1、Lessons Learned Entry: 0292Lesson Info:a71 Lesson Number: 0292a71 Lesson Date: 1993-07-15a71 Submitting Organization: JSCa71 Submitted by: Larry GreggSubject: Selection of Critical Test Parameters and Conditions for Monitoring Hardware or System Performance (Such as Temperature of Computer Electroni
2、cs) Description of Driving Event: During burn-in testing of an orbiter General Purpose (Flight) Computer (GPC) in the JSC avionics engineering laboratory, a “fail to synchronize“ error was detected. This error indicates that the computer is out of synch with other GPCs involved in the test, and may
3、be the result of a computer hardware problem. The GPC in question was discovered to be very hot to the touch, and inspection of internal sensors indicated the GPC had been overheated beyond design limits. An ensuing investigation showed that the unit had operated without cooling for 11.5 hours due t
4、o failure to remove the ventilation system air duct cover prior to installation of the GPC in the test bed. This event was a direct result of not following installation procedures described in the Test Preparation Sheet (TPS). Additional factors contributing to this event included:A. The design of t
5、he cover plate for the GPC cooling duct did not prevent installation of the GPC with the cover in place. Most of the cover plates used in the laboratory were modified with additional foam to preclude installation of the GPC without removing the cover.B. The alarm system used to monitor GPC cooling d
6、epended on vacuum sensing instead of positive airflow indications. With the air duct cover in place, the vacuum sensor on the GPC in question showed a normal range reading even though there was no air flowing through the computer. The monitoring of an indirect parameter (vacuum) rather than a direct
7、 indication of absence of a critical parameter (cooling air flow) or the existence of the hazardous condition of concern (temperature of the GPC) did not guarantee that proper cooling was provided.C. Only one junior technician was assigned to install the GPC. This was a violation of instructions on
8、the TPS, which required two technicians to perform this task. As a result, the quality assurance assignee for the task helped to perform the installation procedure rather than performing his intended monitoring function.D. There was no formal program of training or certification in place for technic
9、ians in the laboratory. The training received was principally an undetermined amount of on-the-job Provided by IHSNot for Resale-,-,-training.Factory evaluation of the GPC involved in this event determined that it was fit for use only as a test or prototype unit and not as a flight unit. The differe
10、nce in cost between a flight GPC and a test GPC is approximately $700,000.Lesson(s) Learned: Measurement/monitoring of indirect indications for critical hardware performance parameters can lead to improper determination of status and equipment loss or damage.Recommendation(s): Mishap board recommend
11、ations: Consider using positive air flow indicators rather than indirect vacuum sensors. For cold plate mounted equipment, consider a pyrometer or thermocouple on the case as a redundant direct temperature indication.Evidence of Recurrence Control Effectiveness: N/ADocuments Related to Lesson: N/AMi
12、ssion Directorate(s): N/AAdditional Key Phrase(s): a71 Test Articlea71 Test & VerificationAdditional Info: Provided by IHSNot for ResaleNo reproduction or networking permitted without license from IHSApproval Info: a71 Approval Date: 1993-09-13a71 Approval Name: Ron Montaguea71 Approval Organization: JSC/NS3a71 Approval Phone Number: 281-483-8576Provided by IHSNot for Resale-,-,-
