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.

Impossible concentration range in a micromixer

Please login with a confirmed email address before reporting spam

I am modeling a simple passive micromixer. I have two inlets, one of some arbitrary concentration (say 100 mol/m^3) and the other zero. The transient solution converges and looks fine qualitatively, but the concentration range goes from -1200 to 2000 mol/m^3. I cannot figure out how this is possible and how I can correct it.

Any tips would be greatly appreciated.


9 Replies Last Post 15 juin 2010, 08:53 UTC−4

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 mars 2010, 01:58 UTC−5
1st hint:
www.comsol.com/support/knowledgebase/952/

I am having these kind of problems too. One thing that *seems* to be working is to run a steady-state beforehand, and then running transient, in addition to the comments above.

please keep me updated with your progress.
1st hint: http://www.comsol.com/support/knowledgebase/952/ I am having these kind of problems too. One thing that *seems* to be working is to run a steady-state beforehand, and then running transient, in addition to the comments above. please keep me updated with your progress.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 mars 2010, 13:57 UTC−5
So I've referred to the link and tried the recommendations.

No change when using a finer mesh. No change when using the flc2hs function.

The next thing I tried was solving the Navier-Stokes equations first and then solving the Diffusion equations, rather than solving simultaneously. This also had no effect.

I didn't try using the log(c) as I'm not sure how to implement this. I'll check out the sample model and update with any success.



So I've referred to the link and tried the recommendations. No change when using a finer mesh. No change when using the flc2hs function. The next thing I tried was solving the Navier-Stokes equations first and then solving the Diffusion equations, rather than solving simultaneously. This also had no effect. I didn't try using the log(c) as I'm not sure how to implement this. I'll check out the sample model and update with any success.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 mars 2010, 17:26 UTC−5
I have attempted the logarithmic definition of the concentration boundary conditions but it doesn't have any effect. I'm not sure how it would possible maintain a positive value just by defining the boundary as log(1) instead of zero, or log(10) instead of one. Is there a way to define concentration as log(c)? In other words, how do I define the dependent variable?

Update: Dependent variable needed to be selected in the Model Navigator. I changed the dependent variable to C and defined a scalar expression c=exp(C). Running the solver now. I'll update if it works.
I have attempted the logarithmic definition of the concentration boundary conditions but it doesn't have any effect. I'm not sure how it would possible maintain a positive value just by defining the boundary as log(1) instead of zero, or log(10) instead of one. Is there a way to define concentration as log(c)? In other words, how do I define the dependent variable? Update: Dependent variable needed to be selected in the Model Navigator. I changed the dependent variable to C and defined a scalar expression c=exp(C). Running the solver now. I'll update if it works.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 mars 2010, 18:29 UTC−5
also try anisotropic streamline diffusion (0.1), in case.
also try anisotropic streamline diffusion (0.1), in case.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2 mars 2010, 13:25 UTC−5
Using log(c) keeps the concentration positive. However the maximum concentration still peaks to 10^5 or higher (with an inlet concentration of 100).

I'm trying the anisotropic diffusion tuning parameter 0.1.
Using log(c) keeps the concentration positive. However the maximum concentration still peaks to 10^5 or higher (with an inlet concentration of 100). I'm trying the anisotropic diffusion tuning parameter 0.1.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 3 mars 2010, 09:42 UTC−5
one more thing that my help is to run a steady-state before a transient simulation, and use it as initial conditions.
one more thing that my help is to run a steady-state before a transient simulation, and use it as initial conditions.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 23 mars 2010, 06:50 UTC−4
Hello.

I am experiencing the same problem: I designed a passive micromixer and want to check its efficiency. Using the Incompressible Navier-Stokes model first, then applying the Convection/Diffusion one seems to work.

But as well, the concentration range is impossible: sometimes -40 to 200 whereas I put respectively 0 and 10 at my inlets. Also, observing the concentration at the outlet subdomain, I find concentrations between 2.05 and 2.10 (something like that), which is absurd: it should be centered around 5.

It seems there is a problem of mass/matter conservation. Any idea to solve that issue?

Many thanks.
Hello. I am experiencing the same problem: I designed a passive micromixer and want to check its efficiency. Using the Incompressible Navier-Stokes model first, then applying the Convection/Diffusion one seems to work. But as well, the concentration range is impossible: sometimes -40 to 200 whereas I put respectively 0 and 10 at my inlets. Also, observing the concentration at the outlet subdomain, I find concentrations between 2.05 and 2.10 (something like that), which is absurd: it should be centered around 5. It seems there is a problem of mass/matter conservation. Any idea to solve that issue? Many thanks.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 15 juin 2010, 08:50 UTC−4
I am having this problem as well. Whenever I choose the 'flux' boundary condition the concentration builds up to a physically impossible level. However, when I choose the 'concentration' condition it works out fine. I've attached my model of pore-scale mixing. It seems to be a fairly common problem... can anyone help to resolve this issue?

Thanks!
-Tim
I am having this problem as well. Whenever I choose the 'flux' boundary condition the concentration builds up to a physically impossible level. However, when I choose the 'concentration' condition it works out fine. I've attached my model of pore-scale mixing. It seems to be a fairly common problem... can anyone help to resolve this issue? Thanks! -Tim

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 15 juin 2010, 08:53 UTC−4
Sorry... problems in attaching my model. Hopefully it works now...
Sorry... problems in attaching my model. Hopefully it works now...

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.