-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
24 hour Water cooling.
- Thread starter jim0ne
- Start date
- Joined
- Dec 26, 2004
jim0ne said:
Excluding yesterday, my comp has been running 24/7 for a good month or two.
The only moving part in your WC loop (pump) should be able to last 24/7. In fact, some are meant to run 24/7 (Laing D4/D5 series, and DDC series, as well as Iwakis, possibly even Eheims). We don't always use pumps made for PC Watercooling, but industrial pumps.
- Joined
- Oct 12, 2001
- Location
- Los Angeles
First, Welcome to the forums! To answer your question, YES, my machine runs 24/7/365 and is only off if I am working on it, which is not that often. If you buy quality components and use PROPER hose clamps (metal worm style) then you should not have a problem.
- Joined
- Dec 26, 2004
jim0ne said:
Probably wouldn't fail if you left it on for 5 years
Good choices man. I was using that same pump up until yesterday, half broken, and it still was pumping through my loop just fine and dandy. Great little pump indeed. 50,000MTBF as well (roughly 5 years until it's expected to fail).
- Joined
- Jun 26, 2006
yep i approve of this thread.
but i always turn off my computer... i don't know why... i just... have that weird urge to hit that power button when im not using it.
its like when your hungry, its like... a felling you can't explain... but you know your hungry...
but i always turn off my computer... i don't know why... i just... have that weird urge to hit that power button when im not using it.
its like when your hungry, its like... a felling you can't explain... but you know your hungry...
- Joined
- Sep 15, 2001
The only way my computer is off is when I'm installing hardware. If not then it's on all day every day. Just keep an eye on the water level in your loop. Even on a closed system there still seems to be some evaporation. I loose maybe a cc every couple months on my sealed T-setup.
- Joined
- Dec 26, 2004
Mr. $T$ said:
Water is permeating through the tubing. Quite normal for a WC loop.
I actually haven't had any permeation for the past month or so. It seems quite strange
- Joined
- Dec 26, 2004
jim0ne said:
I'd expect it to be around the same. Can't say for sure, never used a res (though I really do want a MCRES-Micro). I don't see why it would be any different.
- Joined
- May 1, 2006
- Location
- Virginia, USA
Well when you have "Three Beasts" leaving one running is easy.nikhsub1 said:
- Joined
- Dec 3, 2004
- Location
- VT... or MD
Micro res does a good job. But the higher the flow from the pump, the longer it will take for the bubbles to work themselves out of the loop.
But i am new to water cooling, and thats just what ive seen so far.
But i am new to water cooling, and thats just what ive seen so far.
- Thread Starter
- #13
Sleepy_Steve said:
I'm planning on getting a MCP655, which is suppose to be a "gangster" pump, so i assume that the flow should be high, but how long would it take for the bubbles to work itself out?
I read on another thread on this forum to run the water setup until the bubbles are out of the system and then turn on the comp. So i would probably do that to make sure everything is up and running before i blow up my CPU.
- Joined
- Jul 20, 2004
Psst! http://www.oc-rev.com/forums//showthread.php?t=828SolidxSnake said:
and the standard "safe" leaktest is a 24 hour run with only the pump getting powered...
- Joined
- Dec 3, 2004
- Location
- VT... or MD
Aye, that way the pressure from the pump can show you any leaks that might not have been apparent otherwise.orionlion82 said:
- Joined
- Dec 26, 2004
orionlion82 said:
Well, by half-broken, I meant that the internal speed control was completely broken (the original pot broke, and the replacement I put in ruined the PCB... it wasn't running full speed). Otherwise, the impeller/stator/rotor and such were all intact.
- Joined
- Dec 3, 2004
- Location
- VT... or MD
Micro worked decent for me... I'd say anywhere from 30 min to 2 hours with the pump running and the bubbles go away. You may need to top off the res durring that time though as water displaces the air in the lines.jim0ne said:
I got foam once i added some antifreeze to the loop though, and need to sort that out the next time im home, unlike bubbles, foam was there for over a day... and i had run out of time over my turkey break.
The MCP 655 is rated at 317GPH. My Mag Drive 3 is rated at 350GPH 8) So the flow rate is similar. I would expect the behavior of bubbles as they work out of the loop to be similar as well. Maybe a bit faster if you can drop down the pump's flow rate setting.
- Joined
- Jan 26, 2005
When filling a loop with a D5/MCP655 turn to pump to the lowest speed setting, it will create way less bubbles. With a res I also fill the res, hit the power until i suck all the liquid down from the res, kill the power, refill the res and repete. This will help you avoid bubbles.
QuietIce
Disabled
- Joined
- May 7, 2006
- Location
- Anywhere but there
I've only built 3 loops so far but I've never had a loop take more than ~1 hour to bleed and all were T-lines. Aside from the extra 10-15 min. it takes to fill the loop I can't see any reason for not using a T-line, and the fill time can be reduced by using a long 3/4" T instead of a 1/2".
BTW - All 3 of my computers run 24/7 unless I'm working on them (SETI crunchers) - one 2xMCP655 loop and two DDC+ loops. As long as you use good pumps, rinse before building!, leak test for 24 hours, and keep the res/T-line full you're good to go ...
BTW - All 3 of my computers run 24/7 unless I'm working on them (SETI crunchers) - one 2xMCP655 loop and two DDC+ loops. As long as you use good pumps, rinse before building!, leak test for 24 hours, and keep the res/T-line full you're good to go ...
Last edited:
