Do I need to make any special considerations for the mesh of this geometry?

I am trying to simulate internal flow.

- 180 Views
- Last Post 13 June 2018
- Topic Is Solved

Waleed Ghayas
posted this
11 June 2018

Do I need to make any special considerations for the mesh of this geometry?

I am trying to simulate internal flow.

raul.raghav
posted this
11 June 2018

For meshing the internal flow without considering the TEG, you can slice the geometry based on the following screenshot and just make sure you share topology between the sliced bodies.

Rahul

Waleed Ghayas
posted this
12 June 2018

Thank you very much for your help!

I basically merged all this (there were a total of 578 bodies) into a single part to allow for same number of nodes at interface between two bodies. I have also tried running the simulation, however, temperature divergence occurred. I think that I somehow messed up in the meshing of the TEG, since when I remove the TEG, and instead replace it with a simple block of ceramic material, I find that there is no divergence. I thought that the problem might be caused by the small 0.2 mm distance between the legs and that I might need a finer mesh or a smaller growth rate, but doing both those did not change the temperature divergence.

If I do what you asked, could it solve the problem I am facing?

raul.raghav
posted this
12 June 2018

Waleed, could you explain your model a little elaborately? Do you need to model the TEG or could the model be simplified by appropriate thermal boundary conditions? What is the goal of the simulation?

Rahul

Waleed Ghayas
posted this
12 June 2018

Of course. I need to model the TEG. The goal of the simulation is to get how much power is generated from the TEG. The TEG is placed upon a heat exchanger as shown below.

The heat exchanger has fins protruding inwards and has a fluid flowing through it.

To simplify the simulation, the heat exchanger was segmented into one-sixth section (as shown below), and the simulation was run on that.

Now the model of a TEG is a bit complex. Essentially, you have two cuboid semiconductors connected by copper connections.

Thank you very much, again, for all your time and consideration.

kkanade
posted this
13 June 2018

Hello,

You need to look at following.

1. In geometry, please check if share topology has worked correctly.

2. Get at least 2-3 elements across thickness.

3. Check mesh quality. The min. orthogonal quality should be above 0.1.

4. Check by reducing your inlet velocity.

5. Decrease energy urf

6. Solve only flow equation for say 50-100 iterations. Then enable energy equation.

Hope this helps.

Regards,

Keyur

- peteroznewman 61
- abenhadj 49
- rwoolhou 36
- kkanade 27
- jj77 23
- EngineerMech 23
- FlaviaGelatti 18
- SandeepMedikonda 14
- tsiriaks 14
- yang 12