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 The information provided may be out of date.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.
Which physics module should I use?
Posted Nov 2, 2011, 6:17 a.m. EDT MEMS & Nanotechnology, MEMS & Piezoelectric Devices, Modeling Tools & Definitions, Parameters, Variables, & Functions, Studies & Solvers Version 4.1 5 Replies
Please login with a confirmed email address before reporting spam
Hello everybody,
I'm starting learning COMSOL by myself, I must admit that I thought that it would be easier at the beginning.
I'm currently handing a couple of MEMS sensors for enviroment measuremtn. And in both desings using COMSOL I have always the same question. How can I simulated then? Which physics should I use?
The first MEMS is a temperature sensor based on polysilicon resistor. I thought I must use AC/DC module and Heat Transfer. But then, I don't know which other options should I add.
The second one is and relative humidity sensor base on poliymide layer. At the beginning I was going to use microfluidics but any of the options I found down the microfluidics module were useful to me.
Please I hope you can answer me and help because I have been throught lot of time just trying to find out how to simulated then.
Best Regards
I'm starting learning COMSOL by myself, I must admit that I thought that it would be easier at the beginning.
I'm currently handing a couple of MEMS sensors for enviroment measuremtn. And in both desings using COMSOL I have always the same question. How can I simulated then? Which physics should I use?
The first MEMS is a temperature sensor based on polysilicon resistor. I thought I must use AC/DC module and Heat Transfer. But then, I don't know which other options should I add.
The second one is and relative humidity sensor base on poliymide layer. At the beginning I was going to use microfluidics but any of the options I found down the microfluidics module were useful to me.
Please I hope you can answer me and help because I have been throught lot of time just trying to find out how to simulated then.
Best Regards
5 Replies Last Post Nov 6, 2011, 4:37 a.m. EST
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
there is no one solution there, you must start by sketching our your device, define the materials, the environment, the interactions, all on a piece of paper first , then group this, decide which physics to start with (all in one go will make a mess, add one physics at the time from the moment you mater the first ones)
Mems device are typically mixing of structural/solid dependent variables the deformations "u" or scalar u,v,w in 3D vector format.
You might have some temperature effects, source sink ... you need to add "T", probably not radiousity J
You have some Joule heating and resistance you need electric currents our joule heating and need to solve for "V"
You might have some interaction with air, or humidity ? you need to find out if you need the fluid modules or if you can start to consider the air as a static heat conductor or convective heat exchange (keep it simple if possible). but moisture and adsorbtioon ? => porous media and diffusion, this starts to become really nice complex ...
enough to have fun for some time there :) it will be an interesting model
--
Good luck
Ivar
there is no one solution there, you must start by sketching our your device, define the materials, the environment, the interactions, all on a piece of paper first , then group this, decide which physics to start with (all in one go will make a mess, add one physics at the time from the moment you mater the first ones)
Mems device are typically mixing of structural/solid dependent variables the deformations "u" or scalar u,v,w in 3D vector format.
You might have some temperature effects, source sink ... you need to add "T", probably not radiousity J
You have some Joule heating and resistance you need electric currents our joule heating and need to solve for "V"
You might have some interaction with air, or humidity ? you need to find out if you need the fluid modules or if you can start to consider the air as a static heat conductor or convective heat exchange (keep it simple if possible). but moisture and adsorbtioon ? => porous media and diffusion, this starts to become really nice complex ...
enough to have fun for some time there :) it will be an interesting model
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Thank you very much for your answer.
After read your answer, I am a little bit scared. It seems to be quite complicated.
If I have understood you properly. In both cases I have to add Structural Mechanics/Solid Mechanics just because they are MEMS and it is supossed to have a deformation. Start doing some analysis (I don't know which one), and once it works, start adding a new physics, do you recomended me the Joule Heating?
On the other hand, and once a physics is added on your model, when you make right click on the physics, the transparent/blue boxes represents the boundary conditions right? And what about the purples ones?
Thank you again for your time and for your answers.
After read your answer, I am a little bit scared. It seems to be quite complicated.
If I have understood you properly. In both cases I have to add Structural Mechanics/Solid Mechanics just because they are MEMS and it is supossed to have a deformation. Start doing some analysis (I don't know which one), and once it works, start adding a new physics, do you recomended me the Joule Heating?
On the other hand, and once a physics is added on your model, when you make right click on the physics, the transparent/blue boxes represents the boundary conditions right? And what about the purples ones?
