Thread: UPS ?
View Single Post
  #3  
Old February 21st 18, 01:55 PM posted to alt.comp.hardware
philo
external usenet poster
 
Posts: 1,309
Default UPS ?

On 02/20/2018 10:01 PM, Paul wrote:

snipped for brevity

Ways to turn on the PC:

1) Busted power switch. Front button busted in the ON state.
Â*Â* But the circuits near the SuperI/O can be edge sensitive and
Â*Â* actually a busted switch like that won't turn on. On some PCs,
Â*Â* when you depress the button the PC doesn't come on, then when the
Â*Â* momentary button on the front is released, the PSU comes on
Â*Â* right after that. A modern motherboard is more likely to "behave
Â*Â* the complicated way".

2) The BIOS has a "Power Loss" setting, that can "Restore power to
Â*Â* previous state". If the PC was on when the AC went off, the PC will
Â*Â* try to turn on again, if the BIOS was set that way.

3) The motherboard will also switch on, if the Southbridge (PCH) is under
Â*Â* electrical stress. When an IDE cable on my machine was "half unplugged"
Â*Â* so only one half the pins touched, the machine turned itself on.
Â*Â* The problem corrected itself after I re-seated the IDE cable. This
isn't
Â*Â* clever chip design, and it's just a side effect of the overloading
Â*Â* caused on the I/O pads.

*******

The Patriot would be a bad UPS, if it showed a fault code because of
a couple milliamps of leakage. The filter caps on the front end of the
ATX PSU, they "leak on purpose" into the Safety ground. This is by design.
And also by design, you can't run a PC off the power plug underneath
your shaving mirror in the bathroom (because the leakage detector there
is not intended for that level of leakage).

Components C2 and C3 here, are examples of capacitors that leak 60Hz
from the AC rails, back into Safety ground. They do it this way, to
filter switching noise currents so they don't go back into rail. The
level of current flow is a function of the capacitor value, and the
capacitor value has some tolerance on it (could be -80%, +20% for example).

http://www.pavouk.org/hw/en_atxps.html

Now, if the supply was actually defective, and the magnitude of the
Safety current flow was much larger, then the Patriot would be
a Good UPS :-) It's all a matter of interpretation as to whether
any Fault detection is too sensitive.

I would take my cheater cord, with the three exposed wire loops
sticking out of it, and put my clamp-on AC ammeter on the Safety
ground wire, and check the current flow level when powered from
the wall. My meter isn't sensitive enough to detect a normal
level of current flow (1 or 2 mA), but if the PSU was damaged,
and more than normal current went into Safety Ground, I could
detect that. The clamp-on ammeter means I'm not going to come
in contact with the circuit while I work.

I would prefer not to mess around with "opening" the Safety Ground
and trying to measure the current flow the old fashioned way with my
Simpson. That's an option if it's all you've got.

The PSU will work, even if the Safety Ground is open. However,
when you touch the chassis of the PSU under those conditions,
you'll get a (non-fatal) shock. The leakage level is intended
to be "safe but annoying" for dry hands. The leakage level is
too much for taking the ATX PSU into the bath tub with you.
Anything could happen then.

If you own an outlet tester, you could also test
that the outlet the Patriot is connected to, shows a
"good" code. At times, those can be hard to find,
and in the style you might like. I managed to get one
at a local hardware store, but I had to go all over town
to find it. I know I have another one around here... somewhere.
These can use things like neon bulbs to detect reverse
hot and neutral, missing Safety ground and so on. There's
usually a table of all the codes to help you interpret how
the outlet wiring is screwed up.

Â*Â* Paul





Thanks Paul.

It seems to be something to do with the mobo's behavior with a PSU other
than the original.

The mobo has a 20 pin power connection and the original supply had a 24
pin connection (with the 4 pin connector left dangling of course.)

I replaced it with a simple 20 pin PSU


I know that there the 20 pins of a 24 pin PSU are somehow a bit
different than that of a straight 20 pin PSU because when I test 24 pin
supplies with my 20 pin tester, one of the red warning lights goes on
even though the supply is good.