Pressure-sensitive paint measurements on a rotor disk surface at
Contents
1. ExecuteTriggeredCapture QuitMenu KEY 0x4000 F6 gt MIPXScripts ExecuteMIPXScrptFile Name scripts checkert mpx ExecuteMIPXScriptFile Name scripts checkert mpx ExecuteMIPXScriptFile Name scripts psp_avgt mpx lt QuitMenu lt QuitMenu 35 KEY 0x5900 SF6 ScreenlHeight 30 Screen2Height 0 Screen3Height 0 SetSplitScreen lt QuitMenu gt VideoFormats SetFormat RS 1701ff14 3MHzClock lt QuitMenu lt QuitMenu 36 PSPAVGT MPX lt QuitMenu gt VideoDigitize Display CurrentlmageBuffer 1 QuitMenu gt ImageProcessing gt ImageSequenceOperations SequenceStartingBuffer 1 SequenceEndingBuffer 10 AveragelmageSequence ImageAreaofInterest Fulllmage lt QuitMenu lt QuitMenu gt MIPXScripts CHECKERT MPX lt QuitMenu gt VideoDigitize Display CurrentImageBufferl CurrentlmageBuffer2 CurrentlmageBuffer3 CurrentImageBuffer4 CurrentImageBuffer5 CurrentImageBuffer6 CurrentlmageBuffer CurrentImageBuffer8 CurrentlmageBuffer9 CurrentlmageBuffer10 lt QuitMenu gt MIPXScripts 57 APPENDIX B PSP SETUP AND PROCEDURE The following describes the current procedure for PSP measurements over a rotor disk using the PSP test bed facility The procedures assume all of the test bed elements of the test bed are available 1 Compressed Air Setup Procedures Check that all access valves are closed Check air pressure gauge gt 50 psi is required for sustained runs above 20 000 RPM Switc2. NPS ARCHIVE 1997 GAHAGAN S NAVAL POSTGRADUATE SCHOOL Monterey California THESIS PRESSURE SENSITIVE PAINT MEASUREMENTS ON A ROTOR DISK SURFACE AT HIGH SPEEDS by Shane G Gahagan June 1997 Thesis Advisor Raymond P Shreeve Thesis G126 Approved for public release distribution is unlimited DUDLEY KNOX LIBRARY NAv L POSTGRADUATE SCHOOL MONTEREY CA 939328104 REPORT DOCUMENTATION PAGE OMB No 0704 0188 Public reporting burden for this collection of information is estimated to average 1 hour per response including the time for reviewing instruction searching existing data sources gathenng 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 tne 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 June 1997 Master s Thesis am 4 TITLE AND SUBTITLE 5 FUNDING NUMBERS PRESSURE SENSITIVE PAINT MEASUREMENTS ON A ROTOR DISK SURFACE AT HIGH SPEEDS 6 AUTHOR S Gahagan Shane G 7 PERFORMING ORGANIZATION NAME S AND ADDRES
3. B C D and E Ref 1 Select Image File Load Save from main menu and save file as a TIFF format 46 file i e c 4mip images xxx tiff To save the color image select Save LUT w Image option 47 10 LIST OF REFERENCES Sievwright D L Application of Pressure Sensitive Paint in Shock Boundary Layer Interaction Experiments M S Thesis Naval Postgraduate School Monterey California March 1996 Varner D R Pressure Sensitive Paint Measurement On a Rotor M S Thesis Naval Postgraduate School Monterey California March 1995 Quinn K J Pressure Sensitive Paint Measurement Technique Development for Turbomachinery Application M S Thesis Naval Postgraduate School Monterey California Liu T Torgerson S D Sullivan J P Johnston R Fleeter S Rotor Blade Pressure Measurement in a High Speed Axial Compressor using Pressure and Temperature Sensitive Paints ALAA Paper No 97 0162 35 Aerospace Sciences Meeting amp Exhibit Reno Nevada January 6 10 1997 Schanze K S Carroll B F Korotkevitch S Morris M J Temperature Dependence of Pressure Sensitive Paints AIAA Journal Vol 35 No 2 February 1997 Woodmansee M A and Dutton J C Methods for Treating Temperature Sensitivity Effects of Pressure Sensitive Paints AIAA Paper No 97 0387 35 Aerospace Sciences Meeting amp Exhibit Reno Nevada January 6 10 1997 Schanze K S
4. CCU Oriel illumination system EPIX frame grabber software and associated hardware and software devices required to capture and store freeze frame images of a pressure event generated by a shock generator on the rim of a disk moving at supersonic speeds Varner and Quinn Refs 2 and 3 provided detailed descriptions of the test facility and PSP test bed set up The following provides a brief description of the PSP test bed facility components and the modifications made during the course of the present work A ROTOR DISK A 12 inch diameter aluminium rotor disk designed for uniform stress and machined at NPS was used as the PSP experimental platform Figure 2 shows the rotor disk dimensions and Figure 3 shows the rotor as it was mounted on the turbine driven shaft High centrifugal stress loading was the prime design consideration and led to the rotor s tapered profile Two inch stainless steel Allen set screws were machined and set 180 apart at a radial distance of 5 8 inches from the rotor center The set screws extended 3 32 inches above the rotor face were intended to act as shock generators at l 650 nm XYBION ISG 350 Hamilton Standard ORIEL Z 77 Carrera Fuel Pump Turbine aaa w X Uno an a p a a n a a 5 BE osse 1 SERIE ss WaveTech Function Generator p Wind Off Waveform PIN 3CD Photodiode 28V Rev Bias Input Red LED 4V Rev Input i 1 28V DC Power Wind Off Trigge
5. 255 Pixel Value Figure 13 Pixel Frequency vs Intensity Ratio 20 000 RPM 400 588 686 X Coordinate Figure 14 Pixel Value vs Single Line of Pixels Above Set Screw 20 000 RPM 28 Value Pixel 200 300 400 500 X Coordinate Figure 15 Pixel Value vs Single Line of Pixels Below Set Screw 20 000 RPM Frequency wu Fu v M a ADA mn a N L JP y eo WM 1 k m M NE UNE AS ay a na im 4 z T m 68 80 188 128 148 160 188 3200 226 Pixel Value Figure 16 Pixel Frequency vs Intensity Ratio 23 400 RPM Run 2 29 7 o lt U 7 U 5 bis 48 68 88 188 128 148 168 188 200 220 Pixel Ualue Figure 17 Pixel Frequency vs Intensity Ratio 23 400 RPM Run 3 30 VI CONCLUSIONS AND RECOMMENDATIONS A high speed rotor test facility using pressure sensitive paint to obtain surface pressure distributions has been developed and operated successfully Additionally focused PSP image intensity ratio maps were produced at subsonic and supersonic speeds At supersonic wheel speeds paint stripping around the leading edge of the shock generator occurred A more adhesive paint was required for operation at supersonic speeds The facility itself was shown to be a suitable test bed for developing quantitative PSP measurements before application on a transonic compressor rig To continue the development of the PSP measurement technique for supersonic speeds and transonic rotors th
