Anyone ever blow their mobo's VRMs?

[fffblackmage]

I came across these articles at overclock.net today:

Why VRMs are a big issue & why to choose a "better" motherboard for high-TDP CPUs

A clarification of phases - decent explanation on how VRMs work

AMD Motherboard VRM information and reference list

So now I'm wondering if anyone here blew their mobo's VRMs due to overclocking.

 

[wwswimming]

well, the problem is, how many people go back to test if a VRM is blown ?

most people don't have the test set-up, or the capability.

maybe we could all ship our broken MB's to AT so they can test them and do an article.

 

[fffblackmage]

Overheating and burning an electronic component should be pretty obvious. For one thing, the smell of something burning is rather unmistakable. And if something burned, there's strong physical evidence remaining.

 

[iamthelaw]

Had a few AsRock P55 Extreme and Deluxe VRMS blow up st stock speed for no apparent reason.

 

[Zen0]

No, but then VREG temperatures are very important (perhaps just as important as Core temps) for those of who actually do a lot of overclocking...

 

[Athadeus]

Not yet, but I'm hoping to in the future

 

[john3850]

I came across these articles at overclock.net today:

Why VRMs are a big issue & why to choose a "better" motherboard for high-TDP CPUs

A clarification of phases - decent explanation on how VRMs work

AMD Motherboard VRM information and reference list

So now I'm wondering if anyone here blew their mobo's VRMs due to overclocking.

Thanks for the good links

 

[Arcanedeath]

My Socket A ECS K7S5A w/ Enermax 350watt vintage PSU died of burnt VRM's, I found out later that older Enermax PSU's that i used in that build had horrible ripple and after around 4 years the system just gave up the ghost. It was on a UPS w/ AVR it's whole life too. Only board I've managed to burn the VRM's on, it was an Athalon Xp 1600+ overclocked to 1.8ghz via a pin mod.

 

[pcgeek11]

nm

 

[Rifter]

Yes, all the way back to the socket 7 days. I've blown up many cheap mobo's over the years. I blew up 2 MSI boards in a row with my old Duron that would run %50 over base clock once i found a motherboard able to proplerly power it.

This is why i spend the money on "Enthusiast" boards now that have multiple phase power for both the CPU and the RAM. Since switching to the "enthusiast" boards and staying away from the cheap 4 phase boards i have not had any issues at all.

Oh and cooling the VRM's helps ALOT but at the end of the day a cheap 4 phase board is still a cheap 4 phase board.

EDIT: spelling

 

[fffblackmage]

Oh and cooling the VRM's helps ALOT but at the end of the day a cheap 4 phase board is still a cheap 4 phase board.

Actually, from the reference list, I realized that the Asus M4A785TD-M EVO mobo I have has a "good" 2 transistor + 2 driver design. However, the Gigabyte GA-MA785GM-US2H has a "cheap" 3 transistor design. I don't know what the driver chip is, what the trade-offs are, or why it's preferable. If anyone knows, please do share!

The only thing I do know is that using a transistor in-place of the driver is cheap. The pic I linked a few posts above may be an example of this. That was from an MSI NF980-G65. From the reference list, the specs says it supports up to 140W TDP, has heatsinks to cool the VRMs, it's a 4-phase design, and it uses a transistor instead of a driver. I'm wondering if the failure can be attributed primarily to the poor 3-transistor design (no 'proper' driver), or if it's because it's a 4-phase instead of 8-phase, for example, or maybe the problem is just cheap transistors with a low current rating.

 

[jiffylube1024]

My Socket A ECS K7S5A w/ Enermax 350watt vintage PSU died of burnt VRM's, I found out later that older Enermax PSU's that i used in that build had horrible ripple and after around 4 years the system just gave up the ghost. It was on a UPS w/ AVR it's whole life too. Only board I've managed to burn the VRM's on, it was an Athalon Xp 1600+ overclocked to 1.8ghz via a pin mod.

Ditto. The only board I've had die by VRM failure was also an ECS K7S5A at stock speed (never overclocked!). I've had countless ASUS boards over the years that survived overclocking no problem.

 

[Rifter]

Actually, from the reference list, I realized that the Asus M4A785TD-M EVO mobo I have has a "good" 2 transistor + 2 driver design. However, the Gigabyte GA-MA785GM-US2H has a "cheap" 3 transistor design. I don't know what the driver chip is, what the trade-offs are, or why it's preferable. If anyone knows, please do share!

The only thing I do know is that using a transistor in-place of the driver is cheap. The pic I linked a few posts above may be an example of this. That was from an MSI NF980-G65. From the reference list, the specs says it supports up to 140W TDP, has heatsinks to cool the VRMs, it's a 4-phase design, and it uses a transistor instead of a driver. I'm wondering if the failure can be attributed primarily to the poor 3-transistor design (no 'proper' driver), or if it's because it's a 4-phase instead of 8-phase, for example, or maybe the problem is just cheap transistors with a low current rating.

I dont trust MSI motherboard to begin with, the two i fried with my duron were both MSI boards, i believe K7T model's and both fried the VRM's. When I finally sprang for a Asus is when i got it stable.

All i can tell you is switching to 8 phase or better i have not had issues since. And yes the transistor based ones are cheaper than a true driver. I wouldnt buy a transistor based power circuit board IMO.

 

[skipsneeky2]

i remember my asus p4c800-e deluxe lost a electronic piece on the motherboard not sure exactly what it was but it ended up giving my cpu a bit more voltage and made that processor really hot the piece was almost like a benedryl pill and silver,i know that sounds stupid but back then i didnt know what to make of it that motherboard was a trooper still with that issue will i went core 2 duo

 

[GeezerMan]

I know this is an older thread, but it has some important info in it. In the past, I did not give a lot of thought to VRM design since I don't overclock. I have built PCs with 4 phase boards and 125 watt CPUs, and used good heatsinks with 120mm fans thinking it should last a long time without overclocking. Those threads linked above talk about how the better cooling HSFs often have poor VRM cooling. You might get better VRM cooling with a stock cooler. Interesting reading.

 

[nexusN]

Guys, I am worrying much in this.
ASUS P8B WS is not available in Hong Kong, and I will be required to get one from SuperBiiz and the shipping costs $76 USD, it makes the board end up costing $299 USD.
This is fine but in case of DOA, another $76 USD will be needed and I am not sure such burning will be within the DOA terms.........

Is it really a rare case for VRMs to blow or quite COMMON?
I can see quite a lot of reports these weeks for the 1155 boards, ASUS, Giga........