• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

980Ti fried (hardware gore inside)

What were temps? 1500mhz and increased voltage in summer temps is going to be an issue with what doesn't look to be an ideal application of thermal paste.
 
Did you wipe off the corner for the photo? Or was it actually like that? If it was like that then it looks like you did not cover the corner with thermal paste at all... so that part probably overheated due to having no thermal paste on it...
 
NirK said:
If it was like that then it looks like you did not cover the corner with thermal paste at all... so that part probably overheated due to having no thermal paste on it...
Click to expand...

Exactly this. If you did infact cover the die completely then you must have tightened that corner of the hearsinkfirst therefore squeezing all of the TIM away from that corner.

Put it down to experience and more on.
 
Seems likely that if you had card running fine for 1 year then it dies shortly after you removed cooler that you fitted it wrong and the pics look like it wasnt flat.
 
is it possible that the cable from the fan header was pinched under the heatsink, slightly raising that edge.

Looks like an imperfect contact caused this..
 
Disco_P said:
MSI allow cooler removal.

https://www.facebook.com/msiuk/posts/652634514773229

This was still valid when I asked in late May.

It just looks like one of those things, cooler removal had nothing to do with it. RMA time - be warned MSI are sloooowwww.

Finally I would play it safe and get a different PSU. OCZ have a fairly rubbish rep when it comes to power supplies.
Click to expand...

I remember reading this when the topic raised it's head on this forum before, Just screw it back together properly and send it off, Or if you have another year on the warranty, leave it for 6 months until there Maxwell stocks depleted and then send it in :D Only kidding there not EVGA sadly.
Earlier in the year a gent on this forum sent EVGA a Titan Black and got a Titan X replacement and then when he said I was running crossfire they replaced his second Titan Black free of charge. Anyway back on topic, Send it in and please let us all know how good the customer service was.
 
Let us know please.

Just bought myself a MSI 980Ti 6G from the MM, it should be covered until mid 2018 but don't want this happening :D
 
It's pretty obvious that the corner that went volcanic wasn't contacting the heatsink properly, you can even see the tide marks from each operation as the grease heated, retreated and left that corner TIM-less.

Incorrectly & unevenly tensioning the hold down screws on a CPU/GPU cooler will result in this sort of failure, on the old Athlons you'd crack the core and kill it instantly. Unfortunately you pooched it mate. Next time do 1/2 a rotation on each screw in a diagonal opposing pattern to get a consistent and complete mating of surfaces. It takes time but it's obviously worth the effort.
 
Last edited:
+1 for badly tightening the GPU screws... Also you are running it at the max voltage allowable in afterburner, probably no need for that - be sure to check you dont have it set to performance mode in nvidia control panel next time also to avoid full clock 24/7.
 
Yeah its defo like the others have said because something you did when refitting the cooler, can't see MSI replacing that :(. Using 2 MSI 980ti's myself and the temps are great so never had a reason to change the paste. I run my 2 at the max voltage allowed on afterburner too but not had a problem. You might have even forgot to tighten 1 corner but I wish you luck mate MSI customer service is really good so you never know.
 
Last edited:
Thanks for the reply's everybody. I do agree that it is probably my own fault to be honest. I am experienced with replacing thermal paste having done it on numerous other cards (fitting 3rd party coolers, waterblocks etc). I also delided my 4770k and applied the famous liquid metal direct to the die so this one was a walk in the park in comparison :p

There is always the chance you can mess it up though and its not hard for complacency to creep in when you've done something numerous times so I will put it down to that. Its the risk you take after all.

What does surprise me is how well the card was running after i applied the new paste. On the stock paste i was running the card at its factory settings as it was running at 82 degrees so GPU boost would keep the clocks at 1380mhz or lower no matter what i did in afterburner. It was clearly at its thermal limit. With the new paste, temps dropped to the mid 70s which allowed me to actually overclock the card for the first time since I got it.

I've been running it at around 1500mhz with +87mv since without any stability problems at all. I've put about 80 hours into witcher 3 with them setting so would have expected a problem like dodgy paste application to result in stability issues before a massive failure like this. The card has been reporting temps bellow 80 degrees even through the heat wave whilst overclocked.
 
Last edited:
Well hopefully they do RMA it for you and you never know they may give to an upgrade ! Fingers crossed mate, just not sure what they are gonna say when they see the burn mark! lol
 
I run my 2 cards on air both running stable at 1500 core mate, have setup a fan profile so it changes depending on the load. Great coolers and really for me no need to change them or the paste. Gaming have never seen temps go above 75.
 
I don't believe you were the cause of the burn out. If it had happened immediately after the repasting then it would be more plausible but a month of solid use without issue suggests that it was fine.

The lack of paste wouldn't cause that, even in the corner where it is thin. Heck there is more there than what was originally on my MSI 680. Surely if something is overheating then it would throttle to protect itself?

I would RMA the card. Personally I would have no qualms with adding a little bit of paste where it is thin and saying that I changed the thermal paste 6 months ago and haven't had any issue since. Would help avoid them pointing the finger at you.
 
Hm not sure, I think if there was no paste there, or it was pushed to the other side, that it could have caused it... most of GPU running at 80c for example, so there is no thermal shutdown from the sensor, but that corner running super hot and eventually burning. Honestly I do not know but it looks like that.
 
You must log in or register to reply here.
Back
Top Bottom