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 The information provided may be out of date.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.

boundary conditions

Please login with a confirmed email address before reporting spam

Dear all,

I have an incompressible fluid flow problem. If i have only inlet but no outlet in the model (the mass conservation is fulfilled by expanding volume), which boundary condition (velocity or pressure) is best to be implemented?

In fluid flow (laminar flow) model, is it a reference pressure must be defined somewhere in the model? If yes, is that mean I hv to find a way to define reference pressure if i use velocity as inlet condition?

What if FSI physics - is there any reference pressure needed?

6 Replies Last Post 2012年3月11日 GMT-4 07:50
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 2012年2月28日 GMT-5 07:50
Hi

look at the PDE's if the pressure part of your dependent variables you must fix it somewhere, otherwise you results can slide +/- a constant and your solution is not unique. But this fixed value can also bee seen as a "gauge" value, and your absolute pressure is defined by an arbitrary constant to add to the fixed gauge value

--
Good luck
Ivar
Hi look at the PDE's if the pressure part of your dependent variables you must fix it somewhere, otherwise you results can slide +/- a constant and your solution is not unique. But this fixed value can also bee seen as a "gauge" value, and your absolute pressure is defined by an arbitrary constant to add to the fixed gauge value -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年2月28日 GMT-5 11:46
How were you going to model the expanding volume?

I think you just have to define another wall as an open boundary, or as an outlet and set the pressure there to be 0.

Technically the volume is increased without force and the pressure in that reservoir is zero.

Unless there are some mechanical properties related to the increase in volume of that reservoir. Than you will have to use FSI.

How were you going to model the expanding volume? I think you just have to define another wall as an open boundary, or as an outlet and set the pressure there to be 0. Technically the volume is increased without force and the pressure in that reservoir is zero. Unless there are some mechanical properties related to the increase in volume of that reservoir. Than you will have to use FSI.

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年3月1日 GMT-5 05:10
Thanks Ivar and Jaap!

What I know about open boundary is, it could play roles as both inlet and outlet so I am not sure is the suggested way (“define another wall as an open boundary, or as an outlet and set the pressure there to be 0”) suit in my case or not. Can I have ur opinion? Below is my case description:

The fluid filling into a closed chamber through inlet and there’s no outlet - no fluid should escaped out from the rigid chamber (attached condition.jpeg). In order to obey mass of conservation, I prescribed the volume expansion by using ODE function to define the increment of boundary’s length (instantaneous radius and height change) according to inlet flow.

The information I have are inlet velocity and volume expansion pattern. However, I can’t run simulation without define pressure somewhere. What I do now is, impose a pressure point constraint (constant pressure value) on the model. But it took me days to run the simulation!

Looking forward for your reply. Thanks a lot!
Thanks Ivar and Jaap! What I know about open boundary is, it could play roles as both inlet and outlet so I am not sure is the suggested way (“define another wall as an open boundary, or as an outlet and set the pressure there to be 0”) suit in my case or not. Can I have ur opinion? Below is my case description: The fluid filling into a closed chamber through inlet and there’s no outlet - no fluid should escaped out from the rigid chamber (attached condition.jpeg). In order to obey mass of conservation, I prescribed the volume expansion by using ODE function to define the increment of boundary’s length (instantaneous radius and height change) according to inlet flow. The information I have are inlet velocity and volume expansion pattern. However, I can’t run simulation without define pressure somewhere. What I do now is, impose a pressure point constraint (constant pressure value) on the model. But it took me days to run the simulation! Looking forward for your reply. Thanks a lot!


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 2012年3月1日 GMT-5 05:24
Hi

I believe there is a thread or two about a similar model, a few months ago, no outtlet, but a pressure point, and only an inlet and an exanding volume

My first rection is that your model BC should work if the solid can expand and interact with the fluid pressure, and that you have some velocity driven input

--
Good luck
Ivar
Hi I believe there is a thread or two about a similar model, a few months ago, no outtlet, but a pressure point, and only an inlet and an exanding volume My first rection is that your model BC should work if the solid can expand and interact with the fluid pressure, and that you have some velocity driven input -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年3月5日 GMT-5 21:40
Thanks Ivar. The simulation did run succesfully. However the result (very high pressure) makes me have a thought about the wall boudnary conditions set.

In the model, i used moving mesh coupled with fluid flow physics and the wall is set to no slip condition (i suppose to hv both moving and no slip condition). In the result, I can see wall displacement in spatial frame (although it does not set as moving wall).

Now I am not sure - the movement that I saw in the result is actually the displacemnt of moving mesh only or both (mesh and wall) in this setting? How can I check on this?

In v4.2a, it only allows me to choose either no slip or moving wall under wall boundary condition but not both.
I tried with moving wall only by setting the same velocity as moving mesh velocity, but convergence errors appear.

Anyone could help on this?

Thanks a lot!

Thanks Ivar. The simulation did run succesfully. However the result (very high pressure) makes me have a thought about the wall boudnary conditions set. In the model, i used moving mesh coupled with fluid flow physics and the wall is set to no slip condition (i suppose to hv both moving and no slip condition). In the result, I can see wall displacement in spatial frame (although it does not set as moving wall). Now I am not sure - the movement that I saw in the result is actually the displacemnt of moving mesh only or both (mesh and wall) in this setting? How can I check on this? In v4.2a, it only allows me to choose either no slip or moving wall under wall boundary condition but not both. I tried with moving wall only by setting the same velocity as moving mesh velocity, but convergence errors appear. Anyone could help on this? Thanks a lot!

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 2012年3月11日 GMT-4 07:50
If I can explain your problem with the analogy of blowing up a balloon. (corect me if I'm wrong)
The pressure in the balloon is not homogeneous and the pressure allong the balloon wall will change.
If there is no opposing force to the flowing liquid, it is just like flowing a liquid through a nozzle and I think the slosh tank example should help you. You define a volume force on the liquid and the liquid will change shape using the deformed mesh function.

However, if there is there is an opposing force to the liquid, the liquid is ment to flow in a certain shape, you will have to model the opposing force at that wall on the liquid, and you will need FSI for that I think. Unless you have a good way of describing the pressure of the wall on the liquid as a function of x y and t. I think though that that is very difficlut to describe.
If I can explain your problem with the analogy of blowing up a balloon. (corect me if I'm wrong) The pressure in the balloon is not homogeneous and the pressure allong the balloon wall will change. If there is no opposing force to the flowing liquid, it is just like flowing a liquid through a nozzle and I think the slosh tank example should help you. You define a volume force on the liquid and the liquid will change shape using the deformed mesh function. However, if there is there is an opposing force to the liquid, the liquid is ment to flow in a certain shape, you will have to model the opposing force at that wall on the liquid, and you will need FSI for that I think. Unless you have a good way of describing the pressure of the wall on the liquid as a function of x y and t. I think though that that is very difficlut to describe.

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.