• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Another way not to position your Eheim

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

BeerHunter

Member
Joined
Nov 5, 2001
Location
Oil Country
After hanging the pump by intertube to elminate vibration noise I was wondering why I could'nt fill my system and it would'nt "take" the water into pump..So I searched these forums and found many of you saying "the pump outlet down" is bad. However, As it tunes out both the inlet or outlet pointing down is a negitive since the air sits inside the pump and can't escape. More so with the intake pointing down since the whole impeller cavity inside the pump is above the outake in this position and air is trapped. SUX.
DSCF0010.jpg
 
BeerHunter said:
After hanging the pump by intertube to elminate vibration noise I was wondering why I could'nt fill my system and it would'nt "take" the water into pump..So I searched these forums and found many of you saying "the pump outlet down" is bad. However, As it tunes out both the inlet or outlet pointing down is a negitive since the air sits inside the pump and can't escape. More so with the intake pointing down since the whole impeller cavity inside the pump is above the outake in this position and air is trapped. SUX.
Maybe not ideal, but once filled & bubble free this pump position should work as any other.
I understand that, without a three-tap system, filling the loop and priming the pump is not an easy task in this position...
From your pic I can't understand clearly how you fill the loop.

That's why I use the 3-tap system in my system (not very elegant but terribly efficient...).

pict0157.jpg


Filled & bleeded in seconds, air free in minutes...

CD :)
 
Re: Re: Another way not to position your Eheim

Turbokeu said:

pict0157.jpg


Filled & bleeded in seconds, air free in minutes...

CD :)

Ok, you gotta explain the method you use to fill-bleed with that contraption, and show a slightly further out picture so we can see the water/air path. I've seen the fill-bleed kits sold online before and figured I could make one myself for alot cheaper, but how exactly do you go about filling/bleeding with it?
[/threadjack]

like Contra said, rotating your case around will help get the bubbles out of there.
 
ehh?
Move the T to the top of the system, at the pump's outlet. Should be able to fill it fine then.

Right now you've got a giant air bubble trapped in the lines and the air can't move down to exit through the T.

Course, laying the computer on it's side (so the side shown to the camera is on top) and tiliting it further so that the T is the highest point in the system will make it much easier to fill.
 
Re: Re: Re: Another way not to position your Eheim

TheGhengisKhan said:


Ok, you gotta explain the method you use to fill-bleed with that contraption, and show a slightly further out picture so we can see the water/air path. I've seen the fill-bleed kits sold online before and figured I could make one myself for alot cheaper, but how exactly do you go about filling/bleeding with it?
[/threadjack]

like Contra said, rotating your case around will help get the bubbles out of there.
Sorry... here are some pics.

Overview:
pict0191.jpg

The middle valve is an Eheim 1/2", the two left valves are Eheim 3/8", the other parts are 20mm ID PVC T's.

During filling or bleeding (this is an older pic with a Swiftech MCW462 and a Sicce Idra (big sh*t! :mad: ) pump:
pict0108.jpg


wc.jpg

In normal operation valve 1 is open, valve 2 & 3 closed.
To fill: Close valve 1, open valves 2 & 3, put a hose connected to valve 3 in a recipient containing the coolant, suck with your mouth on a hose connected to valve 2 till coolant enters the pump. Power on the pump and check coolant begins to flow through the loop. If so, put the hose connected to valve 2 in the recipient with coolant too and let run for a couple of minutes till most of the air is gone. Then close valves 2 & 3, open valve 1 and remove the two hoses. The airtrap will take care of the remaining air in the loop. Done...
To bleed: Attach the two hoses to valves 2 & 3 to a jar, close valve 1, open valves 2 & 3, wait till most coolant has dropped into the jar, blow into the hose connected to valve 2 to remove remaining coolant. Done

Recent pic: (cabling and front panel have to be finished)
pict0208.jpg


CD :)
 
That is kick *** setup you have there.:) My issue is when water disaccociates over time turning to gas it's fills the pump cavity which seems to rely on water for lubrication/spinning medium of the impeller shaft..the pump simply stopped spinning after filling with air and air could'nt escape because the outlet is lower than the cavity. When I rotate case so pump is upright no problemo... Need to think about your setup more...Did I mention it was sweet.

PS are there any probelms going from the pump to your first block? Most setups I've seen go from rad to first block?
 
that looks sweet, but quick question, shouldn't the inlet to the air trap be long to be more effective? mines about that long, but my pump is soo strong, bubbles just get sucked back into the outlet.
 
BeerHunter said:
That is kick *** setup you have there.:) My issue is when water disaccociates over time turning to gas it's fills the pump cavity which seems to rely on water for lubrication/spinning medium of the impeller shaft..the pump simply stopped spinning after filling with air and air could'nt escape because the outlet is lower than the cavity. When I rotate case so pump is upright no problemo... Need to think about your setup more...Did I mention it was sweet.

PS are there any probelms going from the pump to your first block? Most setups I've seen go from rad to first block?
Thanks guys, and thanks Beerhunter...

