• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Watercooling: Does Waterflow speed matter?

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
cyclonWizard,

I agree that the Nusselt number will increase as the flow velocity increases in normal flows... which will effect the heat transfer coefficent. My point is that in the reynold calculation that you have shown is overly simplified and does not simulate the actual effects of the jet turbulence.

The flow is laminar exiting the nipple of the waterblock... I think we can both agree with. However, the nipple acts as a trip, increasing the turbulence of the water flow. The boundary layer seperates and the jet of water impacts directly upon the waterblock.

In an impinging jet flow, most turbulation models do not accuratly predict the mean velocity and reynold stress preditions in the near-wall region of the flow. At distances well removed from the point of impingement the Nusselt number predictions are accurate... however due to failure to show acceleration of the flow along the impingement plate, the heat transfer predictions are poor. This is due to the assumptions in 2 equation turbulence models. The jet flow, after exiting the nipple, undergoes streamline curvature and is subjected to strong deceleration as the fluid nears the impingment plate. After the jet impinges on the plate, the fluid accelerates radially, forming a 3 dimensional flow field. Turbulent vortices generated in the jet shear layer will be streatched parallel to the plate after impingement, altering the turbulent length scale of the flow. 2 equation turbulence models assume a linear stress-strain relation and a length scale based on thin shear layer approximations. Therefore, these equations can be expected to perform poorly near the point of impingment where these assumptions are not valid. At locations well removed from the point of impingement, the flow approximates that of a 2D boundary layer and the 2 equation models of turbulence perform reasonable well. [copied without permission from the CFD society of Canada] http://www.cfdsc.ca/english/benchmarks/cfd95/node1.html

I believe, based on the above the although velocity will or can have some effect on the turbulation of the flow and thus the heat transfer... the equations that you are showing are incorrect... and in fact that the effects of a velocity increase are negligible when compared to the rapid decelleration and turbulation caused by the jet hitting the waterblock.

And yes... I am an engineer... even if my spelling sucks... they need a spell checker in this thing!!
 
Originally posted by: kleinwl
And yes... I am an engineer... even if my spelling sucks... they need a spell checker in this thing!!
Click to expand...
First, turbulation is not a word: turbulence. Second, dynameter isn't a word. It doesn't even sound like diameter. You may very well be an engineer, but your contributions don't indicate that your expertise is in the area of discussion. I can readily solve 2-D boundary layer problems analytically. However, that doesn't mean that they're relevant to the discussion at hand. The long and short of this discussion was well summed up by Mday:
No, there would not be a point where the speed of the water would cause any more heat than cooling. However, there is a point where the speed of the water would make the cooling process less efficient. Furthermore, there is also a speed of the water which would cause leaks due to the forces placed at the joints.
Click to expand...
This is all that needs to be said. For modeling efforts, you can make them as overcomplicated as you'd like and achieve an extra 0.01% accuracy - feel free to do so. Maybe you can even get a publication out of it or something. For myself, I prefer models that give accurate, analytical solutions so I can see exactly which parameters are useful to vary to optimize the system. Feel free to use CFD, but there is no guarantee your results are any more accurate than my own. Based on the assumptions that you put forth here, I very much doubt that your results would be as accurate as mine.
 
turbulence:
noun: unstable flow of a liquid or gas
noun: instability in the atmosphere
noun: a state of violent disturbance and disorder (as in politics or social conditions generally)

dynameter = diameter... give me a break!

I think this discussion is done. I explained the practical way of optimizing a system... but you were more interested in showing off with textbook formula. I said, if you really want to technical then you need to treat the system as the 3D system it actually is. But, no... that's too much. so do whatever you want.
 
Originally posted by: kleinwl
turbulence:
noun: unstable flow of a liquid or gas
noun: instability in the atmosphere
noun: a state of violent disturbance and disorder (as in politics or social conditions generally)

dynameter = diameter... give me a break!

I think this discussion is done. I explained the practical way of optimizing a system... but you were more interested in showing off with textbook formula. I said, if you really want to technical then you need to treat the system as the 3D system it actually is. But, no... that's too much. so do whatever you want.
Click to expand...
Turbulence != turbulation. I never said turbulence is not a word.

The method you described is NOT practical for anyone in this forum except probably you and myself. The method I described, while not exact (neither is yours), can be accomplished without the use of software that costs thousands of dollars. I'd also be willing to bet that the results obtained using my methodology would be within 1% of the experimental results. But, even without all my handwaving arguments, I can say with absolute certainty that the trends I stated will be 100% correct: increasing the fluid flowrate WILL increase heat transfer. At some point, it's not worth increasing it any further. The point at which the tradeoff is no longer beneficial is in the eye of the beholder, not in a CFD solution.
 
