Negative Pressure tube buckling

[Cue]

Hi all, I've recently ran into a problem with one of my loops. The inlet tube to a Koolance RP-452X2 (with D5 pump) seems to have collapsed under the negative pressure of one of the pumps (the other loop is fine):



It's a little hard to see but the spring ring at the bottom is what the tube diameter should look like. You can kind of see at the top that the tube has completely buckled and my flow became low enough that the alarm went off. It's worth mentioning that this is the right side loop of a rev.1 RP-452X2 that I have had since early 2011 however I didn't receive the free bleed pipe kit so that's not installed (I did contact Koolance via PM on these forums regarding it, I really should have chased that up). This problem happened straight after trying to bleed it for the first time in a long time but it could have been building up to it for years.

So anyway some advice please, what do you think is the best course of action to prevent this? I looked at some negative pressure resistant tubes (the ones reinforced with helix wire) but it's hard to find them with 10mm inner diameter that match my fittings. The alternative is to change my fittings and get thicker walled tubing. Would that do it or is there an underlying problem I need to fix. Would simply replacing the tubing like for like then getting the bleeding pipe kit for the RP-452X2 mean that it is less likely to happen in future?

[Charles Wirth]

Something like this will work

http://www.frozencpu.com/products/66...tml?tl=g30c289

I see you already have a blue one.

[Cue]

Something like this will work
I see you already have a blue one.

Hi Fugger. I'm not so sure. At the bottom of that tube I have a steel version of an anti-kink coil


When I look inside it-it seems to have collapsed within it too though to a lesser extent.



I think the plastic blue antikink coils I have would help even less than the steel one there, so I don't think that will solve it but I can certainly give it a try since I have a lot of the stuff.

[Rinaun]

Using antikink at this point is just bandaging the problem. Are you able to decrease the pump flow at all? The only ways to fix this issue (as far as I know) are:

1. Find out where you have the bottleneck of flow; is something clogged? Dropping from 1/2 to 1/4g somewhere? eliminate that/round out fitting for more flow. If water can't make it back to the pump it puts stress on the pump and pressurizes the lines on the "out" end of the equation, and depressurizes the side on the "intake" side of the equation.

2. Slow the pump down. The pump is obviously struggling to pull water. Adding a reservoir won't do any good because its just adding more water, not fixing the flow problem. D5 is already overkill for 1/2 the projects they are used in.
3. Stiffer hose. Again, I feel this is just a bandage to the issue of too much flow/restriction issues. It's the easiest fix though.
4. Use a circuit setter and make a small bypass to alleviate pump pressure. This is how I stop my system for vortexing the water due to the D5's strength. Just set a T somewhere on the "outtake" side and have a 1/4 line input that right before the pump intake. Twist circuit setter until pipe stops collapsing.

Oh, it would help to post your entire loop in a picture as well so we can get a rough idea of what you have setup. For all I know you could have connected things in a unworkable fashion and the math was never right in the first place. I know when I did my D5's I had issue running high temps and long amounts of tubing right before the intake. Tygon loses its shape quickly when heated.

[Chilla X]

Been ages since I visited this forum & I couldn't resist posting on this thread. This looks very similar to a problem that I had with tubing to a pump inlet collapsing. The cause in my case was a gunked up cpu block back down the loop. It might not the cause in your case but its easy enough to check if there is any blockage or restriction before the inlet tube.

[Ace123]

all that tubing is so old its yellowed..... ALOT. You might consider just replacing all of your tubing with newer stuff.

[Cue]

Hi everyone, thanks for the replies.

all that tubing is so old its yellowed..... ALOT. You might consider just replacing all of your tubing with newer stuff.

Are you telling me that I can't use beer as a coolant? Yeah the tubing is extremely old now, from 2011 IIRC. I will definitely replace it too but I want to prevent the new tubing from doing the same if I can.

This looks very similar to a problem that I had with tubing to a pump inlet collapsing. The cause in my case was a gunked up cpu block back down the loop. It might not the cause in your case but its easy enough to check if there is any blockage or restriction before the inlet tube.

Good point, my blocks have clear tops and from a quick inspection they don't seem to be blocked, but it may well be my rad. I will flush them when I take my loop apart just for safe measure.

Are you able to decrease the pump flow at all

Not quick access but absolutely. It's a D5 Vario so I can adjust the speed manually if need be. Is that a good idea, will I inevitably see my temps increase?

Find out where you have the bottleneck of flow; is something clogged? Dropping from 1/2 to 1/4g somewhere? eliminate that/round out fitting for more flow. If water can't make it back to the pump it puts stress on the pump and pressurizes the lines on the "out" end of the equation, and depressurizes the side on the "intake" side of the equation.

Not sure what you mean by 1/2 to 1/4g to be honest, can you please elaborate. Do you mean some kind of tube constriction? I have leak-free quick release fittings so I think they do constrict a fair bit at those points, a bit like a venturi pipe.

Slow the pump down. The pump is obviously struggling to pull water. Adding a reservoir won't do any good because its just adding more water, not fixing the flow problem. D5 is already overkill for 1/2 the projects they are used in.