Beerhunter,
I know you're completely right about the pump stopping due to air buildup, but I cant stop wondering if the re-appearing air is not simply due to remaining air in the coolant.
After filling my loop, and after some time, the inside of the tubing is being covered with lots of little air bubbles. The next day the airbubbles have gone away and the coolant level in my airtrap has dropped almost half.
I don't think water dissaccociates, but I think firmly that remaining dissolved air in the water comes free after some time.
As you can see on one of my pics I have a fishtank (in fact several fishtanks).
After every fresh installation of one of those fishtanks (especially with fresh tapwater) the glaswalls become covered with thousants of little air bubbles after a couple of hours...(which will dissappear gradually after several hours, a fishtank being not isolated from the surrounding air like a watercooling loop)
I think also that if you had any kind of air trapping device (an airtrap or a reservoir) your issue would be certainly less pronounced.
(OMG, I never used that much the word "after"....:p)

This doesn't exclude the fact that re-appearing bubbles (due to little or microscopic leaks for example) can prevent the pump from runinng normally...

About the mounting order in the loop...
I'm convinced (and I tried it out by measuring temps at different locations in the loop) that it doesn't matter really...
You will probably notice no difference between the different options, as the temperature gradients through the loop are very small (I'm talking about tenths of degrees C...).
A good WC setup, with a consequent water flow, shoudn't have big temp gradients through the loop (if so it means there is a thermal transfer problem in the rad, or water flow is too low...)

I'd say, use the option that permits you to route your hosing most conveniently and as short as possible...

CD :)
 
LuckMan212 said:
Please Turbokeu, some questions: (sorry if they are dumb!)
first off, really sweet job you did there!!:p

Thanks...
There are no dumb questions, there are only dumb anwers... (as we use to say here in Belgium....:) )
Trying to answer as less dumb as possible ...:p :

Q: How did you mount your power supply down here??
A: I did nothing, its the original location of the PSU in a Lian Li PC-70...

Q: Home made shroud??
A: Yes, see my webpage...

Q: Rad?
A: BeCooling AquaCoil II cube rad, cube rads have very bad reputation (bad performance, big flow restriction), but it suits me very well... ( I could even say, it fits my PC-70 perfectly...) See also my post here.

Q: Where did you get these clamps?
A: These are standard clamps for domestic tapwater tubes, just search for the right diameter.

Q: How is this attached to the case?
A: with a M4 screw and a nut...

Q: What exactly is this part? Is it standard?
A: No, it isn't standard. it's just an L-shaped aluminium part I made for stability, connected to standard fishtank Eheim valves...

Hope this helps.

CD :)

PS: Do not forget this project started in 2001, I had the time to plan it, and also I'm very good with my hands...:D
 
Korndog said:
that looks sweet, but quick question, shouldn't the inlet to the air trap be long to be more effective? mines about that long, but my pump is soo strong, bubbles just get sucked back into the outlet.
When I designed my airtrap, the very simple idea was: air bubbles will rise, water will go down...
This is true for big bubbles but in fact it's not that simple.
Due to water velocity, as you observed yourself, most of the air will get sucked back into the outlet.
In real world conditions my airtrap turned out to work very well.
On special request I made five of them in total, bringing me to write a little howto on my webpage.

I will try to explain the working with a little drawing:

airtrap.jpg


As you probably observed too, after filling the loop, the trapped air in the coolant is "mulched" by the pump in thousants of little air bubbles.
It's more like foam and it's even more pronounced when using additives as Water Wetter, Purple Ice or other.
Coolant coming from the 1/2" ID tubing enters the Tee who has an ID of 20mm (a little more than 3/4"). Due to the increase of diameter the coolant slows down. Most of the flow goes indeed right to the outlet, but a part goes up to the upper reservoir of the airtrap where it creates a very slow vortex (I can only see this vortex with a torch lamp and when there is still air in the loop). This where the air is separated from the coolant by leaving enough time for the bubbles to reach the surface.
At every passage (several times per minute) through the airtrap a part of the coolant is made airfree.

From a milky solution at the beginning it turns to a clear liquid in tens of minutes, only leaving a layer of foam at the surface in the airtrap.
After a couple of hours this foam layer dissapears too and you will notice that the coolant level has dropped quite a bit.
Just add new coolant through the lid in the airtrap till level is halfway again.

During normal use, the lid of the airtrap must not be airtight (I open the lid about half a turn) to allow air bubbles to escape and air to enter if coolant level changes (ex. due to temperature variations).
Only during transportation or during filling the lid has to be closed completely.

Several commercial reservoirs-airtraps (like Innovatek's Tank-O-Matic or Fass-O-Matic) suffer from a big squirk: all the coolant passes through the reservoir making it almost useless with high flow pumps...but they are beauties. If I could find somebody with the needed knowledge and tools I liked to make my airtrap from anodized aluminium too...:D

CD :)
 
I need to know where to purchase those Eheim taps that you have. I want a 1/2" three ball valve fill/drain system like that but I've had no success in locating those taps. Mind enlighting the rest of us as where to procure these elusive ball valves?
 
leadsalad said:
I need to know where to purchase those Eheim taps that you have. I want a 1/2" three ball valve fill/drain system like that but I've had no success in locating those taps. Mind enlighting the rest of us as where to procure these elusive ball valves?
Normally you should be able to find (or order) them in any petshop that carries Eheim brand products and accessories.

Part numbers:
1/2" (12/16mm) valve: 4004512
3/8" (9/12mm) valve: 4003512
Completely made of PVC (no metal parts).

Originally both sides are meant for hose connection with the Eheim type hose clamp (like the original in- and outlet of an Eheim pump). By removing the "nut" part (the clamp) and cutting of the hose barb part (the non threaded part) you can connect it to a 1/2" inner thread (for the 1/2" valve) or to a 3/8" inner thread (for the 3/8" valve).

CD :)
 
Last edited:
Back