New custom water loop issues, please help

budy86

Distinguished
Jan 15, 2012
109
0
18,690
Thanks for reading, This is my first set up.
Everything is all set up and and working fine.. But i think i have bubbles in rad or something.
When the pump is on low i get a bubble that sticks in the top of the CPU block (EK clear). But when i turn the pump up anywhere from 2-3rd setting or above the bubble goes away. (5 speed settings) I also note that there is no temp difference no matter what speed i have the pump or the fans.
When i first boot the comp I'm sitting on temps of about 19, now I'm in the mid 20's (Celsius). When i run prime 95 on stock settings i get up 50 degrees. For a custom loop with good gear i would expect better as this is the same result i was getting with my H100i.

I know you can tilt the computer to try and catch out any bubbles but its not working.. What are the probable causes? and or solutions? I'm not too keen on pulling it apart, or am I expecting too much? abiant room temp is about 25 also.

my rad: Alphacool 480
CPU: EK Supremacy Clean CSQ
Pump: EK D5

Im only cooling the Motherboard and CPU
 

oczdude8

Distinguished
First off, how are you getting 19c idle when your room temp is 25c? maybe your not reading the right temps?

Second, the easiest way to bleed out big bubbles, is to pull out one of the tubes so it pours out of the loop and turn on your pump. make sure to refill your res and that your pump doesn't run dry.
 

budy86

Distinguished
Jan 15, 2012
109
0
18,690
The problem has seemed to have fixed itself after i turned the pump up high for like the millionth time. I think the issue may be my thermal paste. I may have put a bit too much on i think. but I'm happy so far. Thank you.
 


To put it mildly, tilt the computer, more like rock the thing from side to side and front to rear, whatever it takes to get the air out, and always with the pump on it's highest flow setting if it is a variable speed pump, just make sure the reservoir is capped, and keep an eye on the air in the reservoir, so you don't pull purged air back into the pump intake with your tilting and rocking back and forth.

Remember what you cannot see, is, Air always goes to the highest point of whatever it is in, get the air completely out and your temps should improve, and air trapped in your pump could destroy the pump bearing, so be sure all the loop is air free.

 

budy86

Distinguished
Jan 15, 2012
109
0
18,690



Thanks very much,, my temps seem to be great at the moment.. trying to OC and i had the vCore at 1.312 and stress testing it was only at low to mid 50's.. but then it crashed :/ so i think its good now,, and i haven't seen any more bubbles..
 


That can be a problem?

Thermal compounds only purpose is to replace the non conductive and insulating air between the two metallic mating surfaces, even though machined they look very smooth and flat especially the water block base, but microscopically the surface magnified looks like the surface of the moon.

Thermal compound is only to fill in those imperfections and they are very small, too much thermal compound can act as an insulator instead of a heat conductor.

Maximum thermal cooling performance comes from using just enough thermal compound to spread evenly flat covering the IHS barely to the edges, less is best, when it comes to thermal compound.

Many think they need a thick layer of compound and that is so wrong, that's why Artic MX-4 is so good it is a thinner compound and spreads out super thin, and that sir is what you want for maximum thermal conductivity.

You could actually use nothing but metal to metal, but then you'd have air between the two surfaces, and air is bad!

Understand?

 

budy86

Distinguished
Jan 15, 2012
109
0
18,690


Yea, I understand. Temps seem pretty good. fans are only just spinning and my pump is on speed 4 (out of 5) so i think I'm ok
 


You can play around with your pump speed to find the water block cooling sweet spot, I usually run my D5 on 3 as it seems to be the best cooling range with my setup, all setups are different relating to flow restriction so once the loop has reached it's temperature equilibrium, you could try various pump speeds to see if cooling improves or gets worse.