Bit of wire as good as an antistatic strap?

K

kony

...assuming that there's a bleed resistor across the PS caps, and that it
hasn't opened.

No, assuming the system (was) working at all, it had to have
5VSB to turn on... 5VSB does drain the main high-side caps.
I've seen both types of failures, and while I've never
personally been zapped by them I've seen it happen to others -- with much
higher voltages than you will find in today's semiconductor circuits.

If the system "works", it is extremely unlikely there is ANY
cause at all for concern. Remember that we're not talking
about opening the power supply to repair it- only working
inside the system case.

(In one such incident while I was working at a TV station in the mid
1960s was that I learned several Spanish cusswords: the victim was
a refugee from Castro's Cuba, and following the incident he expressed
his opinion of the ancestry and probably destination of the designer
of the circuit that zapped him after it was unplugged.)
Sure, there are plenty of circuits not safe to mess with
until they've been drained. Not so in a PC.
 
A

aleX

kony said:
Beware of kids posting about video cards.
There are NO high voltages, let alone lethal ones, outside
of the power supply casing. I will grant you that someone,
somewhere, might be able to do something really odd to
electrocute themselves with 12V, but it's not at all common
and would require piercing the skin with something
strategicially placed across the heart. Anyone inside the
PC having such an event happen was probably bound to end up
killing themselves one way or another regardless of the PC.
;)


The amperage is fairly high on some cards, but that in
itself isn't a cause for concern, you could literally rub
your hands all over a card and only damage the card.

On the other hand, it is good advice not to fiddle with a
modified card while it's running- too great a change of
shorting something if poking around on it with multimeter
probes or fiddling with a volt-mod potentiometer... though I
admit I've done both of these things myself, without issue,
but still it's a do-so-at-own-risk situation, attention to
detail is most important.





Their caps hold a charge but the onboard power circuit
continues to run- it is not "told" to shut off, it just does
so by running out of power- by draining the caps when the
sytem PSU is shut off.

Even so, there is no voltage on a video card that's over
12V, and the GPU and memory run at much lower, below 2V
these days.

There just isn't any reasonable risk of being electrocuted
by any part in a PC except inside the power supply, not even
a risk of any harm to a human except under very
extraordinary circumstances. The risk is all to the
equipment That is, unless you had some serious power supply
fault in additon to a faulty AC outlet... such things are
already outside of the expected conditions and the power
supply should've immediately shut off- not a situation with
a running system.

Thanks for your in-depth reply, I know a bit more about it all now. The
page URL I referred to is long since forgotten, but was a page about
overclocking a graphics card, possibly one of the fast Radeon's 2 or 3
years ago. I think they were attempting to up the voltage to the card
whilst it was running and they issued that dire warning. It looked like
a respectable site from what I remember, but I'll take it that there's
little danger in graphics cards themselves. Best to switch the power off
before doing anything anyway I'd say...
 
A

AT

Now I'm not TRYING to be insulting - but I've never seen soo much taffy
collectively distributed - much of it coming from computer-wishbook magazines.


Look, I was born with a soldering iron in my hand, my oldest computer is older
than most of you and I started working computers, specifically before a good
2/3rds of you shuffled onto this planet.

Yeh, that was only 1970 and the computer's only a Straight-8, but it sounds like
no one here has any electrical experience and when it comes to electronics, can
build a box, but may well trip over Ohm's Law. I've laid wires from the
microscopic to whole buildings, even read the electrical code.

If you don't buy what I say, please check it out for yourself.


YES and no - the metal's OK, but attach the strap to the GROUND line of the
power/mains *after* assuming the house wiriing is right and you are not about to
incinerate yourself good. Attach any ground strap human-to-case does NOTHING.
Attaching both to a good GROUND works.

The first good computer investment is one of those $2.50 3-neon bulb line
checkers - if you don't get the right lights (2 green, no red) fix the outlet
first or fry everything.> I have seen many a ground line go nowhere, and many a
ground line tied hot or neutral by Bad Electricians. Suspect Every Outlet until
checked.

Advanced APC Back-UPS and better models do this for you, but I wouldn't risk one
of those babies without checking the outlet first - Worst Possible Case - if the
outlet is wired to 2 hots and a ground, you JUST MIGHT hurt the outlet tester,
instead of blowing yourself or your gear to bits.

So test, then connect yourself to a ground, via wire or groundstrap!

