August_EDFA_Digital

4 / 58

edfas.org
ELECTRONIC DEVICE FAILURE ANALYSIS | VOLUME 18 NO. 3
4
EDFAAO (2016) 3:4-8 1537-0755/$19.00 ©ASM International®
HIGH-VOLTAGE CAPACITOR FAILURE
ON A DOWNHOLE OILFIELD PCB
John Bescup, Weatherford
INTRODUCTION
Often in technical discussions within the failure analy-
sis community, passive components are overlooked in
favor of novel analysismethods or emergent challenges to
semiconductor reliability. However, passive components
still occupy a vast amount of real estate in today’s circuit
designs and are not poised to disappear anytime soon.
With that in mind, this article presents a case study of a
failed high-voltage leaded-ceramic-chip capacitor that
met its demise through an unlikely failure mode, which
highlights the importance of well-trained operators
behind the inspection equipment deployed to prevent
latent defects.
DOWNHOLE APPLICATION
The capacitor to be examined in this article was
intended for use in the high-vibration and high-temper-
ature realm of oil drilling, which has a set of reliability
concerns familiar to automotive and aerospace engineers.
The components on these printed circuit boards (PCBs)
will ride behind the drill bit, penetrating deep into the
harsh environment of the Earth’s crust, where they will
help performmeasurements to evaluate the rock forma-
tions around them. That information is communicated in
real-time to engineers on the surface who are guiding the
drilling string toward its intended target. Because of the
abusive operating conditions and the harsh penalties for
electronics failures, all components must be thoroughly
vetted and their failure modes understood.
During the qualification of a new board design, a
high-voltage ceramic-chip capacitor had failed by short-
ing itself. This part is rated for high temperature and has
leads soldered onto its terminations with a high-melting-
point solder. In many circuit designs, a single shorted
capacitor may not influence the overall functionality and
may escape completely undetected; however, this par-
ticular component played a vital role, directly impacting
board functionality. Surprisingly, this part had already
undergone qualification studies on a different PCB with
similar operating conditions and had been in use for some
time. So, why had this particular unit failed?
EXTERNAL ANALYSIS
During electrical troubleshooting, a technician had
identified a suspect appearance on the terminations of
this capacitor. When the component was replaced with
a brand new part, board functionality was restored and
testing continuedwithout incident. The failed component
was given an initial optical inspection, which revealed two
points of interest:
• Metal migration from the termination onto the capaci-
tor face was observed from both sides of the part.
• The high-melting-point solder, which held the leads
in place, had clearly reflowed and was showing heat-
stress discoloration.
Both of these conditions can be seen in the optical
images in Fig. 1.
Metal migration via dendritic growth mechanisms is
a threat particularly where high-voltage conditions exist,
such as for this 2000 V capacitor. However, an examina-
tion of all sides of the capacitor quickly determined that
the dendrite growth had not progressed far enough to
create an external bridge between the terminations. To
“OFTEN IN TECHNICAL DISCUSSIONS
WITHIN THE FAILURE ANALYSIS
COMMUNITY, PASSIVE COMPONENTS
ARE OVERLOOKED IN FAVOR OF NOVEL
ANALYSIS METHODS OR EMERGENT
CHALLENGES TO SEMICONDUCTOR
RELIABILITY. ”