NASA LaRC FIB Multi-Channel Anemometry Recording System
Contents
1. Anders 12a DISTRIBUTION AVAILABILITY STATEMENT 12b DISTRIBUTION CODE Unclassified Unlimited Subject Category 34 Distribution Nonstandard Availability NASA CASI 301 621 0390 13 ABSTRACT Maximum 200 words This report is part of a series of reports describing a flow physics high lift experiment conducted in NASA Langley Research Center s Low Turbulence Pressure Tunnel LTPT in 1996 The anemometry system used in the experi ment was originally designed for and used in flight tests with NASA s Boeing 737 airplane Information that may be useful in the evaluation or use of the experimental data has been compiled The report also contains details regarding record structure how to read the embedded time code as well as the output file formats used in the code reading the binary data 14 SUBJECT TERMS 15 NUMBER OF PAGES High lift Boundary layer transition Hot film anemometer High Reynolds numbers 41 16 PRICE CODE 17 SECURITY CLASSIFICATION 18 SECURITY CLASSIFICATION 19 SECURITY CLASSIFICATION 20 LIMITATION OF REPORT OF THIS PAGE OF ABSTRACT OF ABSTRACT Unclassified Unclassified Unclassified UL NSN 7540 01 280 5500 Standard Form 298 Rev 2 89 Prescribed by ANSI Std Z39 18 298 1022. 2 12 Sampling Groups Data Retrieval and Conversions 2 21 Retrieval Procedures 2 2 1 1 GMT Time Specifications 2 2 1 2 Header Time Adjustment SUN Computer Data Extraction Procedures Decoding Embedded Time 2 2 3 1 Seconds from Base Time 2 2 3 2 Bit Orientation 2 2 3 3 Decoding Example NN NN WN Analysis Software Program EXABYTEREAD version 1 0 3 1 1 Running the Code Verification DC Offset Conversions SUN Identifier Timecode Analysis Sync Verification of Embedded Timecode Buffer Time Analysis 4 5 1 Buffer Time Calculations O1 11 11 12 12 12 12 12 12 12 13 13 15 15 15 19 17 17 17 22 22 22 22 23 23 23 General The anemometry system was developed to provide high frequency hot film data from a large number of surface mounted hot film sensors in flight The system hardware was built with a pallet going into the LaRC Boeing 737 aircraft cargo bay and a workstation mounted in the B 737 cabin The system is flight hardware built and flight hardened to LaRC Flight Specifications per LHB 7910 The hardware is modular and may be used in ground facilities it is tracked and used as a complete system The purpose of the present description is to provide enough information regarding the hardware and associated systems used to allow a potential user of the database to assess the entire system or parts of it as well as to perform data analysis This documentat
3. It is always tab delimited and does not have reference information since it is suited for spreadsheets Linel inserted every dataset Column1 m record number m 1 is the first record in a file Column2 iss dataset number 1 through 8 Column3 irecsun amp record identifier counter in the Sun and the only acceptable way of knowing that records in systems A and B are synchronized Wraps Column 4 recordPos amp location in the file of the first byte of the dataset Then follows kmax lines with the signal data in counts Note that the data is inverted in this file see manual System A Column 1 7 1 sampled hot film data counts for channel i i 1 7 Column 8 8 sampled data counts embedded group information Last line inserted every dataset Column1 m record number m 1 is the first record in a file Column2 iss dataset number 1 through 8 Column3 irecsun amp record identifier counter in the Sun and the only acceptable way of knowing that records in systems A and B are synchronized Wraps Column 4 recordPos amp location in the file of the first byte of the record Column5 mins time in minutes from embedded timecode Column6 Secs msecs time in seconds and milliseconds from embedded time System B Column 1 7 1 sampled hot film data counts for channel i i 1 7 Column 8 sa 8 sampled data counts embedded group information Last line1 inserted every dataset Column1 m record number m
4. Write to NASA STI Help Desk NASA Center for AeroSpace Information 7121 Standard Drive Hanover MD 21076 1320 NASA CR 2002 211440 NASA LaRC FIB Multi Channel Anemometry Recording System User s Manual Compiled by Sherylene Johnson NYMA Inc Hampton Virginia Arild Bertelrud Analytical Services amp Materials Inc Hampton Virginia National Aeronautics and Space Administration Langley Research Center Prepared for Langley Research Center Hampton Virginia 23681 2199 under Contract NAS1 96014 ee ee February 2002 Available from NASA Center for AeroSpace Information CASI National Technical Information Service NTIS 7121 Standard Drive 5285 Port Royal Road Hanover MD 21076 1320 Springfield VA 22161 2171 301 621 0390 703 605 6000 Table of Contents General System Instrumentation 11 1 2 1 3 14 1 5 1 6 1 7 1 8 1 9 1 10 Hot Film Sensors Anemometer Multiplexers Anemometers Dantec DISA Filters DASM Data Acquisition System Management Port Expander Time Code Generator D A Clock Generator System Computer 1 8 4 Flight System Start up Procedures 1 8 2 Flight System Power down procedures Analog Back up System System Modifications for LTPT Exabyte Data 2 1 2 2 Data 3 1 Data 4 1 4 2 4 3 4 4 4 5 Record Structure 2 11 Header 2 1 1 1 DC Offsets 2 1 1 2 Record Counter 2 1 1 3 Channel 8 Identification 2 1 1 4 Gain 2 1 1 5 System A System B Bank Number
5. and corner frequency were kept unchanged 10 2 Exabyte Data In this section the structure of the digital data is given followed by the data retrieval procedures 2 1 Record Structure See Figures 2 3 and 4 The digital data is organized into records Figure 4 shows the overall anemometer data format Figure 5 the record structure and Figure 6 the record header definition Each record contains a Header and 8 Sampling Groups for a total of 131 072 bytes of data 2 11 Header See Figure 3 and 4 The Header consists of 512 bytes The first 4 bytes are dedicated to the SyncWord FrameSync which is FAF3 20B3 and the next 8 bytes represent time in seconds and microseconds The following 98 bytes are the DC offsets for the sensors starting with System A Bank 1 Channels 1 7 then System B Bank 1 Channels 1 7 System B follows System A through each Bank change see Anemometer Header Layout After 2 pad bytes the Record counter Channel 8 identification Gain and System A and System B Bank numbers follow and the remaining 392 bytes are pad bytes The Header is updated with every record but all of its contents are not updated for every record the update rate depends on the parameter 2 1 1 1 DC offsets The traditional way of recording fluctuating signals where high resolution is needed is to high pass filter the signal at some appropriate corner frequency and amplify the remaining AC component In the present case a different solut