Second good idea: Get an old unwanted power cord and cut off the POWER prongs on
the plug, LEAVING THE GROUND - plug this into the machine PS and a tested outlet
to pull charge off the machine - even if you draw a spark touching a grounded
part of the box, you may do no harm.

Better way is to slit the insulation, use a meter to check the ground wire
(usua,ly green) and cut about 4 cm of each of the other two wires and toss.
Tape over with good electrical tape, GREEN if you can get it so you never try to
use it as a plug again and wonder why it's not working. It just plugs in faster
that way, though.
No, because anti-static stuff is only poly that's been given a coat of a
non-hazardous-to-equipment suffactent (detergent-like stuff). Spraying the table
with the same stuff (sold in US as Static Guard for clothing) the floor around
the area and long sleeves if you are working on hot days (try ALL-COTTON
clothing) helps. In really static-sensitive areas, either lay down a static mat
or heavy aluminum foil and, again, run it to ground. Don't pay more than $20 US
for a static mat or you're being ripped off. OR work on a metal work surface,
where you can drill a hole, bolt through a wire and attach to a ground.

Save the pink and metal-flashed bags for storage!

As I said, only if they're part-whole synthetic or fur/leather. Cotton sprayed
with Static Guard is fine.

TRUE - you put the mobo in its bag in its box before doing anything AND you do
not put it on the metal - you CAN put it on top of its bag on a static mat or a
piece of cardstock sitting on one if afraid of shorting the battery, though I do
not *BELIEVE* <so I don't do it myself> that the lines for the small battery
would be run on the lowest layer of the mobo.

But just the contact with free-floating anti-static stuff doesn't drain the
charge - the charge has to have a place to flow, and that, sinto the ground. For
the real paranoid, wrap more grounded copper around your sleves if you need to
keep warm.
Suffactent - good idea, though the spraycan stuff is easier to apply. You end up
with a sticky work surface or floor!

Agreed, BUT most of today's products above the chip level can be handled without
too much care. That's not an excuse for getting sloppy.

Static discharge damage would cause that form of failure known as "infant
mortality" - dead within 2 weeks. All semiconductors (and, for that matter
vacuume tubes) and associated parts suffer from a MTBF (mean time between
failure) that looks like an inverted U with the bottom stretched for about the
length of a page |__________________etc_________| or something.

For the first month or so, everything is prone to failure due to bad
transistors on the dice to bad solder joints. RUN YOUR EQUIPMENT HARD for the
first few weeks, that is, no overclocking, just 24/7 running Seti@Home or
similar apps AFTER 2 days of BURN IN, either via program or, these days, BIOS
setting.

You may blow a memory loclation or even a CPU during that period, and
the seller had better take it back, because it won't be your fault if you follow
the plan. After two day's heavy running (checking regularly for heat spots,
exploding capacitors, the smell of frying epoxy board...) followed by 30-90 days
24/7 just running, your system should last you years, or at least through Total
Obsalescence <some exceptions exist, but, I mean Seagate even guarantees its
DRIVES for 5 years now, and I wouldn't use anything else, save Hitachi DeskStars
designed by IBM>

Trucker Al's the handle, be seeing you around
(And anyone that works around electronics with a copper wire
strapped to a body part is an idiot. )


Yeah, just the thought of that wire wrapped around my wrist and no way for
my buds to get it off quickly, while I flop round the floor like a fish out
of water, sorta scarey.

The number one static producing machine is my wheelchair.....after running
round Best Buy for ten minutes, I could put all thier floor model computers
out of buisness. LOL Once I through an arc of 4 inches to my granddaughter
hand in the airport.
 
J

JAD

whoa speakin bout a WHOLE LOT of TAFFY...............


AT said:
Now I'm not TRYING to be insulting - but I've never seen soo much taffy
collectively distributed - much of it coming from computer-wishbook magazines.


Look, I was born with a soldering iron in my hand, my oldest computer is older
than most of you and I started working computers, specifically before a good
2/3rds of you shuffled onto this planet.

Yeh, that was only 1970 and the computer's only a Straight-8, but it sounds like
no one here has any electrical experience and when it comes to electronics, can
build a box, but may well trip over Ohm's Law. I've laid wires from the
microscopic to whole buildings, even read the electrical code.

If you don't buy what I say, please check it out for yourself.


