Difficulty with Meshing for Heat Flux

[wilson12]

Hi everyone,

I'm extremely new at using Simulation Multiphysics. I downloaded it last week and ran through relevant tutorials. They haven't exactly what I need to know for this project I am attempting. The assembly can be available on Dropbox. I'm attempting to find the steady state heat flux.

This design is basically a case for a hot component inside and I want to determine the Heat Flux.

Some of the problems I am encountering:

I'm assuming I need to use a plate/shell 3d mesh on the outer 2 parts and the inside component is solid meshed.

I don't understand what to do for element type and element definition.

The inside part is going to be a heat source (arbitrary value) and the top and the 2 flat surfaces (front and back) of the case will be acted on through conduction of a heating/cooling plate. Even though it is conduction, should I still set it up as convection?

Part is attached or ask for DropBox access. Feel free to play around with it and send it back.

Thanks for any and all help!

-Andrew

[AstroJohnPE]

Hi Andrew,

One thing to keep in mind is the plate elements do not consider conduction through the thickness. Another way to state this is that the temperature is the same on both sides of the plate. So if your cover fits this condition, then you can use plate elements.

I think that you will figure out the element type and element definition after you mesh the part.

I assume you are doing steady state heat transfer. That analysis type requires a source of heat and a way to remove the heat. So it sounds to me that the convection load would be required.

Just FYI: you attached the assembly file (.iam), not a part file (.ipt).

[wilson12]

Thanks for the reply John, here are the ipt's and the iam that i'm working with. Its a pretty thin casing. I'm trying to find the amount of heat loss.

Hope this helps.

[AstroJohnPE · Accepted solution]

Hi Andrew,

If I measured your model properly, the case is 1.75 mm thick and the heat source is 6.5 mm thick. Since the case is "thick" compared to the heat source, my suggestion is to mesh the entire model as solid (brick elements).

For your question about the cooling plate, you should represent it with whatever load is best suited. For example, if you want to assume that the cooling plates keep the model at a specific temperature, then use controlled temperatures (or a convection with a large coefficient). If the cooling plate removes X watts, then use a heat source. 

You may want to consider making some slight changes to the model to make it more "simulation ready". The attached image shows the ribs on the outer cover. One end of the rib is chamfered which leaves a very small section of the cover exposed. It would be better to remove the chamfer for the analysis and make the rib flush with the top of the cover.

[wilson12]

Thanks John,
You're the best.
Kudos and solved.