[SOLVED] help with deciding if bad psu

Jun 7, 2019
6
0
10
Hello,

For the last 2 days my PC has been shutting down unexpectedly and my GPU is crashing when playing. This only happens when I open a GPU intensive program like a game. I monitored the voltage of my GPU when playing, but it has never exceeded 1200mv. The strange part is that after a few minutes of playing and the voltage fluctuating between ~1083mv to the max I've seen 1202mv my entire system shuts down completely. After this happens, if i try to power on my PC there is no power to the south-bridge where the CPU is and neither the fan on the back or my CPU heat-sink show any power, only the front fans of the case unless i manually unplug the power cord from the PSU and re-plug it in. After that, my PC works fine again until I attempt to start any GPU intensive tasks. Sometimes the GPU merely disconnects, I hear the Windows "disconnect" sound, and upon restart it says the display driver crashed and failed to start up again, and some others the entire system shuts down like mentioned above.

I've seen in some other threads on the internet that this can also be GPU related, but since my entire system shuts down completely, I believe it may be the PSU. My GPU also works properly during the time before it crashes, no strange lines or artifacts usually associated to GPU failure.

What would be the more likely culprit, the GPU or PSU? It's also worth mentioning the PSU is already several years old.

I also tried re-seating the card, cleaning the power cables and even reseatting the RAM to be safe.

Specs:
Mobo: Gigabyte B150-HD3 updated to latest BIOS
CPU: Core i5 7600
RAM: 8GB Kingston (One Stick)
PSU: Corsair CX500
GPU: Gigabyte GTX 970 Windforce

Any advise will be greatly appreciated.

Edit: Forgot to include GPU.
Edit2: I just want to report that it was indeed my PSU. as soon as i changed it, everything returned to normal. thank you all for your help
 
Last edited:
Solution
I have an update. I'm still waiting for my psu, but I decided to do a bit more testing on the matter. I found through GPU Z that when my driver crashes, all power to the card drops to 0. TDP, Wattage, everything. All clocks also stop completely and usage reduces to 0 as well. Could this be proof that it is the psu?
That sounds like a PSU problem, but I wouldn't do anything about it until you try with the new PSU.
There is no definitive proof pointing to a faulty PSU.

You need to replace it, and then see if the system is still unstable or not.

There may also be other things, like overheating of CPU/GPU, bad capacitors on motherboard.
could it be a motherboard issue as well then? my gpu and cpu never went anywhere close the dangerous zone, my cpu rarely goes over 70-75 degrees, my gpu under heavy load does go up to 70-75 but i read that is to be expected from a gpu?
 
Yeah, pretty much.
I have an update. I'm still waiting for my psu, but I decided to do a bit more testing on the matter. I found through GPU Z that when my driver crashes, all power to the card drops to 0. TDP, Wattage, everything. All clocks also stop completely and usage reduces to 0 as well. Could this be proof that it is the psu?
 
I have an update. I'm still waiting for my psu, but I decided to do a bit more testing on the matter. I found through GPU Z that when my driver crashes, all power to the card drops to 0. TDP, Wattage, everything. All clocks also stop completely and usage reduces to 0 as well. Could this be proof that it is the psu?
That sounds like a PSU problem, but I wouldn't do anything about it until you try with the new PSU.
 
Solution