pump cooling?

[Joe Camel]

ill keep this simple,


if i put a bunch of heatsinks on the outside of my pump(s), would this help lower the amount of heat it puts into the loop?

i have GOOD airflow around the pump(s).

is the heat a pump adds from the motor or is it the act of moving the water (friction) that adds the heat?

pumps: Mag 3, little giant (500 GPH 45 watts)


?

[Ares]

just a guess, but Id assume most of the heat is near the impeller, or in some way in that area.

a few fins cant hurt though. doubt it would make a huge difference, but you guys are all about .5degrees and all, so if it gets you .01degrees why not

[AntiBling]

Depends on how hot the pump gets, I think. Hate to give you another "just a guess", but just guessing it wont make any difference except for appearance.

As far as I know, the heat source is both motor heat and friction. If you're really into it, read about Joule and how he came up with his constant. He quantified the energy required to heat water a certain amount by stirring it.

[pdf27]

First of all, is the pump casing actually hot in operation? If not, there's no point!

[SlaterSpeed]

pdf27 make the best point

If it is hot then try it and tell us if it works. If it does then we can all go out and make hsinks for our pumps

[Ares]

hey I got an idea thats not a guess!

I read one con to submersible pumps is that it adds extra heat to the system by being cooled to the water.

so by this, it means atleast SOME heat must escape to the atmosphere. right?

[jlrii]

hey I got an idea thats not a guess!

I read one con to submersible pumps is that it adds extra heat to the system by being cooled to the water.

so by this, it means atleast SOME heat must escape to the atmosphere. right?




With a non submerible least a little always does....how much depends apon the design. A submersible will be cooled completely by the water...no choice in the matter- all the heat has nowhere else to go. Mag or direct drive pumps release varying amounts of heat to the water. I beleive Cathars testing showed that the D4, having a fully encapsulated armature, put almost all of its 15-16 watts of electrical heat into the loop via its face plate, while a pumps like the Little Giant MD series have an open armature and are fan cooled so most of the heat is removed via air.

[krazy]

Pumps always make some heat. Most of the heat comes off of the motor, but the friction of the impeller will make a negligibly small amount of heat, too.

If you run the pump inline, the vast majority of pump heat is dissipated through the pump's housing. It will be dissipated into the air in the case. The small amount of heat made by the impeller will obviously always end up in the water.

Running the pump submerged in a reservoir will capture all of the heat of the pump with the water. there is nowhere else for the heat to go.

Adding heatsinks to the side of an inline pump that has healthy airflow might help the pump run cooler, but it is doubtful that the water temperature will change any measurable amount.

What kind of pump are you using?

[Joe Camel]

i AM using BOTH the Mag3 and the lil giant.

ok so if "most of the pumps heat is removed via air" why the heck are we worried sooo much about adding a 2nd pump :shrug: ..see "dual pump" thread.

ill agree that my theory of 2x flow thru the rads dosnt help (much), but the added head and reduced resistance in the 2 loops will help.

[Cyco-Dude]

most of the pumps heat is NOT "removed via air." mag-drive pumps are generally pretty inefficient, and a lot of the power is converted into heat which warms the water. very little heat is actually transfered from the pump to the surrounding air (this depends on what type of pump you have. i'd suspect a direct-drive pump would get pretty warm and require air-cooling of the motor). putting heatsinks on a mag-drive pump would be pretty useless.

[Joe Camel]

ok, lets take this to an EXTREME, if one were to (crazy i know) submerge the pump(s) in a 3rd loop (no WB just pump(s) heat)....you could just buy a Promy and still have enough $ to get a 6-pack+ and have way better temps....nevermind.

guess it comes down to: if you have a HS sitting around, it wont hurt your temps if you put it on your pump...

[AngryAlpaca]

Do it anyway. A little bit of heat would escape, and it would look awesome.

[feathers]

I don't think you would see any temperature drop at all (unless your pump is hot?). I'm using an Eheim 1048 which gets warm.

The small amount of pump heat that goes into the water can easily be removed by the water radiator!

