user interface On the data acquisition GUI when the SAMPLE DATA pushbutton was depressed the sample data run m function processed the information that was inputted in the GUI and controlled a series of traverses to each radial position for data collection At each radial position the probe would sample the flow field and process the pressure data using the constants previously calculated from the pressure calibration to determine the flow relative to the pressure probe When necessary the sample data run m function would input a new angle to the rotate m function to perform a null yaw routine The data would again be sampled and processed The raw pressure and probe actuator angle data along with processed data including the Mach number sensed and flow angle were outputted to an Excel Spreadsheet The Matlab experimental DAQ and feedback control block diagram is shown in Figure 20 20 Figure 20 Matlab experimental data acquisition and feedback control code block diagram 21 THIS PAGE INTENTIONALLY LEFT BLANK 22 IV CALIBRATION A PURPOSE The purpose of calibrating a multi hole pressure probe was so that 1t could be placed in an unknown flow field to determine the magnitude and direction of the flow In this experiment a three hole pressure probe was calibrated by inserting it into a known
76. ures 1 Center Hole Pressure P1 Pressures 2 Right Hole Pressure P3 Pressures 3 Total Pressure of jet Poj Pressures 5 sStatic Pressure of jet Psj Pressures 4 YoYo Yo YoYo YoYo YoYo Y Ao Vo Yo AoA Vo Vo Vo Vo Temperature in pipe Tp is entered manually in the GUI Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo Yo YoYo YoYo Vo A AAA YoYo YoYo YoYo Vo Vo ALCU LATION SY gt Y gt YoYo Vo Vo Vo Vo Vo Vo Vo Vo A q_probe P1 P2 P3 2 V_jet i sqrt 1 Psj Poj gam_air 1 gam_air sqrt 2 Cp_air Tp Cp1_Probe i P1 P2 q_probe Cp2_Probe i P1 P3 q_probe Cp3_Probe i P2 P3 q_probe At_Probe i P1 Poj q_probe 86 As Probe i 2q probe Poj Psj Mach Number Mach No i 2esqrt Poj Psj gam air 1 gam air 1 2 gam air 1 end Yo YoYo Yo YoYo Yo YoYo YoYo Yo YoYo Vo Vo Vo Vo Vo Vo Vo Vo Yo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo Vo Average the values Yo YoYo Yo YoYo Yo YoYo YoYo Yo YoYo Vo Vo Vo Vo Vo Vo Vo So Vo Vo Vo Vo Vo Vo Yo Vo Vo Vo Vo Vo Vo Vo Vo Vo Yo cur_angle cur_angle 1 cur_angle 2 cur_angle 3 3 Cp1_Probe Cp1_Probe 1 Cp1_Probe 2 Cp1_Probe 3 3 Cp2_Probe Cp2_Probe 1 Cp2_Probe 2 Cp2_Probe 3 3 Cp3_Probe Cp3_Probe 1 Cp3_Probe 2 Cp3_Probe 3 3 V_jet V_jet 1 V_jet 2 V_jet 3 3 At_Probe At_Probe 1 At_Probe 2 At_Probe 3 3 As Probe As Probe 1 As Probe 2 As_Probe 3 3 Mach_No Mach_No 1 Mach_No 2 Mach_N
77. usion Factor As the speed 15 increased or the mass flow rate 15 decreased the diffusion factor increased This demonstrated the increased blade loading for those conditions The general value for the diffusion factor of an axial compressor is 0 45 and the value for the diffusion factor at stall is 0 6 Cumpsty 1989 In Figure 42 which shows the diffusion factor at near stall for speeds 70 100 the diffusion factor approached 0 6 at the tip for all speeds measured The high diffusion factors at the tip indicated that the rotor would undergo tip stall as was tested and described in Measurement of the Unsteady Case Wall Pressures Gannon Hobson Shreeve 2005 48 3 Comparison of Exit Plane Data to Experimental Data The CFD data produced by the SWIFT code was compared to experimental data taken from downstream of the rotor for the equivalent pressure ratios The following graphs demonstrate the high level of accuracy of the for 90 speed At peak efficiency the CFD total pressure ratio Figure 43 exhibited nearly the same form and produced data very close to that of the experimental Both Mach number calculations Figure 44 and flow angles Figure 45 were also very similar and all three graphs are shown below However an inaccuracy was introduced during the production of the CFD data In the experimental model the hub did not rotate downstream of the rotor In the CFD grid a rotating hub was used in the model do