6. by the SUN computer and Port Expander The system selection is dependent on the Anemometer Multiplexer Control Cards which are identical in both systems and differ only in the connection of the jumper hardware Since each group bank has a unique analog voltage which is output through the Anemometer Switching Control Card to the DASM Channel 8 of System B which when analyzed provides embedded bank identification in the data If a large S N signal to noise ratio is desired bypassing the multiplexer may be a consideration The DASM is also a multiplexer See section 1 5 for definition 1 3 Anemometers Dantec DISA See Block Diagram Figure 1 The Dantec units are responsible for maintaining constant sensor temperature only active sensors are heated This is achieved by an internal bridge and servo amplifier that monitors the sensor resistance There are a total of fourteen Dantec DISA units between System A and System B each assigned seven channels which controls multiple sensors switched by one of the two multiplexers For detailed information on the units see Instructional Manuals Dantec 56C17 CTA Bridge and 56B10 56B12 Main record The bridge has a 1 20 bridge ratio with a top resistance of 400 Ohms Maximum bandwidth is 150 kHz with 5 meter cable In the LTPT experiment excessive lead resistance led to modifications in the bridge circuit see section 1 10 The power supply limits the current output from each System A or B
7. to 191 mA for each channel and it is necessary to ensure that the total heat loss conduction to the substrate and convection through air flow from the sensors used stays below this limit The entire system provides slightly in excess of 1 33 Amps and if more than 191 mA per channel is needed fewer than 7 channels can be used 1 4 Filters See Block Diagram Figure 1 For the digital system there is no AC coupling i e the signal contains the DC voltage see 1 9 regarding the backup analog system description The analog low pass filters are of type Maxim Max 274 with capability of providing Chebyshev Bessel or Butterworth filtering programmed by external resistance Pole Frequency range is 100 Hz to 150 kHz The low pass filters currently being used are of 8th order continuous time analog Chebyshev filters providing anti aliasing bandwidth reduction There are two filter units one for System A and one for System B The system has a sampling rate of 50kHz and the filters provide a cut off frequency of 20kHz at the 3dB points The filter cards also have an inverting amplifier with a gain of two Note that the DASM reference 1 5 amp 2 1 2 4 has a computer activated gain control in addition to the filter card gain 1 5 DASM Data Acquisition System Management See Block Diagram Each SCSI based DASM AD14 has an eight channel multiplexer followed by a single input 14 bit A D analog to digital converter The DASM provides a 50kH
8. 1 is the first record in a file Column2 iss dataset number 1 through 8 27 Column3 irecsun amp record identifier counter in the Sun and the only acceptable way of knowing that records in systems A and B are synchronized Wraps Column 4 recordPos amp location in the file of the first byte of the dataset Column 5 ibankA group system A from embedded information Column 6 ibankB group system B from embedded information lt input filename gt cf This file contains the autocorrelation function and the crosscorrelation function determined if kmax _ gt 64 samples The autocorrelation function is based on an asymmetric window of 20 samples gt 0 The crosscorrelation function is evaluated from a symmetrical in window 20 samples wide Column 1 m record number Column 2 iss dataset number Column 3 k sample number Column 4 1 msec tau in msec Column 5 11 ar ar 1 k through 7 k Column 12 18 cf cf 1 k through 7 k lt input filename gt e amp w is an error and warnings file that may or may not be empty Bank indication has changed Timecode sync not found Timecodesync followed by one or two more sync like values 28 APPENDIX 2 The zero offset values are converted to voltage through use of the following calibrations Zero offset Volts DAshift DAslope Zero offset counts Zero offset counts is identical to ha i found in the stat file see Appendix 1 Note that the channels should be ide
9. ME 3 BUFFER RECORDS 528 535 RECORD 528 START TIME 33 840 RECORD 535 END TIME 36 430 36 430 33 840 2 59 BUFFER TIME RECORD 536 START TIME 37 920 33 840 2 59 1 49 BUFFER DELAY TIME RECORD 528 END TIME 34 140 34 140 33 840 3 RECORD TIME RECORD 535 START TIME 36 150 36 430 36 150 28 RECORD TIME 24 APPENDIX 1 EXABYTE OUTPUT FILES As described in Chapter 3 the program generates three output files plus two more signal files if the signal option has been chose All files are in text format The files are lt input filename gt stat A or lt input filename gt stat B if system A or B has been indicated Since the name of an should be 710 _ A or B will indicate if the file has been appropriately interpreted The data is comma delimited default or tab delimited suitable for spreadsheet Runmode information is included to describe how the output file was obtained Line 1 filenameout the filename created with the file Line 2 mstart the first record read in the input datafile Line 3 mmax the last record read in the input file Line 4 kmax number of samples per dataset Header information The following header information printed out initially and then each time the zero offsets is changing i e almost never Column 1 m record number in file Column 2 recordPos amp location in file of first byte of record m starting from zero incrementing by 131 072 pe