YES and no - the metal's OK, but attach the strap to the GROUND line of the
power/mains *after* assuming the house wiriing is right and you are not about to
incinerate yourself good. Attach any ground strap human-to-case does NOTHING.
Attaching both to a good GROUND works.

The first good computer investment is one of those $2.50 3-neon bulb line
checkers - if you don't get the right lights (2 green, no red) fix the outlet
first or fry everything.> I have seen many a ground line go nowhere, and many a
ground line tied hot or neutral by Bad Electricians. Suspect Every Outlet until
checked.

Advanced APC Back-UPS and better models do this for you, but I wouldn't risk one
of those babies without checking the outlet first - Worst Possible Case - if the
outlet is wired to 2 hots and a ground, you JUST MIGHT hurt the outlet tester,
instead of blowing yourself or your gear to bits.

So test, then connect yourself to a ground, via wire or groundstrap!

Second good idea: Get an old unwanted power cord and cut off the POWER prongs on
the plug, LEAVING THE GROUND - plug this into the machine PS and a tested outlet
to pull charge off the machine - even if you draw a spark touching a grounded
part of the box, you may do no harm.

Better way is to slit the insulation, use a meter to check the ground wire
(usua,ly green) and cut about 4 cm of each of the other two wires and toss.
Tape over with good electrical tape, GREEN if you can get it so you never try to
use it as a plug again and wonder why it's not working. It just plugs in faster
that way, though.
No, because anti-static stuff is only poly that's been given a coat of a
non-hazardous-to-equipment suffactent (detergent-like stuff). Spraying the table
with the same stuff (sold in US as Static Guard for clothing) the floor around
the area and long sleeves if you are working on hot days (try ALL-COTTON
clothing) helps. In really static-sensitive areas, either lay down a static mat
or heavy aluminum foil and, again, run it to ground. Don't pay more than $20 US
for a static mat or you're being ripped off. OR work on a metal work surface,
where you can drill a hole, bolt through a wire and attach to a ground.

Save the pink and metal-flashed bags for storage!

As I said, only if they're part-whole synthetic or fur/leather. Cotton sprayed
with Static Guard is fine.

TRUE - you put the mobo in its bag in its box before doing anything AND you do
not put it on the metal - you CAN put it on top of its bag on a static mat or a
piece of cardstock sitting on one if afraid of shorting the battery, though I do
not *BELIEVE* <so I don't do it myself> that the lines for the small battery
would be run on the lowest layer of the mobo.

But just the contact with free-floating anti-static stuff doesn't drain the
charge - the charge has to have a place to flow, and that, sinto the ground. For
the real paranoid, wrap more grounded copper around your sleves if you need to
keep warm.
Suffactent - good idea, though the spraycan stuff is easier to apply. You end up
with a sticky work surface or floor!

Agreed, BUT most of today's products above the chip level can be handled without
too much care. That's not an excuse for getting sloppy.

Static discharge damage would cause that form of failure known as "infant
mortality" - dead within 2 weeks. All semiconductors (and, for that matter
vacuume tubes) and associated parts suffer from a MTBF (mean time between
failure) that looks like an inverted U with the bottom stretched for about the
length of a page |__________________etc_________| or something.

For the first month or so, everything is prone to failure due to bad
transistors on the dice to bad solder joints. RUN YOUR EQUIPMENT HARD for the
first few weeks, that is, no overclocking, just 24/7 running Seti@Home or
similar apps AFTER 2 days of BURN IN, either via program or, these days, BIOS
setting.

You may blow a memory loclation or even a CPU during that period, and
the seller had better take it back, because it won't be your fault if you follow
the plan. After two day's heavy running (checking regularly for heat spots,
exploding capacitors, the smell of frying epoxy board...) followed by 30-90 days
24/7 just running, your system should last you years, or at least through Total
Obsalescence <some exceptions exist, but, I mean Seagate even guarantees its
DRIVES for 5 years now, and I wouldn't use anything else, save Hitachi DeskStars
designed by IBM>

Trucker Al's the handle, be seeing you around
 
R

Ralph W. Phillips

Howdy!

AT said:
Now I'm not TRYING to be insulting - but I've never seen soo much taffy
collectively distributed - much of it coming from computer-wishbook magazines.


Look, I was born with a soldering iron in my hand, my oldest computer is older
than most of you and I started working computers, specifically before a good
2/3rds of you shuffled onto this planet.

