Koolance RP452X2 bleeding - impossible!

[Odai]

OK, so far I've spent about 30 hours in total trying to bleed my new loop with the Koolance res.

I'm using the "P1" mount (so the res on the left side). Photos of my loop here:

http://www.xtremesystems.org/forums/...d.php?t=268381

Problem is, since I started, I've made barely any progress at all. Once the loop was filled, I turned down the pump to minimise the flow rate. Although this seems to have worked, in actuality there are massive bubbles lurking everywhere and as soon as I turn up the voltage on the pumps they fly round the loop.

I then tried leaving it on the minimum flow setting for a while until I could no longer hear air bubbles. Then I would turn it up a bit, leave it, and turn it up again when the air bubble sound had gone. After about 12 hours of trying that, I'm still stuck on the "2" setting on my D5 pump. The bubbles just aren't going anywhere.

I tried tilting the case forwards, so that the inlet/outlet on my roof mounted radiator would be higher (bubbles in radiator would leave), but this causes a massive sudden increase in the amount of noise the pump makes, then it suddenly goes very quiet, with the flow almost stopping. Not good.

I then tried tilting it back about 45 degrees, so the res would be the highest, this quickly gets rid of the air bubbles in the pump, and it becomes quiet. I can still hear water sloshing about in the radiators though.

Any ideas on what I should do? Totally lost here.

[fatherTime27]

OK, so far I've spent about 30 hours in total trying to bleed my new loop with the Koolance res.

I'm using the "P1" mount (so the res on the left side). Photos of my loop here:

http://www.xtremesystems.org/forums/...d.php?t=268381

Problem is, since I started, I've made barely any progress at all. Once the loop was filled, I turned down the pump to minimise the flow rate. Although this seems to have worked, in actuality there are massive bubbles lurking everywhere and as soon as I turn up the voltage on the pumps they fly round the loop.

I then tried leaving it on the minimum flow setting for a while until I could no longer hear air bubbles. Then I would turn it up a bit, leave it, and turn it up again when the air bubble sound had gone. After about 12 hours of trying that, I'm still stuck on the "2" setting on my D5 pump. The bubbles just aren't going anywhere.

I tried tilting the case forwards, so that the inlet/outlet on my roof mounted radiator would be higher (bubbles in radiator would leave), but this causes a massive sudden increase in the amount of noise the pump makes, then it suddenly goes very quiet, with the flow almost stopping. Not good.

I then tried tilting it back about 45 degrees, so the res would be the highest, this quickly gets rid of the air bubbles in the pump, and it becomes quiet. I can still hear water sloshing about in the radiators though.

Any ideas on what I should do? Totally lost here.

You are jumping the 24pin to run the pump correct? Just keep filling/cycling the pumps power and pinching the lines to help get the bubbles out. Also I cannot see a res in the pics you posted.

[Martinm210]

P1/left reservoir is the best one to use for a single pump, so you're good there. I wouldn't let flow rate go slow though, you need higher flow rates to flush the bubbles out.

It helps a little if you can tilt the case on it's back a bit, but I would simply run it full speed and cycle the pump on/off to get the big stuff out. As long as it's moving water and not air locked, the pump is getting cooled and lubricated by water so it's no problem on the pump.

Initially when you have a bunch of air, it can also help to turn the pump on...then off...then on...then off, to sort of let the big air pockets get around the baffle plate without getting drawn back in the loop. As long as you're seeing big bubbles of air come up with the on/off bit, it's a good way to accelerate the process.

You'll get it...hang in there, but definitely speed up the flow. The air needs to get flushed out and broken up and find it's way into the front portion of the reservoirs, high flow will do that.

[Odai]

Hello,

I have tried simply leaving the pump on it's highest flow setting for a few hours, to try and break up the bigger bubbles as you said, but it doesn't seem to have had an effect.

There is one very large bubble at the inlet chamber of the rear radiator. Whenever I switch on the pump, it appears to be forced into the chamber, but not through the radiator tubes. No matter how high the flow rate, as soon as I switch off the pump again, the large bubble simply floats back out.

[Odai]

@fatherTime27, I am simply using a 12V DC power supply for the pump, and the res is in the top 2 5.25" bays.

[Martinm210]

I think you're fighting low flow rates and an inverted radiator. The reservoir probably isn't making it any easier, but air pockets like that can happen with any pump/res setup if you'r flow rates get really low.

That's a tough one....sometimes you can walk those out by rotating the case around at different angles, but it's not easy. A second pump in series would probably increase flow rates enough to push it through, but I would probably suggest just letting it run full speed for a week and see if it works it's way out. I'm not quite sure what happens (some sort of air absorption thing), but I have seen stuck air pockets like that on inverted radiators. If it's just a little air stuck in the radiator plenum chamber, it should affect performance at all.