10. NASA CR 2002 211440 NASA LaRC FIB Multi Channel Anemometry Recording System User s Manual Compiled by Sherylene Johnson NYMA Inc Hampton Virginia Arild Bertelrud Analytical Services amp Materials Inc Hampton Virginia aa ee February 2002 The NASA STI Program Office in Profile Since its founding NASA has been dedicated to the advancement of aeronautics and space science The NASA Scientific and Technical Information STI Program Office plays a key part in helping NASA maintain this important role The NASA STI Program Office is operated by Langley Research Center the lead center for NASA s scientific and technical information The NASA STI Program Office provides access to the NASA STI Database the largest collection of aeronautical and space science STI in the world The Program Office is also NASA s institutional mechanism for disseminating the results of its research and development activities These results are published by NASA in the NASA STI Report Series which includes the following report types TECHNICAL PUBLICATION Reports of completed research or a major significant phase of research that present the results of NASA programs and include extensive data or theoretical analysis Includes compilations of significant scientific and technical data and information deemed to be of continuing reference value NASA counterpart of peer reviewed formal professional papers but having l
11. Volts to counts A E Ez NAD 5 8192 E Ez NAD A 5 8192 Ez E NAD A 5 8192 Ez E NAD 2 5 8192 Ez E NAD 5 16384 4 2 SUN Identifier The SUN Identifier SUN Record Number is produced by the in flight SUN computer and proves synchronization of Systems A and B In order to verify the identifier flight files 710A aa and 710A ba were used along with the EXABYTEREAD program The same settings and record numbers were requested for both systems when running the EXABYTEREAD program The files produced by the EXABYTEREAD program were then called up The file record numbers column 1 and SUN identifier column 3 were the same for both systems thereby proving synchronization 4 3 Timecode Analysis The Timecode recognition limit of 50 counts from the base count of 2730 to be high A counts limit of 20 counts would be sufficient corresponding to a noise level of 6 mvolts in the analog time code signal 22 4 4 Sync Verification of Embedded Time Code A sync verification k some value is provided in the EXABYTEREAD program as a quick identification of records within a buffer When running the EXABYTEREAD program this k value appears directly below the record number and will remain the same for eight records one buffer The k value may not always change to a different value between buffers If the value remains the same an alternate method of determining records within a buffer should be uti
12. cel etc these files can be used to verify the accuracy of data such as timecode channel identification synchronization of the two systems and buffer time The following output files are created all in ASCII format Always stat Provides statistical information from the input file including zero offset voltages expressed as an integer 0 to 255 The statistics includes average deviation from the zero offset standard deviation skewness flatness auto correlation time cross correlation time with the neighboring channel and the value of this correlation 0 to 1 e amp w Includes errors and warnings detected during the execution of the code volts Provides the mean voltages and standard deviation in mV i e including zero offsets Optional E Time series of each of the 7 channels from the particular system analyzed time tagged cf Auto correlation and cross correlation functions describing the functional shape corresponding to the information included in the stat file 311 Running the Code In order to run EXABYTEREAD version1 0 the anemometer flight file must be located on the drive reference 2 2 1 of the computer being used After selecting the software the title page will appear hit return to begin The following selections should be answered based on the type of data requested the italicized print represents computer prompts Input Anemometer system 1 System 2 17 System A selection provides
13. channels 1 7 and time and System B selection provides channels 8 14 and bank identification After the system selection has been made the flight file must be opened A dialog box appears allowing input file selection select file corresponding to system selection Open The program will then list the number of bytes and the number of records in the file Input Desired Run Parameters Graphics Desired 0 1 Yes 0 Defaults to No Graphics Normally no graphics are requested unless visual aids are required see page 25 for graphics request Enhanced graph representations can be obtained by using a graphing software program ex Cricket Graph or Excel Stat file 12comma delimited 2 tab delimited default comma Tab delimited is preferred for use with spreadsheets Comma delimited is often suitable if the file is used as input to a BASIC code Signal output O No 1 0 Defaults to No Signal Signal output may be desirable for frequency analysis but it can create large signal output files Start record no mstart 1 The record start number is limited only by the number of records in the selected file The first record is always number one but the reading may start at any record number in the file End record no mmax 20 Any number after and including the start record number in the selected file Output signal from records msigstart to msigend please input mstart mmax values Defaults to record se
14. data Selection A return Change Group 1 2 or 3 return Enter Bank 1 through 7 return 11 Upper left window on screen hilift uppercase L Lytflight return 12 Lower right window on screen hilift acq4 return Go To BLUE MENU of the DaDiSP for program selection To lift icons Go to outer edge of each icon you wish to remove and click center mouse button Go to center of screen click right mouse holding while moving down to back Release mouse 1 8 2 Flight System Power down Procedures ANEMOMETER POWER DOWN PROCEDURE Go to lower left window click left mouse to activate Hit the ESC key on keyboard At this time the right lower window will begin shut down of the acquisition program When the hilft prompt returns in lower right window meaning acquisition and data storage have completed shut down Go to title bar of screen press right mouse pad down move to quit and release Move to upper left window click right mouse pad down holding and moving cursor to quit and release window should close and disappear from screen Move cursor to lower left window click right mouse hold and move to exit release Move cursor to center of screen move cursor to exit click left mouse right lower will close at left lower window command When hilft appears type sync sync halt Program terminated will appear on screen Turn off circuit breakers in the following order 2 3 4 6 7 10 11 14 15 18 19 13 amp 16 1 9 Ana
