geezer said:
Is static electricity always accompanied by an audible (even if
slightly) spark (arc)? I guess what I am asking is if I ground myself
each time before I touch the computer innards, can I be sure I have
eliminated static electricity because I don't hear or feel anything?
IOW, can static electricity flow from me to a metal computer innard
without my knowing it? While I am simply standing there holding onto
a metal computer part?
If you think this is a dumb question - just ignore it.
Not a dumb question... the details can be a bit surprising and
counterintuitive.
The typical static charge that you get zapped by (i.e. that you notice)
is in the thousands of volts, sometimes tens of thousands. Under dry
conditions and with the right carpet, an evil pair of shoes, and
malicious intent, you can build up a charge of 50,000 V for maximum
shock infliction on an unsuspecting human. I suggest using a housekey
as your contact point if you don't want to suffer the shock yourself.
Not that I'd know anything about this first-hand, really I'm a nice
guy...
The threshold for feeling static shock is about 2,500 V, while it only
takes around 400 V to fry electronics. This is a pretty big margin of
"dangerous" static shock that can sneak under your radar. Similarly,
you can't depend on seeing or hearing the shock if you're not working in
a dark and quiet room.
So the answer to your question is, no, you cannot depend on your senses
to alert you to static shock hazards to electronics.
For the most part you can get away with just a few precautions for
working on computers -- don't wear sweaters, fleeces, anything wool, or
anything else that is obviously static-prone, and don't do computer work
on a carpet. If your chair is cloth, don't scootch around in it while
working, or better yet, don't use it at all, or treat it liberally with
anti-static spray. Ground yourself to the equipment by touching the
metal case or power supply before contacting electronics.
Some manuals (mostly big-iron equipment like Sun, HP, IBM) demand
keeping the power cords connected at all times in order to ground the
chassis while working on it, but other manuals make the opposite demand
in order to ensure there is no power.
Of course, the safest solution is to use a grounding-strap at all times,
connecting your wrist to the computer chassis... but frankly, for PC
work, I've never seen anyone bother. Even for big-iron equipment, this
precaution seems to vary depending on the value of the boards being
thrown around. Just some PCI NIC's or such? Bah... But if it's a
system board worth more than the average Ferrari? Well then, strap
yourself in.
Years ago I had an interesting problem at a small office where I was
sysadmin/netadmin/PC support... we had just bought a couple of cisco
Catalyst 2950G ethernet switches to tie the whole office together, and
this was a major purchase for a small company like this. Soon after
installation, however, I was losing about one port a week, completely
dead. The damage worked out to about $400/port as I recall. Most of
the incidents were associated with one particular employee who used a
laptop and re-plugged his ethernet cable every morning. After much
guessing, cable swapping, and equipment juggling, we finally figured out
that the problem was static discharge -- the whole office was carpeted,
and this employee was unknowingly zapping the end of the ethernet cable
when he grabbed it every morning to plug it in. Given the 400-2500 V
"danger zone" of unfeelable static shocks, this is understandable, but I
was surprised at how fragile the cisco ports were. The solution was a
hack, but it worked: Spray the entire carpeting of the office with 5:1
water / fabric softener solution (re-applied a few times a year, a pain)
and inform laptop users to avoid touching the metal pins on the ethernet
cables. With today's technology, of course, wireless is a better
solution for the laptop workers.
Hope that helps,
-SK
