NASA, Wiring and the Space Shuttle
Contents
1. Figure 1 Connector Pin Module NASA repair procedures require that the pins in connectors be repaired or replaced rather than replacing the entire connector Therefore the connectors were labeled as modules and the pins as replaceable components with the three failure modes discussed previously see Fig 1 The connector was modeled as a collection of pins with the appropriate pin labels see Fig 2 To avoid unnecessary undetected failures only the wired pins were represented in the TEAMS model 547701725 pinR Corgonen Corgoanean S407 70LIZ5_pinT Corgo nan 54U770J25 pin Figure 2 Bottom part of a connector module showing the wired pin components PO C ATEAMS51 COMPLIB NasaAuto mec TEAMS version 5 1 alpha File Edit Analysis Reports Tools Help saj tii afe al elel fo 9 n e aly 54V77WIPIllF FZS 2810 10 Wire l sav 7WiPlllF F2S 2810 10 142 uec1 BadDielectric HardShort DeltaResistance EDT X 4 Figure 3 Failure modes of a wire PA C TEAMS51 COMPLIB NasaAuto mec TEAMS version 5 1 alpha iol xj File Edi Analysis Reports Tools Help ole tH ae al elele w ol len al Qai 54V77WlPlllF F2S 2810 10 142 mecl 54V77WIPIlllF F2 Figure 4 TEAMS model of a two conductor shielded wire 2 Modeling of wires Wire failure modes were defined as follows 1 Open e g due to a broken conductor 2 DeltaResistance e g oxidized or frayed conductor2. t t t t t a A A CA B 8 o o 0 0 6 T6 B b 0 o o o o o are epre EL F F SUI E G zG 2G G G B o 0 n D 0 ee ere tct c a a a a a z o 70 TM cH MH H cH a 8 x F T PP a a un a a 3 3 3 ES Loc t Uc u u Iu u u cv Y7v a oR A ACR R R x x cx r zz TIt tct Figure 11 Segment of MECI J3 TEAMS Model Generated from SCAN data By establishing some basic rules on wire failure modes and test methods we were able to automatically create a MECI wiring assembly model in TEAMS The many features available with the TEAMS tool helped to greatly simplify the task and perform the analysis The resulting model closely resembles the physical structure of the wiring circuit thus allowing updates to be made quickly and easily This project illustrated how TEAMS can be used to assess testing methods and to create a diagnostic strategy tailored to specific needs QSI demonstrated that the run time tools TEAMS RT and TEAMATE can handle the analysis and runtime of the MECI subsystem in a simple laptop computer and feel confident that the solution will scale to about two orders of magnitude higher than the MECI model Based on the work accomplished in the pilot project the following recommendations were provided 1 Implement the TEAMS analysis as an important aspect of spacecraft wiring design and life cycle support from analysis to su
3. www teamgsi com servlet teamateservlet P Go Perform the fo ing test Measure Continuity at 54V77W2P115 pin 3 zi amp x E Internet Figure 10 Screendump of a Web based TEAMATE diagnosis session PROJECT SUMMARY AND RECOMMENDATIONS This project established the methodology for creation of a multi signal TEAMS model of a typical space shuttle wiring circuit of medium complexity The process started creating the individual components of the system and then added dependency paths that closely followed actual interconnectivity A partial block diagram of the MEC1 J3 connector harness as modeled with TEAMS is depicted in Figure 11 S4v23abe 312 E 1v332 345221 Ssv Jala List GHI r S42 3aL_al pint S4v33ale3s iv37234 133 KEN st 1v22234 L SSJ JAL L 3 3 1 1 H H 2 2 i i 1 1 PEREI te H i di is 103 ii L ts os t L D D i ou Hoi 22 2 23 23 24 i4 M 25 25 x 26 36 z 2 5 22 5 31 gt 28 om irit it it 3 HOH lt z 34 T 3535 B si 36 e 5 n T n ocn a nor ib it a LE E aoa i o 0 c 450045 g 46 46 u 0o 0 v dg x sod z it it 2 i12 Barrani as 23L i HOO N gc SS N ETT oo ETE eE 3 zz oo Shv2Inlarshs fbv2223 42500 ShvJonlaesee siat T a o o 2 ans 8 pr t e 6 a B B o k Lj z G 6 B g g nu on fiv2223 42548 Shv2Inlaesae a a i E rink s u u u u u Y 3 of 1 e 3 a p x x x z t
