[SOLVED] My msi mb burned, what may be the cause?

[kemorno]

I have this MSI 970A-G43 Plus for about 3 years now, it had a Aerocool Strike-x 650w PSU that stopped working, so i bought a DA500 500w from Deepcool one since i didnt really need those extra watts. It had a Fx8320e on it, that i had been running in turbo mode for about 1 week (4Ghz, Turbo from the mb Bios). While it was with a medium load in the CPU, it simply turned off, i thought it was the building Power that went off, its not usual but it happens sometimes here where i live, So i Turned the PC back on and it started a little fire in those two pieces, so i quickly turned it bad off and the fire went off, and some smoke came out.
Any guesses of why it could be?

https://imgur.com/tUd6WGy

View: https://i.imgur.com/tUd6WGy.png

 

[drea.drechsler]

I have this MSI 970A-G43 Plus for about 3 years now, it had a 650w PSU that stopped working, so i bought a 500w one since i didnt really need those extra watts. It had a Fx8320e on it, that i had been running in turbo mode for about 1 week (4Ghz, Turbo from the mb Bios). While it was with a medium load in the CPU, it simply turned off, i thought it was the building Power that went off, its not usual but it happens sometimes here where i live, So i Turned the PC back on and it started a little fire in those two pieces, so i quickly turned it bad off and the fire went off, and some smoke came out.
Any guesses of why it could be?

https://imgur.com/tUd6WGy

View: https://i.imgur.com/tUd6WGy.png

CPU VRM FET's shorted. Could be a number of reasons but a good possibility is that 500W PSU was poor(er) quality than the 650 it replaced in terms of build quality and/or power output. It just didn't deliver enough clean current to the VRM, making it work harder to maintain a steady output to an extremely power hungry CPU.

Another possibility is when the earlier PSU failed it heavily stressed the FET's with an excessively high voltage. That left a latent defect which only needed a little more time to go.

In either case, I'm not sure I'd put that (new) PSU in another system, not unless you have the means and know how to test it.

 

[kanewolf]

Did you have an AIO liquid cooler on the CPU? No airflow on the voltage regulators can cause problems like that.

 

[kemorno]

CPU VRM FET's shorted. Could be a number of reasons but a good possibility is that 500W PSU was poor(er) quality than the 650 it replaced in terms of build quality and/or power output. It just didn't deliver enough clean current to the VRM, making it work harder to maintain a steady output to an extremely power hungry CPU.

Another possibility is when the earlier PSU failed it heavily stressed the FET's with an excessively high voltage. That left a latent defect which only needed a little more time to go.

In either case, I'm not sure I'd put that (new) PSU in another system, not unless you have the means and know how to test it.

Now i'm changing it for a ryzen 5 1600 AF 65w so i think i'm good with that cpu no? Its a deepcool 500w, bronze rating iirc, also wont need to put the cpu @turbo mode anymore... i have a rx580, its not power hungry, so, am i good to go or that psu is probably gonna burn my mb again?

 

[kemorno]

Did you have an AIO liquid cooler on the CPU? No airflow on the voltage regulators can cause problems like that.

I did have a liquid cooler, and it did cover that whole area, since now i'm changing for a ryzen cpu i wont use that cooler anymore so less risk of it burning in the new mb

 

[drea.drechsler]

Now i'm changing it for a ryzen 5 1600 AF 65w so i think i'm good with that cpu no? Its a deepcool 500w, gold rating iirc, also wont need to put the cpu @turbo mode anymore... i have a rx580, its not power hungry, so, am i good to go or that psu is probably gonna burn my mb again?

I'd not assume you're good with that PSU. It might have been damaged by connection to the motherboard when the VRM's on it went out. That's an unfortunate thing that happens when power circuits like that go out, they often seriously stress the other circuits connected to them. Even if not outright blowing it they leave a latent defect that can cause it to fail really soon after. Which, of course, then takes out the VRM components on your new motherboard.

It's often just the better part of caution to simply discard everything, especially if you don't know how to check it out for damage. I'd want to put it on a load cell and put a heavy (in spec) load on the PSU for several hours before declaring it good to go.

 

[kemorno]

I'd want to put it on a load cell and put a heavy (in spec) load on the PSU for several hours before declaring it good to go.

Is there any way to do it in home? Because PSU here where i live are pretty expensive, and im moving to portugal in a couple of weeks, so i would like to use this PSU until i arrive there, and then by a new one...
Can you recommend me any forum post on it or something, i searched for but it only returned me a non-load test, as in voltage and wattage...
Also, if i do not use the PSU in stress while using the PC, lets say that i dont render neither play nothing too demanding, would i lower the risk? or is it the same independent of the stress on the PSU?

 

[DSzymborski]

I'm not seeing any 500W Gold Deepcool PSUs. I only see 80 Plus Gold on their DQ series (and no 500W PSUs). If it's a DA-500 or DN-500, which are their 500W PSUs, they're neither Gold-rated (Bronze or Unrated) nor quality PSUs.

 

[kemorno]

I'm not seeing any 500W Gold Deepcool PSUs. I only see 80 Plus Gold on their DQ series (and no 500W PSUs). If it's a DA-500 or DN-500, which are their 500W PSUs, they're neither Gold-rated (Bronze or Unrated) nor quality PSUs.

I miss wrote, my previous PSU was an Aerocool Strike X 650W and my Deepcool DA-500w is a bronze80, will edit now...

 

[drea.drechsler]

Is there any way to do it in home? Because PSU here where i live are pretty expensive, and im moving to portugal in a couple of weeks, so i would like to use this PSU until i arrive there, and then by a new one...
Can you recommend me any forum post on it or something, i searched for but it only returned me a non-load test, as in voltage and wattage...
Also, if i do not use the PSU in stress while using the PC, lets say that i dont render neither play nothing too demanding, would i lower the risk? or is it the same independent of the stress on the PSU?

A load bank to do it, or parts to build one, probably costs as much as a PSU itself. You can take a chance with what you have, but if it does work you have to consider you're going on borrowed time. And if it is one of the many low-cost 'junk' units you can be pretty certain it lacks adequate protection circuitry to protect your motherboard.

 

[kemorno]

A load bank to do it, or parts to build one, probably costs as much as a PSU itself. You can take a chance with what you have, but if it does work you have to consider you're going on borrowed time. And if it is one of the many low-cost 'junk' units you can be pretty certain it lacks adequate protection circuitry to protect your motherboard.

Tbh, it will have to do, it wasnt the cheapest on the store, i tried to get one at least decent, but probably it does not have a circuitry protection, i'll end up buying a EVGA gold80 when i move overseas tho, i know how important a good PSU is... thx very much man, also i sent an email to the deepcool support for further information about the PSU