I've seen some people get those out by unmounting all the components but leaving everything connected. Then they take and shake/invert/turn the rads etc so the air will work out. I've never really done it that way myself, but I have read through several guides suggest that.

[Serpentarius]

Hello,

I have tried simply leaving the pump on it's highest flow setting for a few hours, to try and break up the bigger bubbles as you said, but it doesn't seem to have had an effect.

There is one very large bubble at the inlet chamber of the rear radiator. Whenever I switch on the pump, it appears to be forced into the chamber, but not through the radiator tubes. No matter how high the flow rate, as soon as I switch off the pump again, the large bubble simply floats back out.

Jerk the flow by switching on and off ... the pump will be ok (heck, we all shutdown/restart pc every often)

this is what i do:
1. start pc
2. let pump run
3. identify the plug (molex/ 3-pin)
4. just pull the molex / 3-pin plug
5. wait 1 sec
6. plug it back
7. wait 2-3 sec
8. pull it out
9. wait 1 sec
10. plug it in

repeat till the larger bubbles are gone.

since there's not way to physically move the bubbles/air pockets (in the tubes) ... we'll have to rely on the pump+water surface tension to do the job. But jerking it is like more controlling where the bubble goes

atleast you have a see-through tube .... look at some other modders with no transparency ... they have to do the "guessing" game

[MagisD]

I found if I attached a line to top outlet of res it leaves the air some where to go besides the res and inlet. It's going to take path of least pressure vs suction. Faster easier bleed. When most if air out remove line and top up.

Sent from my X10a using Tapatalk

[NaeKuh]

*takes mental note... leave some lax tubing so u can pull bay res out a bit and tilt it. *

[Odai]

Thanks for all the advice. I'll give it a shot.

Although, after a few more hours of running the pump at it's highest speed, I'm still seeing large bubbles fly around the loop as opposed to the "frothy" aerated mixture I used to see with other loops. It does indeed seem to be a low flow rate. I've got all of these components in a single loop with one D5:

-About 4m of 7/16" tubing
-EK Supreme HF
-EK FC VGA block
-Swiftech NB Block
-TFC 360 radiator
-EK 120 Radiator
-XSPC RX120
-AC Flow sensor

Am probably pushing it with this many components. I might just invest in another pump for my next build and make it into two loops.

[CedricFP]

I have similar issues bleeding my dual DDC set up with an XSPC res top. Every now and then a huge bubble will "whoosh" around my system - I'll hear it.

It's been that way for months. Because of that, my thin-walled Tygon R3402 tubing is experiencing cavitation as well (is that the right word?) - what i mean is that it's being pinched by negative pressure.

Very annoying but I haven't managed a remedy yet. It's just one huge stubborn air bubble.

I kind of like it. I've named it George.

[antiacid]

put your two hands on both ends of the top of the case, gently tilt the case in one of the 4 main directions up to 45 degrees, shake the sh1t out your setup, try one of the other directions, shake again.

Also, what's this shenanigans about putting your pump at low? Keep it at 4-5. Also, if you can actually see bubbles in your tube, pinch that section (and slightly before the bubble) and release it gently, the bubble should disappear.

This isn't rocket surgery, you should be able to figure it out. Two pumps wouldn't solve this issue either. You're using A LOT of tubing, there's going to be a lot of pinching to get everything cleared out.

[Odai]

OK, I've tried all the suggestions, and despite it being almost a week, still have made very very little progress.

It may indeed be due to a low flow rate, does the amount of restriction I have in the loop seem excessive? Would it warrant tearing the whole thing down, and re-assembling with two loops (time consuming, and very expensive as I'd have to purchase an addition flow sensor and pump)?

I just removed the flow sensor at the very bottom due to this issue. With the case on the side, pulling the flow sensor out slightly put it at the highest point, and the amount of air left in the tubes is massive. About 20cm in each tube connecting to the sensor is totally empty. I figure it might help just to top these off before re-attaching them to my flow sensor.

Although, I'm still concerned there will be air lurking in places even after this measure... I did have the idea of maybe running the pump at a higher voltage than 12V (can't recall the limit -14V maybe?). In which case I'm sure it would be easy to source some sort of a transformer, or make one myself, just for the bleeding.

It's gotta be easier than totally re-building the loop...

[NaeKuh]

Odai pull the res out.. and then have it tilt so the face plate is in a diagonal angle looking at the top of your wall.

Then this forces all the holes in the rear to be completely submerged.
Then it will allow all the free bubbles to float up, and since ur res is tilted, the bubbles wont get reintroduced back into the system.

As the Res gets lower, add more water and more water.. and repeat.

[Martinm210]

Not sure this will help the low flow rate/high restriction issue, but I did create thislittle bleeding video that might provide some tips.

