If you run your water pump in a loop...

amdfreak04 · Jan 15, 2007

does your water temp stay the same, or does the water temp go up, this is of course with no load, just the pump circulating the water in a loop....

Reason i ask is i have a submergable pump that i modded a while back and i put cooling on the pump heat sinks and fans to cool the water .. i start of with room temp water.. then i check 15 mins later .. and the water is at 30 degress... with no load on it ...

is this pump crap or is it the same senario with every high end fast flowing water pump .. this pump is 19w/120v./.3 amp/ 190gph flow .. if my memory serves me right ..

please let me know your thoughts

this is what my pump looks like

thanks
Steve

Tyreal · Jan 15, 2007

Every pump will dump a certain amount of heat into your loop. That is why it is ideal to run the radiator after the pump so that the heat is removed.

Omega Destroyer · Jan 15, 2007

Yes it's true, the pumps themsevles generate heat which they transfer into the water.

amdfreak04 · Jan 16, 2007

...

are you guys saying that even any high end pump out there that you can buy for a w/c set up 1/2" will still dispense heat into the water?... Does that mean my pump is good for a w/c setup

Linux_Box · Jan 16, 2007

Yes, all of them will introduce some heat... how much is determined by the rating (watts) of the pump.

For your pump, it seems ok but what's the head on it? Head is the pressure it can exert to push the water through the system. Head is often more important than flow rate.

I've read (in quite a few places) that it's about 50-75% of the rated capacity that is introduced into the cooling system. When you multiply this amount by the thermal resistance, you can get an idea of much you are raising the temp from just the pump. It seems to be less than 1-2 degrees Celsius based on the math I've seen for most situations. I'll try to find a link or two.

ares350 · Jan 16, 2007

going off my aquarium knowledge, that being a mag 3 it looks like? mag something I know.

anyway, those are water cooled, IE water is in the area where there is friction, and near the motor. now I just picked up the swifttech whatever pump, it LOOKS like a large scale pump, IE electric motor attached to a pump device. that SHOULD add less heat, as the motor heat is seperate from the water. of course the pump itself does have some friction, as well as the friction with the water. but it should be less so than submersible mag driven impellor pumps.

Im assuming based on swifttech though, it LOOKED like it was a seperate pump to the motor, could be mistaken.

FYI seperate pumps are always external, as the motor section is air cooled and not sealed, as opposed to mags which can be run either way.

ares350 · Jan 16, 2007

btw, for my aquarium, my mag 9 can heat 40 gallons of water to 90 degrees if I put the top on the garbage can its mixing.

QuietIce · Jan 16, 2007

ares350 said:
going off my aquarium knowledge, that being a mag 3 it looks like? mag something I know.

anyway, those are water cooled, IE water is in the area where there is friction, and near the motor. now I just picked up the swifttech whatever pump, it LOOKS like a large scale pump, IE electric motor attached to a pump device. that SHOULD add less heat, as the motor heat is seperate from the water. of course the pump itself does have some friction, as well as the friction with the water. but it should be less so than submersible mag driven impellor pumps.

Im assuming based on swifttech though, it LOOKED like it was a seperate pump to the motor, could be mistaken.

FYI seperate pumps are always external, as the motor section is air cooled and not sealed, as opposed to mags which can be run either way.

Indeed, some people even add RAMsinks to the back side of the MCP655 to help cool the motor and keep heat out of the water ...

aaronjb · Jan 16, 2007

Tyreal said:
Every pump will dump a certain amount of heat into your loop. That is why it is ideal to run the radiator after the pump so that the heat is removed.

The heat may "be removed" by the radiator, but don't forget that in a cycle, the radiator is removing heat from both the pump and the block(s) - no matter where the heat producers (block and pump) are located in the loop. So if the volume of water were to travel through your loop only once, then having the radiator directly after the pump would indeed be ideal, and may produce a meaningful difference in the effectiveness of the cooling system. But since the water cycles in the loop, there is no meaningful difference. You can try to cheat this effect by placing the radiator directly after the pump, but the radiator is always removing the heat generated by the same two sources - the pump, and the block(s). The cooling loop is a closed system and the water (or other cooling fluid) will reach near equilibrium in temperature.

Yes, there is a measurable difference in water temperature at any two points in the same loop (less than 1C, from what I understand). However, the difference in cooling capacity is negligible, and the difference in CPU temp will be as well. I should also add that putting the radiator directly after the pump may be a poor choice for a block which relies on jet impingement. The drop in pressure may hinder the ability of the block to operate effectively.

Seven · Jan 16, 2007

A lot of pumps now a days have motors that run cooler and dump basically no heat into the loop, the DDC+ is based on this design AFAIK. Cathar mentioned something about it but I'm not sure. I just know it's really not a big deal in any normal pumps we'd be using.

