SOR 518 The Folly of Reliability Predictions
Abstract
Kirk discusses the continued reliance on the misleading approach of using reliability prediction for reliability development.
Key Points
Join Kirk as he discusses MIL Handbook 217 and its prodigy, the Arrhenius equation, and the reality of early electronics components (tubes, discrete transistors) being a major factor overall reliability of a system
Topics include:
- Kirk learned about reliability as a technician in a calibration lab and then a field service engineer troubleshooting complex high vacuum semiconductor processing equipment.
- Kirk was made Manager of the new ESS (Environmental Stress Screening) Processes after Dr. Gregg Hobbs had demonstrated the benefits of HALT and HASS to Storage Tek on a small tape drive loader. Gregg had demonstrated that Storage Tek could find 90% of the reported field failure mechanisms in a 15 minute combined stress screen.
- At a ESS conference in Oregon Kirk met Charlie Leonard, a key development engineer from the Boeing Commercial Aircraft Company. Charlie believed that most of the cooling of electronics on aircraft did not improve reliability and was costly and wanted to learn more about HALT and HASS. To learn more Charlie Leonard brought four other Boeing Engineers developing the Boeing 777 aircraft to Storage Tek in Louisville, Colorado for two days to learn more about HALT and HASS and observe its application and data supporting it.
- Charlie also introduced Kirk to Professor Michael Pecht, the founder and leader of CALCE (Center for Advanced Life Cycle Engineering) Consortium at the University of Maryland and that led to Professor Abhijit Dasgupta from the CALCE joining the trip with Boeing to Storage Tek.
- Almost 20 years later Kirk co-wrote a paper with Professor Pecht, published in the IEEE Design and Test of Computers on an experiment of long term overstressing of PC’s. Professor Pecht asked Kirk to include his association with CALCE as a Senior Collaborator in his public credentials.
- Reliability engineers are still following the old models and predictions that are misleading because real field reliability data is extremely sensitive and most times difficult to find even in companies that you work in. For those reliability engineers who do see actual causes of field failures, they probably will see that most are due to errors in design margins, manufacturing processes, or misuse by the user, not semiconductor wear out modes.
- Please download and read the very important US ARMY and CALCE paper linked in the notes below, “Reliability Prediction – A Continued Reliance on a Misleading Approach”
Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches.
Show Notes
Here is the link to the US ARMY/CALCE PAPER presented at the 2013 RAMS conference event mentioned in this podcast– “Reliability Prediction – A Continued Reliance on a Misleading Approach” . This is a government document that is in the public domain and should be distributed a read by any engineer that believes failure prediction methodology is useful or beneficial during product development.
Here is a link to Kirk’s book co-authored with John J. Paschkewitz available from Amazon “Next Generation HALT and HASS: Robust Design of Electronics and Systems”
Leave a Reply