Watercooling loop leaked twice... (hardline)

Associate
Joined
31 Mar 2019
Posts
29
Hi everyone !

A little while ago (2 months), I built my 1st ever hardline custom waterloop.

1 week ago, I had to move my PC from one room to the other one and the same day after 2hrs of gaming it started leaking. After investigation, it occurred that the end of the tube section was not perfectly round which I assume was the cause.

I bent the tube again with a precision which I'm quite proud of and tonight, 5 days later and again after 2hrs of gaming, it started leaking again, plus in the same spot.

The only obvious reason that comes to me is the pressure, as the tubes were very hot but is that a thing in water-cooling loops ? How should I handle it ?

FYI, when reinitialising the loop, I let it run 2hrs with fill port open then closed it.

Honestly I have strong doubt that my tubes could be wrong again which is why I'm coming to you : what reasons could cause my loop to leak for no obvious reason ?

I'll be glad to provide any helpful info, I'm feeling a bit desperate.


Thanks a lot
Vincent


IMG_1301.thumb.jpg.9845a3ae524f34cf1fdebacb94d684d4.jpg
 
Associate
OP
Joined
31 Mar 2019
Posts
29
It's interesting that it's leaked twice in the same place. Given that you replaced the tubing, what's left is the fitting. Have you got a spare you can put in its place? If not, do you have spare O-rings?
I actually have 1 or 2 spare fittings. I replaced a few of them on the 1st leak. I’ll give it a try !

Any opinion about TT fittings, which I’m using here ? I only read negative reviews about them and get usually bashed when I mention using them.
 
Soldato
Joined
18 Oct 2002
Posts
8,275
Location
Aranyaprathet, Thailand
Sorry, never done hardline so can't comment on TT (ThermalTake?) fittings or any others but I know there can be problems with O-rings in all water-cooling. I just had a major leak in a waterblock because of a pinched O-ring which cost me a motherboard even though I was using Cryofuel.
 
Associate
OP
Joined
31 Mar 2019
Posts
29
EA0D1D9B-C600-4914-B6B3-49143157821D.jpeg.da2321fe22da3fb62657c01f182e01dd.jpeg

Additional info ;



i dismounted all the loop and I noticed this : all of the tubes, even the ones entering te fittings in a perfect straight line like the one on the picture, have partially or totally retracted like compressed or overheated. I guess this is a possible cause.

is that common in hardline loops ? What’s the cause ?



I thought it could help with diagnosing.
 
Soldato
Joined
28 Dec 2017
Posts
8,454
Location
Beds
Is it PETG tubing, or acrylic? PETG tubing can deform at quite low temperatures like 60°C. What was your water and component temperature? PETG tubes have been known to bulge and/or pop out of fittings.

60°C is also typically the highest temperature a pump is rated for, so you should aim to keep liquid much cooler even if using acrylic.
 
Associate
OP
Joined
31 Mar 2019
Posts
29
Is it PETG tubing, or acrylic? PETG tubing can deform at quite low temperatures like 60°C. What was your water and component temperature? PETG tubes have been known to bulge and/or pop out of fittings.

60°C is also typically the highest temperature a pump is rated for, so you should aim to keep liquid much cooler even if using acrylic.


These are alphacool PETG tubing.

The pump is a barrow ddc pump

Rad is Ek 240mm

Rad fans are Corsair ml120 pro

They pull air which is exhausted by case fans on top & rear which means hot air travels inside the case and temps may be high.


Last temp check after 2hrs of apex : cpu 45C and GPU 50C (peaks around 60) but I assume the fluids can be hotter which may explain the issue if they bend around 60.


I can’t really add rads inside this « small » case.
Right now my best solution would be to switch to soft tubing in order to prevent tubes from retracting but if I follow you, the pump could fail too ?

What would you advise ? Any other ideas ?
 
Soldato
Joined
28 Dec 2017
Posts
8,454
Location
Beds
The good news is, your fluid temp couldn't really get higher than your chips. The CPU and GPU are the source of the heat so they would always be hotter than liquid temp. If neither component exceeds 60° you can be confident your fluid was lower.

Still I'd get a sensor of some kind, I'm planning to add one to my next loop, for more accurate data.

Your graphics card is using 300W+ though, so that's a lot of heat to dissipate through a 240 rad even without CPU added. You could move to acrylic hard tubing as you won't have to buy all new fittings.

Have you checked internal case temperatures too? It's worth checking, though I don't know if it would impact the tubing.
 
Associate
OP
Joined
31 Mar 2019
Posts
29
The good news is, your fluid temp couldn't really get higher than your chips. The CPU and GPU are the source of the heat so they would always be hotter than liquid temp. If neither component exceeds 60° you can be confident your fluid was lower.

Still I'd get a sensor of some kind, I'm planning to add one to my next loop, for more accurate data.

Your graphics card is using 300W+ though, so that's a lot of heat to dissipate through a 240 rad even without CPU added. You could move to acrylic hard tubing as you won't have to buy all new fittings.

Have you checked internal case temperatures too? It's worth checking, though I don't know if it would impact the tubing.


