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 The information provided may be out of date.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.

Moving wall boundary condition

Please login with a confirmed email address before reporting spam

Hi,

in fluid flow simulations with a moving wall boundary condition (a rotation of a circular surface region), I get strong pressure oscillations at the edge of the rotating part of the surface (where the rotating part is in contact with the non-rotating part), that even do not disappear for higher mesh densities. Nevertheless, the velocity distribution seems okay.

Can anyone please comment on that effect?

Thanks, Marcel

10 Replies Last Post Mar 25, 2014, 3:03 p.m. EDT
Luke Gritter Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 4, 2014, 4:42 p.m. EST
Marcel,

If the rotating boundary contacts the non-rotating boundary, you may be specifying both a non-zero value and a zero value for one or more velocity components at the interface of the boundaries. If so, this (non-physical) inconsistency will tend to produce non-physical results near this location.

--
Luke Gritter
AltaSim Technologies
Marcel, If the rotating boundary contacts the non-rotating boundary, you may be specifying both a non-zero value and a zero value for one or more velocity components at the interface of the boundaries. If so, this (non-physical) inconsistency will tend to produce non-physical results near this location. -- Luke Gritter AltaSim Technologies

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 5, 2014, 10:15 a.m. EST
Hello Luke,
thanks for your comment. In my case, the non-rotating part is subjected to a symmetry condition (u dot n = 0), so that should be fine. Probably, the high velocity gradients (see attached) cause the fluctuating pressure distribution (also attached). Changing the mesh does not change anything qualitatively, but in pressure quantities.

The question arises if I can trust in the velocities?

Best regards,
Marcel
Hello Luke, thanks for your comment. In my case, the non-rotating part is subjected to a symmetry condition (u dot n = 0), so that should be fine. Probably, the high velocity gradients (see attached) cause the fluctuating pressure distribution (also attached). Changing the mesh does not change anything qualitatively, but in pressure quantities. The question arises if I can trust in the velocities? Best regards, Marcel


Luke Gritter Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 5, 2014, 11:03 a.m. EST
Marcel,

I can't make out what you're doing in the model from the pictures, but the pressure distribution certainly appears to be erroneous, so I would not trust any of the results. Something in the model setup is generating this non-physical behavior. From the pictures, it appears that you may have fluid on both sides of a moving wall boundary. If so, this may be contributing to the strange pressure results.

--
Luke Gritter
AltaSim Technologies
Marcel, I can't make out what you're doing in the model from the pictures, but the pressure distribution certainly appears to be erroneous, so I would not trust any of the results. Something in the model setup is generating this non-physical behavior. From the pictures, it appears that you may have fluid on both sides of a moving wall boundary. If so, this may be contributing to the strange pressure results. -- Luke Gritter AltaSim Technologies

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 5, 2014, 3:45 p.m. EST
Hello,

yes there is fluid also around the rotating part.

I used different meshs and solvers without success. I will try an easier material model to further check that out.

Marcel
Hello, yes there is fluid also around the rotating part. I used different meshs and solvers without success. I will try an easier material model to further check that out. Marcel

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 22, 2014, 9:33 p.m. EDT

Marcel,

If the rotating boundary contacts the non-rotating boundary, you may be specifying both a non-zero value and a zero value for one or more velocity components at the interface of the boundaries. If so, this (non-physical) inconsistency will tend to produce non-physical results near this location.

--
Luke Gritter
AltaSim Technologies


Hey Luke,

I am confronted with exactly this problem. Do you know how to solve it? I tried to solve it with a function for the moving boundary. At the edge, the velocity of the moving wall is 0 and for the rest of the moving boundary it has a velocity, but the inconsistency still exists. Do you have another idea?

Thank you very much!
Sabine
[QUOTE] Marcel, If the rotating boundary contacts the non-rotating boundary, you may be specifying both a non-zero value and a zero value for one or more velocity components at the interface of the boundaries. If so, this (non-physical) inconsistency will tend to produce non-physical results near this location. -- Luke Gritter AltaSim Technologies [/QUOTE] Hey Luke, I am confronted with exactly this problem. Do you know how to solve it? I tried to solve it with a function for the moving boundary. At the edge, the velocity of the moving wall is 0 and for the rest of the moving boundary it has a velocity, but the inconsistency still exists. Do you have another idea? Thank you very much! Sabine

Luke Gritter Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 24, 2014, 9:35 a.m. EDT
Sabine,

In most cases, the best approach is to include a small boundary with a Slip BC between the No Slip BC and the Moving Wall BC.

--
Luke Gritter
AltaSim Technologies
Sabine, In most cases, the best approach is to include a small boundary with a Slip BC between the No Slip BC and the Moving Wall BC. -- Luke Gritter AltaSim Technologies

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 24, 2014, 10:43 a.m. EDT
Hey Sabine and Luke,

I have a slip condition around the moving wall boundary, but the pressure characteristics remains the same. I also tried ramping the velocity at the edge of the moving wall without success.

Any comments would be appreciated!!

Best regards,
Marcel

Hey Sabine and Luke, I have a slip condition around the moving wall boundary, but the pressure characteristics remains the same. I also tried ramping the velocity at the edge of the moving wall without success. Any comments would be appreciated!! Best regards, Marcel

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 24, 2014, 1:36 p.m. EDT
Hey Luke,

Thank you very much. I included a slip boundary between the no slip wall and the velocity wall (not an interior wall, but next to the moving wall, on the surface). Is that what you meant? Now I get the right maximum value of the velocity, but the distribution close to the no slip does not seem correct. You can find a screenshot attached.
Do you know why that happens?

Thanks
Sabine
Hey Luke, Thank you very much. I included a slip boundary between the no slip wall and the velocity wall (not an interior wall, but next to the moving wall, on the surface). Is that what you meant? Now I get the right maximum value of the velocity, but the distribution close to the no slip does not seem correct. You can find a screenshot attached. Do you know why that happens? Thanks Sabine


Luke Gritter Certified Consultant

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 24, 2014, 2:21 p.m. EDT
Sabine,

It is difficult to discern what is going on from the picture. If you can post the model file, I will take a look at it.

--
Luke Gritter
AltaSim Technologies
Sabine, It is difficult to discern what is going on from the picture. If you can post the model file, I will take a look at it. -- Luke Gritter AltaSim Technologies

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago Mar 25, 2014, 3:03 p.m. EDT
I think I figured it out now. My slip-boundary was too small (<0.1mm) and my mesh was too coarse.

Thank you very much for your help
I think I figured it out now. My slip-boundary was too small (

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.