Hi Denise,

We are sure it was corner cracks that caused the 10% increase in resistance. 

The posted picture was from a reliability test on coupons that were on the edge of a production panel. We measured the resistance before testing, during preconditioning (6X260°C) and during testing (thermal cycles to 150°C). We stop testing within seconds of hitting a 10% increase in resistance. Then, since the circuit is not open, we apply a small current while observing the coupon under a thermal camera. In this manor we find the most damaged PTH out of hundreds. The section is processed on the identified PTH and its' immediate neighbors. In this way we are able to observe a failure as it is "ripening". The test method and failure location is elegant and we gain great insight to failure modes using this method. 

The picture posted is a typical knee crack that was observed in many PTHs in the section.

Generally speaking, laminar plating is a reliability liability. The pictures Chris posted are classic knee cracks due to lead free thermal excursion. 

Cracks of that magnitude, in a significant number of PTHs, would exhibit a functional failure if resistance was being measured on the sample before, during and after thermal cycling.


Sincerely, 
Paul Reid 

______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________