Custom Watercooling loop with bad GPU temps all of a sudden. Help?

Good afternoon,

Long time reading, first time poster.

I am sure most people are familiar with Jayztwocents. As a first time builder/custom loop guy I followed his build with a few different tweaks when it comes to the hardware but overall if you picture Skunkworks I've got a copy of that. 2 different loops as one is for the CPU and one is for dual GPUs.

This was a fun build for myself and a friend I asked to join in with me along the way so we both gained some experience doing this sort of thing. When it was all complete everything was fine. Temps were in great shape and the GPU temps were around 20-25 for 1 card and 30-35 for the other (celcius). I read this was common so I didn't think much of it.

The other day I was playing Rainbow Six Siege and noticed some frame lag. I run dual MSI GTX Gaming Z 1080 cards in SLI. I exited out of the game and checked my temps and one card was approaching 87-90 degrees but rapidly dropping in temps since I had exited the game. My computer had been running fine for almost 2 weeks without any issues and nothing had changed prior to the temperature spike. I came back to my system a few days later and booted it up and checked some things and again the same thing happened with the temps on the Heaven benchmark.

I took the card out of the system and took it apart and cleaned up the thermal paste and re-applied some and put the card back together and put it back in my loop making sure to run everything so all the air got out. I am still noticing issues with the temps creeping up. Just running rocket league alone with MSI afterburner in the background showing me an overlay of my temps I'm seeing the card creep up to 80 degrees just on the start menu.


For the time being I'm off games until I get this sorted out and I'm hoping the community can send me in the right direction for what to check/fix.

I have a 560 rad (blackice nemesis) with 4x140mm corsair fans running at 100% on it (I have a commander mini on order to setup some fan profiles), ek waterblocks, distilled water w/red dye, a D5 Revo ek pump and then various fittings in the loop itself which is PETG. My CPU loop is doing just fine and keeping everything cool.

Any thoughts or comments are appreciated.

When you had the block off of the card did you take it apart and clean it? Also when you did the initial build did you flush all of the components well. This sounds line a case of something breaking loose and clogging the cooling fins.

Thank you for the reply, Lochekey.

When I first installed the rads I washed them out with distilled water by adding some into one of the ports and vigorously slushing it around before draining it. I can definitely take that apart again and try again. Would you suggest using any type of compressed air? I have access to that.

I did use some distilled water in a squirt bottle to wash through the inside of the block itself and cleaned off the thermal paste that was on the underside.

On Thursday I plan to take the loop apart again to go over everything since the commander mini will have arrived so I'll have the case on the work bench then.

If that is the only part that is showing issues then yes I would pull it apart. I would avoid using compressed air as the blocks are easy to disassemble. If you are worried about the warranty on the block you could try backflushing the block in reverse. I tend to hook my blocks up to a fountain pump in a bucket of water and flush then in both directions.

Sounds great. I will check into it as soon as I can and report back in case others have the same issue and I end up solving mine or need further assistance.

Yup, I would do a complete tear down and clean.

I have to ask, have you done any maintenance during those 2 years since its a pretty long time if you didn't and not surprised of a issue arising if this is the case.

Give us a list of your PC and loop as well as a pic or two, if you can.

As for Jay, I think he's alright but wasn't up to par with his custom water cooling info a while back. It's nice to see new gear as he gets tons of nice stuff from sponsors but don't think he runs the same loop long enough like your average joe. He completely misses the point of annual maintenance at the least as those with color fluids need to this more frequent with in that time period.

Regardless, you're in good hands now and to OCFs btw.

Edit: I almost forgot. I would steer away from the Corsair Link as it's a nightmare. They've had issues with it for almost 5 years now. I ventured away from it and payed a bit extra for a more consistent and amazing device called Aquaero 6. German engineered, nuff said.

Hi Gtx Jack. Thank you for also weighing in on my issue. I think you may have read my time frame on my original post incorrectly. I've only been up and running for 2 weeks so with such a short time being operational I haven't done any upkeep other than keeping an eye on levels(fluids) and temps

I would also think you have crud in the card block, tear it open and give it a good looking at, remember to look in the fittings and interconnects also.
very pretty build by the way.

nice change in the gpu rad return to reservoir, the hard tube looks better there.

I agree it is a pretty build. I will say that if I was doing it I would configure it as one loop with the pumps in series to better distribute the cooling potential of those rads.

I don't think I'd do that but, I might put a curly Q in each of the pump outlet tubes just to beak up the long runs.

I don't think I'd do that but, I might put a curly Q in each of the pump outlet tubes just to beak up the long runs.

What is the curly Q? Like a pig's tail on a bend?

Thanks for everyone's comments and help so far. Got the system torn down and flushed the radiator with distilled water x10 so far. Will pick up where I left off tomorrow after work. Will break down the cards and waterblocks and do some cleaning.

Anyone mention two years of red dye in the blocks? Just wondering.....

Conumdrum said:

Anyone mention two years of red dye in the blocks? Just wondering.....

Click to expand...

He has only had the loop together for 2-3 weeks.

To be fair, it did say two years and no, I didn't misread what I saw before the edit.

Uhh actually it never said two years. I edited only once because I wrote exited but my phone corrected it to excited.

when you get your rig back together grab a flashlight and look closely at your tubing and especially your bottom gpu for air bubbles. Also listen for gurgling in the rad on the gpu side. If you hear any gurgling tip your case every which way until you dont hear it anymore. same for any bubbles you may find. try to ease it out of the part its stuck in by tipping your case around. Being a new build its likely theres a bubble in your loop that needs help getting pushed into the res. Thats a sweet rig man! That case is a beauty. like Jack said well get you up running in no time. Welcome to OCF Dribbles!

I will indeed do that! No news to report yet but I will have something tomorrow. I'm a shift worker and switched to nights and my last 12 hour is tonight. I have managed to put everything back together and leak tested for a few hours with no issues. I'll do some tilting tomorrow and then reconnect it all and get back to everyone with an update and hopefully good news.

Update: You guys are freaking amazing with your help! I had hoped that my temp issue wasn't going to involve the amount of work that it did (taking everything apart) but that appears to have solved 1 of my 2 issues. When I was cleaning out my rad I did notice some crap in the distilled water with every wash out so hopefully that was exactly the problem I had but it may have also been the thermal paste too... who knows. Just for fun I also swapped out the thermal pads (with OEM replacements). I have managed to run heaven and tested out rocket league. Temps never soared above 50 degrees on 4k ultra settings.

Here is hoping, however, that the amazing people who are following this thread and those that have already helped out can suggest a solution to my last problem (not exactly liquid cooling related though).

When all this happened and I was fiddling around physically touching my video cards etc one of my two cards (the one the monitor was not plugged into) decided to pop a code 43. This has rendered the card useless in the system at this time so I've just disabled it. Research online suggest it has something to do with static build up? One forum I read suggested removing the card and then putting it back into the system etc. When I was solving this temp issue I removed both cards and re-did the thermal applications (pads/paste) and also decided to swap them spots on my motherboard (MSI Z170 Gaming Pro Carbon). When I booted up for the first time my monitor was plugged into the code 43 card. I did manage to get a successful boot however my monitor looked very slow to respond and pixelated.

I have since disabled the card again and plugged my monitor into the known good card (that previously had temp issues) and everything is hunky dory however I'd really like to have both cards functioning. Anyone have any ideas or should I maybe post into a new thread in a different area?

Thank you all again for your assistance with my cooling issue!!

If you've tried each GPU separately and in each other's PCIe lane and the same issue exists (artifacts), then I'd say the GPU isn't looking too good and I'd request a RMA with said company.