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: Laurens
Date posted: Tue Jun 10 11:26:50 US/Eastern 2003
Subject: Modern MTBF calculations
I have recently been assigned to develop a new product. Client's require 25 year MTBF, but do not specify the calc method. Looking through our copy of MIL HDBK 217F and performing calculations shows that the product wouldn't meet this (or even come close). What is considered a more up-to-date method of reliability prediction (esp. for items like ceramic chip caps)?
Subject: Prediction Models
Reply Posted by: bwd
Date Posted: Wed Jun 11 14:12:32 US/Eastern 2003
MIL-HDBK-217 has not been updated for over ten years. This means that many models are not up to the current technology. The RAC's PRISM program is newly developed and accounts for much of the current device technology. Models for chip capacitors and chip resistors are included in the PRISM list. Another source that has chip capacitors is Telcordia. The prediction model, SR-332, has most electronic components listed.
Subject: Credible Reliability Prediction
Reply Posted by: Larry George
Organization: Problem Solving Tools
Date Posted: Tue Jun 17 19:35:37 US/Eastern 2003
The ASQ Reliability Division monograph, "Credible Reliability Prediction," suggests using observed, age-specific field failure rates of comparable parts to predict MTBF and failure rates of new products and parts.
Designs and processes change, but customer and environment factors don't change, much, so the failure rate function may not change, much. Furthermore, that change can be modeled in a relative risk model, using data available at the time a prediction is needed. The monograph also shows how to update predictions as field data become available. There ought to be field data for caps by now to make credible reliability predictions for them.