Boundary condition

Numerical methods and mathematical models of Elmer
yuanxun
Posts: 17
Joined: 20 Feb 2019, 14:04
Antispam: Yes

Boundary condition

Post by yuanxun »

Hi,

I was doing heat transfer simulation of a wall,but the surface temperature is the same at every place.
Since there're different materials and convective transfer,it shouldn't be like this.Now the result looks like the surface temperature was set as a constant value.
I attached my settings and sif file,maybe there're anything wrong?
Thanks a lot.

Yuan
Attachments
case.sif
(4.57 KiB) Downloaded 253 times
boundary condition.png
boundary condition.png (30.84 KiB) Viewed 4672 times
equation.png
equation.png (30.6 KiB) Viewed 4672 times
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

I have read if you use the diffuse gray you have to check the open radiation boundary. You could try the idealized model as well.
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

I have read if you use the diffuse gray you have to check the open radiation boundary. You could try the idealized model as well.
yuanxun
Posts: 17
Joined: 20 Feb 2019, 14:04
Antispam: Yes

Re: Boundary condition

Post by yuanxun »

Hi,

Thanks a lot!I tried to set it to idealized(by the way I may didn't turn on the convection before).But this time warning"WARNING:: HeatSolver: Convection model specified but no associated flow field present?" appears and result didn't change.
Here's my solver log file.

Yuan
Attachments
solver log.zip
(20.03 KiB) Downloaded 250 times
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

I believe convection works with the fluid flow solver
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

You are doing steady state. If the boundaries do not radiate or do not radiate fast enough then eventually everything would get to constant temperature. Might try doing it in a transient to see how the heat conducts in the solids.
yuanxun
Posts: 17
Joined: 20 Feb 2019, 14:04
Antispam: Yes

Re: Boundary condition

Post by yuanxun »

Thank you,it does work now.Outdoor side has temperature change now.
But I'm not sure that if it's right because indoor side with 24°C of air is still the same.
Here's my setup.
Attachments
setup.png
(45.77 KiB) Not downloaded yet
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

Transient should have number of time step intervals and time step sizes set. For example if you want to run the simulation for 30 seconds at a time step of .5 seconds each. Then Time Step Size = 0.5, and Time Step Intervals = 60.

60 steps at 0.5 seconds each = 30 seconds
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

You may also consider setting initial conditions, heat equation, temperature, for each body otherwise each body starts at 0.
kevinarden
Posts: 2301
Joined: 25 Jan 2019, 01:28
Antispam: Yes

Re: Boundary condition

Post by kevinarden »

attached is an example of a multi-material 3D radiating model in steady state and transient. The temps are in kelvin.
Steady_state_column.zip
(106.93 KiB) Downloaded 280 times
transient.zip
(107.14 KiB) Downloaded 257 times
The study files are the salome models, from salome export unv and use file open in ElmerGUI. SInce I added the project files you can go to ElmerGUI and load project and select the folder you put the files in.

I post process in paraview, so they are set up to write vtu files.
Post Reply