Bubbles!

[Da_Rude_Baboon]

I have my system up and running and i'm happy with the temps but is it normal to have bubbles in the system?

I've turned my PC upside down and side to side and had all the large pockets of air out the loop before i turned it on but theres a steady stream of tiny bubbles shooting around the loop.

[Knipex]

There will be a certail amount of air in the system for a while. Let it run for at least 24 hours and they should go.

How are you feeding your pump by the way ?????

[Da_Rude_Baboon]

with electricity? :blink:

Not a 100% sure what you mean but in my loop the reservoir is at the top of the case followed by the pump at the bottom. Soon as i put water in the reservoir gravity does its thing and the water goes down into the tube priming the pump ready for use. I think thats what you asking?

I'll post a pic showing the layout.

http://server5.uploadit.org/files/DaRudeBaboon-side.jpg

[adwhitworth]

That's right. Just tilt the case sideways, fowards, backwards for a while and all the water will bleed through. The amount in the Res will go down so you will need to top it up until it is gone.

[Pug]

Gah, I had a huge post typed up in reply to this then Winamp decided to hardlock my machine, forcing me to reboot and lose all the docs I had open! :angry:

Anyways, I think Knipex was referring to whether you had the sequence of res -> pump for the feed issue.

Couple of points to make sure - outlet from res at lowest point, inlet should be higher (to avoid introduction of air through bubbles being sucked in to start with).

Try to maintain a smooth loop throughout the components. If your loop goes up and down repeatedly, with components at peaks of it which have intricate internals (such as the CPU block), air can be trapped at these points - check for visible signs of air pockets through the walls of the blocks to confirm.

As Adam said, tip the case rather than upturning it. If the water level drops below the res outlet and exposes it to air, you're back to square one.

I've been known to detach blocks/rad in situ (if hose routing allows) and rotate/agitate each one independently in sequence through the loop until all air is flushed through to the airtrap.


One other possibility of introduction of air is through a loose fitting junction on the suction side of the pump - The way I check for this is to detach a hose before filling and blow down it with my thumb over the fitting I took it out of, as a rudimentary pressure test.
If one of your hoses isn't pushed fully home, this should show you that this is the case.


If you've got a pic of the complete loop as it stands now, it may shed a little more light on it for us though. :)

[Da_Rude_Baboon]

lol i'm too embarrased to show pictures of the assembled loop as i thought i may have had to return the aquastream the cablings all over the place. I'll maybe take one tonight and post it so you can see. Currently the loop goes:

Res > Pump > GPU > CPU > Radiator

Its in a nice smooth loop with no up or downs between components. The water goes into the reservoir from the top inlet and exits out one of the bottom holes. There are no large bubbles, and there isnt any catching in the blocks, as there were during the leak testing lol. :p

As i said in the first post i'm happy with the performance and the temps and the pump isnt making any noise. You have to look really closely to see the bubbles they are that small but i was just checking if this is normal or some thing to worry about.

On the temps side i'm getting 42C under heavy load and thats cooling an XP2500 overclocked to XP3200 and a Radeon 9800 Pro. I didnt think that was bad at all for a single fan radiator considering my temps were 54C under aircooling and that was just the CPU. The noise level is also a gentle hum instead of a roar which was the main objective. B)

[Da_Rude_Baboon]

Had a look last night and the stream of bubbles was getting worse. Checked both connections on the pipe between the reservoir and pump it looks like it was drawing air from the pump inlet. Soon as i i tried to push it in further a drip appeared and the tube went in another couple of millimeters.

Thanks for the advice! :D

You need a Thumbs up smilie

[Pug]

B) Cool, sounds like that could well have been your culprit. Let us know how it goes.

*me makes note to do a pushfit q&a thread*

[dutchcedar]

It sounds like you found the problem, but might I add... stationary little bubbles will be absorbed and are normal when a rig is new or even if fresh water is added. Bubbles that are moving can be a sign of trouble.

[Da_Rude_Baboon]

Still have bubbles moving around but i'm going to get a couple more components, drain and rearrange the loop slighty anyways.

It looked good planned out on paper but now its up and running theres a few optimisations i can make.

[Da_Rude_Baboon]

Bubbles all fixed! Noob mistake on my part. :wacko:

My own fault really i had the intake for the aquatube partially above the water. The turbulence and therefore bubbles it was creating were being sucked right out the aquatube and recirculated around the loop. Topped up the water a little bit in a few hours completly bubble free.

[mrplow]

Whilst finding silly errors is embarassing, at least it solves the problem!

Glad to hear it's all workin' again http://www.techangel.co.uk/forums/images/smilies/custom/great.gif


ps I have a very similar setup, only the rad is at the front not the tube.

[Da_Rude_Baboon]

I placed the rad in the middle so it had more space underneath it to draw the air through. Plus the with my case on my desk its nice to have the aquatube at the front as its more intresting to look at then the grill and fan. B)

[Pug]

Happens to the best of us - I put an AP900 in my PC-50 last night to replace the AquaStream I was putting in my ClearPC. Had a couple of free bays, so I decided to baymount my AquaTube and got a bunch of bubbles to start with when I fired it up.

Thinking it was because the pump was too pokey (it shifts some serious coolant compared to my 1046s) I though I'd have to swap it out again until I noticed that the coolant level in the res had dropped so that the inlet was below the waterline. this introduced turbulence and air bubbles into the res to the point where they were being sucked straight down through the outlet into the pump and round the loop.
A quick top-up and problem solved.
It's nice when it works out that easily. :)