Can a PSU damage the motherboard capacitors?

[emy12]

Hi all,

I have GA-K8NF-9-939 motherboard and had a FSP300-60PN(PF) that I replaced it with Antec TPN 750 blue PSU. I noticed a week ago a capacitor leakage on top of it, after a few days one more, today another one, the top of it is getting brown.
When I had first PSU, had no problem at all (was pushing more air down to the pc case too), since replace it, the capps started to fail one by one. There are 5 of NCC KZG 6.3V 3300UF
The system is working fine, no problems.
I have also replaced the cpu's cooler with Freezer 13. The fan is pushing the air from one to other side not down to the cpu like previous cooler. The Northbridge temps is over 60.
Since noticed the first capacitor brown on top of it and swollen, I left the side panel off, thinking that the heat could be a problem.
I have ordered some of Sanyo WG 6.3V 3300UF to replace all five capacitors.
I am thinking to send back the new PSU to get a different one.
Any advice is welcome.

[brethin]

Re: Can a PSU damage the motherboard capacitors?

Yes a bad PSU can damage the caps on a motherboard. More than likely though those caps were going bad anyway, I would replace all the KZG caps before I used it anymore.

[emy12]

Re: Can a PSU damage the motherboard capacitors?

Thanks for quick replay. I will wait to get the capacitors and change them.
The PSU is only two weeks old. I never heard anyone saying that Antec New 750 could go that bad. It runs fine and quiet so far.

[PCBONEZ]

Re: Can a PSU damage the motherboard capacitors?

Yes a bad PSU can damage the caps on a motherboard. More than likely though those caps were going bad anyway, I would replace all the KZG caps before I used it anymore.

I second that. - All of it.

[Perineum]

Re: Can a PSU damage the motherboard capacitors?

It is possible for a bad PSU to kill a motherboard or it's caps. However, the KZG caps are notorious for going bad so it was likely just going to happen.

[bigbeark]

Re: Can a PSU damage the motherboard capacitors?

What was the Northbridge temperature with the old cooler? 60C is way too hot for a socket 939 motherboard.

Did the old PSU fail? Are you overclocking?

Make sure your cooler is fastened properly. I assume you used paste on top of the CPU when you mounted the heatsink?

[brethin]

Re: Can a PSU damage the motherboard capacitors?

What was the Northbridge temperature with the old cooler? 60C is way too hot for a socket 939 motherboard.

Did the old PSU fail? Are you overclocking?

Make sure your cooler is fastened properly. I assume you used paste on top of the CPU when you mounted the heatsink?

Double posting bad info is no help. 60C for a Nforce 4 chipset on a Gigabyte board is low.

The CPU cooler cools the CPU not the Northbridge for the most part. Most of the Gigabyte boards I have seen run as high as 70C on the northbridge and Gigabyte states that under 80C is fine esp if its a passive cooled northbridge.

Anyway none of what you posted has anything to do with his VRM caps going bad other than maybe because the crap KZG caps failed because of the added heat in the case in which adding another case fan to remove heat from the case and keeping the caps cooler may help his northbridge temp is fine.

[dood]

Re: Can a PSU damage the motherboard capacitors?

Yes, a PSU can cause caps to fail. No, I don't think that's what happened here.

KZG are known to fail. Chances are, it was a mix of heat and age that did these in.

[mockingbird]

Re: Can a PSU damage the motherboard capacitors?

If I understad it correctly, if a PSU is not filtering out enough ripple because of lousy capacitors, this ripple now has to be filtered by the motherboard's capacitors...

Actually, I heard of someone replacing an all polymer Asus M3A78-EM... Point is, if these PSUs damage some of the polymer caps, go now and figure out which caps you have to replace without an ESR meter... Either the motherboard had other problems, or rather, like I suspect, polymer capacitors can fail without bloating...

How polymer capacitors can fail without external signs boggles me... It's hermetically sealed, so the only thing I can think of is the polymer drying out inside the can, but without enough pressure that would actually cause it to bulge.

[PCBONEZ]

Re: Can a PSU damage the motherboard capacitors?

If I understad it correctly, if a PSU is not filtering out enough ripple because of lousy capacitors, this ripple now has to be filtered by the motherboard's capacitors...

Correct.

How polymer capacitors can fail without external signs boggles me... It's hermetically sealed, so the only thing I can think of is the polymer drying out inside the can, but without enough pressure that would actually cause it to bulge.

Lytics bloat because liquid goes to a gas.
Lytics can also fail without forming a gas. Just depends on what is going on.

