24 pin power connector burned

[WeThePpl]

Well, just some background on the rig before it had a stroke. It's an Asus Maximus III Formula, i7 860 at 3.66ghz, 8gb ram at~1750mhz, 2 gtx 570s, 1 8800gt, cooled by air, with an 850 watt PSU.

I turned off the computer for the day and turned it back on about 12 hours later. The videocard fans and processor fans were NOT spinning and there was no POST. After pulling the 24 pin power connector, I noticed that 3 of the pins had been burned (it melted the plastic housing on the PSU) and so I think that was the culprit. The power supply is being replaced through RMA, which fixes that problem. BUT! What about the motherboard? All of the hardware still works, including the processor. When I cleaned the three pins on the mobo that had plastic and burn residue still in contact, I put another PSU in as a test and it seems to be working fine. Over-volt protection seems to be victorious in this case. I used 2 molex adapters to pci-e to power the 8800gt, and that is why it only burned those 3 pins. My question is, will I have problems with system stability in the future after just replacing the PSU? If I RMA the mobo Asus has already informed me that I will have to pay to get it fixed. After inspecting the mobo very carefully, I cannot find any other signs of damage, so I imagine they will just re-solder a new 24 pin connector onto the board. What do you all think?

 

[WeThePpl]

HOLY CRAP!!! I'm so sorry for the pictures! I am using fedora right now and I have not used GIMP in a long time! My apologies.

 

[turbohans]

hahaha,, just make a rectangle selection, and click "fit canvas to selection"


As far as the PSU, if you think it may still work you might be able to talk one of these guys into selling you an end off of a dead one.

no idea why it would do that though other than maybe not being plugged in all the way.

 

[WeThePpl]

Yeah, I was going to try that but Antec (the MFG of the PSU) said they were RMA it with no problems. So I would rather not tamper with it just in case.. The motherboard is what I am concerned about... I cleaned it thoroughly, it looks just like it did before this happened. And like I said, there is no other VISUAL damage on the motherboard itself. And when I plug another PSU into it, it seems to work fine.

 

[CompuTamer]

That's odd. I've never seen a motherboard do that... it seems like it got hot, while it was off Maybe something on the PSU or Mobo shorted and heated the socket up?

 

[TheQuadFather]

i have seen this before but i never carried on reading the thread to find out the cause, but it isnt to uncommon as all the ones ive seen have been the same pins as you so im sure if you do your homework you could find out from old threads.

 

[SPreston2001]

Wow thats scary. Im glad it didnt do any real damage to your system.

 

[kayson]

There are a number of reasons this could've happened. A short, as someone mentioned before, would dump a ton of current through the pins and it would get very very hot because of all the excess power dissipation, but something would've likely fried on your mobo. Same logic applies to an over-voltage situation. I've heard that if there's a bad connection between cable pin and mobo header for some reason, you get this result too.

 

[Bobnova]

Decently common with high card count SLI in the benching world. EVGA's 4way SLI boards are somewhat famous for it. The difficulty is that the motherboard connector there supplies the 75w that PCIe cards are allowed to draw through the slot.
Put two cards in, and it's doing 150w.
Put three in and it's doing 225w.

Now OC them, and it's doing >300 watts. It's not rated for nearly that much, and poof! There it goes.

That's why Asus put extra Molex connectors on the R3E, to supply that 12v power to the PCIe slots.


That said, it shouldn't have done that with three cards at stock or stockish clocks.

 

[ChanceCoats123]

There are a number of reasons this could've happened. A short, as someone mentioned before, would dump a ton of current through the pins and it would get very very hot because of all the excess power dissipation, but something would've likely fried on your mobo. Same logic applies to an over-voltage situation. I've heard that if there's a bad connection between cable pin and mobo header for some reason, you get this result too.

This would be my best bet. There was probably a decent connection, but more wattage was being pulled through than physically could. When that happens, the connectors themselves start to act like resistors. Anyone who's dealt with resisitors before knows they get hot when greatly reducing the current.

 

[CompuTamer]

Bobnova has a good point. It could have just been a fluke and something drew more power than it should.

 

[WeThePpl]

Bobnova has a good point. It could have just been a fluke and something drew more power than it should.

So, do you think it would be worth it to RMA? Or save up and get another board? I just finished testing ALL the hardware. So far the only noticeable fault is my RAID config and the power supply is completely dead. Everything seems to still work on the motherboard side, though. Would it be best to sell the processor and maybe RAM and rebuild a new system?

I am concerned that I may acquire a new set of problems down the road because of this. I want to make DAMN sure that the mobo is okay/going to be okay in the future.

 

[Woomack]

I had exactly the same running ~1kW hardware on 750W PSU ( forgot that the same setup was running before on 2x psu for tests ) ... was looking worse than it was in real , board was ASUS P6X58D-E where I just cleaned 24 pin socket and changed 24pin connector in PSU ( was looking almost the same as yours but 12V wires were kinda melted on ~10cm length ). All is running now without problems, both mobo and psu.
Photos are terrible ... but I see that connector on board isn't melted so just try to clean pins and check if it's running stable.