4. for the other harnesses intact 3 Modeling of Mate Demate status of connectors The mate demate status of each connector was modeled using switches see Fig 5 These switches can be programmatically opened and closed in TEAMS and TEAMATE TEAMS RT to simulate the mated and demated states of the connectors When connectors are demated they provide access to the pins and various tests can be performed to assess fault coverage and isolation We have created System modes which should be used when running the analysis The first one is labeled CompleteHarness Selecting this mode will simulate all connectors being mated to their mating connector with the exception of the last connector in each wire path Running the analysis in this mode will generally result in large ambiguity groups caused by the inability to fault isolate paths containing multiple wire segments and connectors The second System mode we have created is called DisconnectAll Selecting this mode will simulate all connectors being demated thus allowing testing on the pins of all connectors Running the analysis in this mode will result in much smaller ambiguity groups Additional system modes can be defined or changed as needed If it is known that some connectors in the wiring harness cannot be accessed then System modes can be defined to allow demating of all but these connectors In the actual application the mate demate status of the connector would be read from SCAN an
5. is a hierarchical modeling methodology Models can be built top down or bottom up In the present effort we pursued a bottom up approach where we first created a library of connector and wire types and then interconnect them as per the SCAN wirelist to generate the complete model To automatically generate the TEAMS model the raw data from the SCAN wirelist was converted into a format more compatible with TEAMS A filter was written to reformat this into a TEAMS model An import function was added to TEAMS to read the TEAMS SCAN model over ODBC and automatically generate the model In making the models of the components we made some basic assumptions regarding failure modes of components level of repair testing methods etc These are described in the following subsections 1 Modeling of connectors For connectors it was assumed that only the pins of the connector can fail and that the possible failure modes of the pins are PushedPin or open circuit caused by pushed pins 2 PinCorrosion or resistive contact possibly caused by corroded pins or poor contact with mating pin and 3 BentPin or shorts possibly caused by a bent pin shorting to an adjacent pin or a piece of metal shorting two or more pins Fol C TEAMS51 COMPLIB NasaAuto mec TEAMS version 5 1 alpha PE Xx File Edit Analysis Reports Tools Help Sjaj tij als aj alle i ol eum Bly 40V77W21P409 pin47 1 40V77W21P409 1 1mecl
6. Model based Testability Assessment and Directed Troubleshooting of Shuttle Wiring Systems Somnath Deb Charles Domagala Roshan Shrestha Venkatesh Malepati and Kevin Cavanaugh Ann Patterson Hine Dwight Sanderfer and Jim Cockrell Qualtech Systems Inc Suite 501 100 Great Meadow Road Wethersfield CT 06109 NASA ARC Research Center Moffett Field CA 94035 ABSTRACT As the space shuttle ages it is experiencing wiring degradation problems including arcing chaffing insulation breakdown and broken conductors A systematic and comprehensive test process is required to thoroughly test and QA the wiring systems The NASA Wiring Integrity Reseach WIRe team recognized the value of a formal model based analysis for risk assessment and fault coverage analysis using our TEAMS toolset and commissioned a pilot study with QSI to explore means of automatically extracting high fidelity multisignal models from wiring information databases The MECI Shuttle subsystem was the subject of this study The connectivity and wiring information for the model was extracted from a Shuttle Connector Analysis Network SCAN electronic wirelist Using this wirelist QSI concurrently created manual and automatically generated wiring models for all wire paths associated with connector J3 on the MECI assembly The manually generated model helped establish the rules of modeling The complete MECI model was automatically generated based on these rules thus sa