[Joe Camel]

Ive been testing the D4 pump (Laing/DangerDen) @ 16V in my 2Loop setup and Ive noticed one thing... IT GETS HOT!

quote from Cathar over on OCUA:
"The middle rear of the motor gets pretty warm to hot when run at 16V continuously, possibly reaching upwards of 25-30C above ambient without any active cooling (say a fan) running over the pump."

after getting my 3/4" D4, i decided to do a little testing B4 installing it into the 2Loop.
the back of the pump housing "looked" (to me) like it should come off. after a few trys, it did. (it snaps in/out nothing broke)
i was happy to find an opportunity to improve what i found


so i whipped up a SIMPLE test (nothing scientific!) to see if this was true.




i ran the loop for 2 hours with the stock pump ( 12V !!). the water got to 86.0F with room temp of 66.6F. (the temp stayed @ 86.0F for 15min)
then i took the back off and removed the board from the housing (just let it hang to the side).




let it run for an hour and the temps were: 82.8F water and 67.1F room.


so by taking the "heat source" out of the housing, it went from 19.4F > room to 15.7F > room. so a 3.7F drop!

needless to say, Ive been modding the pump housing so as to have a fan cool the INSIDE of the pump... a little ghetto right now, but i think i came up with a functional way

its 40 min into the test...going to let it run all night...update soon...MUST sleep!...

UPDATE:

the fan is blowing OUT...



after 8 hours: 77.4F water, 63.9 room 13.5F > room. (5.9F lower than stock)

[BillA]

terrifying mods,
I've forwarded this thread to Laing

have you considered watercooling it ?

[Titan151]

Hey I bet this is all so you can get Far Cry to run smoothly! I have been wondering about pump heat as well. I have a junky via aqua pump and a preliminary water chiller set up. I am going to run it submerged then in-line. I will post my results on the temp difference.

[Joe Camel]

Originally Posted by unregistered

terrifying mods,
I've forwarded this thread to Laing

have you considered watercooling it ?

terrifying? LOL the only part that scared me was drilling the first hole; the drill was like 1/64" (too damn close) from the "guts" of the pump. the rest of the holes are >= 1/8" away.

quote from Cathar:
"The D4 passes approximately 90% of its energy input into the water that flows through it. The other 10% gets emitted out the pump motor body.

The MCP600 passes approximately 70% of its energy input into the water, with the other 30% passing out the motor body."

that was the statement that got me thinking i might be able to improve on the D4 somehow.

EDIT: this is the OC forum thread:
http://www.ocforums.com/showthread.p...hreadid=295201

[Brians256]

Originally Posted by unregistered

terrifying mods,
I've forwarded this thread to Laing

have you considered watercooling it ?

LOL! Thanks for that comment Bill. He could have put pelts on it to see if could ramp up to 20V.

BTW, how are you generating the 16V for the pump, Joe Camel?

[redleader]

Any reason you don't just improve your radiator? If its performance was already so marginal that adding 10 or 20 (or even more) watts from the pump is hurting CPU temps, you'd probably do well to get a bigger radiator regardless.

[pauldenton]

Originally Posted by Brians256

LOL! Thanks for that comment Bill. He could have put pelts on it to see if could ramp up to 20V.

BTW, how are you generating the 16V for the pump, Joe Camel?

he uses these...
http://www.surpluscenter.com/item.as...tname=electric

[Joe Camel]

Originally Posted by redleader

Any reason you don't just improve your radiator? If its performance was already so marginal that adding 10 or 20 (or even more) watts from the pump is hurting CPU temps, you'd probably do well to get a bigger radiator regardless.

hummm, i think 2x Fedco 2-342's (single pass) heater cores, in series, with a 540cfm dual blower is about as good as it gets...yes?

this is more about squeezing EVERY last .1C out of the loop. also... im (nuts ) running 2x of these pumps (series) thru the WB (only) and 1 Rio 2100 pumping thru the rads (only) both empty into a common (custom) res...its a LONG story... :shrug:

here is a quick pic of the (latest) layout: (the "test pump" is hard to see but its in there. ill be drilling the 2nd soon )




lid on...comp sits on the lid






PS im known as the "Ghetto God of Overkill" over on OC.com :shrug: