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 The information provided may be out of date.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.

fluid solid interface boundary

Please login with a confirmed email address before reporting spam

I have a solid domain and a fluid domain and am trying to model FSI. Both the domains have circular cross-section. Solid domain is a hollow cylinder while the fluid domain is a cylinder concentric with the solid domain and located inside the solid domain. Its basically the 3D "pipe flow" problem with FSI.

COMSOL as a default has "fluid-solid interface boundary" listed in the boundary conditions. This boundary condition (BC) selects the surfaces that are at the intersection of the fluid and the solid domains which are the "interface surfaces". This BC infact overwrites the wall BC. So I assumed that it takes over the conditions that signify the wall (no slip BC). Am I right?

The other question is: when I define a "boundary load" at the interface surface, the "fluid-solid interface boundary" is still active and is not overwritten. My boundary load will be the load applied by the fluid on the solid surface.

Since my fluid domain is supposed to respond based on the solid domain that surrounds it, I felt that I should use prescribed mesh displacement at the interface with dx=u_solid etc. Only then the fluid will move in response to the displacement of the solid . Please let me know if I am right in using the "prescribed mesh displacement".

However, when I use prescribed mesh displacement at the interface, then the "fluid-solid interface boundary" condition is overwritten. So how does COMSOL know where the interface is? I was thinking that this is the condition that tells COMSOL about the "wall" location and hence determines the no slip boundary condition. But if interface BC is overwritten, how does it solve the problem?

I am asking because I am having convergence issues and I am guessing that something is wrong in my boundary conditions.

Thanks for the help
Sirisha


8 Replies Last Post 2011年6月1日 GMT-4 14:05
Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月25日 GMT-4 01:55
Hi

I believe you are getting confused between the user model set up way of SOLID+SPF+ALE physics and the FSI module that is a combination of the three "pre-cooked" by COMSOL

I can only advise you to follow the video on the main site + do a couple of the model library FSI examples (do not forget to download the latest items, see the KB)

take some time to read through end experiment the examples, and check the equations, understanding the background takes some time, but once understood it goes so much quicker ;)

Another point, be sure you have upgraded your CM to the latest patch, see the main web site

--
Good luck
Ivar
Hi I believe you are getting confused between the user model set up way of SOLID+SPF+ALE physics and the FSI module that is a combination of the three "pre-cooked" by COMSOL I can only advise you to follow the video on the main site + do a couple of the model library FSI examples (do not forget to download the latest items, see the KB) take some time to read through end experiment the examples, and check the equations, understanding the background takes some time, but once understood it goes so much quicker ;) Another point, be sure you have upgraded your CM to the latest patch, see the main web site -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月25日 GMT-4 12:06
I have the latest patch. I am waiting to get a hold on version 4.2 though.

What is the video you are referring to?

I already did a couple of examples (infact there is only one example that is 2-way FSI and they have one other one that is 1-way FSI). None of these examples use the prescribed mesh displacement. The manual does not explain this concept (prescribed mesh displacement) with an example. It just has a few lines about it but that's all. COMSOL support team has not responded to two of my queries yet. So I am skeptical to email them about this.

Thanks for the suggestions!

sirisha



I have the latest patch. I am waiting to get a hold on version 4.2 though. What is the video you are referring to? I already did a couple of examples (infact there is only one example that is 2-way FSI and they have one other one that is 1-way FSI). None of these examples use the prescribed mesh displacement. The manual does not explain this concept (prescribed mesh displacement) with an example. It just has a few lines about it but that's all. COMSOL support team has not responded to two of my queries yet. So I am skeptical to email them about this. Thanks for the suggestions! sirisha

Ivar KJELBERG COMSOL Multiphysics(r) fan, retired, former "Senior Expert" at CSEM SA (CH)

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月26日 GMT-4 02:43
Hi

I agree there are so many physics and so many pages to read, and still one would want hundred more ...
"My way" is to read, and re-read the doc, for each new version, and check regularly the model library updates, and redo the model library examples, as well as following several of the webinars, one catches always new ways therein

--
Good luck
Ivar
Hi I agree there are so many physics and so many pages to read, and still one would want hundred more ... "My way" is to read, and re-read the doc, for each new version, and check regularly the model library updates, and redo the model library examples, as well as following several of the webinars, one catches always new ways therein -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月27日 GMT-4 03:59

I have a solid domain and a fluid domain and am trying to model FSI. Both the domains have circular cross-section. Solid domain is a hollow cylinder while the fluid domain is a cylinder concentric with the solid domain and located inside the solid domain. Its basically the 3D "pipe flow" problem with FSI.