Originally posted by: CycloWizard
Originally posted by: check
if we are assuming the water is going to be a X temp coming out of the radiator no matter what, then yeah I agree the faster the better.

But the faster you pump the water the faster is is going through the radiator, therefore the warmer the water will be when it gets to your waterblock on your cpu...

it's begining to sound like a calculus max/min problem to me >_<

Maybe my thinking is ass backwards though...
Click to expand...
Actually, the faster you move the water through the radiator, the more it will cool down. Generally speaking, the faster a fluid flows, the more quickly it will equilibrate with its surroundings. This is because as the particles/molecules move past each other at higher and higher rates (corresponding to higher and higher velocities), they have greater numbers of interactions. These interactions cause the energy to be transferred from one particle to another, so the more interactions you have, the greater uniformity of temperature you will have.
Click to expand...


Uhh no. The faster you move water through a radiator the less heat transfer will occur. If sufficent heat transfer does not happen at the radiator the whole system heats up. There has to be a speed that accomidates the size of the cooling system with the amount of fluid present, and the physical size of the radiator.
 
Originally posted by: nailzer
Uhh no. The faster you move water through a radiator the less heat transfer will occur. If sufficent heat transfer does not happen at the radiator the whole system heats up. There has to be a speed that accomidates the size of the cooling system with the amount of fluid present, and the physical size of the radiator.
Click to expand...
Uhh, no. While what you're saying might make sense in a batch system where there is not continuous looping flow, this is not the situation at hand in a watercooling system. The thermal boundary layer within the fluid will decrease as the velocity increases, thereby increasing overall heat transfer. This is day one of transport phenomena - basic film theory. If you disagree, I can't say anything but that you're wrong.
 
WOW! Great! So I can go out to my car, take the radiator fan off, as long as I keep the RPMs up, and the water and antifreeze is moving really really fast, I'll never have to worry about it over heating. Great way to get a little more horsepower too! But if it overheats and blows up I'll send you the bill, OK?
 
Originally posted by: nailzer
WOW! Great! So I can go out to my car, take the radiator fan off, as long as I keep the RPMs up, and the water and antifreeze is moving really really fast, I'll never have to worry about it over heating. Great way to get a little more horsepower too! But if it overheats and blows up I'll send you the bill, OK?
Click to expand...
In almost no auto design does the radiator fan lessen the produced horsepower by any significant margin, even in the instance of a belt-driven fan. In my car, it's electric, so nothing at all would be gained. If you knew much about many cars, you'd realize that the radiator fans on non-belt-driven models doesn't run most of the time - it's thermally controlled to only turn on when the natural convection caused by the car's own velocity is insufficient to remove all of the heat generated by the engine.

You are confounding several different heat transfer resistances and trying to keep them all under the same umbrella. Unfortunately, the fan does not aid the heat transfer from the coolant mixture to the metal. It simply lessens the resistance to heat transfer at the surface of the radiator by, again, decreasing the thermal boundary layer thickness. The fact that you recognize the importance of a fan in creating high velocities for thermal transfer but can't see the mirror image for the hot side of the heat exchanger is clearly demonstrative of your lack of understanding of the situation.

However, since I'm sick and tired of being talked down to by high schoolers, I am going overboard to demonstrate just how right I am. Yes, my modesty is lacking, but ignorant impudence can only be overcome with education. 😀

http://it.che.wustl.edu/~reillyma/watercooling/velocity-heatflux.htm
 
Umm yes there is horsepower loss, from the fan belt and fan, clutch or not, and if you take off the mechinical water pump and replace it with an electric water pump you would see and even greater gain.
I get tired of being talked down to by someone that thinks their technobabble is in any way impressive, and actually has no clue what they are talking about.
 
Originally posted by: nailzer
Umm yes there is horsepower loss, from the fan belt and fan, clutch or not, and if you take off the mechinical water pump and replace it with an electric water pump you would see and even greater gain.
I get tired of being talked down to by someone that thinks their technobabble is in any way impressive, and actually has no clue what they are talking about.
Click to expand...
Um, I answered your jackassery in my first paragraph of my previous post:
In almost no auto design does the radiator fan lessen the produced horsepower by any significant margin, even in the instance of a belt-driven fan. In my car, it's electric, so nothing at all would be gained. If you knew much about many cars, you'd realize that the radiator fans on non-belt-driven models doesn't run most of the time - it's thermally controlled to only turn on when the natural convection caused by the car's own velocity is insufficient to remove all of the heat generated by the engine.
Click to expand...
Simply because you havea belt-driven fan does not mean all cars have them. Simply because it's belt-driven does not mean it significantly impacts horsepower. If you knew much about fan performance, you would know that at most it's putting out 1-2 hp of work, so it might draw up to 3-4 hp. Even on my four-banger, that's negligible.
 
