Overclock.net › Forums › Cooling › Water Cooling › Help Needed - Bleeding Massive Air Pockets from Loop
New Posts  All Forums:Forum Nav:

Help Needed - Bleeding Massive Air Pockets from Loop

post #1 of 25
Thread Starter 
Hi All,

This is my first watercooling build, but coming in I thought I had watched enough setup videos to be proficient. Sorry for the long read, but the whole process might help put things into context.

Basically, my loop was about 50% air, consisting of huge air pockets that, depending on which angle my PC is tilted, can take up entire lengths of tube (or I imagine entire blocks/radiators, if I could see inside).

Here is a rough sketch of my loop (case is Switch 810): http://imgur.com/9MYQzAg

So I started by filling the res, turned on pump to drain about 80% of the fluid, stop & repeat. This worked well for about 4-5 cycles until....

I would fill the res and turn on the pump, pump drains most of the fluid from the res, but when I turn off the pump to prevent it running dry, ALL of the fluid returns to the res (filling it back up). Repeating this process 5-10 times produced the same results. Basically, I believe the air pockets were somehow pushing the fluid back into the res (but wouldn't bleed out). I tried tilting the case on its back and holding up the res so that it was at the highest point, but this didn't change the results.

What I did from here is to tilt my pc from side to side to move the air pockets into the res. When they get to the res, I open it and fill the small space with more fluid.

After releasing a few air pockets (roughly 30-60 minutes of tilting) I can do 1 more pump cycle. I did this about 2 times.

At this point, if I turned on the pump it would quickly drain about 2/3 of the contents of the res and then stop (all indication of flow stops but the pump is still clearly running, and no sound to indicate it's running dry). I believe the air pockets are somehow restricting the flow.

What I did here was, AS THE PUMP WAS RUNNING and the res 2/3 drained, I fill the res and seal it, then turn of the pump. Now, if I was to undo the seal, fluid would erupt out of the res. What I do is repeat the process above (move air pockets to the res). I then open the res cap just a teeny bit to allow the air to escape, and then quickly seal it to prevent fluid from shooting out. After doing this 4-5 times (maybe 45-90 minutes of tilting) the pressure is relieved enough for me to open the res without any spills, and I repeat the process. I've repeated this process about 4-5 times (each time filling the res with about 2/3 capacity of fluid), but the pump still seems to get restricted by the air in the loop and can't fully drain the res.

Basically my question is, do I simply continue this long and tedious process of bleeding the air pocket by pocket? I would estimate the air in my system has dropped from 50% of the total to ~15-20% of the total.

OR, is something wrong with my loop or part(s)?

It's also worth noting I've already completely drained the loop and reconfigured it twice, and I am experiencing the same problem each time.
post #2 of 25
so... you run the pump and no water comes into the res inlet even though the res never runs dry? It sounds like one of your blocks or rads is plugged. Is your tubing clear so you can see where the water moves and doesn't?
post #3 of 25
I had a similar setup in my switch with similar bleeding issues.

It's important that the input on the MCP655 does not run dry while you are trying to bleed the system.

Does your res have another input that you could put a fill tube on? The fastest way to bleed the system is with it running. Switching it on and off puts you at odds with gravity and water pressure as you have discovered.

You would need to relocate your res to the top of the loop to completely bleed it by shutting it on and off.

It also helps if there is an alternate exit for the air in the system as it fills.

Hope this helps.
post #4 of 25
Thread Starter 
Quote:
so... you run the pump and no water comes into the res inlet even though the res never runs dry? It sounds like one of your blocks or rads is plugged. Is your tubing clear so you can see where the water moves and doesn't?

That's what weird about it. The pump will be flooded and sounds like its working, but nothing appears to be moving. Tubing is clear with white liquid. Nothing seems blocked... Liquid has traveled around the loop freely when I tilt and turn it around.
Quote:
It's important that the input on the MCP655 does not run dry while you are trying to bleed the system.

Does your res have another input that you could put a fill tube on? The fastest way to bleed the system is with it running. Switching it on and off puts you at odds with gravity and water pressure as you have discovered.

You would need to relocate your res to the top of the loop to completely bleed it by shutting it on and off.

It also helps if there is an alternate exit for the air in the system as it fills.

The pump hasn't run dry while bleeding. On my first pump, I heard it run dry briefly so I know what it sounds like). I had to RMA it due to a broken speed dial. The new one hasn't run dry.

The res looks like this: http://www.mountainmods.com/images/images_big/MCRES.jpg

The top barb is input, the bottom barb is output, and you fill it from the hole in the top. When the res is 2/3 drained as I explained, it's usually sitting just above the input.

Alternate air exit? Examples?
post #5 of 25
For an air exit, I put a long piece of tubing on one of my radiators at the top of the loop (in your case the RX360). I have an alphacool UT60 so it has a bunch of ports. I looped it over a rafter and into a bucket. Just be careful that when it burps it doesn't spew liquid into the system. Also make sure it's closer to the water return than the intake if you can.

More importantly, I was thinking you could attach a fitting or barb and a long tube at the top of the res and fill it with a funnel. That way as the air bleeds you would not need to stop the system so often because there would be a quantity of fluid to immediately displace the air. Again be careful while filling. Hopefully you are bleeding using a test power supply and the system is not hooked up.

