PSU just "exploded" what's the cause?

[playman]

Hi there.
I was checking out a PSU I was given, as soon as I plugged it in, and used my antec PSU tester, the PSU "exploded" giving crackling sounds sparks and smoke.
The big caps look ok, just alittle burned, then there is busted minicap or sumthing.

What can cause this explotion?
this has hapened few times before, but I've never understood why it does. (not to the same PSU)
it was filthy when I opened it though, perhabs that is the cause?

It's a FSP group PSU
Model AX500-A
460W (500W peak)
230v

[370forlife]

Re: PSU just "exploded" what's the cause?

Looks like one of the metal oxide varistors did their job and blew to save the attached hardware. They blow in the event of a large inrush current, which means the NTC thermistor might be shorted or there was some sort of large voltage spike.

[playman]

Re: PSU just "exploded" what's the cause?

Large voltage spike, as in from the wall plug?
also there could not be any relation to that it had been storaged for awhile and was
completly dry?

[c_hegge]

Re: PSU just "exploded" what's the cause?

I doubt if the storage would have ahd anything to do with it. The only time I've ever seen the MOVs explode like that was when someone put the voltage switch on 110V and plugged it into 230V

[playman]

Re: PSU just "exploded" what's the cause?

I doubt if the storage would have ahd anything to do with it. The only time I've ever seen the MOVs explode like that was when someone put the voltage switch on 110V and plugged it into 230V

That's what I thought at first, but it was selected at 230V.

So like I said I was just wondering why this happens, as I have had few explode like this.
Why do they have to explode like that here, why did they not explode at the previous owner?

[999999999]

Re: PSU just "exploded" what's the cause?

I don't think that is an MOV, maybe a ceramic disc cap. MOV are usually sitting on AC input side before the bridge rectifier not after it at the DC voltage doubling circuit area. Based on another picture I saw which seems the same FSP design, there is an MOV but it's on the opposite side and opposite end of the pictured heatsink... near middle, one inch from the bottom in this pic:
http://www.xbitlabs.com/misc/picture...de_big.jpg&1=1

I can't tell what that charred component is though, you might desolder it and scrub the area on the PCB to see if you can find a silkscreened letter-number part code for it.

What I suspect happened is one of the switching transistors shorted and the resulting overcurrent condition shut the PSU down before there was the common bang/pop/smoke failure evidence you see on cheap PSU. If the (then) host computer had been unplugged from the wall and then restarted it probably would have completed the present failure damage at that point in time, but it was only a matter of when it was next powered on after an AC-disconnect reset.

Being filthy inside... yes it's a contributing factor if the transistors overheated, or it could have been a voltage spike on the incoming AC, or merely a marginally defective discrete component that just took awhile to fail. If you pry away the rubber heatshrink tubing around the failed part, maybe it will be easier to identify then.

[370forlife]

Re: PSU just "exploded" what's the cause?

The blown component is a MOV. The device you are thinking of is a NTC thermistor, or a soft start thermistor if you like. It has a higher resistance when cold, to limit the incoming power to soft start the unit. It's own resistance heats it up, which lowers it's resistance turning it into effectively a piece of wire.

[999999999]

Re: PSU just "exploded" what's the cause?

^ You are right that I was thinking of an NTC thermistor, but MOVs are also usually found on the AC input side, not downstream where other components already had to handle a surge voltage.

I'm not suggesting FSP "couldn't" do it, just that they shouldn't have and I'm not convinced it's a MOV.