6. distributions in accordance with the Stern Volmer relation Paint stripping and temperature dependence prevented the measurement of pressure at transonic speeds The test bed facility and acquisition techniques developed here could now be used to overcome those limitations 14 SUBJECT TERMS l 15 NUMBER OF Pressure Sensitive Paint PSP Rotor Measurements Photoluminescence Imaging PAGES 61 16 PRICE CODE 20 LIMITATION OF ABSTRACT 18 SECURITY CLASSIFICATION OF THIS PAGE Unclassified Unclassified Unclassified UL 17 SECURITY CLASSIFICATION OF REPORT 19 SECURITY CLASSIFI CATION OF ABSTRACT NSN 7540 01 280 5500 Standard Form 298 Rev 2 89 Prescribed by ANSI Std 239 18 Approved for public release distribution is unlimited PRESSURE SENSITIVE PAINT MEASUREMENTS ON A ROTOR DISK SURFACE AT HIGH SPEEDS Shane G Gahagan Lieutenant Commander United States Navy B S North Carolina State University 1986 Submitted n partial fulfillment of the requirements for the degree of MASTER OF SCIENCE IN AERONAUTICAL ENGINEERING from the NAVAL POSTGRADUATE SCHOOL June 1997 EIO EV UAIOY PIRRARY Ol A L DA A R Ager TH ON Jj sl Pi ABSTRACT y Measurement of the static pressure distribution over the surface of a rotor disk was attempted using pressure sensitive paint PSP A uniform stress high speed rotor disk fitted with a shock generator was built installed and operated at speeds in exce
7. ne lis N ead le DAL e CE re ee risar niaa 15 5 Ga EV BIS WES IS c ee 18 V RESULTS AND DISCUSSION BE m E 19 Pie OAC TUSIONS AND RECOMMENDATIONS ruaen eonun 31 APREMIDI yea IPS P SS ge pe m A 33 EM BE ESE SETUR AND PROCEDURES eec roo eine eene 39 E OIE e DE RENO Ser ee noe CNN 49 INTA LE ON 5n vil I INTRODUCTION Recent developments in the use of luminescent coatings on turbomachinery blading as a means for measuring spatial distributions of blade surface pressure and temperature has generated significant interest Current methods for obtaining surface pressure measurements require the installation of individual pressure ports along each aerodynamic surface Although highly accurate the information is spatially discrete and is usually not cost effective for rotor blade analysis due to the complex array of pressure lines that must be installed and led out of a pressure instrumented model Refs 1 10 Pressure sensitive paints PSP however offer a conceptually simple means for determining continuous blade surface pressure distnbutions Such information is critical in the validation of new rotor designs that are based on computational fluid dynamic CFD calculations Originally applied to high pressure gradient external flows supersonic and transonic the wide application of PSP s as a quantitative measurement tool has been slowed as a result of the temperature dependence of the a
8. the center 5 Image Focusing Disconnect image inhibit BNC connector on back of CCU this step is required since an erroneously high setting of image inhibit signal is possible in the non powered state 41 Disconnect BNC Vin from CCU to CP and connect to Video In Line A on back of Sony the on Set trigger source on CCU to VD set camera gain to max set intensity to max and set range to either 100us or higher to see real time image to avoid intensifier damage start range at a lower setting and adjust higher later until image is viewed Increase line voltage to Oriel lamp to 118 Only A RGB position on front of Sony monitor should be illuminated to view real time image Focus image using camera focus and move camera position to center image on monitor focal length of camera can be changed by adjusting position of the entire lens assembly refer to Xybion reference manual before adjusting Once image is centered and properly focused turn off Oriel lamp and reinstall image inhibit BNC connector Wavetek pulse function generator that supplies the image inhibit pulse should be off before connecting image inhibit connector to CCU Connect Vin input to computer and select L RGB on the Sony monitor front panel both A RGB and L RGB should be illuminated Set trigger source on CCU to EXT and select proper range gate range gate will vary depending camera exposures ti
9. wheel to be operated remotely after execution was triggered at the keyboard The scripts developed for image acquisition are given in Appendix A After wind on images were captured averaged and saved as a TIFF file the process was repeated for wind off images Wind off images were captured after wind on images were obtained in an attempt to ensure that the surface temperature of the rotor was as near as possible to those during the wind on tests To remove the thermal noise generated by the camera dark current images were obtained and subtracted from both the wind on and wind off images The two images were first aligned and then the ratio wind off wind on was computed and multiplied by a scaling factor to produce a final image corresponding to the left hand side of the Stern Volmer relation I I This final 0 255 grey scale image represented the inverse pressure ratio of wind on to wind off in accordance with equation 1 A histogram showing frequency of individual pixel intensity values was inspected in order to select the range of minimum intensity to maximum intensity for further processing The pseudo coloring technique reported by Seivweight Ref 1 was applied to obtain the final image Appendix B details the methodology developed for the PSP set up and procedure 17 B PROGRAM OF TESTS In previous studies rotational speeds were limited by available power Ref 2 or by rotational stresses and lubrication Ref 3 In the present s
