Low vs High Flow Radiators: Convince me!

T

Top Nurse

Supreme [H]ardness
Joined
Nov 4, 2003
Messages
7,346
Greetings most exalted ones...

I have sitting in front of me an Aqua-Computer Airplex Evo 240 and a typical 240 Hardware Labs radiator as well. If I blow through both of them then there is noticeably less pressure need to go through the HWL radiator. There is also considerably more radiating surfaces available in the Aqua Computer radiator to dissipate heat and it has round tubes instead of the HWL unit with flat channels. So which one will work the best in a typical low flow situation and which will work better in a typical high flow system. Will there be any differences in actual cooling capabilities if we switch rads to the opposite camp of thought?

Ready, Set, GO!
 
Well, why don't you try this and find out which one is the better performing rad =), lol. I would honestly try for you, buuuut, I don't have the time and or supplies to do it.

Oh. if you are testing for us, tell us your results, hahahaha.
 
as i keep telling one of my friends who is in philosophy: the theory is based on the results of experiment, the results of the experiments are not dependant on theory.

please test it out and let us know. if you can generate some good data for us, then we will have to revise the theory to agree with it.
 
Jeez... I thought I would wake up and find this thread littered with opinions, but it appears that when it comes down to the brass tacks no one really seems to know. Does this mean that all this low vs high religion going on around here is just hot air?

Well once I get my new system built with the A/C gear coming next week that will allow me to measure temps in the coolant and flow on a low flow system we will have some empirical data to go by. Now what I need to find is a flow sensor that I can use in a typical BIB cooling system. Does anyone know where I can get it from at a reasonable price? I can build in the coolant BIB temp sensors that utilize the same thermistor as the low flow types :)
 
I would also think that it would greatly depend on the Water Block and pump that you use. IE if you're building a low restriction system. I wouldn't think that low frow hardware would work all that well in a high flow setup and vice versa. Although I would also like to see some hard data.
 
The flat tube radiator will perform better do to the increased surface area of multiple small flat tubes instead of a few large round tubes. If it is easier to blow through the hwl radiator that is good news. That means the same volume of water will pass through the radiator at a higher preasure than the other. You also want low flow in a large radiator and high flow in a small water block. low flow =higher preasure in the radiator =higher water temp, this is what you want in a your radiator. high flow through smaller space =low pressure in the water block =lower water temp, this is what you want in your water blocks. Sounds odd that high flow = low preasure(at least it did to me at first) but it is true(phy 243). To prove this, hold two pieces of paper 3" apart and blow through the middle. The pieces of paper will come together do to the higher flow of air between them, causing a drop in preasure. The higher preasure outside forces the paper together. I thought my physics prof was full of shit when he was explaining this but after doing the problems and reading up on it's just the way it is.
 
Lord_Hadronous said:
The flat tube radiator will perform better do to the increased surface area of multiple small flat tubes instead of a few large round tubes. If it is easier to blow through the hwl radiator that is good news. That means the same volume of water will pass through the radiator at a higher preasure than the other. You also want low flow in a large radiator and high flow in a small water block. low flow =higher preasure in the radiator =higher water temp, this is what you want in a your radiator. high flow through smaller space =low pressure in the water block =lower water temp, this is what you want in your water blocks. Sounds odd that high flow = low preasure(at least it did to me at first) but it is true(phy 243). To prove this, hold two pieces of paper 3" apart and blow through the middle. The pieces of paper will come together do to the higher flow of air between them, causing a drop in preasure. The higher preasure outside forces the paper together. I thought my physics prof was full of shit when he was explaining this but after doing the problems and reading up on it's just the way it is.
Click to expand...


A water cooling setup is a closed circuit. The water will spend the same amount of time in the rad reguardless of this high pressure low pressure stuff.

Top Nurse, it all depends on how the rest of your system is configured. A TDX block loves flow, so one would want as little restriction as possible. On the other hand a Swifty 6000 block isn't so dependent on flow, and so could give decent performance in a more restrictive setup.
 
Top Nurse said:
Greetings most exalted ones...

I have sitting in front of me an Aqua-Computer Airplex Evo 240 and a typical 240 Hardware Labs radiator as well. If I blow through both of them then there is noticeably less pressure need to go through
Click to expand...


Just blow through the EVO harder and call it a day :eek: but really dont go all ac and not use an EVO or HTF-3 rad (made by watercool) worst case a 1048 will fit on the back of the aquabay if you want. Just use to two loops, put some plug and cools on a chevette core or get a evo 120 and put that on your GPU's.

1048 with a chevette core, cuplex evo, two aquadrives and 90 elbows on the res i get 0.5 at default 550 pulse on the aquaero. Aquastream, HTF-3 rad (360 type) two aquagraphx i get 0.2 at 550 but cools well so its up to you.
 