Originally posted by: nailzer
Umm yes there is horsepower loss, from the fan belt and fan, clutch or not, and if you take off the mechinical water pump and replace it with an electric water pump you would see and even greater gain.
I get tired of being talked down to by someone that thinks their technobabble is in any way impressive, and actually has no clue what they are talking about.
Click to expand...
Because you don't understand it, that somehow means that he's wrong, right?
 
q]Originally posted by: CycloWizardIn almost no auto design does the radiator fan lessen the produced horsepower by any significant margin, even in the instance of a belt-driven fan. [/L][/quote]

Then you should look up "Flex Fan"

"Flex Fans pull additional air through the radiator at low RPMs when the speed of the vehicle is not enough to cool the engine's coolant system. At higher RPMs, when the vehicle's speed forces ample air through the radiator, the fan blades flatten out, resulting in less power being used to spin the fan. Engine performance and gas mileage are thus improved."

http://www.perma-cool.com/Catalog/Cat_page20.html
 
Originally posted by: nailzer
Then you should look up "Flex Fan"

"Flex Fans pull additional air through the radiator at low RPMs when the speed of the vehicle is not enough to cool the engine's coolant system. At higher RPMs, when the vehicle's speed forces ample air through the radiator, the fan blades flatten out, resulting in less power being used to spin the fan. Engine performance and gas mileage are thus improved."

http://www.perma-cool.com/Catalog/Cat_page20.html
Click to expand...
And you actually believe that this will save you more money in gas than the price of the fan? If so, I have a bridge to sell you...
 
Originally posted by: nailzer
q]Originally posted by: CycloWizardIn almost no auto design does the radiator fan lessen the produced horsepower by any significant margin, even in the instance of a belt-driven fan. [/L]
Click to expand...

Then you should look up "Flex Fan"

"Flex Fans pull additional air through the radiator at low RPMs when the speed of the vehicle is not enough to cool the engine's coolant system. At higher RPMs, when the vehicle's speed forces ample air through the radiator, the fan blades flatten out, resulting in less power being used to spin the fan. Engine performance and gas mileage are thus improved."

http://www.perma-cool.com/Catalog/Cat_page20.html
[/quote]



http://www.edmunds.com/reviews/list/top10/103164/article.html
http://www.edmunds.com/ownership/maintenance/articles/105528/article.html
http://www.consumerreports.org/main/det...d=113261&ASSORTMENT%3C%3East_id=333137

I also read a study done by the EPA which stated that there is currently no gasmilage devices that actually work. (can't find it right now though)

With this mentality wouldn't You think that for say $5-10 any car manufacturer would put that fan in just to show a higher gas milage rating for their car? Damn bro with gas prices as high as they are that would be a SURE selling point for the insignificant cost.

If You read these You'll notice that the increased drag on Your car from having the windows rolled down will actually give You worse gasmilage than using the AC as stated in the above article, so I wouldn't go pulling off My AC belt to save that milage either lol

My 2 cents
 
Originally posted by: nailzer
Umm yes there is horsepower loss, from the fan belt and fan, clutch or not, and if you take off the mechinical water pump and replace it with an electric water pump you would see and even greater gain.
I get tired of being talked down to by someone that thinks their technobabble is in any way impressive, and actually has no clue what they are talking about.
Click to expand...

Yes you do lose power from the belts that power the water pump. Of course you will lose power from the belts that drive the alternator.
 
I do alot off racing my son drives I build the engines . They produce right around 600Hp. So you can imagine the heat . We use a High volum pump because a stock pump won't cool it. Than we have to restrict it so as to get the right flow otherwise the water goes threw the rad. to fast and won't cool. Get the point. Zinn2b
 
Originally posted by: Intelia
I do alot off racing my son drives I build the engines . They produce right around 600Hp. So you can imagine the heat . We use a High volum pump because a stock pump won't cool it. Than we have to restrict it so as to get the right flow otherwise the water goes threw the rad. to fast and won't cool. Get the point. Zinn2b
Click to expand...
:roll:

If the point is that anecdotal 'evidence' trumps actual science in this thread, then I got the point a while back.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top