10. Carroll B F Korotkevitch S Morris M J Concerning the Temperature Dependence of Pressure Sensitive Paints ALAA Paper No 97 0386 35 Aerospace Sciences Meeting amp Exhibit Reno Nevada January 6 10 1997 Schairer E T and Hand L A Pressure Sensitive Paint Measurements on Aeroelasstic Wings in Transonic Flow AIAA Paper No 97 0389 35 Aerospace Sciences Meeting amp Exhibit Reno Nevada January 6 10 1997 Shimbo Y Mehta R D Cantwell B J Vortical Flow Field Investigation Using The Pressure Sensitive Paint Technique at Low Speed AIAA Paper No 97 0388 35 Aerospace Sciences Meeting amp Exhibit Reno Nevada January 6 10 1997 Peterson J I and Fitzgerald R V New Technique of Surface Flow Visualization Based on Oxygen Quenching of Fluoresence Rev Sci Instruments Vol 51 pp 670 671 May 1980 49 11 Hamilton Standard Technical Manual Turbine Driven Fuel Pump Model No TPC 13 15 April 1968 12 Oriel Instruction Manual Quartz Tungsten Halogen Lamp Housings Models 66186 Through 66188 1993 13 Oriel Instruction Manual Lamp Controller Model 6405 1992 14 Xybion Electronic Systems User s Manual Electronically Gated Intensified CCD Video Camera Model ISG 350 May 1992 15 Xybion Electronic Systems User s Manual Video Camera Control Unit Model CCU 01 September 1990 16 Epix User s Manual 4MIP 4MIPTOOL Interactive Image Analysis Vers
11. S ES 8 PERFORMING Naval Postgraduate School Monterey CA 93943 5000 ORGANIZATION REPORT NUMBER 9 SPONSORING MONITORING AGENCY NAME S AND ADDRESS ES 10 SPONSORING MONITORING AGENCY REPORT NUMBER 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 12b DISTRIBUTION CODE Approved for public release distribution unlimited i 13 ABSTRACT maximum 200 words Measurement of the static pressure distribution over the surface of a rotor disk was attempted using pressure sensitive paint PSP A uniform stress high speed rotor disk fitted with a shock generator was built installed and operated at speeds in excess of 20 000 RPM by a Hamilton Standard turbine driven fuel pump A once per revolution trigger signal was converted to a transistor to transistor logic TTL format and used to gate an intensified charged coupled device CCD video camera Multiple low intensity level camera exposures were integrated and captured to produce a single usable image Ten captured images were averaged to increase the image s signal to noise ratio and the result was used to produce an image ratio with respect to a static reference condition Finally a pseudo coloring process was used to develop a color image that related intensities to both temperature and pressure
12. TEST pause 1 lt QuitMenu gt SpecialOperations amp Modes gt TnggerlmageCapture gt 6 WaitNfields N 2 lt QuitMenu lt QuitMenu gt MotionSequenceCapture Display gt TnggerOptions Trigger ExternalInput DelayedbyNfields N 1 QuitMenu QuitMemu gt VideoFormats gt CustomInterfaceModes ExternalPixelClock QuitMenu QuitMenu OKEY 0x5900 SF6 ScreenlHeight 10 AScreen2Heisht 9 Screen3Height 9 SetSplitScreen lt QuitMenu gt VideoDigitize Display CurrentlmageBuffer 1 KEY 0x4000 F6 gt SpecialOperations amp Modes gt TriggeredlmageCapture ExecuteTriggeredCapture KEY 0x4000 F6 1 KEY 0x4000 F6 CurrentImageBuffer2 1 KEY 0x4000 F6 ExecuteTriggeredCapture KEY 0x4000 F6 KEY 0x4000 F6 CurrentImageBuffer3 1 KEY 0x4000 F6 34 ExecuteTriggeredCapture KEY 0x4000 F6 KEY 0x4000 F6 CurrentImageBuffer4 KEY 0x4000 F6 ExecuteTnggeredCapture KEY 0x4000 F6 D KEY 0x4000 F6 CurrentimageBuffer5 1 KEY 0x4000 F6 ExecuteTriggeredCapture KEY 0x4000 F6 KEY 0x4000 F6 CurrentImageBuffer6 KEY 0x4000 F6 ExecuteTriggeredCapture KEY 0x4000 F6 KEY 0x4000 F6 CurrentImageBuffer7 KEY 0x4000 F6 ExecuteTnggeredCapture KEY 0x4000 F6 KEY 0x4000 F6 CurrentImageBuffer8 KEY 0x4000 F6 ExecuteTriggeredCapture KEY 0x4000 F6 KEY 0x4000 F6 CurrentImageBuffer9 KEY 0x4000 F6 ExecuteTriggeredCapture KEY 0x4000 F6 UU KEY 0x4000 F6 CurrentImageBuffer10 KEY 0x4000 F6
13. ctive probe molecule and associated chemical binder PSP s temperature dependencies have been cited as the reason for errors in measuring correct surface pressure when compared with conventional pressure taps and probe measurements Attempts to calibrate these temperature effects and to incorporate the results to produce comparable surface pressure measurements with conventional methods is the subject of on going research The present study continues the work done by Sievwright Varner and Quinn at the Naval Postgraduate School NPS to develop the means and capability to produce quantitative temperature corrected PSP measurements of surface pressure distributions on high speed rotors Refs 1 3 The PSP test bed built by Varner Ref 2 and further developed by Quinn Ref 3 was used A 12 inch diameter aluminum uniform stress rotor disk with small shock generators inserted near the rim was designed built installed and operated at transonic tip speeds 220 000 RPM PSP was applied locally in the area where significant surface pressure gradients as a result of the generation of shock waves would exist Images measuring local luminescent intensities were captured using EPIX 4MIP frame grabber software The image intensity distribution over the PSP was then interpreted to determine variations in surface pressure and surface temperature using the Stern Volmer relationship The following section briefly describes the concept and applicatio
14. destination PixB Change value for ratio coefficient c0 to produce a usable image values 45 between 20 80 will produce adequate image brightness Large values of cO will produce pixel intensity values of 255 and produce image blooming Image buffer 2 will contain the final image ratio Select Image File Load Save from main menu and save file as a TIFF format file i e c 4mip images xxx tiff 12 Psuedo coloring of Image Ratio Select Load Save Image from the main menu and load final image into image buffer 1 From Image Measurements menu select Histograms Displays Select Histogram Horizontal Linear amp Logarithmic to display histogram of Pixel Frequency vs Pixel Value Determine the range of pixel values in which 90 of the data are located The pixel value range determines the min max values for image coloring From Sievwright Ref 1 determine values for A B C D and E according the pixel value range For example an image where 90 of the data are located between pixel values 20 and 80 A and B respectively the median value is 50 corresponding to position C while the values for D and E can be varied but usually have values closer to the median such as 40 and 60 respectively Select Contrast and Lookup Tables from main menu Select Numerically Set amp Show Red Table and enter each segment of the Red Green and Blue tables corresponding to the values chosen for A