7. Test amp Maint Wiring Architecture Schematic Generation Data Archiving Failure Analysis Automated Test Generation ATG Model Export Configuration Data Test Results History Logs TEAMATE Smart Test Sequence Wire Harness Test Prompts Setup and Test Next Best Test Request Test Results S d fi Ins menu Intelligent Interactive Test Results DynamicReasoner Wire Test amp Maintenance Bulk Testing e Auto Records Logging 7 78 f Z f L d gt i i Optical E TDR 4 Meer ee m f a L 4L nidi j m jb Figure 12 Fully Integrated Comprehensive Wire Integrity Support Environment WISE provides a network Internet capable thin client software solution for ease of deployment maintenance and use ACKNOWLEDGMENTS Portions of this work were funded by a pilot study on the Space shuttle wiring system commissioned by the Wiring Integrated Reseach WIRe team at NASA Ames Research Center REFERENCES 1 Space Shuttle Independent Assessment Team Report Report to Associate Administrator Office of Space Flight October December 1999 2 S Debet al Multi Signal Flow Graphs A novel Approach for System Testability Analysis and Fault Diagnosis in Proc IEEE AUTOTESTCON Anaheim CA pp 361 373 Sept 1994 3 S Deb et al QSI Integrated Diagnostics Toolset 1997 IEEE AUTOTEST Conference Anaheim CA September 1997 4 S G
8. WIRe team at NASA ARC for review Tele conferences were held to discuss the source information models and assumptions A final review meeting was conducted on May 8 2000 at NASA ARC Peer review and feedback from domain experts are invaluable in improving the fidelity of the models For example based on feedback received in the final review meeting the model was revised to accommodate isolation and Dielectric Withstanding Voltage tests consistent with prevailing NASA testing practices From the review process it became clear that the WIRe team was more concerned with coverage of wires than with fault isolation Several interesting questions were raised that merit further exploration The following captures the essence of these questions and the answers provided e Question 1 Given the mate and demate states of the connectors how can one assess the maximum achievable fault coverage TEAMS analysis computes the percentage fault detection and isolation but it does not enumerate the covered and uncovered wires e Question 2 Connector 5 will be demated Tuesday for repairs If I test all cable runs accessible through connector 5 what percent of all cable runs will I have tested Answer Fortunately this problem can be solved utilizing our existing tools Figure 9 presents a screendump of TEAMS RT which takes a fraction of a second to compute the numbers for the MECI model The top half of the screen shows the list of failure modes left column a
9. d automatically represented with the proper status in the TEAMS model 40V77W21P103 9S0V77WlPlO3 Pinl o0Uy77W1J7493 40 81V77W8P749 ccduut u umEpDbwWw ummoeudumpndgagm e oC HM ota hu wd 5 OI m onc oro amp ccduuwidl d txzpbmsiuuo umpngdgaugug Cj C M ota hd wd 5 IE X ong og rpm o ooi TM 90V77WUl 2S 281 IQ Pinl 40V77W21 A2S 25 Fus A uA N E i Z up i em E Lp r 3 oe B B r r D D E E Fr F G G H H J J K X L L HM HM N N P P R 3 3 T T u u V V ce OC oM ota u d 5 OI D mono go pb Doo i ro e 4 ce OC oM ota u e t5 OI a mono go pb Doo il Figure 5 Partial Wiring Harness Model showing switches to model mate demate status of each connector 4 Modeling of tests The final step was to add tests to address the failure modes The tests defined for the wiring model are as follows l 2 Continuity To detect open wires and pushed pins Isolation Involves shorting all other pins to ground and then measuring resistance between the pin being tested and ground to assure pin is isolated from all other pins and ground Delta Resistance Involves measuring DC resistance through a wire path and comparing it to a predetermined limit It will detect open wires pushed pins and high resistance paths caused by corroded pins poor contact with mating pins and broken or frayed wires making partial contact It will also detect wires shorted to ground How