15. ed from the last record requested has been processed and the graphics has not been requested the execution stops with FINISHED Input choice O finish 1 another run Defaults to finish If another run is chosen using the same datafile the output files from the previous run will be overwritten To eliminate loss of files change output file names prior to additional runs If graphics are requested Input channels to plot ichstart ichstop 20 Any or all of the seven channels may be selected Enter start channel comma stop channel The graph will appear with legend x and y scales and x and y shift Both scales can be changed by enter the following xscale enter scale number change or yscale enter scale number change To exit graphic enter xscale enter any negative number Input selection enter any negative number Signal output is not required to run the graphics portion of this program 21 4 Data Verification This chapter contains an account of how various parts of the data were analyzed to verify the functionality and validity of the procedures 4 1 DC Offset Conversions The DC offset conversions create calibration curves which allows determination of the bridge voltage for further analysis The following formula describes the process E known voltage used Ez zero offset voltage unknown A filter gain 2 NAD known counts after A D conversion 5 8192
16. ensors The sensors deteriorated over time with their electrical resistance increasing In the data reduction this has to be taken into account for quantitative analysis If only dynamic analysis is performed there is no need for corrections except to be aware of the general problems related to the effects of overheat on dynamic response There is no problem operating the anemometry system with ordinary hot wires or films of different resistance geometric layout etc The main problem arises due to the multiplexing which sets limitations on the tolerable resistance variation for the sensor elements 1 2 Anemometer Multiplexers See Block Diagram Figure 1 A total of 98 sensors can be hooked up to the system at any one time Since there are only 14 anemometry channels the prearranged sensors are operated through multiplexers where the user can select which of the two groups of 7 channels each are being used There are two Anemometer Multiplexers one for the System A hot film sensors and one for the System B hot film sensors Forty nine sensors are input to each multiplexer via RG 188 coax cables which are grouped into seven banks of seven sensors The choice of group or bank is done through use of a six bit character the three LSB least significant bits of which control the seven banks of System A and the three MSB most significant bits control the seven banks of System B The character activates one of the systems when transmitted
17. ert 8mm Exabyte tape into tape drive Go to a cmdtool window type dos return key Select directory C Search and select PCFILE From PCFILE select pcf Select Time File strstpa follow commands in program and insert times requested Copy C pcfile strstpa dta to E home dasmad14 hilft xx strstp dta Type quit to get out of dos Go to cmdtool and type cd home dasmad14 hilft prgms there is a space between cd and the first slash To start the tape read type tape_read2 14 data record Insert directory new file will be stored and new file name ex dsk1 btest_709 To check available memory on directories cd dsk1 When work has been completed move the mouse to the blue screen and press the right button and slide down to Utilities then to Lock Screen and release 2 2 3 Decoding Embedded Time The retrieved data can be verified by viewing the file in hex and decoding the embedded time This time should be the same as the time requested in the strstpa file 2 2 3 1 Seconds from Base Time The base time was formulated from January 1 1970 through January 1 1994 24 yrs 365 days yr 24 hr day 3600 sec hr 756 864 000sec Leap Years Addition 6 days 24 hr day 3600 hr 518 400sec Total Seconds from 1 1 70 1 1 94 756 864 000sec 518 400sec 757 382 400 2 2 3 2 Bit Orientation First Byte of Record Byte 1 Time in Seconds Bytes 5 6 7 amp 8 Time in Micro Seconds By
18. ess stringent limitations on manuscript length and extent of graphic presentations TECHNICAL MEMORANDUM Scientific and technical findings that are preliminary or of specialized interest e g quick release reports working papers and bibliographies that contain minimal annotation Does not contain extensive analysis CONTRACTOR REPORT Scientific and technical findings by NASA sponsored contractors and grantees CONFERENCE PUBLICATION Collected papers from scientific and technical conferences symposia seminars or other meetings sponsored or co sponsored by NASA SPECIAL PUBLICATION Scientific technical or historical information from NASA programs projects and missions often concerned with subjects having substantial public interest TECHNICAL TRANSLATION English language translations of foreign scientific and technical material pertinent to NASA s mission Specialized services that complement the STI Program Office s diverse offerings include creating custom thesauri building customized databases organizing and publishing research results even providing videos For more information about the NASA STI Program Office see the following Access the NASA STI Program Home Page at http www sti nasa gov Email your question via the Internet to help sti nasa gov Fax your question to the NASA STI Help Desk at 301 621 0134 Telephone the NASA STI Help Desk at 301 621 0390