15. e following are recommended 1 Install more responsive LED photodiodes to provide RPM counter readings at higher rotation speeds Alternatively enlarge the l rev hole to increase the amount of light passed to the detector 2 Investigate paint and application techniques that provide good PSP adherence at supersonic rim wheel speeds 3 Investigate methods for separating pressure and temperature effects quantitatively 4 In 2 and 3 develop techniques that are suitable for use on the transonic compressor rig where surfaces are curved access is more limited and the rotor can not be removed readily between tests 31 APPENDIX A PSP SCRIPTS Executable 4MIP TOOL scripts developed by Quinn Ref 2 were modified to provide hands off and remote image capture An executable pause was incorporated into the AVG TEST script to provide time to start up the drive turbine and for the RPM to stabilize before image acquisition Additionally the command Trigger External Input was activated during the image acquisition period to synchronise CCU and EPIX software frame grabber board data transfer and receive rates Synchronisation was reverted back to the RS170 14 3MHz rate at image acquisition completion Script AVG TEST provided the complete image acquisition executable file CHECKERT MPX provided a quick feedback showing the 10 images captured and PSP AVGT MPX averaged the 10 images to produce a single captured image 33 AVG
16. h MOV ON Open line to charge manifold Charge lines by opening access valves 2 3 turns Apply power to oil cooling pump adjust setting for oil pressure of 38 psi and control valve set Y turn Open cooling water valve 1 turn Ensure remote activation set up 2 Rotor Disk Attachment To Turbine Driven Fuel Pump Use 3 32 allen wrench to slide LED and photodiode back into attaching plate before attaching rotor disk this clears area to avoid contact with rotor disk preventing dents scratches etc 39 Slide rotor disk onto dowels on turbine drive flange while avoiding contact with LED mounting bracket Dowels are different diameters to prevent improper attachment of rotor disk Press wheel fully onto turbine drive flange and insert attachment bolts all bolts are similar size and tighten using 7 16 wrench Safety wire all four bolts with 032 diameter wire 3 Krylon Paint Application Align rotor disk with shock generator at 12 o clock position 3 32 transparent hole will be aligned with LED and photodiode Clean rotor disk with acetone to remove greasy film and or contamination before applying paint Let dry for 1 2 minutes Attach cardboard mask for paint application LED and photodiode should be completely covered prior to painting Apply a thin layer 1 2 coats of glossy Interior Exterior Krylon paint 1501 Avoid paint drips and or runs along rotor disk Let dry for approximately 30 45 minu
17. h plate used by Quinn and more nearly uniform temperatures would be expected Results in Figure 13 indicated that the significant pixel value range in Run 1 extended from approximately 50 150 across the disk surface Results in Figures 14 and 15 show a slightly negative slope of surface intensity ratio from leading to trailing edge of the image implying a decreasing pressure and or temperature The psuedo colored image of the intensity ratio for Run 2 shown in Figure 9 illustrates the paint stripping effect that occurred at higher rotation speeds Where the paint came off is consistent with the occurrence of a detached and possibly unsteady shock A shock positioned in front of the shock generator would interact with the boundary layer on the disk surface The gray scale image from Run 2 in Figure 11 shows the exact location where the paint came off Intensity ratio distributions for Run 2 showed a high to low intensity gradient horizontally across the set screw and a nearly uniform intensity distribution below and behind the shock generator This intensity ratio distribution would be inconsistent with respect to the pressure distribution from a bow shock located in front of the set screw Possible temperature effects similar to those in Run 1 can be seen in front of the area where the paint was stripped A histogram showing 21 the frequency number of pixels vs pixel value intensity of the image ratio I I is shown in Figure 16 The p
18. hing and less radiation emittance from the PSP The oxygen quenching process has been modelled by the Stern Volmer relation A T BOE 1 where I and P are the reference luminescent intensity and reference pressure respectively and I and P are measured intensity and pressure at the experimental conditions The coefficients A and B are derived from calibration data and generally are temperature dependent In the present study the reference intensity and pressures were measured at ambient static conditions e g wind off and experimental conditions wind on were measured using a phase locked acquisition technique at various rotor speeds The PSP used in the present work was composed of Platinum Octaethyl Porphyrin PtEOP as the active molecule dissolved in an oxygen permeable binder GP 197 Ref 10 The active molecule was excited at an absorption peak of 380nm and photoluminescence occurred at 650 nm corresponding to the colors of violet and red respectively in the visible light spectrum HI PSP MEASUREMENT APPARATUS A PSP test bed facility at the Naval Postgraduate School s Gas Dynamics Laboratory was used for the present study A schematic of the facility and kac acquisition system is shown in Figure 1 An aluminium rotor disk mounted on a turbine drive for a fuel pump was enclosed in an open ended cylindrical steel spin chamber The image acquisition system consisted of a RPM counter CCD camera camera control unit
