Richard said:
While going to college the second time, I worked as a production test tech
in the era of discrete TTL devices, but before much recognition was given to
ESD control. After a while, I began to recognize a common type of failure,
in which a 74LS04 inverter would got sort of half-on. This typically
occurred where the output in question was routed directly to a card I/O
connection finger. I found my time was most efficiently used by getting the
assembler to replace all the 04's that were connected off-board on
assemblies that had that problem.
you too huh?
I have seen exactly that problem, on videogame logic pcbs. Between my
1st and 2nd stints at Uni I spent 3 years as a video game service tech.
Absolutely nobody in the industry used antistatic anything - bubblewrap
was the packaging of choice. A guy I know in Wellington switched to
antistatic bags & handling, and saw his PCB failure rate halve. He then
replaced all the cheap shitty smps with decent 300W PC supplies, and the
failure rate plummeted to near zero - 1-2 PCBs per year, cf 1-10 per week.
Once I got my engineering degree I went to work for a power electronics
company. We were pretty good with antistatic, but when we moved to SMT
and the volume went up, our failure rate stayed constant, and became a
problem. So we went to town on static - conductive jackets, the works -
and watched the failure rate fall almost tenfold. designing products to
greatly reduce handling also helped a lot.
even so, my first control board was susceptible to static. We used a
Samtec bottom-entry connector on the control PCB, which had the PCB pins
folded over the outside of the housing. This turned out to be a
convenient handle, but one pin was directly connected to the micro. The
first build of about 200 units had about a 10% failure rate on this pin,
due entirely to static. despite foot & wrist straps, mats etc. I added
ESD protection to the next run, and the problem disappeared completely.
That was before the conductive jackets though.
It was important there, because zapping a $0.30 part could easily
destroy $20,000 worth of power electronics.
Note: after removing the baddies and soldering on the new parts, the
assembler would clean the board in an ultrasonic tank full of TCE, then
bring it to me still dripping.
yegods! betcha aint so happy about that nowadays.
Cheers
Terry
