Thermodynics Discussion

joshuass · 02-06-2006, 03:59 PM

Hey guys. I really enjoy the forums here at procooling. So much good info is to be had here. My group and I are researching liquid cooling for senior design. This means we are taking a more structured approach to designing blocks and the like. To start, we have to look at a resistance model for the block and use fluid and heat transfer equations to help facilitate the design of our blocks. I guess my question is if any one here knows of good information sources on getting equations for calculating heat conductivity for home made blocks?

Brians256 · 02-06-2006, 05:14 PM

The best modelling I've seen here isn't a simple equation. Well, it is, but it's repeated a lot. Think FEA. Things simple enough to break down into a one dimensional model can be turned into a single equation, but that will not work for a waterblock. TIM analysis, for example, can be modelled in 1D, but the block itself is ... more complicated than being married to ten wives.

If you use some expensive CFD and FEA programs, you'll get something that might be representative of real life. It might not be really accurate, but it will be closer than most of the armchair experts (if your model is accurate).

Edited for ... making sense.

gmat · 02-07-2006, 02:44 AM

Yes the best way to go is finite elements analysis.
If you can find the work done by Zytrahus (his site has disappeared, but Roscal on these forums can help) he's taken this approach exactly when making his prototype block. But remember that FEA software can be tough to get, especially fluid dynamics software...

Also if you're to make a research paper i'd suggest proofreading everything, to avoid for example typos in the word "thermodynamics" :P

bobo5195 · 02-07-2006, 03:24 PM

what level of research paper?

On a padantic point (as this is procooling). CFD is based on finite difference methods (what goes into a cell must come out) whille FEA is somthing completely different. It is surprising how little overlap between the two there actually is.

stev · 02-07-2006, 04:14 PM

One good CFD package is Fluent.

http://www.fluent.com

This is a model based on cooling with air. Just plugging in the density and other numbers for water would generate a liquid model of the same thing basically.
http://www.fluent.com/solutions/examples/x247.htm

Here's one for liquid fluid flow, or again air if need be.
http://www.fluent.com/solutions/examples/x249.htm

Check to see if your technical school can get a student version of this software.
If you can, I would then highly recommend SolidWorks 2006 student version as well since the cost could still be FREE with submitted proof of being a student.

Stev

02-06-2006, 03:59 PM	#1
joshuass Cooling Neophyte Join Date: Jan 2006 Location: Mi Posts: 4	Thermodynics Discussion Hey guys. I really enjoy the forums here at procooling. So much good info is to be had here. My group and I are researching liquid cooling for senior design. This means we are taking a more structured approach to designing blocks and the like. To start, we have to look at a resistance model for the block and use fluid and heat transfer equations to help facilitate the design of our blocks. I guess my question is if any one here knows of good information sources on getting equations for calculating heat conductivity for home made blocks?

02-06-2006, 05:14 PM	#2
Brians256 Pro/Staff Join Date: Oct 2001 Location: Klamath Falls, OR Posts: 1,439	Re: Thermodynics Discussion The best modelling I've seen here isn't a simple equation. Well, it is, but it's repeated a lot. Think FEA. Things simple enough to break down into a one dimensional model can be turned into a single equation, but that will not work for a waterblock. TIM analysis, for example, can be modelled in 1D, but the block itself is ... more complicated than being married to ten wives. If you use some expensive CFD and FEA programs, you'll get something that might be representative of real life. It might not be really accurate, but it will be closer than most of the armchair experts (if your model is accurate). Edited for ... making sense.

02-07-2006, 02:44 AM	#3
gmat Thermophile Join Date: Mar 2001 Location: France Posts: 1,221	Re: Thermodynics Discussion Yes the best way to go is finite elements analysis. If you can find the work done by Zytrahus (his site has disappeared, but Roscal on these forums can help) he's taken this approach exactly when making his prototype block. But remember that FEA software can be tough to get, especially fluid dynamics software... Also if you're to make a research paper i'd suggest proofreading everything, to avoid for example typos in the word "thermodynamics" :P

02-07-2006, 03:24 PM	#4
bobo5195 Cooling Savant Join Date: Aug 2005 Location: uk Posts: 400	Re: Thermodynics Discussion what level of research paper? On a padantic point (as this is procooling). CFD is based on finite difference methods (what goes into a cell must come out) whille FEA is somthing completely different. It is surprising how little overlap between the two there actually is.

02-07-2006, 04:14 PM	#5
stev Cooling Neophyte Join Date: Feb 2005 Location: niagara falls Posts: 96	Re: Thermodynics Discussion One good CFD package is Fluent. http://www.fluent.com This is a model based on cooling with air. Just plugging in the density and other numbers for water would generate a liquid model of the same thing basically. http://www.fluent.com/solutions/examples/x247.htm Here's one for liquid fluid flow, or again air if need be. http://www.fluent.com/solutions/examples/x249.htm Check to see if your technical school can get a student version of this software. If you can, I would then highly recommend SolidWorks 2006 student version as well since the cost could still be FREE with submitted proof of being a student. Stev Last edited by stev; 07-18-2006 at 03:44 PM.

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