Advice on why leak happened.

Associate
Joined
1 Mar 2019
Posts
423
Hey, so I did all my latest upgrades. Leak tested overnight all good, I've been doing watercooling for years and am pretty confident with it at this point. I've been through leaks and things, alway just lept out of my seat and switched off the power supply
:cry:. Not had that happen in a long time though! but anyways, I finished my build, tested for 16 hours just the pump on full power. I moved it upstairs and after half an hour of gaming a pipe came loose and coolant sprayed all on the inside. I cleaned it all up, wipe everything down with alcohol wipes. triple checked every nook and cranny for hints of coolant (luckily my motherboard is shielded and its all dripped down through the bottom on to my desk. I isolated it to a pipe on my CPU block, the fitting was barely even half tight and the pipe just came out easily. I'm guessing this was just a brain fart on my part not double checking and the heat and pressure of full system use is what eventually forced it out. But could it be any other reasons. FYI, I cleaned it all up put the pipe back in, tightened every fitting up as much as I could and have been using it since with no issues or hint of leak. Just want to see what people think before I'm out of the woods. I'll feel confident removing the paper towels after a week if I've had no faults :cry:. I'm usually very relaxed about watercooling but with how expensive my parts are I'd be lying If I said I didn't pap myself.
NzHjJgH.jpg
 
Associate
OP
Joined
1 Mar 2019
Posts
423
I'm using a acrylic tubes, used to use PETG and had exactly that happen once., I have a liquid temp sensor thats saying about 35 degrees. although it could have been warmer at the time as I hadn't properly configured pump speed and dan curves. currently got the D5 set to 30% scaling to 60% at 50degrees plus. Fans also set to a lowish RPM on idle temps rising as temps do. Only thing I'm not sure is normal as I've just upped to a 14900k, my liquid temps are floating around 35 degrees in games, but CPU package can be anywhere between 70 and 85 degrees. 4090 stays fairly stable around 55 degrees or so. If I have to up the fans I will but I prefer silence. Thats why I went for a larger loop this time, but also now realising pump needs to be a touch faster than I was hoping to circulate the coolant properly. Been using it normally for over a day now though with no issues. I wfh and this is my do all PC for work and games. so its on for about 8-10 hours a day usually
 
Associate
Joined
28 Aug 2014
Posts
2,234
I had an issue I noticed a few times that wpuld stop my fans from spinning after a while. Once I didnt notice as i had my headset on and was proper stuck into my game and must have been playing for over an hour and I felt something warm under my mouse hand. It was coolant and it had come out quite bad. It was very late so best I could do was turn it all off and leave it, I couldn't see it running out so i felt safe. Woke up in the morning and got straight to it. It had leaked abit more over night but was from my pump/reservoir area quite a bit away from any parts it could harm.

The only thing I can think of that made it leak was the coolant temps from the fans stopping made everything loosen up abit. It was just loose fittings, a few of them were loose. It had been setup for over a year and not leaked once.
 
Associate
OP
Joined
1 Mar 2019
Posts
423
Yeah tbf It does seem like these days leaks tend to be less catastrophic. Especially if using non/low conductivity coolants and having parts spaced out. I only ever use EK cryofuel, no solid or pastel stuff. Still scary though as you never know.
 
Soldato
Joined
22 May 2010
Posts
12,002
Location
Minibotpc
Never seen acrylic or PMMA do that, PETG i could understand but not PMMA. It takes way higher temps, much higher than your systems coolant could achieve to deform it.

i'd look at the orings on the fitting itself and see what state they are.
 
Last edited:
Associate
OP
Joined
1 Mar 2019
Posts
423
Never seen acrylic or PMMA do that, PETG i could understand but not PMMA. It takes way higher temps, much higher than your systems coolant could achieve to deform it.

i'd look at the orings on the fitting itself and see what state they are.
Yeah I checked them when I put it back in all seemed fine, all brand new fittings. Been running as normal and playing games for hours for almost 2 days now since it happened. When it happened it was fairly soon after starting so I'm hoping I'm in the clear now. Still just checking fittings and pipes for tightness but nothing seems to have budged
 
Soldato
Joined
23 Jun 2004
Posts
4,703
Location
Blackburn
I'm using a acrylic tubes, used to use PETG and had exactly that happen once., I have a liquid temp sensor thats saying about 35 degrees. although it could have been warmer at the time as I hadn't properly configured pump speed and dan curves. currently got the D5 set to 30% scaling to 60% at 50degrees plus. Fans also set to a lowish RPM on idle temps rising as temps do. Only thing I'm not sure is normal as I've just upped to a 14900k, my liquid temps are floating around 35 degrees in games, but CPU package can be anywhere between 70 and 85 degrees. 4090 stays fairly stable around 55 degrees or so. If I have to up the fans I will but I prefer silence. Thats why I went for a larger loop this time, but also now realising pump needs to be a touch faster than I was hoping to circulate the coolant properly. Been using it normally for over a day now though with no issues. I wfh and this is my do all PC for work and games. so its on for about 8-10 hours a day usually
If coolant temp is about 35 degrees and your using acrylic tubes then I'd quite confidently say it was the loose fitting at fault. No more late night buidling for you :p
 
Soldato
Joined
22 May 2010
Posts
12,002
Location
Minibotpc
Yeah I checked them when I put it back in all seemed fine, all brand new fittings. Been running as normal and playing games for hours for almost 2 days now since it happened. When it happened it was fairly soon after starting so I'm hoping I'm in the clear now. Still just checking fittings and pipes for tightness but nothing seems to have budged

Possible chance you may have knocked it when you moved it?

The bend to the cpu block looks okay, doesn't look like its under any stress or it might have been forced in.
 
Soldato
Joined
28 Jun 2013
Posts
3,781
the only thing i can think of is the fittings are not the greatest design and one in so many fail or you simply forgot to tighten it fully.

i always use soft tubes and have never had a leak thank god, my main worry is O ring failure on the fittings
 
Associate
OP
Joined
1 Mar 2019
Posts
423
the only thing i can think of is the fittings are not the greatest design and one in so many fail or you simply forgot to tighten it fully.

i always use soft tubes and have never had a leak thank god, my main worry is O ring failure on the fittings
I took pictures before moving it and I can just about see the culprit not being fully tight at all, albeit not as loose as it was when it leaked but I guess the pressure and heat could have done that.
 
Soldato
Joined
23 Jun 2004
Posts
4,703
Location
Blackburn
Possible chance you may have knocked it when you moved it?

The bend to the cpu block looks okay, doesn't look like its under any stress or it might have been forced in.

the only thing i can think of is the fittings are not the greatest design and one in so many fail or you simply forgot to tighten it fully.

i always use soft tubes and have never had a leak thank god, my main worry is O ring failure on the fittings

Guys read first post, OP states the fitting was barely tight.

@Curtis vincent Seen as you've been using the PC under load for multiple hours since the leaking incident I'd say your good from here on out :)
 
Back
Top Bottom