We have experienced a failure during life testing of an electronic control
unit,
10 units were continuously powered and monitored at a fixed ambient
temperature
of 95degC. After 8.5 weeks of a 10 week test one unit exhibited a failure
which we
believe is attributable to a transient short circuit within the 8 layer
printed circuit board.

The PCB has 95V rails in specific areas and the failure appears to have been
caused
by high voltage getting onto a signal line. So far we have been unable to
locate the
failure site as the leakage path disappeared during the course of
investigation.

We suspect a conductive filament type mechanism, (CFF), but are told this is
normally
associated with formation of copper dendrites under high humidity
conditions.

Has anyone experienced failures under similar conditions ?

Are there any suggestions on the failure mechanism ?

What corrective actions are applicable ?



Best Regards

Stephen Lane
Lucas Automotive Electronics
[log in to unmask]