why do bigger pumps add heat?

lel4866 · 07-01-2003, 09:45 AM

I currently have a swiftech mp300 that I plan to use with Cathar's new cascade (along with a dtek chip cooler and swiftech gpu cooler) on an nf7-s/amd 1700 setup.

The pump is not in the case (separate fan/radiator/reservoir box).

I've been thinking about going to a bigger pump like the iwaki 15 or 20 in order to take better advantage of the waterblock, but posts here say most gain will be negated by heat from the pump. I want to do the best that is reasonable, but not go to extreme lengths for just 1C gain. Bottom line is I'm interested in the max overclock with standard watercooling, then min noise, not min watertemp.

Is this heat from the shaft, or are we talking heated air, or what?

Thanks.

Larry Lewis

bigben2k · 07-01-2003, 10:15 AM

Actually, none of the above

The heat mostly comes from the pumping action. As hard as it is to believe, the water actually heats up, as it's submitted to frictional forces, through the various restrictions within a loop.

The only thing you can do, is stick to flow speeds that are less than 3 fps (feet per second) within the tubing, so that the pumping action isn't wasted. 1/2" tubing usually does the trick.

The pump you have in mind is just fine.

Cathar · 07-01-2003, 06:12 PM

Actually a lot of the pump heat comes from electrical resistances in the motor itself as a result of the electro-magnetive effect that is turning the shaft impeller.

Pumps are anywhere from as little as 5%, to as high as 70% efficient in terms of input power being converted in motive force.

The balance of the input power is typically converted into heat within the pump's body, which can either be radiated out through the pump's surface, or absorbed into the water flowing through the pump, or some combination of both.

What Ben is referring to is also a valid effect, and its effect grows in proportion to the total back-pressure head of the system. However for pumps consuming <40W of power, the frictional effect is rarely large enough to have a major impact on temperatures, but for pumps consuming 100W+, the frictional effect can be quite substantial. Between 40-100W is a grey area where how effective it all is depends on how well your radiators are dissipating the extra pumping heat.

07-01-2003, 09:45 AM	#1
lel4866 Cooling Neophyte Join Date: Jun 2003 Location: Beaverton, OR, USA Posts: 10	why do bigger pumps add heat? I currently have a swiftech mp300 that I plan to use with Cathar's new cascade (along with a dtek chip cooler and swiftech gpu cooler) on an nf7-s/amd 1700 setup. The pump is not in the case (separate fan/radiator/reservoir box). I've been thinking about going to a bigger pump like the iwaki 15 or 20 in order to take better advantage of the waterblock, but posts here say most gain will be negated by heat from the pump. I want to do the best that is reasonable, but not go to extreme lengths for just 1C gain. Bottom line is I'm interested in the max overclock with standard watercooling, then min noise, not min watertemp. Is this heat from the shaft, or are we talking heated air, or what? Thanks. Larry Lewis

07-01-2003, 10:15 AM	#2
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Actually, none of the above The heat mostly comes from the pumping action. As hard as it is to believe, the water actually heats up, as it's submitted to frictional forces, through the various restrictions within a loop. The only thing you can do, is stick to flow speeds that are less than 3 fps (feet per second) within the tubing, so that the pumping action isn't wasted. 1/2" tubing usually does the trick. The pump you have in mind is just fine. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

07-01-2003, 06:12 PM	#3
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	Actually a lot of the pump heat comes from electrical resistances in the motor itself as a result of the electro-magnetive effect that is turning the shaft impeller. Pumps are anywhere from as little as 5%, to as high as 70% efficient in terms of input power being converted in motive force. The balance of the input power is typically converted into heat within the pump's body, which can either be radiated out through the pump's surface, or absorbed into the water flowing through the pump, or some combination of both. What Ben is referring to is also a valid effect, and its effect grows in proportion to the total back-pressure head of the system. However for pumps consuming <40W of power, the frictional effect is rarely large enough to have a major impact on temperatures, but for pumps consuming 100W+, the frictional effect can be quite substantial. Between 40-100W is a grey area where how effective it all is depends on how well your radiators are dissipating the extra pumping heat.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)