You almost sound like me.
Yeh, that was only 1970 and the computer's only a Straight-8, but it sounds like
no one here has any electrical experience and when it comes to electronics, can
build a box, but may well trip over Ohm's Law. I've laid wires from the
microscopic to whole buildings, even read the electrical code.

If you don't buy what I say, please check it out for yourself.



YES and no - the metal's OK, but attach the strap to the GROUND line of the
power/mains *after* assuming the house wiriing is right and you are not about to
incinerate yourself good. Attach any ground strap human-to-case does NOTHING.
Attaching both to a good GROUND works.

Err - Actually, it's not important that you're connected to an EARTH
ground, but that all the equipment AND you are at the same potential level.
So it makes more sense to connect to the case, and use it as the tie point.
Yes, I'd connect it to Earth also ... but that's secondary.

You sure you thought about this? It's not the absolute voltage, but
the RELATIVE voltage, that causes problems.

RwP
 
B

Bob

Err - Actually, it's not important that you're connected to an EARTH
ground, but that all the equipment AND you are at the same potential level.
So it makes more sense to connect to the case, and use it as the tie point.
Yes, I'd connect it to Earth also ... but that's secondary.
You sure you thought about this? It's not the absolute voltage, but
the RELATIVE voltage, that causes problems.

Are you taking into account that your machine is connected to the
other machines on your LAN, and they are tied to Earth ground.


--

Map of the Vast Right Wing Conspiracy
http://home.houston.rr.com/rkba/vrwc.html

If you can read this, thank a teacher.
If you are reading it in English, thank an American soldier.
 
A

Al Dykes

Are you taking into account that your machine is connected to the
other machines on your LAN, and they are tied to Earth ground.


CAT5 Ethernet is isolated and can't provide a ground path.

IMO when working on a PC you should pull the power cord and if one
feels the need to use a power strap it should be clipped to the
equipment you are working on.

When working on equipment large enough to have hardwired power
you obviously can't do this, but OTOH, this kind of equipment and
it's service provisions given much more thought in design.

IMO conecting a body part to any point other than the chassis I'm
working on, and going to touch anyway, is a no-no from a safety
standpoint.
 
B

Bob

CAT5 Ethernet is isolated and can't provide a ground path.

CAT5 is nevertheless connected to the other machines.

What do you think would happen if I isolated my computer from ground
(with an unpluged UPS) and put 1000 volts on it? My machine would be
1000 volts higher than your machine, and I suspect something would
happen.


--

Map of the Vast Right Wing Conspiracy
http://home.houston.rr.com/rkba/vrwc.html

If you can read this, thank a teacher.
If you are reading it in English, thank an American soldier.
 
A

Al Dykes

CAT5 is nevertheless connected to the other machines.


CAT5 ethernet is isolated by transformers at each end. There is no
copper path between the equipment and even if it did provide a DC path
it's up to 300 ft of small gauge copper wire which whould make a
really crappy drain path for a static impulse.
 
K

kony

CAT5 is nevertheless connected to the other machines.

What do you think would happen if I isolated my computer from ground
(with an unpluged UPS) and put 1000 volts on it? My machine would be
1000 volts higher than your machine, and I suspect something would
happen.

Yes I think "something" would happen too if you put 1000
volts on your computer.

HOWEVER, consider the typical notebook. It's isolated from
earth ground and networks to other systems.
 
A

AT

Err - Actually, it's not important that you're connected to an EARTH
ground, but that all the equipment AND you are at the same potential level.
So it makes more sense to connect to the case, and use it as the tie point.
Yes, I'd connect it to Earth also ... but that's secondary.

You sure you thought about this? It's not the absolute voltage, but
the RELATIVE voltage, that causes problems.

RwP
Yes - there's a problem which is sort of like.... Everybody's at relative
voltage 0 to one another, but all that anti-static film and zero connections to
a real-world ground allows a static charge to slowly build and.......then
somebody touches something connecting the high-potential to a real ground and
all those good electrons stream out.....

Again, I actually find the ground case is overstated unless handling a bare CPU
or memory block, but precautions cannot hurt. Assembled gear, on the level of a
disk drive, *usually* contains components that will block any static problems or
incorrect voltageon pin problems within reason.

TrAl
 
A

AT

Are you taking into account that your machine is connected to the
other machines on your LAN, and they are tied to Earth ground.

Of course it is when operating! when a "system" consists of a bare mobo, empty
case and assorted other bits and pieces, it is difficult to assume a system
ground.