10. ever based on the feedback at our review meeting we disabled these tests as they are currently not part of the standard test procedures Complex Impedance This test will detect changes in resistance wire shorts opens and degradations and reactance degradation in insulation leading to capacitive coupling DWV Dielectric Withstanding Voltage Similar to Isolation test except a voltage signal is applied and gradually increased to detect wire or pin breakdown TESTABILITY ANALYSIS RESULTS Testability analysis was performed using TEAMS to produce reports that provide failure mode coverage metrics and generate optimized test strategy The results of the analysis are presented in a number of formats The primary testability report is the Testability Figures Of Merit Summary TFOMS Report The TFOMS Report for the MECI wiring model is illustrated in Fig 6 The failure rates of the individual wires and connectors were set to 1 per million hours However with the sheer number of wires and pins the mean time to first failure is 1406 hours Such parameters along with estimates of time and cost associated with tests can be adjusted based on field MTTF and experience and then used to optimize the troubleshooting strategies of TEAMS PA C TEAMS51 COMPLIB NasaAuto mec TEAMS version 5 1 alpha File Edit Analysis Reports Tools Help ze tHE g al Plea pope ule aly Testability Report for mecl TESTABILITY FIGURES OF MERIT FOR
11. hile helping NASA engineers monitor the extent of wiring failure modes covered by the testing process The TEAMS toolset could also provide the technician with the necessary tools to electronically log all maintenance activities and help automate the maintenance process 4 while preserving all the existing checks and balances All of the wiring information required for creating the TEAMS model was supplied via a Shuttle Connector Analysis Network SCAN electronic wirelist This partial wirelist contained all the wiring information relative to the MECI assembly Using this NASA supplied SCAN wirelist QSI concurrently created manual and automatically generated TEAMS wiring models for all wire paths associated with connector J3 on the MECI assembly The manually generated model helped establish the rules of modeling The automated model was compared against the manual model to verify that the automatically generated model accurately portrayed the actual shuttle wiring Once it was ascertained that the automatically generated model was identical to the one created manually the complete MECI model was generated thus saving significant modeling cost We also performed testability analysis on the system to determine the capability of the resulting wire maintenance system in detecting and isolating faults These parameters help establish the effectiveness of a wire monitoring and troubleshooting program MODELING METHODOLOGY TEAMS multisignal modeling 2 5
12. hoshal et al An Integrated Process for System Maintenance Fault Diagnosis and Support Proceedings of the 1999 IEEE Aerospace Conference Aspen Colorado March 1999 5 S Deb et al Multisignal Modeling for Diagnosis FMECA and Reliability invited paper in 1998 IEEE SMC conference San Diego CA 6 Wiring Integrity Research WIRe Pilot Study Design for Safety Initiative Document Number AOSP 0001 XB1 August 25 2000
13. is indicates a large number of ambiguity groups comprised of three components This is due to the fact that most wire paths in the sub harnesses are comprised of a wire with a pin at either end If it was necessary to break this ambiguity further Time Domain Reflectometer TDR tests could be used to isolate the failure to a single component Such tests can be modeled easily in TEAMS but were left out of the model to reflect current test procedures practiced by NASA 6 TEAMS also generates an optimized test strategy represented in a diagnostic tree Figure 8 illustrates a partial view of the diagnostic tree for the MECI wiring model The optimized strategy generated by TEAMS involves over 1000 steps and would be an enormous task if it were to be generated manually PO DRE ET XCOMPLIBXNasaAutoxmecl TEAMS version 5 1 alpha File Edi Analysis Reports Tools Help ola tidk al ejen x alo SO0Vv77WlPlOSlTCont e yCheck Perform ContinuityCheck at S0V77WlPlO3 pin S S0V77WlJ749l1TContinjityCheck Perform ContinuityCheck at S0V77WlPlO3 RET AMBIGUITY YES NO 90V77WlPlO3 pinlT 3 s 90V77WlP 103 21 9S0V77WlIPlOS3 pins 4 lt 3S0V7701P1 03 21 AMBIGUITY o 9S0v7 7WlPlOSS Bl Z25 YES 90V77W1J749 pinD 4 z 90V77W1J7 49 22 90V77W1J7491T Bl 24 b TREE Figure 8 Partial view of Diagnostic Tree for MECI wiring model NASA REVIEW COMMENTS Two iterations of the MECI J3 harness model were submitted to the
