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.
Obtaining a stable Convection-Diffusion model
Posted 1 juil. 2010, 14:01 UTC−4 2 Replies
Please login with a confirmed email address before reporting spam
Hello all,
I am working on a microfluidic mixer with several channels. A couple channels will be inputting water, while two others will serve as the respective inlets for two different molecules at low concentrations. Further, both of these molecules have fairly low diffusion coefficients.
Since I do not believe (but please correct me if I'm wrong) that there is a good way to apply different diffusion coefficients to the mixer to simulate the eventual mixing of these two species, I have been solving for the diffusion of each molecular species separately while the other inlet is set to be inputting water. Note: I am not using the Chemical Engineering module, I am simply using a combination of Navier-Stokes/Convection-Diffusion modules.
However, COMSOL is yielding unphysical results for such low concentrations and diffusion coefficients (e.g. the entire mixer shows a uniform concentration.) I know that the model itself is sound since it gives reasonable results for higher concentrations and diffusion coefficients, so the problem seems to be with these parameters. I have been toying with various artificial diffusion settings and refined the mesh as much as possible, but to no avail.
Does anyone have any suggestions to get good results from this model?
Thanks for your help.
Regards,
Natalie
I am working on a microfluidic mixer with several channels. A couple channels will be inputting water, while two others will serve as the respective inlets for two different molecules at low concentrations. Further, both of these molecules have fairly low diffusion coefficients.
Since I do not believe (but please correct me if I'm wrong) that there is a good way to apply different diffusion coefficients to the mixer to simulate the eventual mixing of these two species, I have been solving for the diffusion of each molecular species separately while the other inlet is set to be inputting water. Note: I am not using the Chemical Engineering module, I am simply using a combination of Navier-Stokes/Convection-Diffusion modules.
However, COMSOL is yielding unphysical results for such low concentrations and diffusion coefficients (e.g. the entire mixer shows a uniform concentration.) I know that the model itself is sound since it gives reasonable results for higher concentrations and diffusion coefficients, so the problem seems to be with these parameters. I have been toying with various artificial diffusion settings and refined the mesh as much as possible, but to no avail.
Does anyone have any suggestions to get good results from this model?
Thanks for your help.
Regards,
Natalie
2 Replies Last Post 2 juil. 2010, 13:31 UTC−4
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Did you arrange your tolerances enough to capture enough precision on your very low values? I would suggest to avoid artificial diffusion as much as possible and go to finer mesh, if possible.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Did you arrange your tolerances enough to capture enough precision on your very low values? I would suggest to avoid artificial diffusion as much as possible and go to finer mesh, if possible.
I have tried to adjust the "relative tolerance" value in Solver Parameters, which hasn't seemed to help. Are there other tolerances that I might find and change? If so, where, and do you have any recommendations on how to best modify the default values?