****Chrome plated TJ07 WOOT! ****

Attachment 108007

Attachment 108006


.... First of all, I apologize for the shameless way I attracted you to this thread... but now you're here maybe you can help....in essence, this is a cry for help regarding air in my loop.


Here's a picture of my wc loop. When I start the system, the tubes contain blood red clear fluid...after 10-15secs the tubes fill with thousands of tiny bubbles and looks a dirty pink colour. I'm guessing that cooling is impaired by this air, but I'm really struggling to remove it.

Firstly, it stays no matter how long the machine runs for. I do have a reservoir and have filled this to the top. in the beginning, after running for a while I, the level in the res would drop and I could add more, but this has now stopped.

When the machine is off, I can see air bubbles in the loop which, naturally, rise to the highest point they can.

This is my first wc build and I've put a huge amount of time, thought and, yes, love in to it, but I have run out of ideas how to get rid of the remaining air.

This is a single loop, listed in order below.

Pumps (3x 355 with xspc triple acrylic top)
HDD block
Res
Mosfet1
Ram block
CPU block
Mosfet2
NB/sb block
Gfx Block
TFC quad rad
..and back to pump.

I have run the system with this air problem for around a yr now and recently, 2 out of 3 of the swifttech 355's have failed, so now that I have to do the dreaded drain, I'm looking at solving this problem.

Absolutely any comments, on the craziness of so much in 1 loop, the mad order, or anything else for that matter, would be really appreciated.

I have followed this forum for a while as a silent observer and have great respect for the knowledge and helpful attitude of many of the members...hopefully, you'll have some pointers for me...

Thanks in advance
Rob

I'd personally try (gently!) laying the case on its side so the motherboard is horizontal and facing up, try and wiggle the reservoir free to the highest point in the loop and turn the pump on and off repeatedly to alow the bubbles to get to the res and get released.

Remove that aluminum passive res of yours. Replace it with something else like an EK res. It will solve the problem. Have had experience with 2 similar passive reservoirs like yours. In both cases, i had to lower the power of the pump (655) to stop the loop from frothing up. There was no issue once i replaced either res with swiftech mcres/EK res while the pump was at full speed.

How's the flow rate? Do you see fast flowing bubbles in the tube? The reservoir sure looks like it is made of aluminium and I suspect your water blocks are clogged due to galvanic corrosion.

Phil

It's unlikely to be due to galvanic corrosion from the xspc aluminium passive res. I've had one and they're well-plated and the only risk of wearing the anodising off is on the thread on the top away from the water. the barbs go into a plastic bit at the bottom. Also, the tube going up into the reservoir is pretty long - would be surprising if bubbles were getting sucked back into the loop, especially as your loop has so many blocks and must be fairly low pressure/flow. Is the reservoir fully filled with coolant?

Quote:

---

Originally Posted by fatcat

.... First of all, I apologize for the shameless way I attracted you to this thread... but now you're here maybe you can help....in essence, this is a cry for help regarding air in my loop.


Here's a picture of my wc loop. When I start the system, the tubes contain blood red clear fluid...after 10-15secs the tubes fill with thousands of tiny bubbles and looks a dirty pink colour. I'm guessing that cooling is impaired by this air, but I'm really struggling to remove it.

Firstly, it stays no matter how long the machine runs for. I do have a reservoir and have filled this to the top. in the beginning, after running for a while I, the level in the res would drop and I could add more, but this has now stopped.

When the machine is off, I can see air bubbles in the loop which, naturally, rise to the highest point they can.

This is my first wc build and I've put a huge amount of time, thought and, yes, love in to it, but I have run out of ideas how to get rid of the remaining air.

This is a single loop, listed in order below.

Pumps (3x 355 with xspc triple acrylic top)
HDD block
Res
Mosfet1
Ram block
CPU block
Mosfet2
NB/sb block
Gfx Block
TFC quad rad
..and back to pump.

I have run the system with this air problem for around a yr now and recently, 2 out of 3 of the swifttech 355's have failed, so now that I have to do the dreaded drain, I'm looking at solving this problem.

Absolutely any comments, on the craziness of so much in 1 loop, the mad order, or anything else for that matter, would be really appreciated.

