Can a PSU damage your video card?

[netxzero64]

A month ago I bought my new PSU then before I went to turning on my unit, I also cleaned my gpu and reapplying a tim. When I turned on the PC, images where like sh*t.

I tried reseating the GPU then after a couple of attempts, I was able to bring it back and restarted my PC and went on maximum overclock for the gpu as well as CPU. While running 3dmark vantage halfway, the monitor suddenly shut off and I thought that the computer restarted, but to my surprise, it didn't because I hear the sounds of windows when I tried to press the mouse. I restarted my pc and display was no more.

A couple of weeks later I borrowed a 650 from a friend and I got everything back.

I baked the GPU twice but all I get was artifacts. Was the GPU up on its time or was the psu to blame?

 

[TemjinGold]

My bet's on the GPU just being up in this scenario.

 

[cheez]

A month ago I bought my new PSU then before I went to turning on my unit, I also cleaned my gpu and reapplying a tim. When I turned on the PC, images where like sh*t.

I tried reseating the GPU then after a couple of attempts, I was able to bring it back and restarted my PC and went on maximum overclock for the gpu as well as CPU. While running 3dmark vantage halfway, the monitor suddenly shut off and I thought that the computer restarted, but to my surprise, it didn't because I hear the sounds of windows when I tried to press the mouse. I restarted my pc and display was no more.

A couple of weeks later I borrowed a 650 from a friend and I got everything back.

I baked the GPU twice but all I get was artifacts. Was the GPU up on its time or was the psu to blame?

I bolded the items for question. What do you mean by 650? and what baked? Baked cookie or something? I am trying to figure out what you are trying to say. Please rewrite.


My answer to your post is that your GPU is shot after you tried to reapply the heatsink installation for the GPU. It might not have made full contact between the two parts and overheat. Power supplies rarely burn up the video card, usually the CPU first or the motherboard.

You F'ed up your graphics card man sorry. Get a new graphics card.


cheez

 

[imagoon]

GPU was on its way out or you messed up the TIM application and the chip grossly over heated. Baking the card only works if the failure was due to solder separation which it appears not to be in this case.

 

[netxzero64]

I bolded the items for question. What do you mean by 650? and what baked? Baked cookie or something? I am trying to figure out what you are trying to say. Please rewrite.


My answer to your post is that your GPU is shot after you tried to reapply the heatsink installation for the GPU. It might not have made full contact between the two parts and overheat. Power supplies rarely burn up the video card, usually the CPU first or the motherboard.

You F'ed up your graphics card man sorry. Get a new graphics card.


cheez

hahaha! sorry for not clarifying, my card was a GTX 460 then I baked it twice in an oven. GTX 650 was the 650 that I was referring to.

---------------------------------------------------------------------------------

Thanks for your insights on this guys. I think I messed up when I was on the process of reapplying tim, I may have had damaged the card while I was screwing the HSF out.

Well, I am planning to get a 680 on the 15th. I could've have gotten some cash out from the 460 as I am planning to sell it though. geee..

 

[PliotronX]

hahaha! sorry for not clarifying, my card was a GTX 460 then I baked it twice in an oven. GTX 650 was the 650 that I was referring to.

What type of PSU? Is it the EarthWatts in your sig? I would give it a 5% chance an Antec PSU would damage a GPU and that's being cautious. If it was one of those "MaxCool Super Extreme 550W" deals that are physics-defyingly lighter than the copper wire would elude it to be, I'd up that to 20%.

True story like others have said, incomplete coverage of the GPU die TIM have produced similar symptoms in some of my previous cards. X method if you can spare the TIM, line method otherwise.

 

[netxzero64]

What type of PSU? Is it the EarthWatts in your sig? I would give it a 5% chance an Antec PSU would damage a GPU and that's being cautious. If it was one of those "MaxCool Super Extreme 550W" deals that are physics-defyingly lighter than the copper wire would elude it to be, I'd up that to 20%.

True story like others have said, incomplete coverage of the GPU die TIM have produced similar symptoms in some of my previous cards. X method if you can spare the TIM, line method otherwise.

Yup its the one on my sig. Well, I applied a pea-sized TIM at the center of the GPU and nothing more. I thought what damaged my gpu was the way I handled it during unscrewing of the cooler of the GPU. I still have to wait for 8 days til I get a used GTX 680 for $350. Much more expensive than a brand new R280x.

 

[24601]

Yup its the one on my sig. Well, I applied a pea-sized TIM at the center of the GPU and nothing more. I thought what damaged my gpu was the way I handled it during unscrewing of the cooler of the GPU. I still have to wait for 8 days til I get a used GTX 680 for $350. Much more expensive than a brand new R280x.

Brand new 770s are 330 USD, so unless that 680 is 4gB, I would get a 770 instead of the 680.

 

[nightspydk]

To your original question I think that risk is highly overestimated. Even back 5-10 years the nvidia control panel complete was set to downclock the gpu if undervolted. Not many seem to know that, but I've seen that 1st handhand. So not the psu heat is another issue.

 

[netxzero64]

Brand new 770s are 330 USD, so unless that 680 is 4gB, I would get a 770 instead of the 680.

I don't live in the US that is why GPUs here in our country are a lot expensive that you guys in the US or Europe. GTX 770s here cost about $400.

To your original question I think that risk is highly overestimated. Even back 5-10 years the nvidia control panel complete was set to downclock the gpu if undervolted. Not many seem to know that, but I've seen that 1st handhand. So not the psu heat is another issue.

So the culprit really is heat? Or due to my abusive overclocking of the GPU? Temps for my cyclone 460 when overclocked at 900/1000 1.087v reaches 78c at 100% fan speed.

 

[nightspydk]

Lets try do a reassesment can you do that. Remember I did not say you did not run underclocked, I said that you are going to if you push your system beyond means. Besides I have 2 gtx 460 cards and whatever temps etc you throw at me they will be received with a grain of salt.
(That is because barriers have been put on the card from the manufacturer ..and I'm not talking about shaders..)

edit

 

[nightspydk]

Lets try do a reassesment can you do that. Remember I did not say you did not run underclocked, I said that you are going to if you push your system beyond means. Besides I have 2 gtx 460 cards and whatever temps etc you throw at me they will be received with a grain of salt.
(That is because barriers have been put on the card from the manufacturer ..and I'm not talking about shaders..)

edit

Show me a 4xx card around the 100 deg C. That is interesting. Sleep well my friend on that.

I have been ex 128 deg C just so you know.