Old Iwakis though...well those were the days.
7

amdfreak04 · Jan 17, 2007

Linux_Box said:
Yes, all of them will introduce some heat... how much is determined by the rating (watts) of the pump.

For your pump, it seems ok but what's the head on it? Head is the pressure it can exert to push the water through the system. Head is often more important than flow rate.

I've read (in quite a few places) that it's about 50-75% of the rated capacity that is introduced into the cooling system. When you multiply this amount by the thermal resistance, you can get an idea of much you are raising the temp from just the pump. It seems to be less than 1-2 degrees Celsius based on the math I've seen for most situations. I'll try to find a link or two.

i couldn't tell you what the head is on it .. but its a powerfull pump... it can shoot water into the air virtically about 1 an 1/2 to 2 feet virtically.. I think that pleanty of head pressure.

the specs on the outside of the pump at 120v/19w/.3amp/190gph pump

-steve

ares350 · Jan 17, 2007

aaronjb said:
The heat may "be removed" by the radiator, but don't forget that in a cycle, the radiator is removing heat from both the pump and the block(s) - no matter where the heat producers (block and pump) are located in the loop. So if the volume of water were to travel through your loop only once, then having the radiator directly after the pump would indeed be ideal, and may produce a meaningful difference in the effectiveness of the cooling system. But since the water cycles in the loop, there is no meaningful difference. You can try to cheat this effect by placing the radiator directly after the pump, but the radiator is always removing the heat generated by the same two sources - the pump, and the block(s). The cooling loop is a closed system and the water (or other cooling fluid) will reach near equilibrium in temperature.

Yes, there is a measurable difference in water temperature at any two points in the same loop (less than 1C, from what I understand). However, the difference in cooling capacity is negligible, and the difference in CPU temp will be as well. I should also add that putting the radiator directly after the pump may be a poor choice for a block which relies on jet impingement. The drop in pressure may hinder the ability of the block to operate effectively.

you want to maximize the difference in temperature though. the hotter the water in the radiator, the more heat it will dissipate, the colder the water in the water block, the more heat it will absorb. thus, pump-radiator-block makes the most sense.

SolidxSnake · Jan 17, 2007

amdfreak04 said:
i couldn't tell you what the head is on it .. but its a powerfull pump... it can shoot water into the air virtically about 1 an 1/2 to 2 feet virtically.. I think that pleanty of head pressure.

the specs on the outside of the pump at 120v/19w/.3amp/190gph pump

-steve

I think that's 2 feet head... isn't it?

The Laing DDC+ is rated at 20ft IIRC.

aaronjb · Jan 17, 2007

ares350 said:
you want to maximize the difference in temperature though. the hotter the water in the radiator, the more heat it will dissipate, the colder the water in the water block, the more heat it will absorb. thus, pump-radiator-block makes the most sense.

Makes the most sense if the routing of your tubing and placement of your parts dictates it. At most, we're talking about a 1 degree Celsius difference between water temperature at any two points in the same loop. The possible increased cooling capacity of this "colder" water is generally outweighed by tube routing and placement of the pump, reservoir, etc. And I'd always prefer pump -> block when using a jet impingement block to maximize pressure on the nozzles.

We're talking about a closed loop - the radiator always removes heat from the same sources, no matter the order of those sources in the loop. The water isn't going to be "hotter" in the radiator if it is placed at a different point in the loop. I understand what you're getting at: The most heat possible should be removed from the water before it hits the block(s). But in practice, the difference really is negligible.

This is one of those arguments that goes on ad infinitum... many, many posts on this subject can be found on the forums..

orion25 · Jan 18, 2007

ares350 said:
btw, for my aquarium, my mag 9 can heat 40 gallons of water to 90 degrees if I put the top on the garbage can its mixing.

Keep in mind that the MAG9 can also push 900G/h. Plus with the top on the can you are not allowing the heat to be removed from the saltwater. If you remove the lid and have a little surface aggitation it will not get anywhere near that hot.

**Sorry about that, you guys can get back on topic now**

ares350 · Jan 18, 2007

SolidxSnake said:
I think that's 2 feet head... isn't it?

The Laing DDC+ is rated at 20ft IIRC.

not really, if you put a tube on there, it would go far far higher and spout out the top still. without the form of the tube its not really back pressure, its just wide open flinging water in the air.

and yeah, I know Im not letting the heat escape, thats the point, simply demonstrating how much heat is going in.

If you run your water pump in a loop...

amdfreak04

Member

Tyreal

Member

Omega Destroyer

Member

amdfreak04

Member

Linux_Box

Member

ares350

Member

ares350

Member

QuietIce

Disabled

aaronjb

Member

Seven

Member

amdfreak04

Member

ares350

Member

SolidxSnake

Member

aaronjb

Member

orion25

Folding in memory of my dads, Tim B. & Mike B.

ares350

Member

Similar threads