Hello AAC people.

I have here a full modular OCZ ZX-1000W high-end(ish) power supply that trips room fuses when switched on. There's a crack in psu and click in a fusebox. No loose magic pixie smell tho.
Opened it up, no obvious damage, caps not swolen. Brought it to a computer "repair" shop, thay say its not fixable. When asked why they reply fet's shorted. No further info. Didnt get to speak to actual person who tested/diagnosed it.

This power supply is not exactly cheap to replace, i still have hope i can repair it. Repaired some power electronics in the past, but only with obvious visible faults.

Will be doing another thorough inspection of the unit later today or tomorrow and would appreciate any advice on diagnostic and repair of ATX power supplies.

Thanks.

P.S. This psu was running watercooled gaming rig with overclocked i7-2600k and two overclocked GTX480 boards and was pulling about 750-800W from the wall at full tilt. Could heat my room with it

 

High end ATX power supplies are complicated animals with virtually no chance of getting a schematic.

You can end up going round in circles, blowing newly replaced components and getting no where fast.

It doesn't sound like a faulty electrolytic but you could just try changing them but in my experience I'd write it off and replace it.

 

You could check the bridge rectifier, FETs, caps and all diodes.
Do you have a capacitance and ESR meter to measure the caps? The ESR as well as the capacitance is important.
While testing, run with a mains light globe in line to protect the fuses and other bits.
BUT BE CAREFUL!!!
These things are lethal. Don't forget to discharge the caps before touching them.
If you are in doubt on how to proceed, DON'T!

 

You could check the bridge rectifier, FETs, caps and all diodes.
Do you have a capacitance and ESR meter to measure the caps? The ESR as well as the capacitance is important.
While testing, run with a mains light globe in line to protect the fuses and other bits.
BUT BE CAREFUL!!!
These things are lethal. Don't forget to discharge the caps before touching them.
If you are in doubt on how to proceed, DON'T!

Sadly i dont have a capacitance or esr meter. The last big thing i fixed was a power supply for a heavy duty treadmill with AC drive motor, fancy controls etc. I know the risks of messing with mains power though. Took me a while (couple of years on and off ? ) reading, posting etc. But that particular bit of equipment was so out of "standard" and no schematics that i had to improvise on the fly. Went through it part by part, desoldering bits, testing to part spec (with uni-t UT61E dmm), soldering back in... until it worked. Learned a lot. Now i think i have another project. And since i need a beefy power dupply for my lipo charger (hate charging at 1C) figured might as well try to resurrect this PSU. Current ATX PSU i modded into "lab" psu i use for charging 3s lipos wont give enough juice to fast charge - psu is 250W, 15A @12V - and charger complains if i try to charge faster than 1-2C.

So far tested only onboard fuse and NTC inrush limiter - both checked out ok.

P.S. The pic of insides of this psu , off the net. Will do photos later.

 

As well as some "real" test gear, I got one of these...
https://www.ebay.com.au/itm/LCD-Com...703676&hash=item1a354b17b8:g:z5YAAOSw6nNZuKA7
and they are pretty good for the money. The accuracy may be a bit off but for comparison they are great.
Have a look at..

I note the price has gone up a lot!
And these are VERY good value for a meter..

EDIT: The component tester does ESR when it measures the capacitance.
It is a real handy little device.

 

Here is a web site with several Atx circuit diagrams. If it's tripping the mains breaker straight away, sounds like there is a short across the live/ neutral or the output of the bridge rectifier which then feeds the self oscillating fet to produce the 5v standby.

Ideally you need your circuit diagram, but looking at other circuits can help .

Take pictures of your psu and see which IC they use.

http://danyk.cz/s_atx_en.html