It's down the page here:
http://martinsliquidlab.org/2011/02/...x2-rp-452x2/5/

With two pumps in series, it took me about an hour to fill and have circulating fine...so higher flow rates and more pumping power does help.

While cpu blocks may work perfectly fine down to .7gpm, pockets of air getting trapped can be a problem. Not sure you'll fix that with the restriction and pumping power you have? You might need more pump.

What was you flow meter reading before you pulled it out?

[Odai]

Hello,

Thanks for your suggestions guys, will try them as soon as I have my flow sensor sorted.

Unfortunately, I have no hardware capable of reading the flow rate. I only have the sensor installed in preparation for when the new aquaero arrives.

Does the loop restriction look too great then?

I did some searching before deciding on this setup, and found others with more restrictive loops on one pump. I was actually hoping to start using lower pump settings when the system is bled too (maybe 1-2), as the highest setting produces far too much vibration.

[Martinm210]

Can you give us a list of all your parts?
I can't tell just by pictures

Also, what is the pump rpm?

[Odai]

I posted it above.

-About 4m of 7/16" tubing
-EK Supreme HF
-EK FC VGA block
-Swiftech NB Block
-TFC 360 radiator
-EK 120 Radiator
-XSPC RX120
-AC Flow sensor

Not sure how I can determine the pump RPM, as there is no tachometer wire on my sample. I do however have the pump turned up to the highest setting.

[Martinm210]

I posted it above.

-About 4m of 7/16" tubing
-EK Supreme HF
-EK FC VGA block
-Swiftech NB Block
-TFC 360 radiator
-EK 120 Radiator
-XSPC RX120
-AC Flow sensor

Not sure how I can determine the pump RPM, as there is no tachometer wire on my sample. I do however have the pump turned up to the highest setting.

Had to make some SWAGs on the NB and flow meter, but flow rate should be above 1GPM if you have the pump completely primed which "Should" be enough to move any pockets in tubing. Restriction doesn't appear to be that bad, it's getting down there, but I think when primed you should still be above 1GPM when at full speed.

I think you have an air pocket stuck in the pump volute. Try tipping the case on it's side so the pump volute is laying flat, and turn the pump on/off/on/off several times. Shaking if you can too.

If the pump is fully primed, it should sound very quiet compared to when an air pocket is in there.

[Sideroxylon]

Based on the pics you posted in this thread I could see how you might have some problems bleeding the loop. You really need to consider simplifying that loop. You have way too many places for air to get trapped in there.

[Odai]

I had a feeling something would go wrong with this loop.

My luck has been dreadful this time round.

The pump does seem totally primed. It sounds like it used to when the loop is totally bled, except for the odd whoosh of bubbles and the much higher vibration due to the res mounting system.

I really can't get the loop any simpler. I tried almost every single possible combination of components (some simply weren't possible due to tight bends), and this is the best I could do.

I might just go for another pump (two loops)... I saw the naked Alphacool D5 pump (no top) on sale here for about £60, and I guess I could stomach that considering I'll be returning my £40 flow sensor and won't be getting a replacement. It would also mean simpler tubing runs.

I did try Naekuh's suggestion, and antiacid's, but neither have done much. In fact, I think it would be a bad idea to simply tilt the case every other way, considering how unbelievably heavy it is. It's already developed a worrying flex...

[Odai]

Actually, I've just had a look at Swiftech's page for their D5 pump product.

It says the acceptable voltage range actually goes up to 24V (not sure where I saw 14V...).

So, do you think it would be a good idea to simply hook up my pump to 24V, for a number of hours to simply bleed the loop?

EDIT: Never mind, seems my Corsair PSU has a limit of 0.9A on the -12V rail, which makes it pretty useless for this.

[crash5s]

I use a TJ-09 and hit this all the time as my radiator is at the top, second one at the back and the res is lower. Put the case on it's side, and lift the res out of the case so it's up, you won't overheat anything as long as you don't game, run the sucker for 6 hours like this and call it a day.

[penguins]

just tilt the case out it upside down if you need to, get a strong friend to help if you need it :P that's the only way you're going to get the bubbles out.

I flip my tj09 upside down, on it's side, back and forth, god knows if I had a wife she'd kill me for the dining room table scratches

Get your Strong man on.

[Odai]

Thanks for the responses.

Unfortunately, I really don't think anything is going to get rid of some of the bubbles apart from higher flow. Because of the way the tubes are routed, there is no way I can remove one of the bubbles my simply tilting the case, until it rises into the reservoir. When I get it into a good place, switching on the pump simply forces it back to where it was (radiator plenum chamber).

I think my only two options are to either find some way to put 24V across my pump, or get a second pump and re-do the set up.