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
Posted by: Leroy Meyers
Date posted: Mon Mar 6 15:16:54 US/Eastern 2000
Subject: Prediction "Degradation"
I work at a fairly small company. For our main product, a $200,000 unit for military communications, we have always advertised a certain MTBF which was obtained using MIL-HDBK-217F. Now, we are switching to all surface mount components, which hurts the prediction significantly because of the switch from B level or 883 parts to commercial non-hermetic parts. I could make assumptions, and give the commercial parts a better quality factor than 217 indicates. Or, I can switch to a different prediction technique such as Bellcore or Prism, or which I have no experience. Any opinions or other suggestions?
Subject: SMT Prediction Degradation
Reply Posted by: Bruce Dudley
Organization: Reliability Analysis Center
Date Posted: Wed Mar 8 15:47:48 US/Eastern 2000
Your concern is real, as MIL-HDBK-217 does penalize the use of commercial or industrial surface mounted devices. The problem results from the age of the data that was used to model the devices. 1970's and 1980's data was collected and these results indicated that there was a significant difference in the quality. My recommendation is to use more current models that represent the technology of today. Bellcore models could suffice, but the Prism model is more current and has growth application. The Prism product is available from the Reliability Analysis Center.