The point is I can’t do extra measurements before spending money for this fix so I’m trying to determine what would be the best pick.

I have a 280 mm rad and 2 ml140 pro which I was initially planning to use but didn’t fit in my case.

Moving to acrylic would mean buying acrylic and angled fittings (lus extra acrylic as I never worked with it and is hard to bend)

How highly do you think that switching to a 280mm would help ? I mean I would just have to find a case that has enough room and new tubes, no need for new fittings.

Ps. I just realize that I could find a case that would allow me to use both my 240 and 280mm rad lol
 
Last edited:
Soldato
Joined
26 Sep 2010
Posts
7,157
Location
Stoke-on-Trent
Moving to acrylic would mean buying acrylic and angled fittings (lus extra acrylic as I never worked with it and is hard to bend)
Not true. Acrylic is just as simple to bend as PETG, you just need a little more heat exposure (softens at 80 degrees, not 60) and therefore a little bit more concentration. It also doesn't flex like PETG so you have practically zero leeway with your measurements and cutting, but you should be doing that already. just because PETG can flex doesn't mean you should be sloppy and force tube runs to fit.

Personally I think you have a duff fitting or you have lateral pressure from the tube run not being the proper length so you're not making a proper seal.
 
Associate
OP
Joined
31 Mar 2019
Posts
29
Not true. Acrylic is just as simple to bend as PETG, you just need a little more heat exposure (softens at 80 degrees, not 60) and therefore a little bit more concentration. It also doesn't flex like PETG so you have practically zero leeway with your measurements and cutting, but you should be doing that already. just because PETG can flex doesn't mean you should be sloppy and force tube runs to fit.

Personally I think you have a duff fitting or you have lateral pressure from the tube run not being the proper length so you're not making a proper seal.


Thanks for your message !

So if I follow you, it may be caused by not perfect alignment (tension on the tube) combined with high heat.

My plans right now are : switching to a larger case in which I could use both my 240 and 280mm radiators in order to lower temps. Plus, i would have more room for accurate insertion.

Would this solve the problem in your opinion ?
I could combine with switching to acrylic. I think I mixed my memories with actual glass, which is why I thought you could not bend it :)
 
Soldato
Joined
26 Sep 2010
Posts
7,157
Location
Stoke-on-Trent
So if I follow you, it may be caused by not perfect alignment (tension on the tube) combined with high heat.
I reckon so yeah, but as others have stated it's unlikely your coolant has hit 60 degrees and deformed the PETG because your actual components aren't that hot. It's possible that something is misreporting because you said the tubes actually felt hot, which indicates your coolant temp is high.

Do you radiatiors feel hot to the touch?

And yes I would definitely put the time in to ensure your tubing runs are supremely accurate regardless of which material you use. Note that I'm not saying PETG is rubbish just because I personally don't trust it, there are plenty of people who have great experiences with it. I just think you have a fundamental issue or two with your current setup which is exacerbated by using PETG.
 
Associate
OP
Joined
31 Mar 2019
Posts
29
I reckon so yeah, but as others have stated it's unlikely your coolant has hit 60 degrees and deformed the PETG because your actual components aren't that hot. It's possible that something is misreporting because you said the tubes actually felt hot, which indicates your coolant temp is high.

Do you radiatiors feel hot to the touch?

And yes I would definitely put the time in to ensure your tubing runs are supremely accurate regardless of which material you use. Note that I'm not saying PETG is rubbish just because I personally don't trust it, there are plenty of people who have great experiences with it. I just think you have a fundamental issue or two with your current setup which is exacerbated by using PETG.


Ok thanks !

I just realized that I switched 15 days ago from a 60Hz 4K monitor to a 144hz freesync 2k monitor

Which makes me think that it’s detinitely températures related, occurring after long gaming sessions (and therefore with a warmer GPU)
 
Associate
OP
Joined
31 Mar 2019
Posts
29
Update :


I ordered a Lian li o11 dynamic case, in which you can use up to 36 360mm rads (tricky but still).

I'll start using it with my 240 & 280mm rads, which should solve my temps problem, then (money needed) add a 360mm rad in order to lower them even more.


Will be swapping cases this weekend :)
 
Soldato
Joined
4 Dec 2015
Posts
3,221
Location
London
I reckon so yeah, but as others have stated it's unlikely your coolant has hit 60 degrees and deformed the PETG because your actual components aren't that hot.

It looks exactly like this is what happened though, I’ve had it before with a lazy loop it deformed around the fittings in exactly this way. The fitting gets too hot and as a result the tube deforms under the pressure of the fitting, result is a poor seal and corresponding leak...

I think this is what’s happened here, insufficient cooling and potentially less than optimal flow moving the heat combined with the angles on the tubes.

Update :


I ordered a Lian li o11 dynamic case, in which you can use up to 36 360mm rads (tricky but still).

I'll start using it with my 240 & 280mm rads, which should solve my temps problem, then (money needed) add a 360mm rad in order to lower them even more.


Will be swapping cases this weekend :)

I reckon this will solve your problems so fingers crossed!
 
Back
Top Bottom