Overclock.net › Articles › Database of motherboard VRM failure incidents

Database of motherboard VRM failure incidents

Comments (11)

Uh oh, I have a NF750-G55! At least I'm not using too much of a CPU on it. I hope it doesn't break on me!
so... MSI apparently makes cheap VRMs lol. but i hope everyone is safe!
I'm wouldn't be terribly surprised about their lack of quality. This motherboard of mine really can have some issues...but it's usable.
Still surviving on a MSI 760GM-P23 (FX) with my 1090t.
Didn't see one asus board. Did you skip that company or are they just that good?
asus boards throttle when the vrms get hot
they wont burst into flames, you just wont get any good performance from the 3+1 boards with 95-125w cpus
Haha, theres 1 ASUS board there.
Why 99% of them are AMD boards? LOL
amd chip use alot of power vs intel.... ive seen a asus p5k se on this forum friend because the guy put like 1.45 volt and tried 3.6 ghz..........

People need to get infrared thermometer to realise that mobo run too hot
Gigabyte GA-G31M-ES2L - set on fire that you can add if you like.

I was playing with an old Q6600 rig with after market TTower style cooler. Didn't expect much from a G31, so voltage set to stock and FSB clocked @ standard multiplier to give a 2.8ghz clock speed knowing Q6600's can run close to 3ghz on stock voltage, DRAM was ran at standard voltage and downclocked to run at 800mhz for compatibility - didn't think this would kill it.

OCCT power test saw the Q6600 started to reach mid 70's on some cores after around 20 mins - could be bad cooler seating, I was about to stop the test and inspect cooling when I assume something died or OCP set in for some reason.

Although the CPU was hot, it was no where near thermal shutdown or even throttling. The Northbridge heatsink was VERY hot to the touch though.

After everything had cooled the machine was dead, and even disconnecting the 4pin CPU power connector would not allow the system to switch on, so I thought the VRM should be ok - no smells no evidence of damage, perhaps the PSU shut off. I did a PSU test and pulled the volt meter to see if something odd went wrong on it, although it's a half decent EVGA single rail 12v 48A PSU which is only five months old, but hey things can happen. The PSU was showing 12.1V with bridging the green and black on the 24 pin and no devices connected - no biggy, but it didn't have load so I thought maybe there was a problem or short somewhere. Tried to power on again and heard a 'POP', quickly pulled the PSU cord out and smelt burning wafting around the PSU area, at this point I thought perhaps the PSU was at fault.

Popped in a spare Corsair PSU to rule out PSU, powered on and the system came on right away, went straight into windows, CPU temps were ok (30-40), everything was running fine - no dodgy smells. I ran some basic tasks in windows for about 20 mins with the cover off and then heard a 'pffft' and a mosfet set on fire - quickly pulled the PSU plug out of the back and put the fire out.

I've ordered a replacement board (the same) for the system as I don't have time to do a full rebuild ATM + it's a mini ATX setup, a better board won't fit in there without a full re-shell, but I need to figure out why it popped, I know 3 phase VRM uncooled is cheap and horrible, but I don't see why it cooked on a Q6600 at stock voltage :/

I'm going to chop the old NB and SB heatsink and apply to mosfets and VRM's with thermal tape - bit tricky given the mis-mash layout on the board, then stick a side fan in and see how we go, but needless to say, I will be cautious this time with this board!
Gigabyte GA-EP45T-UD3LR. Confirmed VRM failure (board will power-on without 4-Pin power connector attached, but fails to power on with an initial power blip and all 4 phase LED's illuminated). Board running a Q9550 over-clocked to 3.57GHz @ 1.1875V (basically stock voltage but with improved Vdroop BIOS setting enabled), gave up the ghost during a 6h Prime95 run. Sad.
Overclock.net › Articles › Database of motherboard VRM failure incidents