19. ide a reference intensity 43 condition for the wind on image captured above For the present work wind off images are referenced to static rotor conditions Proper initial alignment of the rotor disk will facilitate proper post processing of images execute image focusing procedures in step3 above through image focusing step rotate rotor disk until set screw is centered around black marker on video monitor Complete step 3 Disconnect l rev trigger from rotor disk and connect second Wavetek pulse function generator Set pulse function generator frequency to the identical value as wind on frequency to simulate wheel rotation Increase line control voltage to Oriel lamp to 118 and select script AVGF MPX execution file File is identical to AVG MPX with the exception of an execution time delay Averaged image will be stored in image buffer 1 Select Image File Load Save from main menu save file as a TIFF format file i e c 4mip images xxx tiff 10 Dark Current Image Capture The dark current image captures the internal thermal noise of camera system Dark current intensities are subtracted from wind on and wind off images to increase signal to noise ratios Repeat procedure for wind off image capture but with the lens cover attached to the camera to isolate noise intensities Select Image File Load Save from main menu and save file as a TIFF format 44 file i e c 4mip images xxx tiff 11 Post Proce
20. ier gain were controlled manually from the CCU 11 A 300 mV 1 REV Trigger AAA A A A EE EA B 4V Amplified 1 REV Trigger e 9V Active Low 1 REV D SV High TIL 1 REV E 5V Trigger Pulse Je Signal 5V Active High Monostable Pulse G Image Inhibit 5V Active High Monostable Pulse H SV Low TTL Signal Figure 6 Image Capture Timing Sequence Source Quinn K J Pressure Sensitive Paint Measurement Technique Development for Turbomachinery Application M S Thesis Naval Postgraduate School Monterey California liz 3 Once Per Revolution Trigger The image trigger was supplied by a light emitting diode LED illuminating a photodiode The photodiode when illuminated once per revolution triggered a 50 millivolt pulse that was amplified and converter to 5 volt TTL active low format required by the imaging circuitry Ref 2 4 Imaging System A high low transition of the 5 volt TTL signal was used to expose the camera intensifier and image the rotor disk PSP area Multiple exposures were integrated only during periods when the image inhibit function was activated 5 VDC and were captured to a single image when the function transitioned from high to low Activation and duration of the image inhibit function was controlled by a Wavetek function generator and separate resistor capacitor RC time constants respectively Ref 2 Additionally the duration of the camera exposure was set u
21. ion 2 8 1993 50 Defense Technical Information Center INITIAL DISTRIBUTION LIST 9929526069259206060609090099092896942060000200260900020090099 022959 292908299 499999999 929222929 9 8725 John J Kingman Road Ste 0944 Ft Belvoir VA 22060 OOUC O a A A Naval Postgraduate School 411 Dyer Rd Monterey CA 93943 5101 A bannen nn nee Department of Aeronautics and Astronautics Code AA Co Naval Postgraduate School 699 Dyer Road Room 137 Monterey CA 93943 5106 oi so Ki Je N T A tn Department of Aeronautics and Astronautics Code AA SF Naval Postgraduate School 699 Dyer Road Room 137 Monterey CA 93943 5106 kemere GL ekl eke ge NN genuineness Department of Aeronautics and Astronautics Code AA Hg Naval Postgraduate School 699 Dyer Road Room 137 Monterey CA 93943 5106 CUA CUENCA Naval Postgraduate School Monterey CA 93943 5002 Contes a Naval Air Systems Command Code Aur 4 4T 1421 Jefferson Davis Hwy Arlington VA 22243 51 8 Naval Air Warfare Center Aircraft Division ee Code Air 4 4 3 1 S McAdams Propulsion and Power Engineering Bldg 106 Patuxent River MD 20670 5304 9 LCDR ShaneGahagan nete UTILI D DD 7209 Madiera Ct Raleigh NC 27615 32 DUDLEY KNOX LIBRARY HAY POSTGRADUATE SCHOOL MONTEREY CA 6294323104
22. mber 0 9 a shock would occur just behind the maximum diameter of the shock generator as the airflow velocity is first increased to supersonic then decreased to subsonic Mach number conditions The resultant intensity distribution in Figure 8 if interpreted to represent the surface pressure distribution is not consistent with what would be expected for transonic flow over the shock generator The plotted intensity ratio distribution should show a low high variation across a shock No such transition is evident Varner Ref 2 discusses the viscous pumping effect of a rotating disk in which air is sucked along the centerline axis and pumped radially outward along the surface of the disk Qualitatively with respect to Figure 8 stagnation from the pumping effect would not be consistent with the red yellow areas indicating lower pressure along the inner portion of the shock generator However since a portion of the set screw threads were slightly 20 exposed above the disk surface stagnation areas could occur which would be consistent with the dark blue areas along the leading semi circle of the set screw An alternate explanation for the intensity ratio distribution radially along the disk surface is that it results from the increased stagnation temperature as wheel velocity increases with radius A strong temperature gradient effect was reported by Quinn Ref 3 however the present disk had a much higher thermal capacity than the 0 25 inc
