Break the law by being a daredevil? (Curiosity questions o..

External Since: Dec 08, 2004 Posts: 310

mike3 wrote:

> http://groups.google.com/group/alt.comp.hardware.pc-homebuilt/msg/cb4d...b7145dd
> " Working on high voltage equipment is always dangerous due to the
> possibility
> of fire and or electrocution. That's why its sealed with a warning and
> that's why only those with the proper training should work on such
> devices.
> ************* There's also a criminal and financial responsibility
> involved. *************
> So even if it doesn't seem like much the consequences can be
> *tremendous*."

Check www.repairfaq.org for information about electronics repair and
safety. It's great.

The danger of shock isn't that bad if the PSU is unplugged from the AC
cord. High voltage capacitors can remain charged if the PSU has
failed, because the failure may have blown a component or copper
trace, removing any path to discharge the capacitors. But a normal
working PSU will discharge in 30 seconds or less due to its circuitry
loading down the high voltage capacitors, but even if that doesn't
happen, there are almost always bleeder resistors that will discharge
them in 30 minutes.

Never open a PSU unless it's unplugged from the AC.

Never touch anything inside a PSU without first measuring the voltage
on it because any exposed metal can be at high voltage, including the
heatsinks. I've measured 170V DC on PSU heatsinks. Discharge all
voltage to the case by connecting a wire or resistor between the case
and the device, being sure to make the first connection to the case.

Never plug a PSU to the AC unless its cover is securely installed
(tighten all screws). The cover provides protection against electric
shock and exploding capacitors.

Never plug a PSU into an AC outlet that isn't grounded or protected by
a GFCI, and don't assume that an outlet is grounded just because it
has 3-prong sockets because some people replace broken 2-prong sockets
with 3-prong ones, as they're easier to find. The only situation
where it's acceptable to use an ungrounded outlet is when it's
protected by a GFCI, but beware that GFCIs can go bad without any
indication (IOW press the GFCI's test button, and verify that the wall
outlet has no voltage.

> Also, I heard that PSUs are simply not meant to be
> repaired, period, not even by a computer repair professional:
>
> http://pcsupport.about.com/od/safetyconsiderations/qt/safety_tips.htm

It's just not economical to have a computer shop repair a PSU because
the typical minimum labor rate is $60-100,.or more than the cost of a
typical brand new PSU. Also the average person working in computer
repair doesn't know how to do many actual repairs and just replaces
whole assemblies (PSUs, motherboards, cards, etc.). But it's silly to
say PSUs aren't meant to be repaired because I've repaired several,
and I'm certainly no technician. In fact a PSU may be the easiest
computer device to fix because so many of its components are generic
and are soldered in conventionally, not surface mount soldered.

Typical PSU failures are from bad fans, shorted transistors or diodes,
bad solder connections, and rotted capacitors (almost always the low
voltage ones, but sometimes a high voltage one in the +5Vstandby
circuit). The power supply forum at www.badcaps.net has detailed
information about the latter and how to fix it.

>
> "Never Service the Non-Serviceable
> When you come across labels that say "No serviceable components
> inside" don't take it as a challenge.
> Some parts of a computer are just not meant to be repaired,
> ********* even by most professional computer repair persons
> ********* .
> You will usually see this warning on power supply units but you may
> also see them on monitors, hard
> drives, optical drives and other dangerous or highly sensitive
> components."
> (emphasis mine also.)
>
> How accurate is any of this information I've been accruing?
>
> (PS. I suppose the warning about hard drives only applies if you care
> about the data on the drive, not,
> for example, experimenting with an old junker that was acquired
> specifically for that purpose.)

External Since: Jan 20, 2008 Posts: 20

On Feb 16, 7:42 pm, Paul <nos... DeleteThis @needed.com> wrote:
> mike3 wrote:
> > There's also a criminal and financial responsibility involved.
>
> This refers to what happens if you repair an ATX PSU for your
> friend. Your friend goes home, plugs in the computer, and
> goes to bed. Later in the night, the supply catches fire.
>
> Much later, the fire marshall and the insurance company, want to know
> who modified/played with the supply.
>
> Your friend's mom decides to sue, to help get money to rebuild the
> house or pay for medical treatments for your friend. She has to
> do something, after all, because the insurance was cancelled.
>
> Even if you do stuff to your own power supply, you still
> might run the risk of invalidating a fire insurance claim,
> if the insurance company investigates and decides a change
> to the PSU caused the fire.
>

Ah. This makes more sense.

> Anything in the computer which is less likely to
> catch fire, and burn down your house, has a lesser
> risk associated with it. (For example, changing
> sticks of RAM.)
>
> Note that getting advice from USENET, you're going to
> get lots of goofy, slightly off answers. Respondents
> don't really care what happens to you. Remember
> that, before you reach for that screwdriver...
>

Well, I wasn't planning on messing around with a PSU
like that anyway: like I said, these are curiosity questions.
Furthermore do you know of a source other than USENET
that is more accurate and is affordable to someone who
has less than $15000/year income?

> Paul