Modelling fluid flow plus membrane deflection

Pablo Garcia De Madinabeitia

Please login with a confirmed email address before reporting spam

Hi all,

I am new to comsol community and I am facing a simulation of a membrane/wall deflection over a microfluidic channel. At this moment, I have a microfluidic channel which is formed by a domain filled with fluid (no wall thickness, just the path the fluid follows). At some point of the channel (rectangular cross section), I have a 200 micron layer which I would like to deflect upon mechanical load.

I have always simulated microfluidic chips being the chip as the domain where the fluid flows, not considering any wall thickness. It is now, that I want to simulate mechanical actuation over a 200micron thickness wall that I wonder how I should set the walls (thickness yes or no?). In addition to this, is it possible to do it with the basic functionalities of COMSOL multiphysics 5.4? Or should I get the specific structural mechanics or MEMS module?

Thanks in advance, Pablo


4 Replies Last Post 1 sep. 2019 13:53 GMT−4
Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 4 months ago 12 juli 2019 08:52 GMT−4
Updated: 4 months ago 12 juli 2019 08:45 GMT−4

Hello Pablo,

What you are talking about here is referred to as a fluid-structure interaction (or FSI, for short) problem. FSI is something COMSOL is very well suited for. There are several different possible configurations, starting with whether the two physics are one-way coupled or two-way coupled. In your case, if the deformation of the membrane is small, you may be able to neglect the impact its deformation has on the fluid flow, making it one-way coupled (In your case, on a fixed geometry). If on the other hand the deflection of the membrane is large, or if the point of your analysis is to determine how much the flow is affected by the membrane deflection, then you will need to treat it as two-way coupled. Next is the question of how to treat that membrane: can you/do you want to model it as a shell/plate (possible if it has a high aspect ratio) or is it thick enough that you model it as a solid (i.e. accounting for its thickness in the Geometry section of your model).

In the situation where you neglect the impact of the membrane deformation on the flow and you model the membrane as a solid, you should be able perform this simulation without the Structural Mechanics Module or the MEMS Module (assuming that there is no other complication involved that you have not mentioned, such as could come from a material behavior, large strains, etc). It would be conceptually similar to this model of fluid flow over a solar panel (But your situtation is simpler because you likely have a much lower Reynolds number so you presumably don't need to account for turbulence): https://www.comsol.com/blogs/efficient-solar-panel-design-improves-pv-industry/ . If you go for the two-way coupling, or if you want to use a plate or shell representation for the membrane, then you will need the Structural Mechanics Module

Here are a couple of webinars that you may find helpful to get going:

https://www.comsol.com/video/intro-to-modeling-fluid-structure-interaction-in-comsol https://www.comsol.com/video/modeling-fluid-structure-interaction-with-comsol-multiphysics

Best regards,

Jeff

-------------------
Jeff Hiller
Hello Pablo, What you are talking about here is referred to as a fluid-structure interaction (or FSI, for short) problem. FSI is something COMSOL is very well suited for. There are several different possible configurations, starting with whether the two physics are one-way coupled or two-way coupled. In your case, if the deformation of the membrane is small, you may be able to neglect the impact its deformation has on the fluid flow, making it one-way coupled (In your case, on a fixed geometry). If on the other hand the deflection of the membrane is large, or if the point of your analysis is to determine how much the flow is affected by the membrane deflection, then you will need to treat it as two-way coupled. Next is the question of how to treat that membrane: can you/do you want to model it as a shell/plate (possible if it has a high aspect ratio) or is it thick enough that you model it as a solid (i.e. accounting for its thickness in the Geometry section of your model). In the situation where you neglect the impact of the membrane deformation on the flow and you model the membrane as a solid, you should be able perform this simulation without the Structural Mechanics Module or the MEMS Module (assuming that there is no other complication involved that you have not mentioned, such as could come from a material behavior, large strains, etc). It would be conceptually similar to this model of fluid flow over a solar panel (But your situtation is simpler because you likely have a much lower Reynolds number so you presumably don't need to account for turbulence): https://www.comsol.com/blogs/efficient-solar-panel-design-improves-pv-industry/ . If you go for the two-way coupling, or if you want to use a plate or shell representation for the membrane, then you will need the Structural Mechanics Module Here are a couple of webinars that you may find helpful to get going: https://www.comsol.com/video/intro-to-modeling-fluid-structure-interaction-in-comsol https://www.comsol.com/video/modeling-fluid-structure-interaction-with-comsol-multiphysics Best regards, Jeff