23. mes required 6 Software Set up The main EPIX program script is called AVG MPX and incorporates LOOK and AVERAGE scripts AVG MPX which incorporates a execution delay time is used to 42 capture images for wind on conditions AVGF MPX is developed for wind off and dark current image capture Software script can only be used for EPIX 4MIP but can be modified for use with 5MIP version Open EPIX 4MIP program and select the MIPX Scripts option from the main menu enter file name c 4mip scripts avg mpx in the execute MIPX script file prompt Use a stop watch to monitor elapsed time Set Onel lamp control voltage to 118 to illuminate rotor disk and turn off lights when exiting room and close all access doors 7 Rotation of Rotor Disk From remote activation area slowly turn control valve CW for air supply to turbine CAUTION Avoid over speeding rotor disk with excess air supply Monitor RPM and stabilize at desired reading Throughout image capture execution monitor oil temperature RPM counter and proper software execution Remove air supply after capture sequence is complete and once rotor disk is at a safe rotation speed 80 RPM enter test bed facility and zero out control voltage to Onel lamp 8 Saving Wind on Image Select Image File Load Save from main menu save file as a TIFF format file i e c 4mip images xxx tiff 9 Wind off Image Capture The reference images wind off are used to prov
24. n of pressure sensitive paints and the governing relationships between pressure and luminescence Section III gives a brief description of the image acquisition systems developed to capture pressure distribution images en IV describes the test methodology while Sections V and VI discusses the results and conclusions and provide recommendations for further analysis II THEORY OF PRESSURE SENSITIVE PAINT PSP uses a photoluminescence and oxygen quenching process by which the molecular energy state of an active molecule excited by radiation from its ground energy state S to a higher energy state S I gt 2 returns to its original state through a series of naturally occurring processes A molecular internal conversion process instantaneously 10 107 secs converts the molecule to its lowest excited state S The molecule is then returned to its ground state through the emission of light or via a radiationless process Oxygen quenching is one radiationless process in which the transfer of energy is the result of molecular interactions collisions with an oxygen molecule As a result the partial pressure of oxygen that is diffused in a PSP paint layer medium is inversely proportional to the amount of emitted radiation Ref 1 and 11 Since the mole fraction of oxygen in air is 0 21 as the air pressure over the paint surface is increased more oxygen molecules are diffused and interact with the PSP resulting in increased oxygen quenc
25. psi air supply was used to operate the turbine However only 10 20 psi air was needed to operate the turbine at speeds in excess of 25 000 rpm Ref 11 Turbine bearing lubrication was provided by a Brown and Sharpe oil pump providing 40 psi gauge pressure with the oil returned to the reservoir via a separate scavenge pump Scavenged oil was cooled via a heat exchanger before flowing into the reservoir The lubrication system is shown in Figure 4 The temperature of the oil exiting from the turbine an indication of the turbine bearing operating temperature was toral throughout the run cycle Details of the fuel pump are given in Reference 11 Figure 4 Lubrication System C IMAGE CAPTURE AND PROCESSING ASSEMBLY 1 Rotor Illumination System The rotor was illuminated through an open hole in the center of the cover plate of the spin chamber using an Oriel 1000 Watt quartz tungsten halogen lamp Model 66187 with a F 0 7 condensing lens assembly Ref 13 The lamp was fitted with an Oriel blue gel and interference filters 66228 and 575 respectively which provided illumination with the wavelength centered at 380nm The lamp was supplied with variable AC voltage via an Oriel lamp controller Model 6405 M The control voltage was adjusted to 11896 of line voltage to ensure ample illumination uniformly across the disk surface Control voltage to the lamp was turned off between image capture periods to preserve the luminescent characteris
26. r Pulse Waveshaper SV DC mum Power IIITTITIITEITIIIITTIIII CH TTL Loaic Gate S Sy o o to EXT TRIGGER 9 to IMAGE INHIBIT oO __ Voltage Monostable Pulse Gate Divider Inverter SV DC Camera Trigger to EXT IN to VIDEO IN to PIXEL CLOCK 14 318 MHZ SONY Trinitron 120 MHz 586 PC Video Monitor 1 REV Trigger Amplified 1 REV Trigger 9V Active Low 1 REV 5V High TTL 1 REV 5V Trigger Pulse 9 5V Active High Monostable Pulses 00 00000 SV Low TTL Signal Figure 1 Image Acquisition System Source Quinn K J Pressure Sensitive Paint Measurement Technique Development for Turbomachinery Application M S Thesis Naval Postgraduate School Monterey California 6 yore Ae RADIUS Bas OR GREATER No reece MAKKIS See TY RORE 0 500 PIR Figure 2 Rotor Disk Dimensions Figure 3 Rotor Mounted on Turbine Driven Fuel Pump supersonic rotor speeds Two 1 8 inch holes were located 5 8 inches radially from the rotor center One hole was used with an optical system as a RPM pick up and the other was partially filled with a non transparent epoxy The rotor assembly turbine drive and rotor disk was dynamically balanced as a single unit before experimental runs in the transonic region were attempted B TURBINE DRIVEN FUEL PUMP The rotor was driven by the turbine of a Hamilton Standard fuel pump Model TPC 13 mounted and enclosed in the spin chamber An 8000 cu ft 300
