EESemi.com Forum Archives

Equivalence of Reliability Test Life to Operating Life

          

A common question among reliability engineers is how to predict the actual operating lifetimes of devices in the field based on the observed rel test lifetimes of samples in the rel lab.  The archived forum thread below ponders this question.

  

The answer to this question is not simple given the many different ways in which the many different failure mechanisms of semiconductor devices may respond to the many different environmental factors such as temperature, humidity, and electrical excitation.  Predicting operating lifetimes from reliability data is referred to as reliability modeling.

  

Posted by Mike_balbuena: Thu Jul 13, 2006 8:12 am    Post subject: accelerated Temp and Humidity Test

 

Hi Anyone know if the parameters in this test such as 85 deg/ 85 rh at 1000 hr provide us with time equivalent. For example testing at certain degrees, temp and hours is equivalent to 1 or 2 years.

Any information is appreciated.

Thanks
_________________
Mike B.

 

Posted by FARel Engr: Thu Jul 13, 2006 10:32 pm    Post subject:

 

Hi Mike,

The equivalent field lifetime for a given reliability test is failure mechanism-specific, i.e., there is a unique equivalence for every failure mechanism. This is because every failure mechanism  is accelerated by a given set of conditions in a unique manner, i.e., the activation energy for each failure mechanism is different.

The question you are asking is very difficult to answer accurately. It is something that can not be easily explained in a forum. The starting point, however, is to choose a specific
failure mechanism that you are interested in.

You will then be needing Arrhenius equation to be able to answer your question. You need to find an acceleration factor (AF) between rel lab test conditions and field operating conditions in order to estimate the expected life of your devices based on your rel data. This may entail running many rel tests to estimate the activation energy of the
failure mechanism you are interested in (if this is not yet known). I suggest that you consult the rel experts of your company or your suppliers to get a more definitive answer.

FARel Engr

 

Posted by Mike_balbuena: Fri Jul 14, 2006 9:02 pm    Post subject: Thanks

 

I got it and found a scientist who can help. Appreciate the idea.

Regards
_________________
Mike B.

      

Back to the 'Best of Forums' Archives

      

HOME

 

      

Copyright © 2008 EESemi.com. All Rights Reserved.