Thank you again for your time and for your answers.
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
No reason to be scared, just start one step at the time.
Now, if you have the mems module, it covers addition to several physics (in v4, this was organised differently in v3) so normally you can do most of the combinations mentioned previously
One way start with a structure (solid), add some current (JC) and temperature (HT) or choose a combination joule heating (JH) or all three joule heating and thermal expansion (TEM) provided that you have access to these combination physics, if not you must link the others.
If you choose TEM or JH, b no forget you have 3-2 physics, so you need to define enough properties for each physics, and or limit the different physics to certain domains only, by adding in new sub nodes. And you need to define sufficient BC (boundary conditions) FOR EACH physics to impose a unique solution to the set of ODE you are preparing. With these pre-cooked multiple physics you have most of the linking already set up for you by COMSOL
--
Good luck
Ivar
No reason to be scared, just start one step at the time.
Now, if you have the mems module, it covers addition to several physics (in v4, this was organised differently in v3) so normally you can do most of the combinations mentioned previously
One way start with a structure (solid), add some current (JC) and temperature (HT) or choose a combination joule heating (JH) or all three joule heating and thermal expansion (TEM) provided that you have access to these combination physics, if not you must link the others.
If you choose TEM or JH, b no forget you have 3-2 physics, so you need to define enough properties for each physics, and or limit the different physics to certain domains only, by adding in new sub nodes. And you need to define sufficient BC (boundary conditions) FOR EACH physics to impose a unique solution to the set of ODE you are preparing. With these pre-cooked multiple physics you have most of the linking already set up for you by COMSOL
--
Good luck
Ivar
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi Ivar,
Again I just can say thank you for your help and effort.
I think I finally get the physics works. I mean, I have used a Joule Heating and Thermal Expansion module, and I used boundary conditions that fit with my proposal. I have a variable called T_enviroment that I want to use to change the temperature of the enviroment.
My question, is how to perform and study, I guess it should be stationary, but I'm not completely sure, in order to obtain how a parameter (in my case the potential) changes within the temperature. Do you understand what I want to do?
EDIT. I have been able to see how the potential changes within temperature, doing a parametric analysis, but then, I have one plot for each temperature value, when I need just one.
Thank you very much. I think I'm finally starting to understand a bit about how COMSOL works.
Regards!
Fernando
Again I just can say thank you for your help and effort.
I think I finally get the physics works. I mean, I have used a Joule Heating and Thermal Expansion module, and I used boundary conditions that fit with my proposal. I have a variable called T_enviroment that I want to use to change the temperature of the enviroment.
My question, is how to perform and study, I guess it should be stationary, but I'm not completely sure, in order to obtain how a parameter (in my case the potential) changes within the temperature. Do you understand what I want to do?
EDIT. I have been able to see how the potential changes within temperature, doing a parametric analysis, but then, I have one plot for each temperature value, when I need just one.
Thank you very much. I think I'm finally starting to understand a bit about how COMSOL works.
Regards!
Fernando
Please login with a confirmed email address before reporting spam
Posted:
1 decade ago
Hi
If your T_env is constant in time, then use stationary. If you want to see the effect for different constant values of T_env, then turn on the Extension Continuation in the stationary solver node, add a predefined Parameter (top Definition node) i.e. Param, and set Param to a range i.e range(0,0.1,1). Then wrtit in your temperature field(s) T_envStart + T_Step*Param (check the units, COMSOL is by default in Kelvin).
However if you want to see the time behaviour, you need to use the transient solver, and define correct time steps, as well as then you might use time dependent values (functions) for your Temperature entries.
Do not forget to check your meshing sizes (search the replies of Nagi, he gives some usefull indications here on the forum in several threads)
--
Good luck
Ivar
If your T_env is constant in time, then use stationary. If you want to see the effect for different constant values of T_env, then turn on the Extension Continuation in the stationary solver node, add a predefined Parameter (top Definition node) i.e. Param, and set Param to a range i.e range(0,0.1,1). Then wrtit in your temperature field(s) T_envStart + T_Step*Param (check the units, COMSOL is by default in Kelvin).
However if you want to see the time behaviour, you need to use the transient solver, and define correct time steps, as well as then you might use time dependent values (functions) for your Temperature entries.
Do not forget to check your meshing sizes (search the replies of Nagi, he gives some usefull indications here on the forum in several threads)
--
Good luck
Ivar