19. ion is not meant to be an exhaustive description and the details regarding hardware and software for engineering analysis have been filed in separate volumes This description is organized in the sequence of the signal flow starting from the sensor end going through the signal conditioning data acquisition data storage and subsequent data retrieval The system consisted of the following hardware built into two physical entities A pallet with anemometers and analog multiplexing A control rack with signal conditioning A D conversion a SUN workstation and two digital Exabyte tape drives for data storage plus a 14 channel analog FM tape recorder Anemometers There were two seven channel Dantec Constant Temperature Anemometers without temperature compensation providing a total of 14 simultaneous channels The maximum current per channel is 191 mA and maximum hot probe resistance lead plus sensor is 30 Ohms It is necessary to match lead lengths 5 or 20 meters for inductance and total resistance due to the analog multiplexing described below The anemometry was completely controlled by software on a SUN workstation and physically 98 channels 2 systems x 7 channels x 7 groups are available through analog multiplexing If more channels are required the anemometry may be switched to standby and the connectors 4 changed Traditionally transition and turbulence data were obtained through use of high pass filtering AC coup
20. ion was chosen utilizing individual zero offsets for each anemometer channel This has the advantage that the entire signal can be reconstructed In a separate control sequence the amplitude of this offset is determined in the following manner The system is cycled through all 98 films and the average bridge voltages for each of the 14 channels are determined using 8 bit resolution When the anemometer system is operated in run mode the system senses which film groups are present and 14 D A circuits provide stable filtered voltages These voltages are subtracted from the signal allowing subsequent amplification of the remaining portion 11 The digital value for the zero offset thus introduced is stored in the header for all 98 channels Only 14 different values exist since the zero offsets are associated with the average of the bridge voltage for each anemometer channel 2 1 1 2 Record Counter A sequential numeric method of tracking data records and synchronization of records between the two exabyte recorders This is generated by the SUN computer 2 1 1 3 Channel 8 Identification Depending on which system is being recorded this channel can represent either time or bank identification When System A is being recorded Channel 8 represents time but when System B is being recorded Channel 8 represents Bank ID see Figure 5 and 6 2 1 1 4 Gain This is a DASM programmable computer set gain By setting the gain a
21. ions If GMT Zulu hours are less than four 04 example 00 26 53 one 1 must be added to the Julian date when entering Start Stop times EXAMPLE Time Requested 5 13 94 00 26 53 00 26 55 Julian Date for 5 13 94 133 Start Stop Time 133 00 26 53 133 00 26 55 Hours Requested 04 Time Adjustment Add one 1 to Julian Date 133 1 134 New Start Stop 134 00 26 53 134 00 26 55 2 2 1 2 Header Time Adjustment The Header Time which is used to retrieve data and changes every second buffer is a time tag of when the data was stored to the exabyte not when the data was taken Because there can be a difference in the storage and data times subtraction of eight 8 seconds from the start time is required to ensure that the requested time is received EXAMPLE Time Requested 5 13 94 09 26 53 09 26 55 Start Stop Time 133 09 26 53 133 09 26 55 8 sec Adjustment 09 26 53 00 00 08 09 26 45 New Start Stop 133 09 26 45 133 09 26 55 Do not add one to the Julian date unless the hours requested is less than four 2 2 2 SUN Computer Data Extraction Procedures This procedure begins by logging on to the computer and the steps for extracting data and storing it on the SUN computer NOTE Underlined print computer prompts Italicized print operator responses Plain print instructional notes Start of Computer procedures 13 LOG IN root PASSWORD Will be given upon request To open windows type Openwin Ins
22. itable for frequency analysis The data are checked for consistency and synchronization between the two Exabyte tapes and the time code on channel 8 is read to verify when the time data were actually taken not when it was recorded Output from the EXABYTEREAD code is in the form of ASCII files that use counts or volts for the dynamic data and volts for the time averaged data The statistical information is rms standard deviation skewness and flatness as well as autocorrelations and cross correlations between neighboring sensors The statistics can be based on anywhere from 16 to 1024 data samples per data set but there is a compromise between processing speed and accuracy The output files may either be tab delimited or comma delimited depending on the desired post processing In the past customized output file layouts have been provided for quick dissemination of results A graphics capability exists allowing signal examination but the capabilities have been limited to keep the code as general as possible A custom interface exists for wind tunnel work This consists of a simple graphics display of output statistics as a function of sensor location and allows indication of the transition location through examination of time segments of moderate lengths after each tunnel run The time segment length is determined by the time needed to switch through the sensors The system covers the entire process hot film anemometry from sensor to s
23. lection given mstart and mmax If signal output was selected a start and end dataset will be needed Data sets per record isets 1 Defaults to one data set 18 There are eight data sets per record and isets determines how many sets from each record that is desired Output signal from dataset isetstart please input 1 Defaults to first data set Input the first desired data set of each record selected for signal output If the number of datasets requested and the first set selected would mean reading past the eighth data set the number of datasets is reduced Example if isets 5 and isetstart has been input as 6 the code will start with set 5 and only provide 4 datasets Samples kmax 16 Defaults to 16 samples For statistical analysis a larger kmax should be chosen kmax 16 to 1024 but the run time increases dramatically OK To continue select return Selection of any other key will return to graphics prompt This starts the execution of the program and the Results window provides the information on some of the statistics data set by data set The first line is information taken from the header Another window illustrates the data selected from each record and after the first record has been processed the approximate processing time remaining is displayed m recordPos amp Counter Sys A bank Sys B bank zero offset m Record number starting at 1 at the beginning of the file recordPos amp The
