So was investigating my loop last night

stuie

stuie

stuie

Soldato
Joined
29 May 2012
Posts
3,395
Location
Dorset
After having a few water level thoughts and issues I spent an hour investigating my loop.

Firstly I squeezed all the tubes at around an inch away from my barbs with the system off, when i squeezed the outlet on my cpu block a big bubble appeared if you can imagine the size of a squashed penny and youl get the idea of what came out plus a couple pea sized bubbles on the inlet tube I frantically checked the area for signs of water but there were non. Thinking this was the problem i took the top off my res and fired her up the bubbles proceeded to enter my top rad and never to reappear then disaster 2 drips from the inlet of my top rad coming from my cpu, quickly powering the system off I attached a second zip tie to the barb and put the res top back on, after wrapping everything up in tissue paper fired the pump up, all was well again fortunately 2 minor marks on the tissue paper (less than a grain of rice all together) on the first test and then all clear on the second test after leaving it priming for half an hour.

Finally I then removed the cap off my res again and all was ok no leaks. So once again ive got a couple of questions.

1. When i run the res without the top off the water level in the res drops about 5mm over what it drops with the top on could this be an indication of a problem?
2. Would you expect now the bubble has cleared from my cpu block that the temperatures should be noticeably cooler as the dont appear to be, my radiators are getting warmer though.
3. The minor leak after shifting the air bubble again is this an indication of a problem in the loop as ive not had a leek before or is it cause of the pressure differences caused by the movement of the bubble.
4. Lastly the now missing bubble will it have managed to free itself with the scuffle ive gone back round the loop squeezing the tubing again without any air being moved about although this is all with the res top on should it be done with it off so the air has an escape route?

Once again thanks guys
 
Last edited:
Seems like your loop has not been bled properly - some air bubbles may have been trapped in the blocks/RAD.
Also, squeezing the tubes(no pun intended) may be the cause for the water to leak!
By your post - it seems you have barbs with zip ties? Have you actually done a leak test before?
1.When the the top is on, loop will have some pressure(closed loop system) - which is normal.
2.If the air bubbles are nearer to the micro fins/channels it would have a bigger impact on temperature. What are your load temps now? also what is your room temp?
3.Most probably because you were playing with the tubes.
4.I usually run the leek test with the res top off - making it easier.
 
This system is around 2 months old and was fully leak tested then without any problems up until i played last night although this is my first full system. Yes im running fatboy barbs and cable ties.

Load temps vary on room temps but with both cards at 1215 clocked and my cpu @ 4.5 1.2v loaded the cpu doesnt go over 75c on the hottest core and the cards max at around 55c, this is in a normal bedroom after a few hours of gaming door closed but a window open id probably say mid to high 20s due to the system kicking out heat the rads are getting up to about 37c, this is with a stock 3770k no delidding but lots of heat off the full imc and pcie lanes, with a fresh cool rom at the beginning of the night no cores are over 70
The res top is firmly screwed on would it be worthwhile just backing it off a few turns so and just blitzing for a few hours on crysis 3 / starcraft 2?
 
Last edited:
I can confirm that squeezing the tubing near to the barbs using fatboys allows a small amount coolant to seep past the ribs and out the end of the tubing. When i first filled and bled my loop at the start of april i brimmed the reservoir. I just had to top it up as it had fallen by about 6-8mm. (started making watery sounds as it sloshed about) There are no leaks in my system, it's just residual air bubbles that slowly find their way out of the blocks and radiators and back to the reservoir. I've brimmed it again and will keep an eye on it. (same XSPC setup as yours) I keep the cap on tight, removing it sometimes makes a small hiss and that's all, the level doesn't seem to change. Your GPU temperature seems to be virtually the same as mine (57 after a few hours) although you have 2 cards I only have a single 6950. (but my cpu tops out at 52)

Have you got 1/2" tubing or 7/16" as this may affect your performance with barbs.
 
1/2 inch tubing what temps do you get on your cpu and is it clocked at all? Im hoping now ive cleared the big bubble things should start to drop once the system re bleeds itself. I also get hissing off my xspc i assumed this was pressure related
 
75C on water? :eek: Looks like a bad mount.
I would reseat the CPU block(with fresh TIM)

Tubing diameter effects are small after 3/8", so 7/16 or 1/2 would not make any difference.
 
what would you expect a fully loaded overclocked system to achieve temp wise? Also when reseating what method would you recommend corner to corner tightening?
 
Last edited:
1/2" tubing on 1/2" barbs may explain why you got coolant to leak out quite easily. I use 7/16" on 1/2" barbs meaning i don't even need jubilee clips or cable ties to keep the tubing tight and leak free.

My cpu isn't comparable to yours as it's an AMD phenom II X4, which really shouldn't exceed 62. (although when i ran only one 240mm rad i saw 71 under heavy stress, which is why i quickly added a second radiator)
 
Not familiar with your CPU block - does it have a specific Inlet and outlet? if so, have you correctly connected the tubes to inlet and outlet?

Ivy does get hot(more so when over clocked) but I wouldn't expect that much.
Yes I do corner to corner but my block is a koolance CPU370.
 
It does have specific in and out labeled on the block and they are correct

Ive even spoken to 8pack about this and he thinks 75c with the system fully loaded is perfectly fine although i wasnt sure looking back at my air cooled temps i was hitting 75c at 4ghz with less volts so maybe ive just got a hot one I am running 32gb and full pcie lanes with sli so plenty of extra heat, might try reseating anyway if nothing else dropping the cpu block will dislodge any possible air pockets
 
Last edited:
Well ive just reseated the cpu block and its made no difference ibt still hits the hottest core at 75c did let out (or in) another pocket of air from the cpu block though
 
Last edited:
temp on cpu sounds about right to me tbh, with 3770k only having thermal paste between the dye and the ihs.. i had this problem until i delid my cpu now the hottest core is 72c(other 3 are under 70c) @ 4.7ghz 1.4v.

it does not matter what you have cooling your cpu if the heat transfer is not good between ihs and cpu dye, i recommend delidding and using Coollaboratory Liquid Pro Liquid instead of the ****y tim used by intel

edit - at 4.5ghz 1.27v hottest core is 58c on my set up with cpu delidded
 
Last edited:
vonex said:
temp on cpu sounds about right to me tbh, with 3770k only having thermal paste between the dye and the ihs.. i had this problem until i delid my cpu now the hottest core is 72c(other 3 are under 70c) @ 4.7ghz 1.4v.

it does not matter what you have cooling your cpu if the heat transfer is not good between ihs and cpu dye, i recommend delidding and using Coollaboratory Liquid Pro Liquid instead of the ****y tim used by intel

edit - at 4.5ghz 1.27v hottest core is 58c on my set up with cpu delidded
Click to expand...

how do u delid it?
 
You must log in or register to reply here.
Back
Top Bottom