I remember there was a blood FSI model in the comsol library, which should be similiar to yours.


COMSOL as a default has "fluid-solid interface boundary" listed in the boundary conditions. This boundary condition (BC) selects the surfaces that are at the intersection of the fluid and the solid domains which are the "interface surfaces". This BC infact overwrites the wall BC. So I assumed that it takes over the conditions that signify the wall (no slip BC). Am I right?


Yes, it does. FSI condition has two boundary conditions, one is that the velocities of the wall and liquid are equal (no-slip condition, and displacement in time is equal), and second that the Cauchy stresses of fluid and structure on the two domain are in equilibrium, so that Sigma_F = Sigma_S.



The other question is: when I define a "boundary load" at the interface surface, the "fluid-solid interface boundary" is still active and is not overwritten. My boundary load will be the load applied by the fluid on the solid surface.

Since my fluid domain is supposed to respond based on the solid domain that surrounds it, I felt that I should use prescribed mesh displacement at the interface with dx=u_solid etc. Only then the fluid will move in response to the displacement of the solid . Please let me know if I am right in using the "prescribed mesh displacement".

However, when I use prescribed mesh displacement at the interface, then the "fluid-solid interface boundary" condition is overwritten. So how does COMSOL know where the interface is? I was thinking that this is the condition that tells COMSOL about the "wall" location and hence determines the no slip boundary condition. But if interface BC is overwritten, how does it solve the problem?


You should not apply anything else that FSI boundary. See above.

Danial



--
Comsol 4.1
Ubuntu 10.04.1
[QUOTE] I have a solid domain and a fluid domain and am trying to model FSI. Both the domains have circular cross-section. Solid domain is a hollow cylinder while the fluid domain is a cylinder concentric with the solid domain and located inside the solid domain. Its basically the 3D "pipe flow" problem with FSI. [/QUOTE] I remember there was a blood FSI model in the comsol library, which should be similiar to yours. [QUOTE] COMSOL as a default has "fluid-solid interface boundary" listed in the boundary conditions. This boundary condition (BC) selects the surfaces that are at the intersection of the fluid and the solid domains which are the "interface surfaces". This BC infact overwrites the wall BC. So I assumed that it takes over the conditions that signify the wall (no slip BC). Am I right? [/QUOTE] Yes, it does. FSI condition has two boundary conditions, one is that the velocities of the wall and liquid are equal (no-slip condition, and displacement in time is equal), and second that the Cauchy stresses of fluid and structure on the two domain are in equilibrium, so that Sigma_F = Sigma_S. [QUOTE] The other question is: when I define a "boundary load" at the interface surface, the "fluid-solid interface boundary" is still active and is not overwritten. My boundary load will be the load applied by the fluid on the solid surface. Since my fluid domain is supposed to respond based on the solid domain that surrounds it, I felt that I should use prescribed mesh displacement at the interface with dx=u_solid etc. Only then the fluid will move in response to the displacement of the solid . Please let me know if I am right in using the "prescribed mesh displacement". However, when I use prescribed mesh displacement at the interface, then the "fluid-solid interface boundary" condition is overwritten. So how does COMSOL know where the interface is? I was thinking that this is the condition that tells COMSOL about the "wall" location and hence determines the no slip boundary condition. But if interface BC is overwritten, how does it solve the problem? [/QUOTE] You should not apply anything else that FSI boundary. See above. Danial -- Comsol 4.1 Ubuntu 10.04.1

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月27日 GMT-4 03:59

I have a solid domain and a fluid domain and am trying to model FSI. Both the domains have circular cross-section. Solid domain is a hollow cylinder while the fluid domain is a cylinder concentric with the solid domain and located inside the solid domain. Its basically the 3D "pipe flow" problem with FSI.


I remember there was a blood FSI model in the comsol library, which should be similiar to yours.