24. ling in addition to anti aliasing filtering low pass In the current system both the time averaged and the time dependent hot film signals are made accessible for transition detection through an analog zero offset system that removes the main portion of the time averaged signal This is done through running the anemometry in a zero offset mode in wind off conditions The system measures the time averaged voltage converts it to an 8 bit integer which in turn determines a compensation voltage for each channel The compensated voltage can then be amplified to the appropriate level SUN control This value is part of the information available in the recorded frames of data and thus the true voltage at any instance can be recreated without the low frequency cutoff The sampling rate is currently 50 kHz per channel with a 14 bit A D with sample and hold filtering is done with an 8 pole Chebyshev at 20 kHz If needed this can be changed e g to Bessel filters for true signal recording and the corner frequency may also be adapted The binary data are recorded in parallel on two digital Exabyte recorders 8 channels per recorder on 8 mm tape with a total capacity of 2 5 or 5 gigabytes Channels 1 through 7 are anemometry the eighth is time code on system A and group identification on system B Each record on both systems contains a record counter value given by the SUN computer that ensures synchronization between the two tapes The time c
25. lized Another method of identifying a buffer change is to look for a change in the header time see stat files Appendix 1 but this is not a completely satisfactory since the header time changes every two buffers The time code is read described in Figure 1 starting with two hex sync words 0AAA repeating every 16 samples e g if the first sync word is found at k 4 the next should be at k 20 but this is not displayed in the exabyte output 4 5 Buffer Time Analysis The EXABYTEREAD program and System A files were used to verify record times buffer times and delay times between buffers The total buffer time should be approximately 2 62 seconds with an approximate 1 5 second delay between buffers See attached calculations and Figure 10 4 5 1 Buffer Time Calculations 1 BUFFER RECORDS 136 143 RECORD 136 START TIME 15 810 RECORD 143 END TIME 18 40 18 40 15 810 2 59 BUFFER TIME RECORD 144 START TIME 19 92 15 810 2 5999 1 52 BUFFER DELAY TIME RECORD 136 END TIME 16 14 16 14 15 810 33 RECORD TIME RECORD 143 START TIME 18 11 18 40 18 11 29 RECORD TIME 2 BUFFER RECORDS 256 263 23 RECORD 256 START TIME 16 530 RECORD 263 END TIME 19 110 19 110 16 530 2 58 BUFFER TIME RECORD 264 START TIME 20 570 16 530 2 58 1 46 BUFFER DELAY TIME RECORD 256 END TIME 16 850 16 850 16 530 32 RECORD TIME RECORD 263 START TIME 18 810 19 110 18 810 30 RECORD TI
26. log Back up System The purpose of the analog back up system was to have a confirmation of the functionality of the digital system but it was added as a separate system on a non interference basis The signals were multiplexed immediately after the output from the anemometers AC coupled and subsequently amplified a factor of 5 6 Recording was done using FM technique standard WB I Wide Band I i e 30 ips tape speed 20 kHz frequency response At any time the 14 channel tape contains Timecode PCM Bank Group selection and 11 channels of analog data 1 10 System Modification for LTPT The modifications to the system were kept to a minimum The flight system utilized a time code generator on board the airplane For the Low Turbulence Pressure Tunnel LTPT high lift flow physics experiment a Time Code Generator was added to the system and its output was simultaneously provided both to the hot film system and LTPT s Modcomp computer that was used to handle reference and pressure data Since the anemometry system hardware had to be located outside the pressure shell of LTPT 60 ft long coax cables fed through the pressure shell had to be used Due to the high lead resistance of the leads on the model the anemometry bridge was modified Three sets of connectors were used to accommodate a total of 294 sensors at a time A separate power supply was needed to provide 28 Volts instead of 400 Hz 3 phase Data acquisition mode filter type
27. nd if the DC component is nulled by the DC offset control the remaining portion of the anemometer signal is magnified preventing out of range voltages in the A D converter The gain can be set as one or five 2 1 1 5 System A System B Bank Number Intended to identify which Bank is in use during recording This is not a reliable source for bank identification because the bank can change within the buffer at which point the computer would not be able to update the header until the next buffer Also note that the embedded bank information may be contaminated by analog noise 212 Sampling Groups See Figure 7 The first seven 7 groups or datasets contain 1024 samples or 16 384 bytes of data taken over approximately 20 msec The data is written in words two bytes long using the two s complement format Due to the initial 512 bytes used for the Header the last group group eight 8 contains 992 samples or 15 872 bytes of data The total samples per record equals 8 160 or 131 072 bytes of data Header plus Anemometer data 2 2 Data Retrieval and Conversions 2 21 Retrieval Procedures The data is recorded on 8mm Exabyte tapes using the SUN computer The date and time of recording are necessary to retrieve the data from these tapes The data can best be retrieved using blocks of start stop times such as 2 10 minute increments The programs required to extract this data are available on the SUN computer 12 2 2 1 1 GMT Time Specificat
