Reminds me of a problem we had years ago. We were building circuit cards for
GE Medical and they were seeing similar cracked solder joint caused field
failures.

They loaned us a shake table and the software they wrote specifically for
the boards we were building. We loaded boards into the fixture and ran the
profile and some parts were actually popping off the boards! Funny thing, it
wasn't the larger heavier parts we thought would be the problem. GE had to
come in and "tweak" their profile and we experienced no more failures as
long as we ran the program.

GE still had the same fractured solder type field failures & they finally
attributed them to equipment vibration and not manufacturing defects. I
guess it's all about resonant frequencies.

Also, I had a laptop at work and was required to take it home every day for
security reasons. So it rode home & back, 45 miles each way in the trunk of
my Harley-Davidson. I had 4 keyboards fail and every one had fractures in
the ribbon cable right where it plugged into the motherboard. I never did
explain resonant frequencies to IT.

Tom

From: David Beierl [mailto:dbeierl@gmail.com] On Behalf Of David Beierl
Sent: Monday, February 28, 2011 12:39 AM
To: Tom Hargrave
Cc: vanagon@GERRY.VANAGON.COM
Subject: RE: Unbelievable, was Re: Spare ECU

At 01:08 AM 2/28/2011, Tom Hargrave wrote:
>The solder fatigue you took pictures of is common for electronics
>that have been in a vehicle for a long time.

Ayuh. And not by any means just automotive gear.  So why aren't there
more images out there?

>  usually the parts that fatigue are heavier power resistors, caps
> or inductors. They seem to be more susceptible than smaller, lighter
parts.

To amplify on that somewhat:  heavy parts, parts that stand away from
the board, parts subject to thermal cycling and most particularly
parts with pronounced thermal cycling that are mechanically connected
to a stiff external structure (like the devices connected to the heat
sink in our ECUs).  But the curious thing I'm finding in the ECUs is
bad joints in small light parts that sit right on the board, in
low-power DIP leads, in the flattened-ribbon connections between
boards.  I find that surprising.

Yours,
David

  _____

No virus found in this message.
Checked by AVG - www.avg.com
Version: 10.0.1204 / Virus Database: 1435/3472 - Release Date: 02/27/11