14. ition of a wire from the day of manufacture through its entire service life complete with appropriate testing and maintenance records Included is the need for safety assessments to document the practical trades that will be required in terms of periodicity of inspection inspection methodology and functional testing This type of program would not only substantially improve safety but also has the potential to reduce no fault found component removals labor inspection time aircraft downtime and therefore aircraft readiness and other airline cost drivers The Wire System Safety Interagency Working Group IWG at the direction of the White House Commission for Aviation Safety and Security issued D 181 SN470351 on 29 June 2000 requesting information to assist government agencies in dealing with the challenges surrounding Wire System Safety This Commerce Business Daily announcement requested information on ongoing activities that include a a better understanding of degradation mechanisms b detection inspection techniques c methods of mitigation and d improved wiring systems In response to this solicitation GRC International assembled a team of companies that have been involved in the issues surrounding wire safety for many years and formed the Wire Integrity Program to develop a comprehensive program for managing wire integrity The teaming of these companies allows for immediate implementation of existing technologies specifically designed t
15. making partial connections 3 HardShort e g a short circuit to ground or adjacent conductors due to damaged insulation and 4 BadDielectric e g worn or degraded insulation allowing arcing when a high voltage is applied possibly causing intermittent and potentially hazardous electrical discharges under normal operation The wires were modeled based on their type conductor twisted pair twisted pair shielded etc A library of wire components was created based on the wire types listed in the SCAN wirelist Consistent with NASA repair procedures a twisted wire was modeled as a single wire component since it would be replaced as a single unit The Wire Type and Cable Descriptor fields of the SCAN database uniquely define bundles For each wire or conductor in the wire type we inserted a sub component with the appropriate failure modes The shield however has only one failure mode representing a broken shield or ground path Fig 3 presents an example model of a wire showing the failure modes Fig 4 presents the model of 2 conductor shielded wire of gauge F wires F2S The model can easily be updated to represent actual wiring repair practices used by NASA or changes in the current practices If for example it is determined that a particular harness cannot be repaired and must be replaced as an entire assembly then the repair label of that particular harness would be revised leaving the previously defined repair labels
16. mecl 96 34 9 33 63 05 4 28 4065 61 4065 61 4045 30 4045 30 38 37 W UW UW 97 95 10 00 TEST OPTIONS Percentage Fault Detection Test Algorithm NEAR OPTIMAL i Breadth l Depth l memory big il te tl aE hare citi System modes DemateAll elastic LA UNE Test cost weightage 100 00 roep DEGUD Sass Fault Isolated to LABEL Component System OK probability 1l amp Mean time to first failure Mean Cost To Isolate Mean Time To Isolate Mean Time To Detect Mean Cost To Detect Lambda Search l Fault Iso 1406 hours 39 58 SYSTEM STATISTICS Number of failure sources 2290 in 567 Components Number of tests 980 HISTOGRAM OF AMBIGUITY SIZE HISTOGRAM OF TEST USAGE Number of switches 245 100 Number of dependencies 6029 Number of modules at level 1l 234 30 Number of modules at level Z 11 ae Number of modules at level 3 2290 60 TEST ALGORITHM STATISTICS 40 Number of tests not used IZ Number of nodes in tree 1741 20 Number of backtracks Efficiency of Test Sequence 4 5 2 1 1 1 1 1 1 1 1 1 O 1 1 now ow ou o to Ld 2 97 5 3 12 15 1B 21 24 27 27 NUMBER OF TESTS luz Jg 4 5l 7g 9x9 a 6 AMBIGUITY GROUPSIZE Y gt TFOM Z Figure 6 TFOM Summary for the MECI wiring model Ej mecl_amd Notepad O x File Edit Search Help Node in Diagnostic Tree 335 No Go Path oul Total Probability of Group 6 661276 T
