CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   heat transfer on dimpled plate (http://www.cfd-online.com/Forums/fluent/43497-heat-transfer-dimpled-plate.html)

mech January 26, 2007 12:37

heat transfer on dimpled plate
 
1. Abstact : heat transfer analysis on dimpled sufaces

(heat sink) 2. Experiment 1) sample : copper plate and dimpled copper plate 2) experiment condition

: I made an air flow channel.

Like heat sink, I put the plate inside of a channel

Re3000,

constant heat flux (heat applied on top of a plate such as heat sink fin) 3) Results : the temperature of dimpled plate is 0.5degree less than flat plate

( I belive that this is reasonable because the surface area is increased by dimples)

3 Fluent ananlysis

1) Solver: Segregate, implicit, steady, 3D Double Precision 2) Viscous: Laminar 3) 2nd order (pressure, momentum, energy), SIMPLEC 4) Boundary Condition a) Constant Heat flux b) Mass flow inlet c) Pressure Outlet ( I fixed Target mass flow)

5) Results : the temperature of flat plate is 0.6degree less than the temperature of dimpled plate.

At first, I thought that the number of mesh is the problem. So, I checked the results as I increase the number of mesh.

For flat : mesh number ( 1st 567000, 2nd 900000, 3rd 1620000) For dimpled plate : mesh number ( 1st 1219488, 2nd 1995296, 3rd 2607776)

But, the results are almost samle. Flat plate is cooler than dimpled plate.

The flow is laminar flow. So, I expect that it will be simple to get numerical date . But it is not.

I iterated the model a lot so that the residual and temperature reach steady condition.

For Re1500, the result is also same.

Do you have any opinion about this? Please help me. Thank you.


Frank January 31, 2007 09:09

Re: heat transfer on dimpled plate
 
If you keep your heat flux (W2/m) constant, you increase the integral power (W) for your dimpled plate, don't you?

Frank

mech January 31, 2007 20:19

Re: heat transfer on dimpled plate
 
Dear Frank

Thank you for your response.

This test is for heat sink. If you see heat sink for CPU, heat is applied to the bottom of the plate. (Not dimpled plate).

Heat flux is constant for flat plate and dimpled plate. Also, total amount of heat is also same.

Thank you

Frank February 6, 2007 07:55

Re: heat transfer on dimpled plate
 
Dear mech,

I set up a 600.000 cell case with 0.3mm deep dimples. As expected in simulation the dimpled plate is cooler than the flat plate.

Either re-check your case or give some more information about boundary-condition etc.

Best regards Frank

mech February 6, 2007 15:15

Re: heat transfer on dimpled plate
 
Dear Frank

Thank you for your concern.

My conditions are Heat flux (Input 11516.5w/m2) Inlet Temp(294.58K) Mass flow rate (0.001842885 kg/s) Channel size ( 33mm x 103.51mm)

You can check Other informations in my first question.

What kinds of scheme do you use for discretization?

And how much temperature difference between flat and dimple?

I also have many questions. If you don't mind, could you tell me your email address?

Thak you very much



All times are GMT -4. The time now is 17:14.