Aviddigi said:
Just blow through the EVO harder and call it a day :eek: but really dont go all ac and not use an EVO or HTF-3 rad (made by watercool) worst case a 1048 will fit on the back of the aquabay if you want. Just use to two loops, put some plug and cools on a chevette core or get a evo 120 and put that on your GPU's.

1048 with a chevette core, cuplex evo, two aquadrives and 90 elbows on the res i get 0.5 at default 550 pulse on the aquaero. Aquastream, HTF-3 rad (360 type) two aquagraphx i get 0.2 at 550 but cools well so its up to you.
Click to expand...

Well I was just looking where I could put that beast in there. Only two usable places. Either in the top or by cutting a slot in the main compartment floor and sitting it so that 1/2 is using the upper exit case fan and using the area above the PSU to exit the air from the case on the bottom of the radiator. The latter method appeals to me as it sounds more difficult and would have a rather unique look to it. :D

So what is your take on A-C's claim that an Aquastream will OC to the equal of a 1048?

I had also thought about using two loops on different setups. A 120mm HWL just fits between the fan and the front HD rack with no modifications, though you then can't use the dubious L-L fan screen that normally sits in front of it. Been considering using a small pump and that radiator to cool the three Raptors going in along with these Koolance HDD coolers. Then I would have the option to put in a h20 cooled PSU at a later date if I can find one that works as well as my PCP&C 510 SLI. ;)
 
As for the aquastream being able flow like a 1048, that would depend on your blocks and elbows etc. but it did not make a difference in my GPU loop. Having two loops though is probably the best idea if you think your case is big enough
 
plywood99 said:
A water cooling setup is a closed circuit. The water will spend the same amount of time in the rad reguardless of this high pressure low pressure stuff.

Top Nurse, it all depends on how the rest of your system is configured. A TDX block loves flow, so one would want as little restriction as possible. On the other hand a Swifty 6000 block isn't so dependent on flow, and so could give decent performance in a more restrictive setup.
Click to expand...

The human circulatory system is also a closed system but differences in pressure still occur. Top Nurse, there are some reviews at www.overclockers.com on radiators that you will find helpful.
 
Lord_Hadronous said:
The flat tube radiator will perform better do to the increased surface area of multiple small flat tubes instead of a few large round tubes. If it is easier to blow through the hwl radiator that is good news. That means the same volume of water will pass through the radiator at a higher preasure than the other. You also want low flow in a large radiator and high flow in a small water block. low flow =higher preasure in the radiator =higher water temp, this is what you want in a your radiator. high flow through smaller space =low pressure in the water block =lower water temp, this is what you want in your water blocks. Sounds odd that high flow = low preasure(at least it did to me at first) but it is true(phy 243). To prove this, hold two pieces of paper 3" apart and blow through the middle. The pieces of paper will come together do to the higher flow of air between them, causing a drop in preasure. The higher preasure outside forces the paper together. I thought my physics prof was full of shit when he was explaining this but after doing the problems and reading up on it's just the way it is.
Click to expand...


Temperature is not affected by pressure in any measurable way, we are dealing with an uncompressible fluid which doesn't follow the PV=nRT law. Also, in a watercooling loop, pressure is not a function of flowrate- flowrate is a function of pressure. The aerodynamics analogy isn't really relevant. The higher flowrate through the smaller orifice in the waterblock is due to a change in cross-sectional area while pressure remains fairly constant, assuming the block is much more restrictive than the radiator and those are the only two components in the loop. Lower the pressure applied by the pump and you will see a decrease in flow.
 
zer0signal667, you are correct about the uncompressible fluid. And here I thought I could apply what I learned. Too eager I guess.
 
Lord_Hadronous said:
zer0signal667, you are correct about the uncompressible fluid. And here I thought I could apply what I learned. Too eager I guess.
Click to expand...

Don't be discouraged, it was a novel attempt :D There are plenty of other principles of thermodynamics and heat transfer that apply, so keep learning.
Surface area plays a role, but it cannot be effective if fluid flow characteristics suck. This depends on design of heat transfer devices and pressure applied by the pump. For example, small passages through a radiator will create a large pressure drop and lower flow, but will create more turbulence (good for heat transfer) within the channels than a radiator with channels of 10x larger diameter. A pump with lower pressure will essentially be the same situation. Either a reduction in pressure or an decrease in restriction will lower the efficiency of the system.
Anyways, Top Nurse, there are too many factors to be able to predict which will work better in your specific application, just try them both. And of course, let us know how it goes.
 
I was just curious if anyone knew if there are and radiators for the PC that are tube in fin. Like a good intercooler on a car. where it's just like most of the flat tubed radiators, but it also has fins inside of the radiator tubes thenselves which greatly infrease the radiator/intercoolers efficency.
 
You must log in or register to reply here.
Back
Top