[Martin_Sz]

Hello
I have problem with overflow error

So i need to make heat transfer simulation in two cases
1) cooling from 2400 C to 70 C in 1 bar abs pressure
2) cooling from 2400 C to 70 C in 0,01 bar abs pressure

I have three gas domains with initiial 2400 C
I have two graphite cages with initiall 2400 C and 40 C external wall temperature . And i need to determine a cooling curve of graphite block over the time.

I suyppose that closed domain do overflow error

Do U have any ideas ??
Sorry for my english

Best regards

[Gert-Jan]

Really no idea what your problem is. Please share your output file.

[Martin_Sz]

Which solver do i need to setup for the case with closed domain and high temperature. Solvers for heat transfer, turbulence and most important radiation.
These solvers i need to define to transient simulation for cooling process.

[Martin_Sz]

In my case i dont have any inlets and outlets

[Gert-Jan]

So, the isolated volumes do not lead to problems. You only need to make sure you set the initial pressure at a location in each volume.

Further on, it is difficult to give general advice on how to setup a problem and give advice on models to choose. Please check the tutorials or the support of ANSYS.

[Martin_Sz]

And what about high initial temperature 2400 C which radiation and heat transfer models use??

[ghorrocks]

As long as you are using material models which are appropriate for the temperature range you specify you are fine. A simple ideal gas might be OK depending on how accurate you want to be. Real gas models will improve accuracy but are more complex to use.

Read the documentation on what the radiation models mean before deciding. You really need to understand the models so you can make an informed choice. But to give you a starting point: There are really only two models which allow heat transfer from one surface to another, Discrete Transfer and Monte Carlo. And as your surface to surface heat transfer looks quite straight forward the Discrete Transfer model would be the one to choose. Read the documentation to work out why that is.

If these domains are closed then the pressure is going to change with the temperature. Is this what the actual device has? Getting something to stay airtight and rigid at very high temperatures is not easy. Are you sure it does not have any leaks? Or pressure relief valve? Or that it deforms under pressure? Any of these factors will have a significant effect.

[Martin_Sz]

code below

[Martin_Sz]

code


sef

[Martin_Sz]

I have very high values of i-radiation on closed domains above 100.

[ghorrocks]

If you have never done radiation modelling before I strongly recommend you do a benchmark simulation first. For instance, model two plates separated by a gap. The heat transfer you should get here is known from analytical equations. So model this case and check you can get it to converge and can get it accurate. Then you will know the settings required to make your model work.

And please put large files (eg your CCL) on the forum as attachments, not as huge long posts. It makes it very hard to read the thread.

[Martin_Sz]

code in attachments

[Martin_Sz]

now everything seems ok - imbalances value is below 1 but when simulation reach about 0,5 physical time i have overflow

[ghorrocks]

I repeat, if you are using a new physical model which you are not familiar with then you should do some benchmark simulations to check you can use it correctly and accurately. Just proceeding and using it in a complex simulation with no understanding is foolish.

[Martin_Sz]

So i do simplified case
and optimum is discrete transfer surfTOsurf inmy case but
after about 40 - 80 iterations i get error overflow.
I make many variants and problem is on ideal gas. Without it simulation works fine.

But I need make two types of simulations
On atmospheric pressure and nearly vacuum 0.01 bar so i need compressible flow .
Any ideas ??
Best regards

[ghorrocks]

Please attach the output file of your simplified model.

[Martin_Sz]

Hello Ghorrocks
Boundary conditions and solver settings are the same like in the macro were i sent earlier .
Attachment macro




p.s
In later simulations i only simplify geometry

[Martin_Sz]

i have one idea

can i make two variations with solvers thermal energy and air at 25

- for 1 bar i give standard gas parameters
- for 0.01 bar i decrease density of gas

will be this correct ??