Water Cooling Flow Problem

WorLad

WorLad

WorLad

Associate
Joined
12 May 2012
Posts
31
Hi

I'm having a problem with my water cooling loop. I have a D5 pump which is pushing water through 3 GPU waterblocks, a CPU waterblock and a quad rad. The water seems to flow fine through everything except for the outlet of the radiator, going to the res, which seems to just trickle very slowly. I have flushed the res and there appears to be no blockage and have bled the system for about 2 hours and have tipped the system in all different directions to correct the problem. The only thing I can think of now is that it's the pump, but I wanted to post here to see if any pros out there can spot something I've overlooked.

I've attached pics to help resolve the issue.

Thanks guys!

 
The pump is turned up to max and there was no resistance during the flush.

[EDIT] It should be noted that the flow of water goes upward from the pump, through the GPU blocks -> CPU -> Rad -> Res, so it is pushing against mavity, however it never did this on my last loop and I cannot find any good way of reversing the flow...
 
Last edited:
a D5 shouldn't have an issue with a loop that size. If water was happiy flowing through when you flushed it I can only presume there is an airlock in there but airlock normally stops water flowing at all rather than trickle in my experience. I presume something didnt come out in the flushing or migrated from another component. If its reduced to such a trickle at the end it could well be blocked.

Is the pump sound like its struggling? Eg. higher pitched?

I wouldn't of thought youd even get this far if the pump was knacked.

Can you give a side on photo showing the whole loop?
 
No, the pump sounds the same as it always does, nice and quiet and seems to push the liquid through the loop nicely when I fill it.

The water that left the rad was as clear as it was when it went in, nothing floating around and no grime. It's also worth noting that the radiator is brand new.

As you can imagine, I'm scratching my head with this one! :confused:
 
Hmm so am I.

The D5 has a head height of 3.35m so I doubt your loop direction is the problem.

Only thing I can honestly think of is something blocking in the rad. :confused::confused:

Only other thing I could suggest would be to dismount the rad if tubing allows and rotate it freely without the case see if it finds a stubborn airlock. I always found the rads at the top of the case a nightmare.
 
Last edited:
Going through three GPU's and a cpu block - along with a 120.4 rad is defo going to cause some loss of flow mate, how much though is going to be debatable.

Have you tested by taking the rad out of the equation? - does it still just trickle then?
I had a similar problem with a 120.3 Thermochill rad - wasn't much I could do about it tbh......
 
Sounds like your rad is totally full of air. My d5 did This for my top rad until I got it properly bled. Was the same positioning as well up top. I played around opening and closing the res while the pump was Going. But other than that I am stumped why yours didn't sort itself out
 
Last edited:
Massive colour difference from what goes into the gpu compared to what comes out.
Click to expand...

The tubing going into the GPUs, the tubing coming out of the bottom of the res and all the tubes at the bottom from the pump and on the tap are from the previous loop, the rest is new tubing. I did it because I didn't see the need to change all of that since I was only doing the top bit... Just impatient I guess. It was clean though and didn't have any grime in it, but the tubing is a little pale compared to the new stuff.

I'm going to flush the rad again and test the loop without it to see if I can find what may be causing this...
 
UPDATE!

I've been up late conducting tests to isolate this issue and I think the problem is with the CPU water block.

I set up the loop with only specific parts to diagnose and locate any blockages. First I tried the radiator only, then the CPU and GPU blocks together and then the GPU blocks.

Here are some pictures of the tests...


RAD ONLY



GPU AND CPU BLOCKS



GPU BLOCKS ONLY



As you can see in the second test, it couldn't fill the tubes properly, however on both the rad and GPU blocks tests it ran perfectly so it must be the CPU block.

I'm going to isolate the CPU block and test it just to make absolutely sure...
 
Last edited:
CPU BLOCK ONLY



So it's definitely the CPU block! Let's open it up and see what's wrong...


INSIDE THE BLOCK



So apparently this is what happens when you leave your CPU block for years without cleaning it! Lesson learned folks!!

Thanks for the help guys!
 
You must log in or register to reply here.
Back
Top Bottom