I just checked my (bad) power supply and it doesn't seem broke

[Metspitzer]

My computer woudn't start. I used another power supply and it started
fine.

I took a multi meter to the "bad" power supply and everything checked
out fine.
I get :
red 5.1
orange 3.3
blue -12
yellow 12
purple 4.1
grey 5.1

My eyes are bad. I didn't see a way to be able to test the SATA
power.

How can I tell if the PS is broke?

 

[Loren Pechtel]

My computer woudn't start. I used another power supply and it started
fine.

I took a multi meter to the "bad" power supply and everything checked
out fine.
I get :
red 5.1
orange 3.3
blue -12
yellow 12
purple 4.1
grey 5.1

My eyes are bad. I didn't see a way to be able to test the SATA
power.

How can I tell if the PS is broke?

The thing is a power supply that can't deliver enough oomph will show
good at low power. It only shows as bad when under sufficient load. I
have never encountered a power supply tester that you can set the load
you want and see what happens.

 

[Metspitzer]

The thing is a power supply that can't deliver enough oomph will show
good at low power. It only shows as bad when under sufficient load. I
have never encountered a power supply tester that you can set the load
you want and see what happens.

The thing is, with nothing but the mobo connected, it twitches. Every
thing is pointing to a bad power supply, but my mistrust in my trouble
shooting will not let me toss the PS.

I have hard drives marked with a sharpie marked "bad"
They have to be marked "really bad" to get me to toss them.

 

[SC Tom]

My computer woudn't start. I used another power supply and it started
fine.

I took a multi meter to the "bad" power supply and everything checked
out fine.
I get :
red 5.1
orange 3.3
blue -12
yellow 12
purple 4.1
grey 5.1

My eyes are bad. I didn't see a way to be able to test the SATA
power.

How can I tell if the PS is broke?

If you get a PS tester, it applies a load to each voltage and gives you a (fairly) accurate assessment as to how it
would work in a normal load situation.

I agree with Loren's reply- even a "dead" D-cell battery will show ~1.5v with a multimeter, whether it's digital or
analog.

 

[Paul]

The thing is, with nothing but the mobo connected, it twitches. Every
thing is pointing to a bad power supply, but my mistrust in my trouble
shooting will not let me toss the PS.

I have hard drives marked with a sharpie marked "bad"
They have to be marked "really bad" to get me to toss them.

Great. Take that hard drive marked as "bad" and connect it to the
power supply. Now you have a 6W to 12W load for testing. Start
the supply using the paper clip trick (PS_ON# to COM). Does
the 5V and 12V rise to full level now ?

I can give you an example, from my very first failed ATX power supply
(which I still have). Lying on the bench, with the paper clip in
place, I get the full 12V. If I connect even one cooling fan (maybe
100mA load), the 12V level collapses under that small loading.
The 12V in that supply is as weak as can be.

I got a "hint" from that supply, before it failed. I had some
constant speed cooling fans on the computer. Then, one day,
I could hear the fans "warbling" a bit. The fan speed would
slowly drift, from its "normal" sound, to a slightly lower
pitch, and then back to normal again. That's the hint that
the supply wasn't entirely healthy. It took a couple months
after that, before it collapsed into the useless wreck it is today.
It does put out good readings under no load, but
that's all it can do. Once you load it, it collapses.

And yes, the red switch on the back, is in the 110V position

Paul

 

[Metspitzer]

Great. Take that hard drive marked as "bad" and connect it to the
power supply. Now you have a 6W to 12W load for testing. Start
the supply using the paper clip trick (PS_ON# to COM). Does
the 5V and 12V rise to full level now ?

I can never find a paper clip. I may actually have to buy some. I
ended up using a bread twist tie.

I can give you an example, from my very first failed ATX power supply
(which I still have). Lying on the bench, with the paper clip in
place, I get the full 12V. If I connect even one cooling fan (maybe
100mA load), the 12V level collapses under that small loading.
The 12V in that supply is as weak as can be.

I got a "hint" from that supply, before it failed. I had some
constant speed cooling fans on the computer. Then, one day,
I could hear the fans "warbling" a bit. The fan speed would
slowly drift, from its "normal" sound, to a slightly lower
pitch, and then back to normal again. That's the hint that
the supply wasn't entirely healthy. It took a couple months
after that, before it collapsed into the useless wreck it is today.
It does put out good readings under no load, but
that's all it can do. Once you load it, it collapses.

And yes, the red switch on the back, is in the 110V position

Paul

I have already ordered another PS. I will try your suggestions.

Thanks

 

[Loren Pechtel]

The thing is, with nothing but the mobo connected, it twitches. Every
thing is pointing to a bad power supply, but my mistrust in my trouble
shooting will not let me toss the PS.

I have hard drives marked with a sharpie marked "bad"
They have to be marked "really bad" to get me to toss them.

I'm paranoid about hard drives. They act up, they're out. I would
only mark a hard drive "BAD" if I was planning to disassemble it for
the magnets or something.

 

[Loren Pechtel]

I agree with Loren's reply- even a "dead" D-cell battery will show ~1.5v with a multimeter, whether it's digital or
analog.

Actually, I had an interesting experience with that a few days ago.

I found a battery laying around that I suspected was dead from age.

As it wasn't an ordinary cell that I could throw in my little tester
that actually puts a bit of load on them I grabbed my meter (digital)
and read the voltage--and watched it tick down, down, down! Apparently
the trivial load of the meter was enough to drain the last bits of
power at an obvious rate.

 

[Metspitzer]

Actually, I had an interesting experience with that a few days ago.

I found a battery laying around that I suspected was dead from age.

As it wasn't an ordinary cell that I could throw in my little tester
that actually puts a bit of load on them I grabbed my meter (digital)
and read the voltage--and watched it tick down, down, down! Apparently
the trivial load of the meter was enough to drain the last bits of
power at an obvious rate.

You can take a sharpie and mark that battery "really bad"

 

[Metspitzer]

I'm paranoid about hard drives. They act up, they're out. I would
only mark a hard drive "BAD" if I was planning to disassemble it for
the magnets or something.

I always kept the drive with the intention of low level formating it,
but I have never tried.