Menu

Discussion Forum

New Discussion
Filter

Note: This discussion is about an older version of the COMSOL Multiphysics® software. The information provided may be out of date.

Discussion Closed This discussion was created more than 6 months ago and has been closed. To start a new discussion with a link back to this one, click here.

convection/diffusion problem

Posted Jan 19, 2012, 8:36 p.m. EST Chemical Reaction Engineering, Modeling Tools & Definitions, Parameters, Variables, & Functions Version 3.5a, Version 4.1 4 Replies

Saeed Miramini
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Hi everybody,

I'm modelling convection/diffusion mass transport in a porous media. The diffusion coefficient of the model is very small (4e-12 m^2/s) and the velocity field is quite small as well (maximum 1e-5 m/s).
The boundary condition is a constant concentration at a side of the geometry.
After I solve the model for 15 min, the concentration at some parts of the model is greater than the source concentration which doesn't make sense. These areas have higher velocity magnitudes.
I would appreciate it a lot if anybody help me to solve this problem.
Can I use artificial diffusion to solve it? Does it affect the result of my model?

Many thanks in advance
4 Replies Last Post Feb 7, 2012, 8:53 p.m. EST
Pranay Ghosh
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 20, 2012, 8:42 a.m. EST
This seems to be a meshing issue. Did you check your minimum mesh quality ? I would suggest to use boundary layer meshing at the boudaries. Also try playing around with the time step and absolute tolerance value for the time dependent problem.
This seems to be a meshing issue. Did you check your minimum mesh quality ? I would suggest to use boundary layer meshing at the boudaries. Also try playing around with the time step and absolute tolerance value for the time dependent problem.
Saeed Miramini
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 24, 2012, 12:36 a.m. EST
Thank you for your response.
I changed the mesh, tolerances and time stepping but the problem persists.
In very small area of my model, the velocity magnitude is very high and it causes instability in the convection/diffusion equation and leads to very high concentration in this area that doesn't make sense.
I tried to add artificial diffusion (crosswind, ordo type). It helped somewhat but there is still some areas with high concentration. On the other hand it adds extra diffusion and changes the result.
Can I change tuning parameter in crosswind artificial diffusion (ordo type)? In the documentations, it is mentioned that this value is calculated for SI. I'm using SI units so I think I can not change it
Thank you for your response. I changed the mesh, tolerances and time stepping but the problem persists. In very small area of my model, the velocity magnitude is very high and it causes instability in the convection/diffusion equation and leads to very high concentration in this area that doesn't make sense. I tried to add artificial diffusion (crosswind, ordo type). It helped somewhat but there is still some areas with high concentration. On the other hand it adds extra diffusion and changes the result. Can I change tuning parameter in crosswind artificial diffusion (ordo type)? In the documentations, it is mentioned that this value is calculated for SI. I'm using SI units so I think I can not change it
Pranay Ghosh
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Jan 25, 2012, 10:35 a.m. EST
Do you have sharp edges in your geometry where the velocity field is high ?. This can also cause absurd artifacts in your solution. I can have a look at your model if you want.
Do you have sharp edges in your geometry where the velocity field is high ?. This can also cause absurd artifacts in your solution. I can have a look at your model if you want.
Saeed Miramini
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Feb 7, 2012, 8:53 p.m. EST
Thank you so much for your hint.

I had sharp edges in my model and it caused the instabilities.
I created fillet at the sharp edges and then the instability reduced significantly. However after creating fillet, the velocity at the edges is like before. I think since I have finer mesh at the filleted region, the sharp gradient is reduced and it causes stable concentration result.

Cheers,
Saeed
Thank you so much for your hint. I had sharp edges in my model and it caused the instabilities. I created fillet at the sharp edges and then the instability reduced significantly. However after creating fillet, the velocity at the edges is like before. I think since I have finer mesh at the filleted region, the sharp gradient is reduced and it causes stable concentration result. Cheers, Saeed

Note that while COMSOL employees may participate in the discussion forum, COMSOL® software users who are on-subscription should submit their questions via the Support Center for a more comprehensive response from the Technical Support team.

Suggested Content