Overclock.net banner

980Ti Hybrid pump broke?

1620 2
Two days ago I started Ark and the temp of my 980Ti Hybrid went from 25c to 90c in seconds. My two nf-f12's that are on the radiator where still spinning, but they where not expelling any heat. I feared that the pump broke or the mounting got loose somehow. To check for the latter I took the shroud off, but everything seemed fine. I put the card back in my system and could hear the pump again. I was relieved until I fired up Ark (99% gpu load) and saw that my temps where above 70c instead of 50c like before. So it seems that the pump kinda works, but not the way it should?

Thinks I tried or thought of:
- I have thought about reapplying the pump to make sure that it is making good contact. I have not tried it before posting here, because there is no reason that it could have come loose, since my system hasn't moved in months.
- I have tried putting my case on it's side to see if a air bubble is the cause of my problem and that didn't help.
- I ran the nf-f12's at max speed (~1400rpm) and the gpu temp reached 65c in Ark. Before they ran at ~800 rpm and it reached 50c like I mentioned before.

Stuff that might be important:
- Mr-Dark (a helpful user) made me a custom bios which I have used for several months now (thread). The only thing that was changed was the power limit.

GM200.zip 152k .zip file

- I changed the shroud (the free promotion by EVGA one) when I saw that the pump 'worked' again and I figured I might as well take the opportunity. Other then the nicer look, the shroud has a led that plugs in the pcb. EDIT: turned the led off in the Geforce Experience software and nothing changed.
- Before I noticed this problem, I had several cpu related blue screens because I was testing a higher overclock. This seems very far fetched, but is it possible that a gpu bios can get corrupted when cpu related blue screens occur?
- My system seems a lot more quiet then before, which substantiates the theory that something is wrong with the pump, since that was the loudest part in my build.
- I have noticed that the gpu temps jump up and down more, instead of changing gradually like before.

Any help is much appreciated!


1 - 3 of 3 Posts

5,062 Posts

sound like the pump not working at the normal speed, that the reason for the normal IDLE temp and high load temp..

the same happen to me with corsair H110I GT cpu cooler.. the bios have no control over the pump, the pump pull the power only from the card while the fan setting in the bios for the blower fan nothing else..

simply flash the stock bios back and RMA the card, EVGA is very good so no worry

109 Posts
Discussion Starter · #3 ·
Thank you for the reply. I probably end up rma'ing the card like you said. The crazy thing is that I have thought about undervolting the pump and now the it has done it for me. Should not keep it like this though, because the pump has already stopped working once and that can happen a second time. Since it is currently usable I can at least schedule the rma when I don't game as much.
1 - 3 of 3 Posts
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.