78. ward xls Sheetli column 52 604 1 10 Degree pitch reshape A size A 1 size A 2 1 105 THIS PAGE INTENTIONALLY LEFT BLANK 106 APPENDIX J UPSTREAM PROBE DATA STAGNATION PRESSURE 70 Rotor Speed mm e e anum m mm Ww Choke Peak Hear Stall 0 06 0 08 Probe Position m 90 Rotor Speed Stagnation press Choke Peak amp Hear Stall 0 06 0 08 Probe Position m 107 90 Rotor Speed ut ut qi eL Stagnat Choke Peak Stall 0 06 0 08 0 1 Probe Position m 100 Rotor Speed 100000 95000 90000 85000 80000 stagnation Press ure 75000 Choke Peak 0000 Hear Stall 0 06 0 08 0 1 Probe Position m 108 APPENDIX K UPSTREAM PROBE DATA YAW ANGLE DISTRIBUTION 70 Rotor Speed m Chole Peak 3 Hear Stall cn Y aw Angle 0 06 0 08 Probe Position m Choke Near Stall Yaw Angle 0 06 0 05 Probe Position m 109 90 Rotor Speed Choke He ar Stall Yaw Angle 0 06 0 08 Position m 100 Rotor Speed Yaw Angle BT 0 06 0 05 Probe Position 110 APPENDIX L DOWNSTREAM PROBE DATA STAGNATION PRESSURE RATIO 70 Speed hoke
79. wer Mach numbers The Scanivalve system produced data with more variations when the flow was below Mach 0 5 Three dimensional graphs of the Scanivalve and PX 138 calibration surfaces are shown in Figures 24 27 The calibration curves for the transducer data are visually smoother than that of Scanivalve system However in comparing the calibration surface graphs consideration must be given to the fact that the Scanivalve system was calibrated to a Mach number of 0 05 and the PX 138 Transducers were only calibrated to a Mach number of 0 066 The data was not calculated as low for the PX 138 transducers because the Scanivalve data was inconsistent below Mach 0 1 and it was not believed the transducers could be calibrated for very low flow fields as well Further calibration details including calibration calculations and variables software programs used calibration data and calibration curves are shown in Appendices 1 32 p e 2o Mach Angle Figure 24 Scanivalve As Calibration Surface es Mach pote E Angle Figure 25 138 Transducer As Calibration Surface 33 m CER T T 20 j 5 a nn e D Mach T n Fe n Angle Figure 26 Scanivalve At Calibration Surface At Mach ie rs E md Figure 27 138 Transduce
80. wnstream of the rotor producing discrepancies in the flow close to the hub This 15 more evident in the flow angle graph shown in Figure 45 Additional graphs for all settings at 90 speed can be found in appendix M 0 7 0 75 Las 0 9 0 95 1 ho Tip Ami kn Figure 43 vs experimental data 90 speed peak efficiency total pressure ratio 49 0 65 0 75 6 6 as 0 5 1 F ub to Tip Fai Asin Figure 44 vs experimental data 90 speed peak efficiency Mach number Hub ioe Tips Radhus Ratin Figure 45 CFD vs experimental data 90 speed peak efficiency flow angle 50 4 Comparison of CFD Diffusion Factor Data to Experimental Results The diffusion factor results were graphed and compared to the SWIFT code results The CFD data also predicted a tip stall Along the outer half of the blade radius the CFD and experimental data exhibited the same characteristics and obtained very close results for both near stall and peak efficiency The choke data however showed the largest discrepancy In addition in the inner half of the blade radius the experimental and CFD data exhibited different characteristics although the magnitude of the data was similar This also can be attributed to the rotating hub used in the geometric grid in the CFD calculations Traveling radially inward the experimental data decreased towards zero While the CFD data decreased by the same magnitude and then i

05Sep_Villescas - Calhoun: The NPS

Contents

Download Pdf Manuals

Related Search

Related Contents