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 The information provided may be out of date.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 Mesh vs. Rotating Machinery

Please login with a confirmed email address before reporting spam

Hi,

Modeling 2D pipe flow with some kind of flap in the middle of the tube.

I am experimenting with Moving Mesh vs. Rotating Machinery. I believe I'd like to have the control Moving Mesh appears to allow (the kind of mesh smoothing), but Rotating Machinery is the only that already includes a specific interface for rotation. If it is necessary to use Moving Mesh, how can I prescribe rotation only using dx, dy ? Can I make dx a function of the distance to a certain point? (effectively creating a rotation of some sort)

When using Rotating Machinery, the mesh doesn't look pretty (see attached image, with Rotating Domain on the left, identity pair being the line in the middle). The mesh doesn't line up properly, is that a problem? I added boundary layers and the solution improved (no longer an abrupt change of regime), but how much should I trust this?

Thank you all for input


2 Replies Last Post 1 juin 2012, 06:45 UTC−4
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 1 juin 2012, 05:08 UTC−4
Hi

moving mesh and rotating Machinery uses different concepts, the moving mesh allows for squeezable flexible mesh domain shapes, while rotating machinery uses a "fixed" circle to make a rolling" virtual contact boundary.

I'm not sure how to mix both, but its probably also possible, I have never tried

I use cylindrical coordinates to look after the RMMcase, but if the RMM case is moving around via an external moving mesh scheme I'm not sure one can get hte cylindrical coordinate system to follow, ... interesting anyhow

--
Good luck
Ivar
Hi moving mesh and rotating Machinery uses different concepts, the moving mesh allows for squeezable flexible mesh domain shapes, while rotating machinery uses a "fixed" circle to make a rolling" virtual contact boundary. I'm not sure how to mix both, but its probably also possible, I have never tried I use cylindrical coordinates to look after the RMMcase, but if the RMM case is moving around via an external moving mesh scheme I'm not sure one can get hte cylindrical coordinate system to follow, ... interesting anyhow -- Good luck Ivar

Please login with a confirmed email address before reporting spam

Posted: 1 decade ago 1 juin 2012, 06:45 UTC−4
Thank you, Ivar.


moving mesh and rotating Machinery uses different concepts, the moving mesh allows for squeezable flexible mesh domain shapes, while rotating machinery uses a "fixed" circle to make a rolling" virtual contact boundary.


I haven't mixed the two Physics yet, just looking into how each one works. The 'fixed' circle looks to me like a problem, if I am to embed the rotating circle (containing the flap) inside pipe flow. I would have to provide a fine mesh, else the resulting wake of the flap would smash into a 'regular' mesh upon exiting the rotating domain. And a meaningful flap would require a circle with a radius ~= the pipe height (and thus, the circle would almost touch the pipe boundaries). This is my fear, as I don't really know how it would work out.
Thank you, Ivar. [QUOTE] moving mesh and rotating Machinery uses different concepts, the moving mesh allows for squeezable flexible mesh domain shapes, while rotating machinery uses a "fixed" circle to make a rolling" virtual contact boundary. [/QUOTE] I haven't mixed the two Physics yet, just looking into how each one works. The 'fixed' circle looks to me like a problem, if I am to embed the rotating circle (containing the flap) inside pipe flow. I would have to provide a fine mesh, else the resulting wake of the flap would smash into a 'regular' mesh upon exiting the rotating domain. And a meaningful flap would require a circle with a radius ~= the pipe height (and thus, the circle would almost touch the pipe boundaries). This is my fear, as I don't really know how it would work out.

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.