17. nd the mate demate status right column The lower half of the screen shows the components that will be covered Good and still untested Unknown if testing were to be performed utilizing the current configuration of connectors The Modes or Mate Demate status can be set programmatically from the SCAN database Also if the tests were performed and pass fail tests results submitted to TEAMS RT it would also be able to compute the Bad and Suspected components still within a second of processing time TEAMS RT when combined with TEAMS KB can also retain the state of the system and quantify incremental test coverage as more and more connectors are demated and tested In addition TEAMATE can be used to guide the technician s and expedite the testing process A web based version of TEAMATE see Fig 11 was also demonstrated in the review meeting Harness1 TEAMS RT Qualtech Systems Inc Iof xi Eile Options Total Faults Inserted Total Modes 152 Aspects Inserted Faults HighResistance 2 40V 7 40V77W21P103MATE Open 1 40V7 7W2 1P409 HighResistance 2 40V 7 40V 7 7W85P 13 1MATE Open 1 40V77Vv2 1P409 HighResistance 2 40V 7 40V 7 7W21P409MATE 50Vv77W106J409MATE 50V 7 7W33 419MATE Select Tests Simulate Run Monte Calro System Reset Cancel Harness1 TEAMS_RT OUTPUT ioj x Suspected Unknown Suspected Unknown Fi I Fi I Ry Processed Frames Run Number Faults amp Instants Repair No C
18. o address wiring system problems The companies participating in this endeavor are Honeywell Inc GRC International Inc DIT MCO International Inc Lectromechanical Design Company Phoenix Aviation amp Technology and Qualtech Systems Inc A resultant Wire Integrity Program includes vehicle modeling test planning and monitoring testing and health assessment and data management The Wire Integrity Support Environment WISE see Fig 12 brings together the advanced modeling and analysis features of TEAMS and MultiLinx to perform wiring system modeling failure analysis diagnostic analysis data logging automatic test generation optimized test and maintenance strategy wire system diagrams architecture and intelligent dynamic reasoning for wire testing and maintenance Honeywell is currently committed to leading the integration of QSI s TEAMS toolset and GRCI s Multilinx tool to develop the first iteration of the WISE software by mid 2001 Graphical Models Diagnose Tcking amp Trending Wire System Source Data Di mdr R ma On Line Wire Test Plans Wiring table M Metrics _ Mating table AX z Parts List Ls om 85s E aa Hifi it System Architecture FD FI Assessment DFT Analysis Mgmt FMECA Parts Removal and e Optimized Diagnostiq neas ae nen MultiLinx History Strategy Test Plans FD Model and Test id Coveraee Data Exchange Test Metrics Trending Prognostics Test Schedules
19. olors Simulate Don t Display Good Moules Dismiss Figure 9 Screendump of TEAMS RT assessing fault coverage for MECI system e Question 4 Given a wire network and given enough time to demate test 2 connectors only which connectors do I demate and test to maximize the number of cable runs tested e Question 5 I need to test the circuit containing run E Which connector pair do I demate to access E but use opportunities to test maximum number of other cable runs e Question 6 Suppose connector 8 is hidden and inaccessible What is the greatest number of wire runs I can possibly tests How many connectors must I demate e Question 7 What are the fewest number of connectors demated to test all wire runs Answer All of the above questions can be easily formulated as set covering problems subject to constraints e g inaccessible connectors cost time budget and efficient search algorithms can be developed to solve the problems The models and TEAMS RT s ability to evaluate coverage will be essential components required for evaluation of the cost function to be optimized by the search process While there is no off the shelf solution to these questions we can develop comprehensive solutions to these problems given the opportunity in the near future Sj teamate Microsoft Internet Explorer File Edit View Favorites Tools Help 2 9 2 2 84 3 8 2 Back Forward Stop Refresh Home Search Favorites History Mail Print Address ei http