COMSOL as a default has "fluid-solid interface boundary" listed in the boundary conditions. This boundary condition (BC) selects the surfaces that are at the intersection of the fluid and the solid domains which are the "interface surfaces". This BC infact overwrites the wall BC. So I assumed that it takes over the conditions that signify the wall (no slip BC). Am I right?


Yes, it does. FSI condition has two boundary conditions, one is that the velocities of the wall and liquid are equal (no-slip condition, and displacement in time is equal), and second that the Cauchy stresses of fluid and structure on the two domain are in equilibrium, so that Sigma_F = Sigma_S.



The other question is: when I define a "boundary load" at the interface surface, the "fluid-solid interface boundary" is still active and is not overwritten. My boundary load will be the load applied by the fluid on the solid surface.

Since my fluid domain is supposed to respond based on the solid domain that surrounds it, I felt that I should use prescribed mesh displacement at the interface with dx=u_solid etc. Only then the fluid will move in response to the displacement of the solid . Please let me know if I am right in using the "prescribed mesh displacement".

However, when I use prescribed mesh displacement at the interface, then the "fluid-solid interface boundary" condition is overwritten. So how does COMSOL know where the interface is? I was thinking that this is the condition that tells COMSOL about the "wall" location and hence determines the no slip boundary condition. But if interface BC is overwritten, how does it solve the problem?


You should not apply anything else that FSI boundary. See above.

Danial



--
Comsol 4.1
Ubuntu 10.04.1
[QUOTE] I have a solid domain and a fluid domain and am trying to model FSI. Both the domains have circular cross-section. Solid domain is a hollow cylinder while the fluid domain is a cylinder concentric with the solid domain and located inside the solid domain. Its basically the 3D "pipe flow" problem with FSI. [/QUOTE] I remember there was a blood FSI model in the comsol library, which should be similiar to yours. [QUOTE] COMSOL as a default has "fluid-solid interface boundary" listed in the boundary conditions. This boundary condition (BC) selects the surfaces that are at the intersection of the fluid and the solid domains which are the "interface surfaces". This BC infact overwrites the wall BC. So I assumed that it takes over the conditions that signify the wall (no slip BC). Am I right? [/QUOTE] Yes, it does. FSI condition has two boundary conditions, one is that the velocities of the wall and liquid are equal (no-slip condition, and displacement in time is equal), and second that the Cauchy stresses of fluid and structure on the two domain are in equilibrium, so that Sigma_F = Sigma_S. [QUOTE] The other question is: when I define a "boundary load" at the interface surface, the "fluid-solid interface boundary" is still active and is not overwritten. My boundary load will be the load applied by the fluid on the solid surface. Since my fluid domain is supposed to respond based on the solid domain that surrounds it, I felt that I should use prescribed mesh displacement at the interface with dx=u_solid etc. Only then the fluid will move in response to the displacement of the solid . Please let me know if I am right in using the "prescribed mesh displacement". However, when I use prescribed mesh displacement at the interface, then the "fluid-solid interface boundary" condition is overwritten. So how does COMSOL know where the interface is? I was thinking that this is the condition that tells COMSOL about the "wall" location and hence determines the no slip boundary condition. But if interface BC is overwritten, how does it solve the problem? [/QUOTE] You should not apply anything else that FSI boundary. See above. Danial -- Comsol 4.1 Ubuntu 10.04.1

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月27日 GMT-4 11:28
Thanks Danial!

Sirisha
Thanks Danial! Sirisha

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年5月27日 GMT-4 11:28
Thanks Ivar!

Sirisha
Thanks Ivar! Sirisha

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2011年6月1日 GMT-4 14:05
"
Yes, it does. FSI condition has two boundary conditions, one is that the velocities of the wall and liquid are equal (no-slip condition, and displacement in time is equal), and second that the Cauchy stresses of fluid and structure on the two domain are in equilibrium, so that Sigma_F = Sigma_S"

Cauchy stresses are equal in magnitude but opposite in direction. In other words, Sigma_F = -Sigma_S

sirisha
" Yes, it does. FSI condition has two boundary conditions, one is that the velocities of the wall and liquid are equal (no-slip condition, and displacement in time is equal), and second that the Cauchy stresses of fluid and structure on the two domain are in equilibrium, so that Sigma_F = Sigma_S" Cauchy stresses are equal in magnitude but opposite in direction. In other words, Sigma_F = -Sigma_S sirisha

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.