Accelerated testing to Service life conversion| 6 August, 2001
I am having some SMT assemblies undergoing thermal shock testing per MIL-STD-883, Method 1011, Test Condition B. I was wondering if anyone has a chart, equation, article, etc. that will give me an idea of the service life of the assembly after it survives X number of cycles. I have seen a table from the IPC-SM-785 standard that gives this information for temperature cycling but I haven't seen anything similar for the MIL standard. Does anyone have any suggestions?
Accelerated testing to Service life conversion| 7 August, 2001
First, we know that MIL types are wont to use charts like that. Give �em Weibull distribution tables and then they�ll be happy, happy. A good book is Nelson �Accelerated Testing: Statistical Models �� Wiley
Second, we don�t use that very nice stuff coz it makes our heads hurt. Here�s two angles that you can play with and see how they compare to other things you find:
1 From G Grossman over at Swiss Federal Institute for Materials Testing and Research � A rough estimate of the lifetime in a real time environment by: Nfr=Nft*(dTt/dTr)exp(1.7) ... where: Nfr = Number of cycles to failure reality Nft = Number of cycles to failure test dTr = Temperature swing reality dTt = Temperature swing test
2 From Process Sciences [512-259-7070], Arrhenius Product Life Calculator/Estimator: Tm2=Tm1*(10^(Ea/2.303R*(Tp2-Tp1)/(Tp1*Tp2)) � where: Tm1 = Component's original life span in days Tp1 = Component's original operating temperature in Celsius (�C must be less than or equal to 95�C) Tp2 = Component's elevated test temperature in Celsius (�C must be less than or equal to 95�C) R = Universal Gas Constant 8.314 J/mol*K Ea = Activation Energy for corrosive reaction in Sn/Pb/Cu substrates 50,000 J/mol Tm2 = Component's calculated life span at the elevated temperature
Finally, we are aware of IPC-9701 "Performance Test Methods and Qualification Requirements for Surface Mount Solder Attachments", but don�t have a clue of what�s in it. [If you get it, but two and send us one.]