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: Jon Ilseng
Organization:Schlumberger Oilfield Services
Date posted: Mon Aug 9 15:54:12 US/Eastern 1999
Subject: MIl-HDBK-217F Stress Analysis Predictions
I have a question concerning the MIL-HDBK-217F Stress Analysis Predictions for Commercial Fixed Film Resistors (Style RN) vs. Commercial Resistor Networks containg Fixed Film Resistors. The PCB I am doing a MIL-HDBK-217F stress analysis prediction contains these parts. The PCB operates in a Ground Mobile environment at 40 Degrees C ambient temperature. For a 0.125W Fixed Film Resistor operating at 2% of its rated power, the predicted failure rate is 0.0984. For a resistor network containing 9 Fixed Film Resistors, the predicted failure rate is 0.037. Why does a resistor network containing 9 Fixed Film Resistors have a smaller failure rate than one Fixed Film Resistor? My only guess is because of the difference in quality factor values. For a resistor network with commercial resistors, the quality factor value is 3.0. For a commercial Fixed Film Resistor, the quality factor value is 15.0. I would appreciate anyone's help.
Subject: Film Resistor-Mil-217F
Reply Posted by: Bruce Dudley
Organization: Reliability Analysis Center
Date Posted: Thu Aug 19 9:15:37 US/Eastern 1999
After reviewing the failure data bases that were used in the 217F model development, I found that the basic rates were 0.0037 and 0.0019 failures per million hours for film and film networks respectively. The current data now shows a change to 0.0015 and 0.0033 failures per million hours, film to filn networks. The original cause was a large data base for film resistors and a smaller base for the networks resulting in the difference. With more information, the situation is reversed and the networks are showing more failures. The model develovment suggests that for the networks the number of resistors in the package is independent of the overall expected failure rate. So the answer to your question is, base failure rate data caused some of the apparent discrepency along with the quality factors. If we used the "new" RAC RATE reliability model for this part type the failure rate for the film resistor would be 0.0072 failures per million hours and the film network would be 0.011 failures per million hours for commercial parts in a ground mobile environment.