Just trying to brainstorm how you can get this done quicker with a small res. If I had to guess, all of the air pockets are in the rx360 as the position of the rx240 should bleed pretty quickly.

thumb.gif

P.S. - it took me hours to bleed my first system build. Even after most of the air was out, I'd say it took more than 48 hours for the pump to quiet down completely and for air bubbles to stop appearing. Don't get discouraged. smile.gif
Edited by pgdeaner - 5/31/13 at 12:50pm
post #6 of 25
Thread Starter 
Yes, most of the air builds in the RX360.

Using an old PSU and only the pump is plugged in.
Quote:
P.S. - it took me hours to bleed my first system build. Even after most of the air was out, I'd say it took more than 48 hours for the pump to quiet down completely and for air bubbles to stop appearing. Don't get discouraged.

Herein lies the problem. I have most of the air out but when I turn on the pump, it doesn't appear as if there is any flow. It now sucks out about 15% of the fluid from the res and continues to hum quietly. If I feel the tube from the res to the pump or the pump to the 240, it's apparent that water is moving through it. However if I grab or observe any other tube there is no indication of flow. Just dormant fluid. I can't bleed the system traditionally because there is a lack of ANY movement in the rest of the loop.

I'm going to take a video to demonstrate.
post #7 of 25
how many angles and how far have you tilted your case? i laid mine almost completely on its back, face and sides. i also had my pump free hanging so i could tilt it also to keep the pump from sucking air
w00t
(18 items)
 
   
CPUMotherboardGraphicsRAM
ph2 x20be asus m4a785-m sapphire vapor-x 4850  geil black dragon 1066 5-5-5-15-20 
Hard DriveOptical DriveCoolingOS
wd  asus dvd multi-write arctic cooling freezer 64 pro win 7 ultimate x64 
MonitorKeyboardPowerCase
samsung 42" lcd logitech wireless antec basic 500watt old dell crap 
MouseAudio
logitech wireless pioneer 7.1 
  hide details  
Reply
w00t
(18 items)
 
   
CPUMotherboardGraphicsRAM
ph2 x20be asus m4a785-m sapphire vapor-x 4850  geil black dragon 1066 5-5-5-15-20 
Hard DriveOptical DriveCoolingOS
wd  asus dvd multi-write arctic cooling freezer 64 pro win 7 ultimate x64 
MonitorKeyboardPowerCase
samsung 42" lcd logitech wireless antec basic 500watt old dell crap 
MouseAudio
logitech wireless pioneer 7.1 
  hide details  
Reply
post #8 of 25
Thread Starter 
post #9 of 25
Why do you feel like fluid is going to erupt out of the res? I always bleed with the res open and pulsing the pumps on and off ... it takes only minutes. You can't get water in without getting air out.
I find in mine air comes out better just after you switch the pump off for each time for some reason.

Even though the res is lower than the top rad it should not overflow, for water to flow down into it from the top rad it would have to draw the same water uphill on the other side from the GPU's ... normally air locks prevent anything serious happening.

If you are really worried screw a spare fitting onto the top port of the res and add a piece of hose you can hold up high.

You are making it a million times harder than it needs to be.
Kusanagi
(21 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3570K Asus Sabertooth Z77 EVGA GTX 780 SC Corsair Vengeance Pro CMY16GX3M2A2133C11 16GB 
Hard DriveCoolingCoolingCooling
Samsung 840 Pro 256GB X2 EK Supremacy Aqua Computer Aquaero 6 Pro Aqua Computer D5 USB 
CoolingCoolingCoolingCooling
Aquacomputer MPS 400 flow meter EK-CoolStream PE 360 EK-Coolstream PE 240 Noiseblocker Eloop B12-PS 
CoolingCoolingCoolingCooling
EK Vardar F2 EK-FC Titan Acetal & Nickel EK-RAM Monarch X4 Clean CSQ - Acetal+Nickel EK-D5 X-RES 140 CSQ 
OSMonitorMonitorPower
Win 8.1 Pro 64 bit Samsung S27A750d LG E2290V-SN Silverstone Strider Plus 850W ST85F-P 
Case
Corsair Obsidian 750D 
  hide details  
Reply
Kusanagi
(21 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3570K Asus Sabertooth Z77 EVGA GTX 780 SC Corsair Vengeance Pro CMY16GX3M2A2133C11 16GB 
Hard DriveCoolingCoolingCooling
Samsung 840 Pro 256GB X2 EK Supremacy Aqua Computer Aquaero 6 Pro Aqua Computer D5 USB 
CoolingCoolingCoolingCooling
Aquacomputer MPS 400 flow meter EK-CoolStream PE 360 EK-Coolstream PE 240 Noiseblocker Eloop B12-PS 
CoolingCoolingCoolingCooling
EK Vardar F2 EK-FC Titan Acetal & Nickel EK-RAM Monarch X4 Clean CSQ - Acetal+Nickel EK-D5 X-RES 140 CSQ 
OSMonitorMonitorPower
Win 8.1 Pro 64 bit Samsung S27A750d LG E2290V-SN Silverstone Strider Plus 850W ST85F-P 
Case
Corsair Obsidian 750D 
  hide details  
Reply
post #10 of 25
Thread Starter 
Watch the video. I do that but the pump just stops pumping or getting any flow past a certain point.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Water Cooling
Overclock.net › Forums › Cooling › Water Cooling › Help Needed - Bleeding Massive Air Pockets from Loop