20. otal Unweighted Probability 6 661369 Number of Modules in Group 3 List of modules in this group 1 54U77W1P118_pin Q 22 lt 54U77W1P118 48 Module Probability 6 666464 2 SSU77W1J11_pinCC 3 lt 54U77W1J11 147 Module Probability 6 666464 3 55U77W1P118 S DAT hT1 5 158 Module Probability 6 666348 Node in Diagnostic Tree 93 No Go Path Total Probability of Group 6 661276 Total Unweighted Probability 6 661369 Number of Modules in Group 3 List of modules in this group 1 58U77U88P159 pin J 58U77U88P159 178 Module Probability 6 666464 2 56U77W88TB36_pin1 1 lt 56U77W88TB36 263 Module Probability 80 888565 3 58U77U88P159 J D1 204 Figure 7 Ambiguity Groups Report for MECI J3 connector wiring model The Test Options section of the TFOMs report lists the options used for the analysis System Statistics provides the model details The Test Algorithm Statistics provides a list of information about the resulting test strategy The TFOMs box presents the Percentage Fault Detection or Fault Coverage and Percentage Fault Isolation metrics The most important information provided by the TFOMS Report is the bar graph entitled Histogram of Ambiguity Size The histogram provides a graph of the relative number of ambiguity group sizes The list of specific components comprising the individual ambiguity groups is provided in the Ambiguity Groups dynamic test report see Fig 7 The analys
21. pport 2 Research the concept of using the optimized dynamic model based reasoner TEAMATE to drive the wiring test equipment 3 Move forward to link wiring technical manuals Interactive Electronic Technical Manuals IETM to TEAMATE for class V IETM 4 Setup a web based remote wiring diagnostic system to enhance troubleshooting and maintenance across NASA centers and contractors 5 Research the aspect of embedding the run time model based diagnostics TEAMS RT on the vehicle to diagnose wiring problems in real time A copy of our project report along with these recommendations were incorporated in the final wire team report 6 THE ROAD AHEAD The TWA 800 and SwissAir 111 accidents both involved electrical wiring failures in commercial aircraft Further it has been shown since these accidents occurred that they might have been prevented if a suitable wire testing program had been in place Industry practice then and even now relies primarily upon visual inspection methods to identify damaged and degrading wiring prior to failure These inspection methods have been proven to be inadequate in identifying wiring anomalies In addition this experience is not limited to commercial aviation but is also true for the military services and even the NASA space shuttle program As understanding of the scope of this problem improves the need for comprehensive wire management programs becomes self evident Such a program would track the cond
22. ving significant modeling cost The methodology is easily extensible to the entire shuttle wiring system This paper presents our modeling and analysis results from the pilot study along with our proposed solutions to the complex issues of wiring integrity assessment problem Keywords wiring arcing chaffing insulation breakdown multisignal Wire Integrity Program TEAMS INTRODUCTION We have recently completed a pilot study on the Space shuttle wiring system commissioned by the Wiring Integrity Reseach WIRe team at NASA Ames Research Center As the space shuttle ages it is experiencing wiring degradation problems including arcing chaffing insulation breakdown and broken conductors 1 A systematic and comprehensive test process is required to thoroughly test and QA the wiring systems The NASA WIRe team recognized the value of a formal model based analysis for risk assessment and fault coverage analysis using our TEAMS toolset However wiring systems are complex and involve over 50 000 wire segments Therefore NASA commissioned this pilot study with QSI to explore means of automatically extracting high fidelity multisignal models from wiring information databases The intent of the pilot study was to investigate the feasibility of automatically creating a TEAMS model 2 for a subset of the space shuttle wiring The model could be used by the TEAMS toolset 3 to guide the technician in the wiring diagnosis and quality assurance process w
Download Pdf Manuals
Related Search