28. ntified using the header map of Figure 6 System Channel DAshift DAslope A 1 0 060679 0 024139 A 2 0 13823 0 024144 A 3 0 055841 0 024126 A 4 0 050429 0 024135 A 5 0 043334 0 024729 A 6 0 0049609 0 024711 A 7 0 075968 0 024720 B 1 0 015992 0 024678 B 2 0 069607 0 024656 B 3 0 16136 0 024672 B 4 0 066739 0 024650 B 5 0 0047358 0 024772 B 6 0 088540 0 024750 B 7 0 022924 0 024740 Based on calibrations May 1995 29 Figure 1 Block diagram of the anemometer system 30 Bn 1FFTIS a B CHANNELS a IR DALE TE Figure 2 Record structure for systems A and B 31 VILIPOFIIN VH WOM LIUC XO UL CLAIRE UNIDAD PLA IO T axem PNT ue aeren cal Ls ety Mum mam ean o re Pies om LETT Figure 3 Header structure found in every block EJ H Figure 4 Anemometers and switching between banks 33 ANEMOMETER DATA FORMATS FILE FORMAT Figure 5 Anemometer data formats 34 TIME FORMAT cop ms sc ms Lou o l Figure 6 Time and Bank ID formats Conversion of IrigB to BCD 35 MEC ee SEQUENCE REPEATS 130 560 BYTES OF DATA HAVE BEEN RECORDED LI eee Lo LLL Figure 7 Anemometer data record Miscellaneous 36 REPORT DOCUMENTATION PAGE Public reporting burden for this collection of information is estimated to average 1 hour per response including the time for revie
29. ode is generated in the SUN computer and allows time correlation with tunnel or flight parameters Recording is normally done in bursts of 1024 samples over 20 msec then 20 msec non recorded another 1024 samples etc until one buffer consisting of 64 x 8 x 1024 samples has been filled over 2 62 seconds When the complete buffer has been transferred to the Exabyte tape 1 5 seconds later the buffer begins to fill up again It is also possible to fill the buffer through continuous sampling if a long continuous signal trace is desired In this case the buffer is filled in 1 31 seconds With the 20 msec bursts the minimum frequency high pass is 50 Hz while the continuous has 1 6 Hz Switching from sensor group to sensor group switches both bridge feed i e heating the sensor and the signal output It is initiated via the SUN workstation but the system will make sure that no switching occurs while filling a buffer Switching is done by specifying the groups to be used in system A and B respectively 1 through 7 each One may also have an automatic roll through of all channels over a preset time to simplify data reduction The data reduction is done by specifying the desired start and stop times for data files to be read and typically the files are moved over to a Macintosh computer where the EXABYTEREAD code runs This code allows stripping out selected data segments either as statistical values only or also as signal samples su
30. position in the file of the first byte of the record starting at 0 at the beginning of the file SYNC This should be FAF3 and verifies that the start of the header has been identified day hr min sec This is the header stamp of time may be off from the correct embedded time by a few seconds SUN counter The record identifier assigned to the record in the SUN computer This is used to make sure that the A and B files have synchronized records in case of recorder problems occasionally this happened SysA bank Group 1 through 7 for System A SysB bankit Group 1 through 7 for System B 19 zero offset Average of the zero offset values across all channels This value should not change more than at most once or twice per flight Statistics of the anemometry channels iss 1 ave i sigma i skew i kurt i autotime i crosstime i crossvalue i iss Dataset 1 through 8 depending on the selections made i Channel number 1 through 7 for both System A and System B 1 Average over kmax samples sigma i Standard deviation over kmax samples skew 1 Skewness over kmax samples kurt 1 Kurtosis flatness over kmax samples Gaussian shape is 3 autotime i Auto correlation time to zero correlation crosstime i Cross correlation time shift channels 1 2 2 3 3 4 4 5 5 6 6 7 7 1 crossvalue i Cross correlation function value of maximum correlation between 0 and 1 After the last dataset request
31. r record Column 3 hr time hrs from header Column 4 min time minutes from header Column 5 sec time seconds from header Column 6 ibank 1 group number 1 7 for system A from header Column 7 ibank 2 group number 1 7 for system B from header Column 8 irecsun amp record identifier counter in the Sun and the only acceptable way of knowing that records in systems A and B are synchronized Wraps The following 98 lines are the actual zero offset information In the output file i is the anemometer number 1 through 98 corresponding to A11 through B77 although the actual info in the header starts with byte 13 Column 1 i anemometer number 1 98 Column 2 ha i zero offsets in counts 0 255 25 Statistical data The statistical data contains the features as determined of kmax individual samples of kmax samples length I e there will be one line output per dataset irrespective of the sample length Depending on whether system A or B is analyzed there will be a difference in content of the first columns Note that system B has a repetition of information regarding bank to make the format the same as A System A Column 1 Column 2 Column 3 Column 4 Column 5 Column 6 Column 7 Column 8 Column 9 Column 10 Column 11 System B Column 1 Column 2 Column 3 Column 4 Column 5 Column 6 Column 7 Column 8 Column 9 Column 10 Column 11 iss hrs min
32. s secs msecs ave sigma skew rkurt m iss ibankA ibankB ibankA ibankB 1 sigma skew rkurt lt input filename gt volts record number data set number 1 through 8 hrs from embedded time minutes from embedded time seconds from embedded time milliseconds from embedded time channel number 1 through 7 average value in counts standard deviation in counts but with decimal point since it is taken over kmax samples skewness kurtosis or flatness 3 for Gaussian record number data set number 1 through 8 group system A from embedded signal group system B from embedded signal group system A from embedded signal repeat group system B from embedded signal repeat channel number 1 through 7 average value in counts standard deviation in counts but with decimal point since it is taken over kmax samples skewness kurtosis or flatness 3 for Gaussian This file contains data of use for absolute sheer determination Column 1 m record number 26 Column 2 iss data set number 1 through 8 Column 3 ibank bank identification for the 7 channels may be either System A or B Column 4 10 volts 1 7 Bridge voltage time averaged over kmax samples Column 11 17 stdev Standard deviation over kmax samples in volts lt input filename gt s is an optional signal file that will print out the actual signal Since it may get large it should be used sparingly
