SRC Forum - Message Replies

Forum: Reliability & Maintainability Questions and Answers

Topic: Reliability & Maintainability Questions and Answers

Topic Posted by: Reliability & Maintainability Forum ( )
Organization: System Reliability Center
Date Posted: Mon Aug 31 12:47:36 US/Eastern 1998

Back to message list Show all replies Topics List About this forum
Original Message:

Posted by: andrew Comons
Date posted: Thu Mar 14 20:25:29 US/Eastern 2002
Subject: Reliability With degraded Mode of Operation
How does the industry handle a MTBF calculation whereby, a mission (single system)may permit operation in a degraded mode? Assuming that a failure did not prevent the degraded mode operation, how would the MTBF calculation be made? For the designer's specification, do I have to specify two different MTBF's - one MTBF for the entire system and another MTBF for degraded system? Or do I specify my maximum mission reliability as the minimum for all hardware relevant to the degraded state? I cannot find much written on this degraded scenario.


Subject: Degraded mode of operation
Reply Posted by: Dave Reinert
Organization: Harris Corp.
Date Posted: Thu Mar 28 11:01:04 US/Eastern 2002
Our practice is to specify the reliability relavent to the degraded state (either mission reliability or MTBCF) and develop the models accordingly. A good example of this would be a phased array antenna with distributed power amp modules, one for each array element. In the "no failures" condition, there is a specific EIRP, sidelobe level, beam width, and maximum useable data rate. Should one or more power amps fail, EIRP, sidelobe level, and beamwidth degrade and the data rate must be scaled back to maintain link closure. If this reduced data rate is still sufficient to pass the minimum amount of data per unit of time acceptible to the customer, then, while performance is degraded, a system failure has not occured. We develop the system spec's and requirements based on this minimum acceptible data rate and sidelobe degradation level and that defines the maximum number of module failures that are allowed. Also, we only specify a single reliability value based on this criteria, not one value for "full up" and another for "degraded." In this example, assuming an array of 16 distributed elements, if the minimum acceptible data rate and sidelobe degradation requirements can be met with 3 element failures, then the array would be modeled as 13 of 16 active redundancy. The ability to perform maintenance, or lack thereof, will dictate the use of R(t) or MTBCF as the overall requirement. Hope this helps.

Reply to this message