Zero defect

[Stephane]

Hello,

I have zero defect with a sample of 150 units after 1000 cycles of T/C between -55°C and 150°C.

The problem is that I need to see at least one defect.
I cannot increase the stress and the number of cycles.
I think that the only solution is to increase the sample size. In this case, I wonder if it's possible to determine the sample size needed to obtain at least one defect.

Thanks for your help.

Stephane

[RS Support]

Yes, one could determine the sample size needed, however one must know (or assume) something about how these units fail. Without that knowledge, or assumption then there is no answer to your question. What do I mean knowledge or assumption? At best, if one knew the life distribution of the units in that environment (i.e., Weibull with Beta=2 and Eta=X) then obtaining the answer is trivial, (see http://www.weibull.com/LifeDataWeb/test_design.htm and also see Reliability Edge Article at http://www.reliasoft.com/newsletter/1q2001/simumatic.htm).

As shown in this article, different knowledge could be utilized to arrive at the answer.