I am curious if anyone can help explain to me the cause of an effect we are seeing with some failed caps.
The parts have been found to be cracked and in most cases are hard failures (capacitance well out of spec and DCR less than 100 ohm or shorted), but in a few cases we have seen parts which exhibit these failure modes, however, after applying a voltage across the cap it returns to within spec. What would cause this effect?
Any help you could offer will be very much appreciated.
The type and location of the crack give hints about the cause. What does your failure analysis say?
Continuing, this a relatively common complaint. While you're waiting for other reponses, searching the fine SMTnet Archives will probably give previous discussions and may even provide links to MLCC fabricator sites and their white papers on the topic.