33. tatistics output Some of the features are readily changed while others are fixed The system was developed for use under flight conditions by the NASA LaRC Flight Instrumentation Branch with the main contributions made by Carroll Lytle Carl Mills and Doug Taylor NYMA Inc and Sharon Graves Lockheed Martin Engineering and Sciences Corp and Keith Harris FIB NASA LaRC 1 System Instrumentation Figure 1 shows the system block diagram The various parts of the system are described in the following section 1 1 Hot Film Sensors The hot film sensors used in the high lift experiment were manufactured in the NASA LaRC Microelectronics and Sensor Development Section The films were manufactured from Nickel coated Upilex film from Inc To obtain desirable resistance values and a small sensor size needed due to the power supply limitations 0 5 Ohm square film was used The sensors were etched with a 0 003 inch line width and a 0 060 inch length giving the films a nominal cold resistance of 10 Ohms With the substrate heat conduction applicable to the B 737 High Lift system the total area of the sensing elements should be less than 200 mils due to the power limitations of the anemometry based on the desired overheat A thin layer of Aluminum Oxide was applied to stabilize the films Even so the sensors turned out to be pollution humidity sensitive and in the flight test series it was necessary to replace entire sheets of the s
34. ter is a SUN SPARC Station 2 with two Exabyte 8mm tape drives and two DASM AD14 SCSI analog to digital converters refer to 1 5 above one DASM per Exabyte The data is retrieved from the DASM and recorded on the Exabyte tape The tape transport mechanism is a product of Exabyte which along with the SCSI interface is packaged and sold by Contemporary Cybernetics Any brand of 8mm tape can be used provided it is data quality and has a storage capacity of 2 5 or 5 gigabytes 1 8 1 Flight System Start up Procedures ANEMOMETER SYSTEM START UP SHEET 1 Master Power On 2 Press in circuit breakers 2 3 4 6 7 10 11 allow 10 amp 11 Exabyte tape drives to complete start up cycle before pressing 14 amp 15 DASM breaker 14 15 18 19 13 CB 16 should be last one on 3 Before logging in make sure Caps Lock is off 4 Log in root return 5 Password __ return 6 hilift openwin return To activate any window put the cursor in the desired window by using the mouse then click the left mouse button The triangle following the hilift prompt will become darker and will accept typed commands 7 Lower left window on screen hilift home dasmad14 hilft_prgms return 8 Upper left window on screen hilift home dsp return 9 Lower right window on screen hilift cd home dasmad14 hilft_prgms return 10 Lower left window on screen hilift cntrl6 return Enter filename of log file create filename for test
35. tes 9 10 11 amp 12 2 2 3 3 Decoding Example Hex time in seconds Byte 5 2D Byte 6 E6 Byte 7 23 Byte 8 EC 15 Hex time converted to decimal time 2DE623EC 770 057 196sec Decimal conversion Base time Record time 770 057 196sec 757 382 400sec 12 674 796sec 12 674 796sec 86 400sec day 146 6990278 146 days 6990278 86 400 60 396 sec 60 396sec 3 600sec hr 16 77666667 16 hr 77666667 3 600 2 796 sec 2 796sec 60sec min 46 6 46 mins 6 60 36 sec Decoded Time 146 days 16 hrs 46 mins 36 sec 16 3 Data Analysis Software 3 1 EXABYTEREAD Version1 0 The purpose of the program EXABYTEREAD is to extract and manipulate hot film data in a convenient manner It uses as input the SUN files containing the data in digital form and as output it provides statistical data along with verification parameters to ensure that the data is correctly read and analyzed Together with the DATAREAD code which reads the thinned pressure data and the DATAC reference parameters the EXABYTEREAD provides a means of accessing validated hot film data This program which is run on a Macintosh can be used to select specific records from any flight data file for either System A or System B It produces two output files lt input filename gt statA or B and lt input filename gt sA or B The A or B suffix specifies which system the data was extracted By using a spreadsheet like Cricket Graph Ex
36. wing instructions 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 February 2002 Contractor Report 4 TITLE AND SUBTITLE 5 FUNDING NUMBERS NASA LaRC FIB Multi Channel Anemometry Recording System User s Manual NAS1 96014 AUTHOR S WU 706 31 11 80 Compiled by Sherylene Johnson and Arild Bertelrud PERFORMING ORGANIZATION NAME S AND ADDRESS ES 8 PERFORMING ORGANIZATION NYMA Inc REPORT NUMBER Hampton VA 23681 Analytical Services amp Materials Inc 107 Research Drive Hampton VA 23666 SPONSORING MONITORING AGENCY NAME S AND ADDRESS ES 10 SPONSORING MONITORING AGENCY REPORT NUMBER National Aeronautics and Space Administration NASA CR 2002 211440 Langley Research Center Hampton VA 23681 2199 11 SUPPLEMENTARY NOTES Johnson NYMA Inc Hampton VA Bertelrud Analytical Services amp Materials Inc Hampton VA Langley Technical Monitor J B
37. z sampling per channel the sampling frequency is flexible 50kHz was used in both the 737 High Lift and LTPT projects the eight analog sample and hold outputs are strobed simultaneously to the multiplexer The samples are then sequentially digitized by the multiplexer at a rate of 2MHz and stored in the DASM until a buffer is completed the DASM emulates a read and write hard drive When prompted by the SUN computer the buffer is read from the DASM over the SCSI bus and written to the Exabyte tape 1 6 Port Expander See Block Diagram The purpose of the port expander is to take a digital command from the single output port of the system computer and expand it into three outputs which can address five devices simultaneously One output controls both anemometer multiplexers another output controls both filters and the remaining output addresses the time code unit The port expander is a NASA designed and fabricated subsystem 1 7 Time Code Generator D A Clock Generator See Block Diagram This device generates the data sample clock which is sent to the DASM and encodes the parallel time code The digital parallel time is converted to an analog signal amplified and is output to DASM Unit 1 System A Channel 8 and is recorded to one of the Exabyte tapes along with every data sample The Time Code Generator D A Clock Generator are NASA designed and fabricated subsystems 1 8 System Computer See Block Diagram The system compu
Download Pdf Manuals
Related Search