H75 High Idle Temps on R9 290 GPUs

[SiuKingBon]

Hey guys, 

I got two Corsair H75 water coolers for my two R9 290 GPUs which have been working great for over a year since I got them. Just recently, I believe after the Windows 10 Anniversary update, the temperatures have gone completely out of wack. The main GPU idles at 57C and the secondary at 41C, under load they both hit 94C which I don't think I've ever seen them that high. 

I've got them both plugged into Chassis Fan plugs on my ASUS Z77 motherboard and the BIOS reports them at ~1550RPM. 

I've cleaned out the dust on the radiators, updated the GPU's drivers and removed the anniversary update but the temps are still that high. 

I also tried swapping the pump's power to the OPT FAN but there wasn't much of a difference there. 

Anyone have any idea as to what might be the cause here? Any help would be appreciated!

Thanks! 

[GalacticRuler]

Do the H75s have indication LEDs? If not, you may have dead pumps. However it seems unlikely that they both died at the same time. Otherwise it's probably the monstrosity known as Windows 10. 

[SiuKingBon]

There's no lights that I can see on the H75s but I did pull out the main card altogether just to see if the pump was dead. 

I plugged it in on its own and I could hear the pump running, albeit I'm not exactly sure how loud or how much vibration would indicate proper functionality. So it's not completely dead. 

I might try Windows 7 if I really have to but it was working fine on Windows 10 for quite a while. Theres certain programs I have that require Windows 10 though.

 

 

[unholy_walrus]

I really don't know how a windows update could've made your load temps go up. What were they like before?

 

I'd just try to run the pumps at full power directly from the PSU just in case there are issues with the mobo fan headers. While you're at it, make sure the radiator fans are nice and good. If the temps still look that bad, I don't know what to say. I never ran my R9 290 with an AIO, but a single 120mm should be enough to handle the load and not produce 90c+ temps. The only thing that comes to mind really is a pump issue. While it's very unlikely that both units would start having issues at the same time, it's not impossible and I'm having a hard time finding an alternative explanation. 

 

In any case, your first priority should be to make sure that the pumps and fans still work.

[SiuKingBon]

Depending on the game it would go up to 70 at the highest. But even now at idle its hovering around 59-61 just sitting on the desktop. 

 

I'm thinking this may be a pump issue or maybe, since both of the GPUs are doing this, that the headers I'm plugging into on the motherboard for the pumps isn't providing enough voltage after the update? I'm not sure what would have caused this. 

The pumps and fans still work and I tested them individually already though.

 

The BIOS shows that those fan headers have RPMs of 1550 but I've never really paid attention to them before this so I'm not sure what value they were at previously. If I had a reference on the RPM then maybe we could get a lead somewhere. 

[unholy_walrus]

If the fans work fine, try running the pumps directly off of the PSU for example and see how the temps are. I mean everything here is more or less pointing to a pump issue.

[SiuKingBon]

Yeah thats what I'm thinking as well, I'll need to find a way to connect them directly to the PSU though, my PSU doesn't have a 3 pin fan connector. Probably have to get myself a 4-pin molex to 3-pin header to get that tested out. 

 

EDIT: I found some 4-pin molex to 2-pin, I'm assuming this will still work? 

[unholy_walrus]

Yeah, those connectors should be really cheap and very easy to find. I hope we can diagnose and fix the problem soon enough!

[SiuKingBon]

I managed to dig up a few out of my spare parts box, I'll give them a go later tonight or tomorrow morning. Hopefully it'll help us pinpoint the issue, or even better, maybe it'll fix it lol 

Thanks for the help

[SiuKingBon]

Alright, so I got the main GPU plugged in directly to the PSU using a molex to 3 pin converter, the idle temperature is now 47C. Its dropped a whole 10C just by changing away from the motherboard header. I have no idea why the header isn't providing enough power all of a sudden. But if its a problem with the motherboard then that would explain why both GPUs suddenly started having a problem with their pumps. 

 

Under full load I'm hitting 84 which is hot but at least its not at 95 like previously. 

[SiuKingBon]

Ah bummer, after some additional stress testing it seems like the temps are back to how they were previously. I guess its not a problem with the power then. There must be an issue with the pumps then I'd imagine?