Pumps. In-Tank or In-Line: which is the quietest?

The alternative to direct air cooling

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply

Pumps. In-Tank or In-Line: which is the quietest?

In-Tank
2
40%
In-Line
3
60%
 
Total votes: 5

DynamicalSystem
Posts: 16
Joined: Thu Apr 15, 2004 9:19 am

Pumps. In-Tank or In-Line: which is the quietest?

Post by DynamicalSystem » Wed May 05, 2004 9:38 am

So I've got my suspended Eheim 1250 pumping through a bunch of stuff. The rad is outside the box, the pump is inside the box.

Temps are OK and it's quietish, but if I suspend it submerged in a reservoir will it get quieter? I've seen posts for and against.

On the temp question, I've seen arguments that say if you submerge your pump, you put the pump's heat through the water and this is a bad thing. But I challenge this for my setup. If the rad is outside and can handle more, then putting the pump's heat into the loop, rather than the case, has got to be a better option.

Anyone got any theories or experience on this?

DS

Gooserider
Posts: 587
Joined: Fri Aug 01, 2003 10:45 pm
Location: North Billerica, MA, USA
Contact:

Post by Gooserider » Wed May 05, 2004 2:12 pm

Putting the pump inline is better as the heat will cause less of an issue for the case than it will for the loop. How much of a problem this is depends in part on your rad, if it's overkill for the existing hardware then adding the pump heat probably won't hurt that much. If it's on the small side, then you are better off keeping the pump heat out of the loop.

Noise is a mixed issue, some folks have reported it helps, others say it doesn't.

Gooserider

apocalypse80
Posts: 90
Joined: Mon Feb 09, 2004 4:16 pm
Location: Greece

Post by apocalypse80 » Wed May 05, 2004 3:46 pm

From what I've seen , submerged works better for silencing.
However some means of vibration dumpening is still required (again from what I've seen , sponges seem to work OK).

As for heat , when inline , most of the pump's heat is passed to the water , when submerged , all the pump's heat is passed to the water.

However the difference should be negligible , your pump consumes 30Watts (or 28? can't remember) , whether 20 or 25 or 30 of those are passed to the water should make no difference.
10Watts are nothing for any respectable radiator.

One thing to note though , since I have the same pump , you'll need a BIG tank (nah , HUGE).

DynamicalSystem
Posts: 16
Joined: Thu Apr 15, 2004 9:19 am

Post by DynamicalSystem » Wed May 05, 2004 9:57 pm

50/50 so far. I'm guessing the answer, as ever, is it depends.

Apocalpse80 wrote : One thing to note though , since I have the same pump , you'll need a BIG tank (nah , HUGE).

Just so I can get some perspective here, are you talking bucket size? I was thinking of fabricating a lexan reservoir (box shaped) to enclose the pump. The pump would be suspended inside using Stretch Magic (or summat). Clearance around the pump would be in the order of an inch all round (after fittings), and this space would be filled with water. The tank would be fixed inside the case in the same space that the pump curently occupies.

apocalypse80
Posts: 90
Joined: Mon Feb 09, 2004 4:16 pm
Location: Greece

Post by apocalypse80 » Thu May 06, 2004 6:21 am

I suppose you're going to remove the inlet barb (it's useless when submerged) and the plastic feet.
Without them , the pump is around 5.5"L x 3.5"W x 3.5"H (without the outlet barb).
Since you want all of it to be submerged + the 1 inch clearance + the thikness of the tank , I suppose you'll need a tank of ~ 7"L x 5"W x 5"H (probably higher to allow for suspension).

In a super midi case (compucase 69U6) , I find that the 1250 (inline) is a tight fit , but for some huge tank like that , you'd better have a cube case.

Anyway , measure a lot before you decide.

One note : I'm a metric guy , so there might be some errors in my cm -> inches conversions.

DynamicalSystem
Posts: 16
Joined: Thu Apr 15, 2004 9:19 am

Post by DynamicalSystem » Thu May 06, 2004 8:26 am

Thanks for the replies, Gooserider and apocalypse80.

The question, I guess, is if I go to the bother of fabricating a 5x5x7" box, will I see any noise benefits? I'm not convinced I will, and I'm not sure I want to go to the bother of fabbing if there isn't going to be any benefit.

BTW, Gooserider, the rad is a generic heatercore from Over-Clock. It's the 240x120mm model - not sure what car it came from but it fits nicely on the back of my 4U rackmount case.

Gooserider
Posts: 587
Joined: Fri Aug 01, 2003 10:45 pm
Location: North Billerica, MA, USA
Contact:

Post by Gooserider » Fri May 07, 2004 4:37 pm

Single or dual pass type rad? (Single pass has one I/O on each tank, dual has both I/O's on the same tank) Singles are lower restriction, but both offer similiar cooling potential. However a dual 120 rad should be able to handle any reasonable heat load without any trouble.

Gooserider

afrost
Posts: 141
Joined: Thu May 01, 2003 9:42 am

Post by afrost » Fri May 07, 2004 7:33 pm

do you have stiff hose connected directly to the pump? you might want to go with a couple inches of very soft hose in and out so that the hose doesn't vibrate.

DynamicalSystem
Posts: 16
Joined: Thu Apr 15, 2004 9:19 am

Post by DynamicalSystem » Sun May 09, 2004 2:37 am

Gooserider, The rad can be seen at Over-Clock. It's the second one down and I think it is single-pass, but your description belies that.

afrost, the tubes are long lengths of 1/2" tygon. Inlet is connected with a straight plastic barb, and it curves smoothly. The outlet is dead straight and connects using a straight plastic barb.

I shall try and get some piccy's up.

Gooserider
Posts: 587
Joined: Fri Aug 01, 2003 10:45 pm
Location: North Billerica, MA, USA
Contact:

Post by Gooserider » Mon May 10, 2004 6:20 pm

That rad is definitely a dual pass, however it is big enough not to be terribly restrictive. OTOH, I've seen test numbers where they compared two rads of about the same size, one a single pass 2-342 (arguably the gold standard of single pass rads) and the other a dual pass. The single pass showed less than 1/4 the resistance at any flow rate.

Gooserider

Post Reply