Help-GB MB Shorted Out and has burn marks-GPU related?

pittguy578

Distinguished
Jul 19, 2010
37
0
18,530
System specs
GB MB Z77x d3h MB
3770k
GTX 960
Corsair PSU 750 W

Here is what happened last night.
System has been running fine for years. My old water cooler an Antec 620 stopped working well-never leaked. Just had high temps than normal.

I went to Best Buy last night. Got an H100 at 8PM

I had it installed by midnight.

Booted system up. Everything fired on all cylinders. Temps were REALLY low. Booted up Core temp. Idled low 30s.

Played an hour and a half of Rainbow 6 Siege that taxes both CPU and GPU. Max CPU temp was 52.

Latest insider build prompted me to install. I hit install and was letting it install while I went to bed.

Woke up this morning to play. System not turned on at all. I tried to turn it on. Didn't post plus there was a weird plastic smell.

Opened up case. I saw the issue. Literally the left side of my MB had burn marks starting right around the PCI express slow where the GTX 960 was sitting. Then I took card out and it appears the front pin seems to have melting around it. See pictures.

I took Power Supply out. Did the paperclip test. PSU boots up fine.
Took the 3770k out of socket to inspect that. Processor seems fine. No burning anywhere near CPU or socket

Can anyone shed any light on what may have caused this to happen? Usually if I didn't have my graphics card set in slot correctly it wouldn't boot up at all. I have installed plenty of graphics cards in my life and think I know what I am doing.

Any assistance is appreciated. Just seeing where I went wrong. Pic of entire MB

https://goo.gl/photos/27DULPr2bZ1AbFWP7

Pic of front pin of GTX 960

https://goo.gl/photos/btGWyrJwZxef9bpeA


7wrNad5eRHwqJk1V8



 
Solution


It might have been in just barely enough to make contact and work for a bit.
Later, vibration makes it go between connect/disconnect/connect/disconnect..all in the space of 0.25mm.

Overcurrent...sparks...poof...cooked parts.

I'm not saying this is 100% the issue, but it is a possibility.

In any case...cooked...

pittguy578

Distinguished
Jul 19, 2010
37
0
18,530


Yes I did take the GPU out temporally since I had to lift the MB up out of the case to take old backplate out and install new backplate for Water Block.

But then I re-seated it I thought correctly. Is it possible to put one in incorrectly and have it work?
 
The socket and card burnt pins are the 12V supply lines. Definitely had a current surge to cause this. I'm thinking it's unlikely that the cooler change out was the cause. More likely its a component failure on the mobo and/or GPU.
 

USAFRet

Titan
Moderator


It might have been in just barely enough to make contact and work for a bit.
Later, vibration makes it go between connect/disconnect/connect/disconnect..all in the space of 0.25mm.

Overcurrent...sparks...poof...cooked parts.

I'm not saying this is 100% the issue, but it is a possibility.

In any case...cooked parts.
Time to break out the wallet.
 
Solution

Karadjgne

Titan
Ambassador
From the looks of it, when replacing the gpu, that older z77 mobo busted a pin under the pcie and did exactly as USAFret described, basically the 12v hot running through the slot acting just like an automotive spark plug. It's really hard to not seat that end of the board correctly as that's also right where the end cap is and it won't screw/close if not seated fully. While it might or might not be your fault, I'm seriously doubting there's anyway you could have known it would happen.. Condolences on your loss..