A question!

[vnigam]

I have cycles to failure and suspension data. My data looks like;
F 106017
S 116000
S 116000
S 116000
S 116000
S 116000
S 116000
F 162608
F 170816
S 200000
S 200000
S 200000
had I continued the run the suspensions to 250K my B10 value at 90% confidence turns out be be lower than when suspending at 200K and or 116K. Intiuitively I would expect the other way round. Can you explain please?

Thanks

[David]

Hi vnigam,

So all of your suspensions are run to 116K, 200K or 250K? Or just the suspensions at the end?

[David]

Hi vnigam,

OK, I think I figured out what you are referring to. When you increase the suspension time, the estimate for Beta is lower. The probability plot line gets pushed out to the right a bit and you end up with a lower slope. In this case it causes the B10 life to be smaller for 250K than at the other suspension times. However, if you consider another metric like B50 then the data set with suspension times to 250K will have a larger value in this case. It just depends on what point in time you are looking at.

I hope this helps.