A natural convection questionà 1

[Tobalcane]

I’m doing some natural convection heat transfer analysis on a plate. I know how to find the “Q” due to natural convection, but I’m only interested in a part of the plate. For question sake, if the plate is 5 inches by 5 inches in the vertical orientation (the whole plate is exposed to air) and I am only interested in a 1 inch by 1 inch part of the plate. The air is hotter than the plate. Do I calculate “Q” for the whole plate (5”x5”) or just the part I’m interested in (1”x1”)? If I calculate 1”x1”, the “Q” is much bigger than “Q” calculated for 5”x5”. In my opinion, I should calculate the whole plate because the whole plate is what “engages” the air, so I have to take it as a whole system.

Thanks in advance for your insight.






Go Mechanical Engineering
Tobalcane

 

[sailoday28]

You must specify much more info to set up conditions to solve the problem.

If this is a steady state problem, then one must know where the heat is flowing to after contact with the plate.

For example, if the plate is maintained in an isothermal condition, then the thermal resistance must be calculated as a function of position on the plate. With this info, the heat flux is then calculated.

 

[IRstuff]

The answer is that you must use the entire plate to do the analysis. But, if you need to know the temperature at a specific spot on the plate, you'll need to crank through some level of discrete or distributed analysis, since the heat flow is a function of the local temperature differences.

The available commercial thermal analysis programs should be able to do this problem. There are a couple that have live demo versions with limited capability that should also be capable of doing this.

TTFN

 

[Tobalcane]

Thank you sailoday28 for responding.

To answer your question, the plate is fasten (on the edge)to a heat source and exposed to hot air. This is more of a transient question. The temperature of the plate will be less than the heat source and air at beginning. I’m trying to find out how much heat will be induced into the 1”x1” lets say that it is in the middle of the plate. I’ve calculated the conduction Q from the heat source to 1”x1”. I want to add what the air is doing too.


Thank you IRstuff or responding.

I agree with you. I was just not sure if this was practiced this way. I’m just doing some ball park calcs now, but later (when I get better in using TAS) I’ll do some FEA.


Go Mechanical Engineering
Tobalcane

 

[IRstuff]

One of my heat transfer books would model it as a thermal resistance network, with additional resistances for the convection. Then iterate until convergence.

Essentially, this is what TAS is doing, but at a much finer granularity.

TTFN

 

[sailoday28]

Tobalcane (Mechanical)WRITES
To answer your question, the plate is fasten (on the edge)to a heat source and exposed to hot air. This is more of a transient question

If the plate is relatively thin, then you could model it as a fin. The major problem is determination of the convective and radiative heat transfer coefficients, which will vary with location and time.
In a finite difference scheme, the coefficents could be modeled for each node.

A simpler method would be to assume Hr +Hc, thermal conducitivity and specific heat as constant. And then solve the transient fin equation which then is a closed form solution and is probably available in a good heat transfer text.

 

[Tobalcane]

Sailoday28

Yup that is what I had in mind to do. Do you concur with IRstuff that the Hc and Hr should be calculated for the whole plate versus the localized area that I am interested?

Thanks Sailoday28


Go Mechanical Engineering
Tobalcane

 

[sailoday28]

A closed form or finite difference solution provides solution for the governing equations and specified boundary conditions. At any location, the flux that you are after will then be known.
I agree with IRstuff in that the whole plate must be used in the analysis.