At 04:24 AM 11/4/98 , you wrote:
>Bob
>While I agree with the general principle that a power supply connected to
>a motherboard will tend to discharge its capacitors when unplugged, there are
>a few caveats:
>  1.  The power supply may shut down before all the power is drained from
>      ALL the capacitors.
>  2.  Large capacitors that hold high voltages (in this case over 500 Volts)
>      for long periods of time tend to develop a "set" - this is one
>      aspect of the "electret effect".  I have seen a capacitor that was
>      shorted (via a 150 Ohm resistor) to reduce its potential to
>      essentially zero, (this was observed on a meter connected
>      across the capacitor) which was then removed from the circuit,
>      and left standing overnight in a locked room.  The next morning,
>      it had "recovered"  over 60% of its original voltage.
>  3.  "Good" power supply designs will have a small, high resistance
>      resistor across the large, high-voltage capacitors to "bleed"
>      the voltage down to safe levels within a minute or two.  This is
>      NO guarantee of safety, as not all power supplies have a bleeder
>      resistor, and I once saw one that originally had a resistor, but
>      it burned out (OPEN).
>
>Your only assurance of safety is to use a BIG, low-value resistor
>(say 25 Watt, 100-150 Ohm), securely grounded, with an insulated
>handle, to ground EVERY terminal of EVERY significant capacitor
>just before working on the power supply.  Depending on the capacitance,
>it may take several seconds to reduce the voltage to a safe level.
>One time constant ( = Resistance x Capacitance ) will reduce the
>voltage by approx. 63%. [Example: 100 Ohm x 0.01 Farad (10,000 microFarad)
>= 1 second. 1,000 Volts drops to 370 Volts after 1 second, to about 137
>Volts after 2 seconds.]  If the resistor gets hot, it should have been
>bigger.  You definitely DON'T want it to fail!
>
>A lower resistance would be quicker, but runs the risk of causing arc
>scars or damaging other parts.
>

Boyd Ramsay, because this discussion is only about PC power
supply's and the potential for harmful, dangerous or fatal electrical
shock I stand by my words "Please have no fear of a power supply's
internal parts as long as the AC supply is disconnected. This means
UNPLUGGED FROM  THE WALL OUTLET."

I have just examined 3 PC power supply and the largest capacitors
found were 470 micro farad and max voltage measured 175 VDC. This
was in a 250 WATT AT style.

With the power supply sitting on the bench with no load on the output.
The voltage across the 470 micro farad cap fell to 20 VDC in 1 minute,
7 VDC in 2 minutes and 2.5 in 3 minutes. With a load of the mother board,
hard disk, floppy drive and cdrom the times were less than 1/2 of that.

I would find it difficult to belive that a power supply connected to a
working computer going through shutdown, power shut off and
being and unplugged from the wall. Then removing the case removing
the power supply and finally disassembling the power supply so a fan
could be replaced that you would find any dangerous voltages.

What I had to say was only about a PC power supply as indicated
in the subject line. It was expressed because of a message of
indicating that capacitors in the power supply could be dangerous
if not lethal. In this case it is simply not true.



Bob Thrasher

                                  -----
                PCBUILD mailing list -  http://nospin.com
         Bob Wright:[log in to unmask] - Drew Dunn:[log in to unmask]