Hard tubing help

[SonLux]

Perhaps some watercooling veterans can help me here.

 

Not long ago I upgraded my soft tubing system to hard tubing and was rather happy with the results despite having to resort to some unconventional angles and routing. 

 

 

Some time passed and all was going swimmingly, as time went by I started noticing liquid pooling around the seal on the tube running from the top of the GPU to the top radiator, the best I could do with it however was to dry it off and pray that it would not carry on. Unfortunately one evening before my eyes and behind the windowed side panel i watched the GPU drop away from the tubing and liquid spray up under pressure onto the motherboard, the PC instantly shut itself off and I pulled the power on it, whipped the panel off and managed to cram the tube back into the fitting and stop the leaking. I have bought and fitted a new Motherboard, new pump, and new fans as a repair and all is working again, while replacing the tubing I noticed that the offending connection was crushed and warped unlike any of the others.

 

 

 

The new system is currently working with improved temperatures but I am beginning to see a very similar warping on the same connection as before.

 

 

What am I doing wrong here? I assume it is a combination of the GPU heat and the weight of the waterblock just loosening and warping over time but it has only been about three weeks since the replacements, is there anything i can do to remedy this?

 

Thank you all.

[Shiv78]

Might be because the tube is not perpendicular to the GPU where it enters the fitting. 

[For Science!]

4 hours ago, SonLux said:

-snip-

Did you use EK Acrylic/PETG tube or was it another company?

Assuming the tube wasn't under tension I can't think of any other reason. I can't imagine your water temperatures being so hot to melt the tubes (given how much rad space you have and based on your setup).

[SonLux]

On 07/10/2017 at 3:52 AM, For Science! said:

Did you use EK Acrylic/PETG tube or was it another company?

Assuming the tube wasn't under tension I can't think of any other reason. I can't imagine your water temperatures being so hot to melt the tubes (given how much rad space you have and based on your setup).

It's all EK PETG 16mm tubing, I expect it to be the heat warping it combined with the weight of the waterblock, I get GPU temperatures of around 50 at idle and all the way up to 90 after a few minutes at full load, I have no idea why it runs so hot, the GPU is not overclocked, however the CPU overclocked to 4.5ghz runs 30 at idle and a max of 60 under load.

[For Science!]

11 hours ago, SonLux said:

It's all EK PETG 16mm tubing, I expect it to be the heat warping it combined with the weight of the waterblock, I get GPU temperatures of around 50 at idle and all the way up to 90 after a few minutes at full load, I have no idea why it runs so hot, the GPU is not overclocked, however the CPU overclocked to 4.5ghz runs 30 at idle and a max of 60 under load.

clearly there is an issue with your block installation. i would advise you to drain the loop and take the block off the card and take photos of where you've put the thermal pads etc on here so we can confirm everything is ok.

[THFourteen]

Looking at this thread seems like the melting point of PETG is 85 degrees so it’s probable your high GPU temps have warped it. 

 

Also how on earth did you dry up the liquid and hope that everything would be ok?! You should have dismantled the look and investigate straight away

 

 

[For Science!]

14 hours ago, SonLux said:

-snip-

Please provide photos similar to the ones in this thread, where I was able to pin down the problem of a overheating card to an extra thermal pad applied by the user. 

[SonLux]

8 hours ago, For Science! said:

clearly there is an issue with your block installation. i would advise you to drain the loop and take the block off the card and take photos of where you've put the thermal pads etc on here so we can confirm everything is ok.

The next available day I'll be able to strip it all down is this coming Saturday so I'll be sure to do that then.

[SonLux]

6 hours ago, THFourteen said:

Looking at this thread seems like the melting point of PETG is 85 degrees so it’s probable your high GPU temps have warped it. 

 

Also how on earth did you dry up the liquid and hope that everything would be ok?! You should have dismantled the look and investigate straight away

 

 

By dry up the liquid and hope it's okay I meant overnight, it happened about 1am and I wiped it all down as best i could and left it unplugged overnight, I didn't try and turn it back on until I had disassembled, dried, and reassembled. 

[SonLux]

6 hours ago, For Science! said:

Please provide photos similar to the ones in this thread, where I was able to pin down the problem of a overheating card to an extra thermal pad applied by the user. 

I will provide photos when I re-apply the paste on Saturday, I hope it's not to do with the thermal pad positioning as they're where the instruction manual specified.

[Limecat86]

Well, seems like another reason to go with acrylic instead of PETG

[SonLux]

On 12/10/2017 at 11:19 AM, For Science! said:

Please provide photos similar to the ones in this thread, where I was able to pin down the problem of a overheating card to an extra thermal pad applied by the user. 

Thermal pads positioned as specified in the EKWB manual for the block

 

[For Science!]

2 hours ago, SonLux said:

Thermal pads positioned as specified in the EKWB manual for the block

 

 

You cleaned the silicon before taking the photo

But I assume you had a good amount of thermal compound on there? Did you follow EKWB's instructions for this part too? I've seen people being too conservative on the GPU thermal paste on a few occasions. Here is mine for reference.

 

[SonLux]

2 hours ago, For Science! said:

You cleaned the silicon before taking the photo

But I assume you had a good amount of thermal compound on there? Did you follow EKWB's instructions for this part too? I've seen people being too conservative on the GPU thermal paste on a few occasions. Here is mine for reference.

 

I used the same, if not a touch more compound when refitting today, while i dont have pictures of how much I applied I'm confident that I did it as best I could, I think the heat is transferring much better as the block itself is now much hotter. However the idle temperature is still sat at close to 60 degrees.

[For Science!]

22 minutes ago, SonLux said:

I used the same, if not a touch more compound when refitting today, while i dont have pictures of how much I applied I'm confident that I did it as best I could, I think the heat is transferring much better as the block itself is now much hotter. However the idle temperature is still sat at close to 60 degrees.

Grasping at straws now, but technically the block will perform better if you swapped the inle/outlet (I know in the instructions it says eitehr can be used. However, the water path would behave better if it hit the jet plate before hitting the microchannels. So the port on the left is better suited for inlet. Although I doubht this would bring the temps down more than a couple of degrees at best.

[Mick Naughty]

Were you monitoring the water temp along with system temps under load?