• 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.

Calling all Gurus, I'm in desperate help here, please come in and help regarding GPU issues.

GUYS, HUGE UPDATE

decided to turn off pump and try without pump, no black screens, pump on black screens immediately, pump off no black screens. GPU gets bloody hot really quick, but in a dozen or so tests for like 10s, not one black screen with the pump off. What does that mean?
 
yeah i think so mate, will have to wait till tomorrow now though. Will have to be bloody careful thought because hardline tubing all connected so wont be able to move it far

I just wish there was a way to test it for longer period with the pump off. cant believe it gets to 80c and 90c within 10s..wow
 
Last edited:
yeah i think so mate, will have to wait till tomorrow now though. Will have to be bloody careful thought because hardline tubing all connected so wont be able to move it far

I just wish there was a way to test it for longer period with the pump off. cant believe it gets to 80c and 90c within 10s..wow

Is there a way to run the pump off another psu and isolated from the case ( out of the case ) To rule out grounding issues?
 
ive got hardline tubing, I can run it from a PSU outside the case, and might be able to detach it from the case, but it'll have to remain inside the case as i wont be able to move it very far with the hardline tubing

edit: just tried using external PSU with pump connected to that with the pump still attached in case, black screens. pump off again, no black screens. Tomorrow I'll get a voltmeter and try and detach the pump from the case then try again. I have a very good friend (mynight) who seems to think the pump might be charging the liquid when its on.
 
Last edited:
I'm far from 100% but if the pumps off and it works kinda leaves two options. One the pumps leaking voltage/current which we should be able to detect on the waterblock (not sure how to test for voltage/current in water) and the second being the pump is producing much more of a magnetic field than it should.

The latter is quite a stretch but other than introducing an abnormal current I can't think of what else a pump can contribute.

Then again if the D5 is anything like the DDC there is zero mettalic parts in contact with the water making any leak unlikely.

Basically it seems the pumps at fault but as to why it's somthing Id never considered a factor before.
 
Last edited:
no advice to give (solid advice and help all round lads) but best of luck sorting this Tony. I had a 4870x2 under water before but luckily avoided these issues.

I instantly thought grounding issue after reading the thread.

Long shot but got another pump you can try? Used this pump before? What liquid you use? Distilled can quickly lose its less than electrical properties just in contact with metal.

EDIT: Removed me from Origin O.o was gonna msg you this
 
Mayhems past my time of dabbling with water but far as I know its low conductive compared to normal distilled and i never had issues with that.

It's a proper random problem, but signs pushing towards pump somehow. I have never heard of this happening though and I was a prominent member on another watercooling site with ~4k posts
 
Hey Kaap, I know we live within 10 minutes of each other, fancy trying to help me out a little here?? I'm really struggling man.

I have been following this thread and if there is anything I could help with I would do so.

Unfortunately some of the things that have come up are just out of reach for me. For example you wanted to work on the 290X coolers, I have 4 but they are all reference and of no use as your ones are MSI non ref.

I also have a spare monitor but again it is a DVI DL not DisplayPort.

If you have not done so already make sure you have enabled trust.
 
OK, still pointing to a ground loop or noise issue. DP is incredibly sensitive to both. Any significant ripple on the mains local to the computer can also kill it.

Looks like you're real close to overcoming this. Get a piece of flexible tubing and run just to the pump, run the pump outside the case and see if you can get it to work by keeping the pump physically distant from the card. This will prove if the pump is creating significant EMI that's disrupting the DP stream. You can throw some makeshift shielding together with tin foil and whatever to help prove the point.

TBH I'd just be looking to try a different pump at this point...
 
Wonder if OCUK can test for EMI issues (assuming Tony doesn't find current on the blocks/water). If not would there be issues with RMA as its most likely a very specific set of variables needed to reproduce said interference.

I agree with swapping out the pump at this point it's the only component that makes a difference thus must be at fault. I'm sure Tony's beyond caring what's wrong only what can be done to fix it.

(Selfishly I am intrigued to know what exactly is causing this)
 
Hi! I saw the Twitter post of Kiwi! Sorry again about all the issues you had so far. :(

About the pump, I would be surprised if it was a manner of magnetic field or even the fact that the liquid as become charged. It might very be a ground problem or simply the pump being faulty.

One thing you could do to isolate the pump from the case without dismentling all the hard tubing, would be to detach the pump and decouple it from the case with a piece of foam or even a cloth towel. Something not thick so you could still have every tubing in place. Also make sure that the molex connector is correct and does not create a short by itself and one other thing I would do is to disconnect the rpm signal (or pwm if it's not a vario) from the motherboard. If your pump is PWM, it will run at a very low rpm since it won't be connected to the board but will still work.

I know it looks like you haven't heard from our CS team and i'm sorry about that but I will do my best to help here.

Also, about the fact that you don't understand why we didn't do a full block for it since it looks to be the most popular 390 on the market, all I can say is that even if it's a popular GPU doesn't mean it's a popular GPU for watercoolers. It's sadly the truth for the vast majority of the AMD line. This is why we tend to stick to the reference design compatibility for the new AMD's. Also, I will discuss with my superiors about the idea of removing the compatibility of it with the Thermosphere. The same issue is currently happening with the 980 Ti Lightning and the Thermosphere.

Keep us inform! I'm usually mainly in the Watercooling threads but will keep in touch on this thread too.
 
Thx very much guys for the continuing help here, you've all been helping greatly.

I've not said anything publicly in detail, but this build has been particularly emotional due to the passing of my father very recently. He loved seeing how I could build a pc, he put some money in his will to help me with this as he knew my intentions before he passed. Along with some really beautiful words, I felt this build was very important to get right. Last two weeks have been pretty rough, just the thought of it not working has been tearing me up a little due to its state and what my late father said to me in his will....kinda felt like a failure, so apologies to all at times if I come across a little raw.

Hopefully tonight when I get home from work, I'll take the pump out and try running it not attached to the chassis. Now I just need to find a fix.
 
Sell it all and buy a 980Ti with Gsync - duh!

;)

Seriously feels like you/we/us are getting closer to getting it fixed. Good replies from all the WC experts in here, it's nice to be part of a community that actually helps people with more than just "spec me" threads.
 
Back
Top Bottom