cooling speed

[Gemnic]

While reading the forum, this question hit me because I would like built myself a watercool system. I am gathering data so it's effective.
Is the speed of the water traveling in the wb affect the temp. exchange rate?
Ex. If you have 150 gallons of water going thru you wb, by making it going faster would you get your block cool faster or not (keeping same water temp coming in)?
Is there a chart for speed/exchange ?

Thanks

[Jester]

I have often wondered the same thing, but I often go different answers. My thoughts on it are this: There is a fine balance that needs to met. To fast and it won't have time to absorb the heat, to slow. To slow, and I am not quite sure what sort of problems you might encounter.

[Voodoo Rufus]

Actually the efficiency of a water block increases with diminishing returns as the flow rate gets higher because the block is getting closer and closer to the water temperature. The point at which the block starts to have decreased efficiency is around the order of 1000 PSI, where cavitation develops, causing air bubbles in the block. Not good.

Radiators are more efficient with lower water speeds so that it can have more time to cool the water that is inside the radiator. The faster it goes, the less the water gets cooled. So you need to reach an equilibrium where heat dissipation is maximized in the radiator while heat absorption is maximized in the water block.

[Hoot]

I've seen some math that suggests anything above about 130gph is a waste. The problem is that unless you lay out your setup carefully, you are not going to get the rated flow of your pump. I think that is why you read about guys running some alleged 600gph pump and swearing by it. I really wonder if they are achieving a true 600gph through their system. If you look at some of the major vendors of water cooling systems, they sell the Eheim 1048 pump. There is a reason they choose that pump, besides its high quality and quiet operation.

Hoot

[Voodoo Rufus]

All Eheim pumps are good. The 1048 should be considered the standard. The 1250 IMHO is good if you've got the money and then you can tweak the flow rate to find the sweet spot for temperatures in your system. :beer: :burn:

[Patchmaster]

I really wonder if they are achieving a true 600gph through their system. Those ratings are for zero lift and with tubing that's the same size as the pump fittings. Pushing the water through the tiny passages in most waterblocks, there's no way you're going to get the full rated capacity. I'd guess maybe a quarter to a half of the rated capacity, assuming no lift.

[Voodoo Rufus]

I bet with a Low Pressure Drop heater core or radiator, as well as a high-flow rate block like the MCW462, Spiral, or new DD Maze IIs you could easily see flow rates over half of the pump's speed with big size tubing.

[Kaese]

Basic Thermodynamics is the answer to all questions.

Increasing flowrate will increase heat transfer, immediately after the flow increase. But as the system equilibrates the temps will adjust to new levels It is hard to say without experimentation and without thermal probes at various points of the system. Then a bunch of calculations. But you could find the best flowrate for your system.

Another key point was brought up in an earlier post. Flow rates of the pumps in a system are vastly different from the advertised flow rates. An actual flow meter in you system would tell you what it is actually pumping and allow you to make educated decisions.

Measure twice, think about cutting and then measure again. Then cut.

Kaese

[stool]

Rise and length must be considered when trying to figure your actual flow. My external reservoir and 24pass radiator system combined with a 2 foot rise made for what has seemed a terrilby slow flow rate. I am now reconsidering either a new pump or completely redesigning the system

[Voodoo Rufus]

I think your radiator is holding you back trmendously. 24 passes is a huge length it its own.
:beer:

[deathstar13]

well i think i may be a example of the higher flow rate is best thinking. just a few facts ive noticed with mine.
350 danner mag
5/8"id barbed heater core
5/8" id hoses
1/2"id Z4 this thing does pump alot of water.how much? no flow moniter. id guess 200 gph is close. becouse im sure the 1/2" id of my block is slowing it down a tad

but thats what i wanted actually. was high flow rate , that would accelerate when it hits the smaller block.

hows it work? well i do the "1 one thousand" thing in my head and did a few guestiamtion tests.

27c@idle and id run a burn in for 30 mins. i figured this would get my h20 to a max temp as that i only have 1 quart in the whole system. it tops out at 39c now. thanks to higher oc

anyhow i get my system to the 39c for A few mins then take away the burn. and them measure how long it takes me to kool the water and get back to 27c
it goes from 39c-29c in 10 seconds. the last 2c take 15 seconds more to drop.

this high flow rate is what i set out to do when i first built it. my goal was to cool the h20 as fast as i could once it heated up. and a resivour woulda just killed that idea.

its all a matter of preferance. but if u want a quik turn around in water temps my setup is the best.

[Koooler King]

Originally posted by Gemnic
While reading the forum, this question hit me because I would like built myself a watercool system. I am gathering data so it's effective.
Is the speed of the water traveling in the wb affect the temp. exchange rate?
Ex. If you have 150 gallons of water going thru you wb, by making it going faster would you get your block cool faster or not (keeping same water temp coming in)?
Is there a chart for speed/exchange ?

Thanks

Yours is a very good and relavent question. It is the effectiveness of the heat exhanger surface that determines the rate of heat transfer and the falloff with increasing flow rates. A plain water block (e.g. just an open space) will quickly have diminshining returns as flow increases. To make it more effective, it needs more surface internally and further, a better surface form beyond that (e.g very fine fins). The best design for a water block would be a hybrid between an automotve oil cooler and a cold plate. This kind of stuff falls into the commercial / academic level of expertese, but can be attained if you are willing / able to experiment. I'm not sure where to suggest you start other than buy an auto type oil cooler and cut / weld to suit. After that, it is all trial and test. Good luck.


Koooler King

[BBigJ]

I'd like to expand on kaese's post. The rate of heat transfer is proportional to the gradient in the temperature. In other words, the faster the temperature changes, the fast heat will flow to balance things. When the water hits your wb, it is as cool as it will be, so the heat transfers quickly. Soon the water is warmer, so the gradient is reduced, and the heat transfer is less right before the water leaves the block. If you increase the flow rate, then the water has less time to absorb heat, so the temperature gradient is large from the beginning of the wb to the end. However, because the water hasn't been warmed much, there isn't much of a gradient at the radiator. In other words, increased flow increases the efficiency of the wb, but decreases the efficiency of the radiator. Only experimenting will determine the optimum flow rate for your system.