Just to clarify, I already have a reservoir, the RP-452X2 has/is a reservoir. The problem with it is that I cannot open the reservoir to alleviate negative pressure on the inlet tubing because the rev 1 has a problem where the water level falls below the pump intake as soon as it's opened and the pressure equalizes. The pump ends up running dry very quickly. The bleed kit would really help in this situation.

4. Use a circuit setter and make a small bypass to alleviate pump pressure. This is how I stop my system for vortexing the water due to the D5's strength. Just set a T somewhere on the "outtake" side and have a 1/4 line input that right before the pump intake. Twist circuit setter until pipe stops collapsing.

Interesting, do you have a picture of this? as I've not come across this before I'm finding it hard to follow what you said. I should probably also point out that after your advice with a fully stopped pump I opened the res to equalize pressure. The tube does not revert back to it's original form at all. It seems to be permanently deformed due to creep.

Oh, it would help to post your entire loop in a picture as well so we can get a rough idea of what you have setup.

My apologies. Here are a couple of pics of the particular loop that I'm having a problem with. If there is any additional information you need don't hesitate to ask.


That is the loop and the relative positions of the components.
The actual pic is kind of messy as there are two loops and my wire management sucks.

[Rinaun]

Hi everyone, thanks for the replies.


Not quick access but absolutely. It's a D5 Vario so I can adjust the speed manually if need be. Is that a good idea, will I inevitably see my temps increase?

No, temps should not increase. Flow is not always related to cooling performance. If you can lower the pump speed you lower the amount of heat dumped into the loop (the pump adds heat, it's an electrical device after all). Even if they did rise, you could easily benchmark this via a few hours and OCCT. For MY rig, I have the flow slowed 20% on a D5 and I see no difference in cooling ability. The only concern with running a low-flow loop is making sure your tubing size is smaller for a better scrubbing action on the tube sidewalls compared to higher-flow and larger tubing.

Not sure what you mean by 1/2 to 1/4g to be honest, can you please elaborate. Do you mean some kind of tube constriction? I have leak-free quick release fittings so I think they do constrict a fair bit at those points, a bit like a venturi pipe.

Well honestly my first "attempt" at solving this would be disassembling the loop and purging everything with some diluted vinegar (1:1 with water is fine). If it worked before and suddenly it stops working, then this is a major indicator that something is clogged. I'd also check the quick-disconnects as depending on what brand they are or what size Inner Diameter, they could be causing a choke point in the loop (doubtful though). The best indicator of whats causing this problem is the fact that it suddenly started to collapse, which means something occurred over time (tube weakening, algae/crud) and not nessecarily bad plumbing.

What I meant by 1/2 to 1/4 is that when you go from a large tube size to a small hole somewhere else (for example, lets say something "blocks" the radiator intake so it's HALF open) it causes a bottleneck, or water "traffic jam". Judging by the color of your loop, I'd bet it's a contaminant in the loop blocking off some part of the radiator or blocks rather than bad math. I've never had issues with loops being blocked because I use 1/3-1/2 cup of antifreeze in my loop to prevent anything growing inside.

Just to clarify, I already have a reservoir, the RP-452X2 has/is a reservoir. The problem with it is that I cannot open the reservoir to alleviate negative pressure on the inlet tubing because the rev 1 has a problem where the water level falls below the pump intake as soon as it's opened and the pressure equalizes. The pump ends up running dry very quickly. The bleed kit would really help in this situation.

I think I remember this issue from SVC; it made bleeding this unit a nightmare and IIRC it's almost impossible without some sort of trickery (prefilling it, etc). I'd highly suggest the bleed kit or eventually you will end up with a broken 80$ pump when the impeller impacts incoming water after spinning incredibly fast in a air pocket.

Interesting, do you have a picture of this? as I've not come across this before I'm finding it hard to follow what you said. I should probably also point out that after your advice with a fully stopped pump I opened the res to equalize pressure. The tube does not revert back to it's original form at all. It seems to be permanently deformed due to creep.

I don't have a picture of this, but with a D5 you should be able to avoid this and just use the pump's variable speed. I remember people saying not to go below a certain speed, but that may be just old-school voodoo talk. I think 3-5 are the recommended speeds. If the tube is deformed, you can heat the tube up in boiling water and it SHOULD rebound back to a solid opened form as a quick fix. If the tube is deformed though you need to replace the tubing (a must). I'd also like to remind people that a ANTI-KINK brace won't do anything for a pressure issue.

http://www.anony.ws/i/2014/08/16/Ideanote_20140816_031608_01-1e6f7c.jpg[/IMG]
That is the loop and the relative positions of the components.
The actual pic is kind of messy as there are two loops and my wire management sucks.

Honestly if you have the cash I'd either get the bleed kit or ditch that res. I'd bet 9/10 that your math overall will be much better without that. Pumps + reservoir combos have always been a nightmare for me and that unit is probably the worst offender. The bleed kit makes it a bit more enjoyable though. In case you are wondering, for my loops I run D5 PUMP > Thermochill 3x120 RADIATOR > CIRCUIT SETTER > Raystorm CPU > EK580 GPU > cylindrical aluminum/acrylic RES > back to D5 pump.