Sign in with
Sign up | Sign in
Your question

7970CF Overheating(Fan at 100%)??

Last response: in Graphics & Displays
Share
February 22, 2013 10:37:14 AM

I got a whole new system yesterday, that includes two XFX 7970 BE(@1000MHz).
My problems started when I plugged it on my three monitors, 10 seconds after windows loads, the fan goes up to 100%, and in 20 seconds, the system shuts down. However, If I plug it into a single monitor, I can use the computer easily, as long as I don't run any demanding application.

The drivers are updated, ULPS is off, tried turning crossfire off, the top card overheats anyway.

If I run a demanding application like FurMark, the system jumps to 103ºC in 3 seconds, and If I leave it at that, it turns off.

I Got a HAF 932 Advanced, with 3 extra fans installed.
The cards are ~2 inches away from each other.

i7 3770K STOCK - Zalman 9900MAX Cooler
Asus SABERTOOTH Z77
8GB DDR3 1600MHz G.Skill Sniper
PSU HX1050
2x XFX 7970 BE

Can you guys help me?
February 22, 2013 1:14:25 PM

If its happening as quickly as you say then it sounds like the heatsink on the cooler isnt properly mounted on the GPU. As a test, remove the top card that is overheating and only run the second card. As long as you dont have any problems with the 2nd card then I would say thats a clear sign card 1 is defective and you should do as the BigMack70 suggested and return the pc/card.
m
0
l
Related resources
February 22, 2013 7:37:32 PM

Guys, I opened the PC while using it, although the thermometer said 100 degrees, it was cold.
I disabled crossfire and took that card out, only using the secondary now.
I was able to push it to 1200MHz with no problems, its at 70ºC
The guy at the store that sold me the GPU said that I must take it to him next monday, so he can test the obviously broken card, and give me another one.

Even with one GPU, the 7970 BE is Killer, I am getting 30FPS on Far cry 3 totally maxed out/dx11 on Eyefinity, its pretty playable.
I only need the other to get ~55 FPS, without having to overclock it.

Just one last thing... If the card is not actually heating up, isn't it a software problem?
Is there some BIOS update or something like that to solve the problem?
m
0
l
February 22, 2013 9:09:52 PM

I'm also curious about this what case are you using? Could this also just be a case of there being not as much room of airflow between the two cards bigmack? What happens when you place the other card in do you get similar temps as you did with the card you had in on its own?
m
0
l
February 22, 2013 9:43:47 PM

BigMack70 said:
He says he's using a HAF 932 and has space between the cards, so I doubt it's the issue.

Even if he did have two cards sandwiched together, I wouldn't expect the top card to get as hot as fast as he is describing without some sort of crazy overclocking going on. 90C load temps wouldn't surprise me in that case, but near-immediate 100C temps leading to constant GPU throttling? Yeah something's wrong there...

The case the OP is using is not going to make any difference, it's the motherboard that dictates how far apart the cards are.
m
0
l
February 22, 2013 9:45:25 PM

bigshootr8 said:
I'm also curious about this what case are you using? Could this also just be a case of there being not as much room of airflow between the two cards bigmack? What happens when you place the other card in do you get similar temps as you did with the card you had in on its own?


I'm using the "Good" card ~1200MHz as we speak, its peaking at 75ºC with furmark.

So no, the airflow is no problem

My question is that if it is a Hardware fault or some kind of crazy BIOS Corruption that's causing that.
m
0
l
February 22, 2013 9:45:57 PM

Yea, mouse though I think at the end of what he said though he understands that. I was just trying to make the point of trying both cards to try and rule out any possibilities. But yea there is no reason for a graphics card of any series to reach the 100 C mark.
m
0
l
February 22, 2013 9:53:45 PM

Right, the same sort of thing goes with CPU coolers if its not properly mounted there could be not enough pressure or too much pressure leading the temps to be off. I agree with you.
m
0
l
February 22, 2013 10:24:50 PM

BigMack70 said:
It's almost certainly a problem of how the cooler is mounted on your defective card. It's really a simple issue.


Just tried the old Card, ran furmark for a few secs(15s before crashing)
MSI Afterburner Showed me 102ºC, but Furmark said it was at ~50ºC, which is the most probable, since the card is not hot when I touch it.
Still, the fan maxes out and it turns off...
m
0
l
February 22, 2013 10:48:41 PM

I mean maybe probable for a short duration but over time that card will probably get in the 70-80 C mark.
m
0
l
February 22, 2013 10:50:01 PM

but Is intsta- 100ºC Possible? Even If I completely removed the coolers, could it go up in less than 4 seconds?
m
0
l
February 22, 2013 11:05:03 PM

I'll go to the store on monday, already called the guy, he will get me another one from the supplier.
m
0
l
February 24, 2013 12:23:12 AM

Update:
I'm monitoring the defective card right now.
Strangely, when I plug in more monitors(more than one) it starts to heat up, even though the usage is low(from 1% with one monitor, to 3% with two, it stats heating up at an alarming rate.
m
0
l
February 24, 2013 7:47:30 AM

Then the culprit has been found the card must go back sir :p 
m
0
l
February 24, 2013 4:12:09 PM

Best answer selected by carcassdevourer.
m
0
l
!