I have followed this forum for a while as a silent observer and have great respect for the knowledge and helpful attitude of many of the members...hopefully, you'll have some pointers for me...

Thanks in advance
Rob

---

For starters, your loop order is bad. You should always go: res > pump > everything else.

Secondly. . .yes, get rid of that res, it's a time bomb just waiting to go off. It will cause galvanic corrosion, it's just a matter of when.

Thirdly, sorry for your loss on the pumps but it may have been do to excessive flow rate (it is/was likely the cause of your air problem too). 3 pumps, IMO, is just silly overkill for nearly any loop. Please do me a favor and head over to this topic and take part in the survey (please also provide the info requested).

I agree with Waterlogged in that your loop order is poor, let alone the poor order causing the tubing to look like a jungle mess as opposed to a much cleaner look you could achieve.

Nothing to add to the comments already posted re loop order and layout, must say loving the chrome though :)

Is your loop full? Because it seems like there's not enough fluid in the loop on the first picture. The tubes that go to your 5.25" bays. This could probably cause the fluid to mix with air hence the bubbles. A reservoir is def not causing this but switch it anyways ;)

Quick thoughts: I thought those XSPC res's had a plastic liner inside (which basically borks any meaningful cooling action, of course)? Maybe I'm remembering wrong. But they definitely don't have any sort of internal baffling to help keep bubbles from being sucked back into the loop again. If you decide to stay with all those pumps, at least make sure you're only running the one in front of the reservoir until the loop is bled, to make sure the other ones aren't run dry. It also looks like you've got an air lock, to judge by the air bubbles that TJTom pointed out above. Tilt the case all around to get that out of there.

Also, how do you fill the loop? I can't see a drain/fill tube, or a plug on top of that reservoir.

Thanks for the replies and pointers everyone, you have convinced me to re-order the loop. Monkey Puzzled, when I do I'll certainly use your method of putting the case on its side.

Regarding the possibility of corrosion; I don't have plain water in the loop, its clear primochill fluid with a primochill 'dye bomb' (the little syringe).... does this negate the danger of corrosion?

Nahsaz, I inclined to agree with you re the passive res, and mine has fins on the inside, which my head tells me won't help. So, I'm still pretty convinced to change the res; thought I'd go for http://www.watercoolinguk.co.uk/p/EK...nced_1388.html Any thoughts?

Since the pumps have gone, I'd like to change to a pair of laing d5's with a dual top..... I thought that having some control over the pump speed would help....any thoughts people? Would the 2 D5's be a better option than the triple 355's? Does anyone think it would be a good idea to consider a dual loop top for the D5's and try and split my single monster loop into 2 more manageable ones?

Finally, when I'm refilling would it help to have an in-line 'T' section with a fill port or something similar at the highest point of the loop..... I mean, is this likely to make getting rid of bubbles easier? Do you guys always install the loop in the case before filling?.... I ask this because getting air out would be a hell of a lot easier just handling tubing, instead of a bloody heavy case all around it too!

Last, but not least, how on earth do those of you with opaque tubing manage air pockets?... I'd hoped to go for plain black or white tubing at the outset of this project but revised this once I faced the practicality of air pockets....

Thanking you all for your attention, comments and suggestions

MPG.... the top screws off the passive res for filling... and Waterlogged, I will drop by the 355 fail thread and supply the details in my failure situation; mine failed just inside a year and my pumps were swifttech branded ones...do I therefore have manufacturer's warranty cover?

Thanks agains guys!

the primochill fluid will help negate corrosion, but will not fully do it, as WL said, it's a timebomb, and do i have it right that you haven't drained your system for 2 years?with the primochill fluid i'm sure you would want to change it every 6-8 months.

unfortunately since you used an aftermarket top, i understand your warranty is void, make sure you take the pump apart and see if there was anything in there too that could have contributed to the failure.

Sick chrome job! but as Waterlogged pointed out, get rid of that res!

My vote for the dual 655 varios. 355's have a nasty tendency to fail prematurely. Esp those 18watter versions. Ive got 4 655 varios in my rigs altogether and have no regrets about replacing the 355's i used to have.

And that ek res is a good choice! Just got myself an X2 a short while back too... :up:

Should be in the worklog section IMHO.