Question Graphics Card Fried - Which part is to blame?

Sep 21, 2021
Yesterday evening, I was playing Skyrim, and suddenly both of my monitors went dark, and were no longer detecting output. I could still hear the game music, and my OS was still running, but I couldn't see anything. Naturally, I powered down my machine. When I went to turn it back on, it wouldn't start. The power LED on my case would light up briefly, but immediately dim again and go dark, like it wasn't getting enough power to boot.

I unplugged everything from my PC and brought it to another power outlet in my house that I knew was good and didn't have anything else connected to it - same problem. Then, I re-seated all of the internal connections from my PSU, with the exception of the graphics card, and it booted. I then tried hooking the GPU back up, and it booted again - there were lights on my GPU to indicate that it was getting power. However, when I brought the PC back to my office, it would again no longer start up. I suspected at this point that my PSU was bad.

This morning, I went and bought a new PSU, brought it home, hooked it up, everything looked good. However, as soon as I powered on the PC, sparks and smoke flew from the top of my GPU, and the PC shut itself down. I removed the GPU entirely, and the PC still starts and otherwise works just fine, but the GPU is now fried. I haven't dared to try connecting another GPU to my machine yet for fear of the PSU frying that one, too.

At this point, I don't know what caused the problem. I'm an IT specialist by trade, and I consulted a few of my friends who are also in the field - the going theory seems to be that my PSU was what died, and it negatively impacted the GPU on its way out as well, which is why it cooked when I plugged in the new PSU. I'm skeptical to believe that given that the GPU did not fry itself when still connected to the old PSU after the initial crash, but I suppose that could be because it wasn't receiving sufficient power to have been toasted. At this point, I am not sure what to do.

I want to connect a replacement GPU, but until I know what the problem is, I'm pretty gun-shy. Another friend of mine recommended that I get my new PSU tested to make sure it wasn't a faulty power rail that toasted my GPU, which I will look into, but is there anything else that I should do? Did I miss anything? I'm 99.9% certain that I didn't connect the GPU to my new PSU wrong - the connector was 6+2 pins, and I connected all 8 right away - should I have just used the 6 initially?

Any advice or answers would be appreciated. Here's a list of the relevant components. I can provide the rest of by system build if necessary:

Old Power Supply: OCZ ModXStream Pro 600W
New Power Supply: Corsair RM850x
GPU: EVGA GeForce 1070 TI


Update your post to include full system hardware specs and OS information.

Old PSU: how old? Heavy gaming use - correct?

Did you reuse any cables from the ModXStream PSU when the Corsair PSU was installed?

Do you have a multi-meter and know how to use it?

If so:

Test both PSUs. Not a full test but any voltages out of tolerance make the PSU suspect.

Also use a bright flashlight and even a magnifying glass to examine all cables and connectors for signs of damage: melting, bare conductor showing, signs of browning/blackening, etc..

Also take a close look at the applicable User Guides/Manuals for motherboard, PSUs, and case. Double check that all connections are both correct plus fully and firmly installed.