27. rred in the local area surrounding the shock generator In an attempt to prevent the paint from coming off in Run 3 PSP was applied to the rotor disk without the Krylon undercoat While the test showed that sufficient intensity could be accumulated without using the Krylon the PSP again came off around the shock generator Increasing blade surface roughness using a bead blasting technique before applying PSP only was attempted in Run 4 In Run 4 as well in all experimental runs over Mach 1 0 a loss of paint occurred in the local area surrounding the shock generator in Rd aa ll Table 2 Rotor Disk Configuration 19 Pseudo colored images of the intensity ratio 1 1 for Runs 1 through 3 are shown in Figures 8 10 respectively A grey scale wind off image of Run 2 is shown in Figure 11 Figure 12 shows a grey scale wind off image of Run 4 The pseudo colored image of the intensity ratio 1 1 for Run 1 shown in Figure 8 clearly shows a non uniform intensity distribution varying from red lowest intensity ratio lowest pressure to blue highest intensity ratio highest pressure A histogram showing the frequency number of pixels vs pixel value intensity of the image ratio 1 1 multiplied by a factor of 50 is shown in Figure 13 Pixel value plots for a horizontal line 10 pixels above and below the set screw are shown in Figures 14 and 15 respectively In a two dimensional uniform flow at a speed of 1047 ft sec Mach nu
28. seudo colored image of the intensity ratio I I for Run 3 for which PSP was applied without Krylon is shown in Figure 10 The intensity ratio distribution is seen to be similar to that of Run 2 with the exception that the paint was stripped above the set screw The paint stripping above the set screw was not repeated in two similar experimental runs Runs at similar conditions to Run 3 produced paint stripping only in front of the set screw similar to Run 2 A histogram showing the frequency number of pixels vs pixel value intensity of the image ratio 1 1 for Run 3 is shown in Figure 17 Finally the result of the attempt made to keep the paint from stripping by roughing the surface using the bead blasting technique is shown in Figure 12 Paint stripping occurred in the local area in front of the set screw similar to Runs 2 and 3 22 23 Figure 8 Image Intensity Ratio 1 1 for 20 000 RPM Z UnA Wide 00r ec 103 1D obey Ay susyuj odeum 6 NG p Yd E pos UR a EN Tm NA DM NV Y e 24 e opor a f y Ku a 25 Figure 10 Image Intensity Ratio 1 1 for 23 400 RPM Run 3 Figure 11 256 Grey Scale Intensity Ratio 1 D for 23 400 RPM Run 2 26 p UNU WAN 00r E7 103 ID oney Ay susyuf AVIS 4919 967 ZI ANSIA Zu 6000 34 5888 yc nl gt z 4060 3888 GAY eee Dow ww m Frequency 2808 1008 8 20 40 eo 88 188 128 140 168 188 288 220
29. sing the CCU An exposure time of 500 nanoseconds was required to adequately focus and freeze the moving image A trial and error procedure was used to adjust the intensity of the captured image at different wheel speeds As the exposure time required for focusing was decreased longer integration periods were required to be set using the image inhibit function Images were captured using an EPIX 4MEG Video Model 12 integrated circuit board and EPIX 4MIP V3 2 software installed in a 120MHz Pentium personal computer Ref 15 Software scripts were developed so that the image acquisition was completed autonomously Ten 13 images were captured each for wind off wind on and dark current image These images were then processed by taking the ratio between them to develop a single image 14 IV EXPERIMENTAL PROGRAM A PROCEDURES The PSP test methodology was similar to that discussed in references 1 and 10 In order to produce significant static pressure variations on the plate surface tip Mach numbers of the order of 0 9 1015 2 ft sec and above were required The rim area surrounding the shock generator was painted with an initial coating of glossy white interior exterior Krylon paint PSP was air bushed onto the surface after the Krylon paint had dned Subsequently experimental runs with PSP only applied to the rotor disk were also attempted Figure 7 shows the painted areas of the rotor disk Both paint applications were done with the ro
30. ss of 20 000 RPM by a Hamilton Standard turbine driven fuel pump A once per revolution trigger signal was converted to a transistor to transistor logic TTL format and used to gate an intensified charged coupled device CCD video camera Multiple low intensity level camera exposures were integrated and captured to produce a single usable image Ten captured images were averaged to increase the image s signal to noise ratio and the result was used to produce an image ratio with respect to a static reference condition Finally a pseudo coloring process was used to develop a color image that related intensities to both temperature and pressure distributions in accordance with the Stern Volmer relation Paint stripping and temperature dependence prevented the measurement of pressure at transonic speeds The test bed facility and acquisition techniques developed here could now be used to overcome those limitations u des TABLE OF CONTENTS I INI DA MCT e T 1 MTT HEORY OF PRESSURE SENSITIVE PAINT nete nnne en 3 IES ME ASUREMENTAPRARATOS nce ees erue rre rre sees se eed oge ede ee 5 w RO ROJR e RM DD DTE 5 E IU EDE SEGE POMP netter ees oce ree nro nnne es ed ee ede 8 C TMAGE CAPTURE AND PROCESSING ASSEMBLY 9 EG Gr MED TE A ELO DES ACME 9 2 KERE eyy tO CC CO TINTE 10 S OME A LU IL E 15 s MAGII SAGA ROZ 13 DORE RE INIENTACE PROGRAMIE am ede sesse gee se oe se see susleme
31. ssing Processing of the captured images results in a single image of the ratio of wind off to wind on images To increase the signal to noise ratio of the images thermal noise generated within the acquisition system is subtracted from wind on and wind off images Select Load Save Image from the main menu and load wind on wind off and direct images into image buffers 1 2 and 3 respectively Select Image Processing from main menu and then Two Image Arithmetic select Subtract Images P xB lt Abs PixB PixA from menu options The Subtract Images operation forms a single image from the difference of the corresponding pixels of two original images PixA is the source image dark current and PixB is the image to be replaced wind on and wind off images Select image buffer 3 as the source Pix A and select image 1 as the destination image buffer Repeat procedure with image 2 as the destination image After image subtraction is complete image buffer 1 and 2 should contain the noise free images of the wind on and wind off images respectively The final image ratio of wind off image with respect to wind on image forms the left hand side of the Stern Volmer relation shown in Equation 1 Select Image Processing from main menu and then Two Image Arithmetic select Ratio Images PixB lt c0 PixB cl c2 PixAt c3 where image buffer 1 wind on is the source PixA and image buffer 2 wind off is the