Pablo Garcia De Madinabeitia

Please login with a confirmed email address before reporting spam

Posted: 4 months ago 24 juli 2019 06:21 GMT−4
Updated: 4 months ago 24 juli 2019 06:20 GMT−4

Dear Jeff,

Thanks a lot for your response. I found the links very helpful and clarifying.

However, I still wonder if we would need to purchase structural mechanics module in order to compute a 2D/3D simulation of this FSI simulations (or simply, could we just get through with the basic functionalities of structural mechanics?). I need to simulate how much force (mechanical load) I need to close a pipe at a fixed flowrate, and the other way around: how much flowrate I have to set to overcome pipe closing mechanical load. I believe my problem would be then a two-way coupled problem. I would be thankful if you could confirm my suspicions.

Thank you very much!

Best regards, Pablo

Dear Jeff, Thanks a lot for your response. I found the links very helpful and clarifying. However, I still wonder if we would need to purchase structural mechanics module in order to compute a 2D/3D simulation of this FSI simulations (or simply, could we just get through with the basic functionalities of structural mechanics?). I need to simulate how much force (mechanical load) I need to close a pipe at a fixed flowrate, and the other way around: how much flowrate I have to set to overcome pipe closing mechanical load. I believe my problem would be then a two-way coupled problem. I would be thankful if you could confirm my suspicions. Thank you very much! Best regards, Pablo

Jeff Hiller COMSOL Employee

Please login with a confirmed email address before reporting spam

Posted: 4 months ago 24 juli 2019 08:26 GMT−4
Updated: 4 months ago 24 juli 2019 08:48 GMT−4

Hi Pablo,

Indeed, it does not sound like the simple case where the flow can be evaluated on the undeformed geometry as in the solar panel example mentioned in my earlier post. In fact, since you mention that the membrane may deflect so much as to close the pipe, it makes me think that the membrane may experience large strains, which would be another reason for needing the Structural Mechanics Module.

Best regards,

Jeff

PS: BTW, the pipe completely closing would lead to a change in topology for the fluid domain. This blog presents a way to handle that difficulty.

-------------------
Jeff Hiller
Hi Pablo, Indeed, it does not sound like the simple case where the flow can be evaluated on the undeformed geometry as in the solar panel example mentioned in my earlier post. In fact, since you mention that the membrane may deflect so much as to close the pipe, it makes me think that the membrane may experience large strains, which would be another reason for needing the Structural Mechanics Module. Best regards, Jeff PS: BTW, the pipe completely closing would lead to a change in topology for the fluid domain. [This blog](https://www.comsol.com/blogs/tips-using-wall-distance-interface/) presents a way to handle that difficulty.

Pablo Garcia De Madinabeitia

Please login with a confirmed email address before reporting spam

Posted: 3 months ago 1 sep. 2019 13:53 GMT−4

Dear Jeff,

Back again with this FSI issue, I understand now that is a 2-way coupled problem.

Indeed I have had the chance to face the problem. When I define deforming domain (a membrane moves) then, I cannot define a force in the structural mechanics physics.

Could you provide any guidance on how to proceed?

Best, Pablo

Dear Jeff, Back again with this FSI issue, I understand now that is a 2-way coupled problem. Indeed I have had the chance to face the problem. When I define deforming domain (a membrane moves) then, I cannot define a force in the structural mechanics physics. Could you provide any guidance on how to proceed? Best, Pablo

Reply

Please read the discussion forum rules before posting.

Please log in to post a reply.

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.