Menu

Discussion Forum

New Discussion
Filtrer

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 The information provided may be out of date.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.

Set up concentration for a domian in diffusion model

Posted 14 août 2011, 10:09 UTC−4 Parameters, Variables, & Functions Version 4.0a, Version 4.2 3 Replies

Yang lina
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

In diffusion model, how to set up concentration for a domian and the concentration is a function of change over time?



Thank you very much!
3 Replies Last Post 15 août 2011, 01:34 UTC−4
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 août 2011, 11:07 UTC−4
Hi

any field in COMSOL (at least 99% of them) accept not only a value or a parameter, but also a function of the type

myConcentration_function(t[1/s])

note the "t[1/s]" to define the time t in seconds, but as a unitless value (just incase your t was in minutes or hours). normally all operators using variables need to be entered "unitless" hence should be multiplied by a [1/specific_unit]

So if you use the time dependent "transient" solver, define the conentration as a function of time, you do not need to define it as a "Definition- function", but you can also write it out as a equation such as:

Concentration = MyBasicConc + MyIncreasingConc*t + My2ndConCentr*t^2 ...

Note: check the units for your formulas: if flagged in orange, there is something missing, red there is an error, but sometimes you need to go out of the node and back to get the error flaging reset

Most important, be sure your functions are at least 2*derivables/ continuous as COMSOL uses the gradients for its solving optimisation, and if these are spiky the solution diverges easily. Xou have the step and puls, or Gausssian predefined functions, that smoothes in a "Heaviside" way the sharp steps. Dirac functions are to be avoided.

Lat advice: get hands on the latest v4.2 and patch it to latest update, te 4.0 was a first release and is not really ironed out

--
Good luck
Ivar
Hi any field in COMSOL (at least 99% of them) accept not only a value or a parameter, but also a function of the type myConcentration_function(t[1/s]) note the "t[1/s]" to define the time t in seconds, but as a unitless value (just incase your t was in minutes or hours). normally all operators using variables need to be entered "unitless" hence should be multiplied by a [1/specific_unit] So if you use the time dependent "transient" solver, define the conentration as a function of time, you do not need to define it as a "Definition- function", but you can also write it out as a equation such as: Concentration = MyBasicConc + MyIncreasingConc*t + My2ndConCentr*t^2 ... Note: check the units for your formulas: if flagged in orange, there is something missing, red there is an error, but sometimes you need to go out of the node and back to get the error flaging reset Most important, be sure your functions are at least 2*derivables/ continuous as COMSOL uses the gradients for its solving optimisation, and if these are spiky the solution diverges easily. Xou have the step and puls, or Gausssian predefined functions, that smoothes in a "Heaviside" way the sharp steps. Dirac functions are to be avoided. Lat advice: get hands on the latest v4.2 and patch it to latest update, te 4.0 was a first release and is not really ironed out -- Good luck Ivar
Yang lina
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 14 août 2011, 22:34 UTC−4
Hi

Thank you for your nice~

Firstly, I have been set up the paramters which are related to concentration variable and added the concentration to "concentration" edge. As a result , the edges are the more deep color in diffusion model as attached file, but I want to set up the concentration variabe for the domain which is in the edge that i have mentioned. What should I do? Add the PDE equation or something else?

Secondly, now i use 'diffusion' model to simulate the process that the bigger molecular diffusions from the vessels through vessel walls (the much deeper edges). Should i set up the 'permeability' for vessel wall and what should i do? I have saw the 'permeability' properties in material, but it is in 'Output properties',so I think it is not right.


Thank you~
Hi Thank you for your nice~ Firstly, I have been set up the paramters which are related to concentration variable and added the concentration to "concentration" edge. As a result , the edges are the more deep color in diffusion model as attached file, but I want to set up the concentration variabe for the domain which is in the edge that i have mentioned. What should I do? Add the PDE equation or something else? Secondly, now i use 'diffusion' model to simulate the process that the bigger molecular diffusions from the vessels through vessel walls (the much deeper edges). Should i set up the 'permeability' for vessel wall and what should i do? I have saw the 'permeability' properties in material, but it is in 'Output properties',so I think it is not right. Thank you~

Attachments:

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)
Send Private Message Flag post as spam

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 15 août 2011, 01:34 UTC−4
Hi

Indeed I would not define a concetration on the edges, but only on/for the domain(s), this should work at least for v4.2 (rather forget the earlier releases, and make your model fully from scratch in v4.2 or later, I do not relly trust 100% the tranlations between these earlier v4 versions, even if the COMSOL programmers, I admit, have done a hard job to get them compatible)

Normally, if the walls/edges/boundaries do not have any specific properties by themselves, you should "just" leave them with default "continuity" properties and work on the domains.
However if they represent "thin features" i.e. a semipermeable membrane you must adapt the boundary physics and paramaters accordingly

--
Good luck
Ivar
Hi Indeed I would not define a concetration on the edges, but only on/for the domain(s), this should work at least for v4.2 (rather forget the earlier releases, and make your model fully from scratch in v4.2 or later, I do not relly trust 100% the tranlations between these earlier v4 versions, even if the COMSOL programmers, I admit, have done a hard job to get them compatible) Normally, if the walls/edges/boundaries do not have any specific properties by themselves, you should "just" leave them with default "continuity" properties and work on the domains. However if they represent "thin features" i.e. a semipermeable membrane you must adapt the boundary physics and paramaters accordingly -- Good luck Ivar

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