32. tes or until dry 4 PSP Application Clean airbrush and PSP paint container thoroughly with acetone and let dry this avoids contamination of PSP before application REN airbrush hose to N bottle and set pressure to approximately 40 psi check for proper operation Half fill PSP container and attach to airbrush 40 LIMIT PSP EXPOSURE TO LIGHT AS MUCH AS POSSIBLE TO AVOID DEGRADATION OF THE PAINT Spray PSP 8 10 inches from rotor disk surface in a sweeping motion again avoid any drips runs or excessive concentration areas of paint Apply paint until a smooth even finish is evident approximately 1 2 minutes of airbrush time Turn lights off in the room and let dry for approximately 10 minutes Remove cardboard mask and check area for smooth paint application Excessive paint application will cause weight imbalances on the rotor disk Adjust LED and photodiode to as close as 1 16 inch trial and error procedure is required to obtain proper RPM pick up 1 16 inch is too large for speeds of 30 000 RPM Hand spin wheel to ensure proper clearance and that the wheel rotates unhindered Attach spin chamber cover plate and lock down with washer and bolts Slide Oriel light source into position markings on floor provide most effective lamp positioning for uniform and sufficient illumination over rotor disk Position camera using markings on the floor Turn on LED and rotate disk until 3 32 hole is located in
33. tics of the PSP The lamp and camera set up are shown in Figure 5 Camera access was through a separate hole in the cover plate Figure 5 Camera Lamp and Rotor Disk Housing 2 Camera System Rotor disk images were acquired using a Xybion ISG 350 externally triggered and gated CCD camera A remote Xybion Camera Control Unit CCU was used to control the camera s tnggering and gating functions through a 23 pin multipurpose connector Ref 14 The camera was fitted with a 75mm f1 4 Cosmicar television lens and an Oriel interference filter 453590 to limit the camera s spectral response to the desired 650 nm wavelength The PSP illumination system and camera set up on the spin chamber are shown in Figure 5 A timing diagram for the camera control 10 is shown in Figure 6 A 5 volt external trigger pulse D was used to trigger camera exposure while the range gate function of the CCU determined the length of exposure The exposure period was varied depending on the rotor disk RPM and was set so that the captured image was focused Multiple exposures were obtained to form a single image through the use of an externally applied image inhibit transistor transistor logic TTL low signal The image inhibit TTL high low transition allowed the integrated exposures to be captured as a single image Similar to the procedure for setting exposure duration times image inhibit signal periods were based on the rotor disk RPM Camera and intensif
34. tor installed on the turbine This procedure reduced the Figure 7 Rotor Disk Paint Area 15 PSP degradation that results from exposure to room light Floor markings for the camera set up were used to ensure repeatability Minor adjustments of the camera position were used to center the shock generator in the image The camera focal adjustments were used to optimize the image clarity Ref 14 Camera gate durations were set to optimize the image focus and then the image inhibit signal time was adjusted to maintain adequate image intensity Since rotor wheel speed was maintained at 1 Hz gate duration time was calculated to limit rotor travel to approximately 007 inches n order to prevent blurred images Ref 2 Table 1 shows the gate duration times used for the reported run conditions Image Inhibit Number of Gate Signal Exposures Duration secs nsecs 20000 333 104720 033 80 2668 360 33 Table 1 Gate Duration Times for Rotor Travel of 007 inches Since gate duration times could not be set exactly a trial and error procedure was used to vary the image inhibit and gate duration times to optimize image intensity and focus Table 1 shows the length of the image inhibit signal used with two different gate durations and the corresponding number of integrated exposures necessary to produce a usable image 16 An image acquisition script was written with a time delay included to allow the
35. tudy the construction of a new high speed rotor was intended to extend the test capability to 30 000 RPM After mounting the rotor to the drive turbine which was fitted to the shaft with dowels the unit was dynamically balanced installed in the spin chamber and operated at progressively increased speeds Using the lubncation system described by Quinn Ref 3 rotational speeds were found to be limited to about 20 000 RPM When the hydraulic pump was replaced with a combination lubrication supply and scavenge pump system oil temperatures again reached 175 F after minutes of operation A shell and tube water supplied heat exchanger was installed in the oil return line and was found to keep oil temperatures steady below 135 F In order to extend testing above 20 000 RPM the acquisition software was modified so that the operator could move to a protected control room and remotely operate the drive turbine A series of tests was then conducted using PSP at increasing rotor speeds Results for 23 400 RPM are described and discussed in the following section Higher speeds were not attempted when it was found that the present paint could not withstand supersonic tip speeds and that the RPM indication was lost 18 V RESULTS AND DISCUSSION The rotor disk configuration for each in a series of four tests is shown in Table 2 Complete images were acquired successfully in Run 1 In Run 2 at rotor tip speeds above Mach 1 0 paint stripping occu
Download Pdf Manuals
Related Search