Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

[myaccount]

I have a Fortron FSP400-60PFN power supply that ran for 11 years before it died recently.

Actually, I didn't think it had died because I tested it with a cheap PSU tester and it powered on fine. So I thought it was the motherboard that had died. Well turns out the motherboard is fine, and the PSU died.

So if the PSU works when it's hooked up to the cheap tester, but it doesn't work at all when it's hooked up to a PC, what does that mean? Any guesses as to what's wrong with it, and if it'd be simple to repair?

I opened up the PSU and didn't see any bulging or leaking caps.

[eccerr0r]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

Well, it's worth repeating those cheap PSU testers are merely a quick go/no go test with really light loads that are nowhere near characteristic of an actual motherboard. So a lot of them really mean nothing other than basic functionality. It does not tell how the PSU responds to heavy or transient loads, nor tests whether your power supply produces noise that wreaks havoc on data integrity.

You'll need more sophisticated tools to tell what's wrong with it, such as an oscilloscope. If you have an ESR meter, it can expose some bad capacitors that you can't tell by simply looking at it.

A lot of the times even though capacitors may look good, they really aren't. The ESR meter or oscilloscope can sniff out a bunch of these.

[Stefan Payne]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

First guess would be +5VSB circuit.
Put the ATX connector onto the Motherboard and measure the +5V Standby Rail.
If it is too low, nothing will happen.

[Dannyx]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

Just yesterday (at the time of writing this), I managed to fix 5 PSUs with this same exact symptom and I can definitely say the capacitor(s) on the 5VSB bus are faulty. Replace them and you should be good to go. Cheers

[kc8adu]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

at 11 years all the caps in that unit should be replaced.
they were junk new and it gets worse with age.

[RJARRRPCGP]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

at 11 years all the caps in that unit should be replaced.
they were junk new and it gets worse with age.

Of course, that would be true if it hypothetically was a mid-2000s CWT-based Antec.
(Especially 2003-2005) (Ironically, earlier Fuhjyyu caps were less likely to take a shit.)

[CapLeaker]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

at 11 years all the caps in that unit should be replaced.
they were junk new and it gets worse with age.

That gets my vote too.

[Dannyx]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

The question now is SHOULD you bother to invest in a PSU like this ?

[kc8adu]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

if the op needs such a unit yes.
you should see some of the old oddball stuff i repair because its the only type that fits and the machine needs to be kept going forever.
how about a packard bell 100w?
yes i repaired one in an industrial application.
say what you will about them but it has been running since early 90's and simply failed to start after a power failure.
it didnt owe its owner anything at this point.

[Dannyx]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

I'm not the one to talk, since I too often (try to) salvage theoretically useless pieces of junk that nobody in their right mind besides us techies would ever give them second thought...

[stj]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

this is no junk, it's a good design.
i have a few FSP units in my spares box, and wont scrap them.
they arent re-capped, because i dont need them yet - but i will re-cap them if i need one.
no point re-capping psu's that may sit for a year or 2 afterwards!

[Quaddro]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

FSP usually lose PG signal.
I have dozens of fsp/amacrox/any psu based fsp with no PG signal, and until now i still can't find the culprit.

Btw, old psu is always useful.
i still running 9 years old tagan pipe rock tg1100-bz for my main system.
And well, i got it for free last year. Dead 5vsb circuit. Get it the ic from benq monitor.

[Dannyx]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

FSP usually lose PG signal.
I have dozens of fsp/amacrox/any psu based fsp with no PG signal, and until now i still can't find the culprit.

Btw, old psu is always useful.
i still running 9 years old tagan pipe rock tg1100-bz for my main system.
And well, i got it for free last year. Dead 5vsb circuit. Get it the ic from benq monitor.

I too came across several PSUs which have PG 0ms on my tester....can't track it down :|

[stj]

Re: Any idea what's wrong with a PSU that runs with a tester but not a motherboard?

some help:
http://www.danyk.cz/s_atx_en.html