Once a system is up and running, it is virtually in a Faraday Cage and a nearby
lightning bolt will have to put out one hell of an EMP to effect it.

TrAl

<btw> on your sig - WW II, mebee, Korea, Viet Nam, Guatamala, Hondouras,
Niquaragua, Yougoslavic bits, Iraqq? bases from pole to pole and an empire where
the sun never sets that cost trillions of taxpayer dollars to *maintain*??

Soldiers deserve honor for giving their time and risking their lives when, no
matter how misguided, their leader sez GO!

But don't throw around jingoistic memes to compress something deserving LOOOONG
thoughts into a Madison Avenue line.
 
A

AT

I think it's always a good idea to do the 'extra press' thing with the
power button to drain any charge out through the switched-off wall
socket as someone stated earlier. Not sure about computer components,
but things like guitar effects pedals and televisions can hold charge
for a long time (many days in my experience).
Anything since a Pentium I has a problem - instead of asking PS makers to add
reliable1.2, 1.5, 2 volts, the idiot ATX standard calls for the lower voltages
to be produced by on-board regulators which 1) throw off half the waste heat in
the box and 2) leave large caps on the board that are not self-draining.

Additionally, many cheap power supplies (why I insist on PC Power&Cooling though
I don't get anything for it but higher bills) do not install bleeder resistors
across the 100 MFD high-voltage caps necessary in a modern switcher.

That's why mobo makers started putting LEDs on boards and telling you to unplug
and then turn "on" until the led goes out.

TrAl
 
A

AT

I did read on a graphics-card modding page that high-end cards can build
up a lethal voltage (or should I say ampage, not sure) and shouldn't be
messed with when the PC is on - a bit like TV sets. Of course, you
wouldn't want to go messing with the mains transformer when the power is
on either! I assume there is no charge stored by these cards, and hence
no risk when the power is off, even fleetingly?

CRAP - to put it politely.

The MONITOR can, if it's a CRT.

There *is* a lot fo danger plugging and unplugging any card when the machine's
on. I think what the person might have misinterpreted is the need of some
soper-cards for a direct feed from a separate PS rail (a Molex 4-pin power plug)
because it could not be run off the wattage supplied a standard AGP or PCI-EX
socket.

TrAl
 
A

AT

It very much is a concern. I remember when I first started work, my
boss warned me what had happened to my predecessor. We had a PDP8,
which was rack-mounted with wire-wrapped connections all over the
backplane accessible at the front of it. Only low voltage - 5 & 12 iirc,
so perfectly safe. Well, no. He was working on the panel, live, when
his wedding ring shorted the 5V rail to the neighbouring earth. Lots
and lots of amps welded his ring in place; and it was getting hot.....
Apparently his finger did survive the mishap. But the point is that
high-current supplies still need respect, even at low voltage.

Letsee, the original PDP-8 PS, linear of course, weighed 125 lbs and drew 12 A
or so.

The main PS for the PDP-15, similar arrangement, only put out 5V for the logic,
and a few small random voltages for the core memory.
The main PS for the CPU and up to 16 K -words of memory (maybe 32 if you
squeezed it) weighed @150 lbs and took in a minimum 20 amps at 110 volts, so
we'll call that 2,000 watts available at 5V DC - only 400 amps at each the
1-per-card-rack power regulator, more than enough to cause a slow painful death
running through your body.

Whereas the old game of using a 9V battery reversed speaker micro transformer,
buzzer (to turn DC into ac and some metal tape will produce a cute box
delivering 15-20,000 volts at such lo amperage that you can't put it down, but
it sure does sting harmlessly.

TrAl
 
A

AT

yes there should be an inverter circuit though most flat
displays are not user servicable, a technician hoping to
service one ought to already be able to assess the risk else
they probably aren't capable of the service at all.
I think that's the balast/inverter for the flourescent lamp that back-lights the
panel, nothing to do with the panel itslef.
 
K

kony

I think that's the balast/inverter for the flourescent lamp that back-lights the
panel, nothing to do with the panel itslef.


Yes, and no... the whole product is ALL the parts in it, and
the lamp is one of those more likely to need serviced at
some point due to fading.
 

Ask a Question

Want to reply to this thread or ask your own question?

You'll need to choose a username for the site, which only take a couple of moments. After that, you can post your question and our members will help you out.

Ask a Question

Top