There is no liquid in polymers. - Little chance of gasses.
One of the bad things about them is they have a smaller ability to self-heal the oxide layer. [Relative to Lytics.]
To counter that it's much more difficult to damage the layer in the first place, hence the higher Ripple ratings.
- But, if you do manage to damage it, it isn't going to heal over time, it's permanent.
.

[PCBONEZ]

Re: Can a PSU damage the motherboard capacitors?

Visualize Ripple damage to the oxide layer as chunks of oxide being ejected into the electrolyte due to the current direction in the 1/2 cycle of the Ripple that is opposite to the DC Bias. [In other words an electron going the wrong way knocks it loose.]

In a liquid:
The oxide chunks can float around and are inclined migrate back to the hole [or some other nearby hole] due to the 'pull' of the DC bias -and- the fact the hole is a thin spot so more leakage current will 'want' to go through the hole.
[The healing is quite similar to what happens in electroplating.]

In the plastic:
The oxide chunks aren't likely to travel very far in the first place and most will move right back to where they came from, but a few will travel far enough to 'get stuck' in the plastic and will never travel back to a hole.
That amount of oxide material is just lost permanently.
.

In reality these 'chunks' are molecules, but 'chunks' is easier to visualize.
.

Anyway: If you exceed the Ripple rating but only for a short time.
A Lytic will probably repair itself completely.
A Polymer won't.
.

[mockingbird]

Re: Can a PSU damage the motherboard capacitors?

Fascinating... Thanks for the thorough explanation.

[Agent24]

Re: Can a PSU damage the motherboard capacitors?

I think if a polymer was going to fail it'd probably go bang like a Sacon FZ... wouldn't it?

[japlytic]

Re: Can a PSU damage the motherboard capacitors?

On one motherboard I have seen, it had several bulging capacitors (of which some were connected directly to the ATX power supply rails). These capacitors took up the ripple current from a poor quality PSU, causing them to heat up and bulge.

[c_hegge]

Re: Can a PSU damage the motherboard capacitors?

I've seen that too. It was a foxconn board and a cap near the ATX connector (Rubycon IIRC) got Antec'd.

[PCBONEZ]

Re: Can a PSU damage the motherboard capacitors?

I think if a polymer was going to fail it'd probably go bang like a Sacon FZ... wouldn't it?

Not sure but probably not unless it's a catastrophic failure.
[Like in backwards or got put on line voltage or similar.]
To go pop something has to be making pressure inside the can.
I suppose plastics could/would off-gas something when they melt but I don't know if it would be enough to pop the can.
'I think' it would have to be so hot inside the plastic wants to burn before it would actually explode.
.
A while back I posted a defective OSCON that exploded on me.
What came out was definitely smoke.
.

[emy12]

Re: Can a PSU damage the motherboard capacitors?

I have just replaced the capacitors with Sanyo WG 6.3V 3300UF. So strange, all the other capacitors are Rubycon, Nichicon and one Sanyo. I do not know why the bigger once were 5 of NCC KZG 6.3V 3300UF. Everything went smoothly.

By the way, I could not wait for the capps to arrive and i went to a pc repairing store if i can get the capps faster. For € 5 tried to pull out 5 capps, (Rubycon) from a motherboard, gigabyte ga-k8n pro-sli. I gave hip €5 more for the whole motherboard.
I tested, updated the bios through Live XP CD, ( copied from usb stick to floppy disk the latest bios), booted from floppy drive and followed the steps of updating the bios.
It is working just fine, only the on board audio chip no, I could not make it going. I think is faulty. I ordered a sound card, a basic one.
I do not know which one to use, the GA-K8NF-9 or ga-k8n pro-sli ( this one has all the small capps, NCC KZG and big once Rubycon.
I will keep in eye on PSU. I do not think is something wrong with it. I think that the capps failed because of heat. I will put two 120mm fans, front, side and 92 mm at the back for better air flow.

[Agent24]

Re: Can a PSU damage the motherboard capacitors?

You shouldn't really use 2nd hand\used capacitors. Best to get brand new ones. Being impatient with electronics is a bad idea.

[emy12]

Re: Can a PSU damage the motherboard capacitors?

I did replaced all 5 with new once. I got and replaced them yesterday, but a week ago I went to a pc shop asking for some capacitors ( i could not wait any longer for the ordered once). He said to pull out some from the old motherboard showing me a ga-k8n pro-sli he had in a box. I got the whole motherboard. Tested it and is working fine, only the sound not.
So, I have two motherboards, and I do not know which one to use, I might go for the sli one, though the GA-K8NF-9 (with new caps) was working fine too.

[ss627]

Re: Can a PSU damage the motherboard capacitors?

Sell one motherboard and mod the sli one with polymers.