MotionSmoother

micha · May 30, 2007, 06:54

Hi,

actually I'm experimenting with some optimization in Foam. I realized a motion of my boundary points with the help of the motionSolver and the direct editing of boundary movements in cellMotionU. now I've got some problems with angels over 90degree where the face to point interpolation causes some natural errors. to fix this I thought of using the motionSmoother but I found no documention or example. did anyone know how to use it, or knows how to fix this a better way?

best regards
micha

virginie_e · March 20, 2009, 06:08

Hello,

I am also interested in the use of motionSmoother. So if anyone would be ready to provide some information, I would be very interested too!

Thank you in advance.

Virginie

elorriaux · March 20, 2009, 08:28

Hello,

I'm also getting strong problems with patch interpolators around corners. For the moment, i've found a solution by removing all the interpolators and by using coincident meshes with vertex based motion solvers. I just copy the needed values from one patch to the other. If you can use coincident meshes, it's surely the easiest way.

I'll take a look at motionSmoother, i've never used it at the moment.

Etienne.

virginie_e · March 27, 2009, 05:24

Hello,

I am implementing an algorithm for mesh smoothing based on the article of S. Giuliani (1982) "An algorithm for continuous rezoning of the hydrodynamic grid in arbitrary lagrangian-eulerian computer codes". I have implemented it so far in 2D in interTrackFoam but I think it can easily be transferred to other dynamic mesh solvers.
It is not very "clean" programming in the sense that I did not check my code to make it faster and so on, but it seems quite robust.
So if some people are interested, you can e-mail me. There might be some bugs left that I would not have seen but if you are interested, I would gladly handle you the sources.

I think I will release the sources on the forum when I have the 3D version ready too.

May 30, 2007, 06:54	Hi, actually I'm experiment	#1
micha New Member michael boehm Join Date: Mar 2009 Location: Karlsruhe,Germany Posts: 3 Rep Power: 17	Hi, actually I'm experimenting with some optimization in Foam. I realized a motion of my boundary points with the help of the motionSolver and the direct editing of boundary movements in cellMotionU. now I've got some problems with angels over 90degree where the face to point interpolation causes some natural errors. to fix this I thought of using the motionSmoother but I found no documention or example. did anyone know how to use it, or knows how to fix this a better way? best regards micha

March 20, 2009, 06:08		#2
virginie_e Member Virginie Ehrlacher Join Date: Mar 2009 Posts: 52 Rep Power: 17	Hello, I am also interested in the use of motionSmoother. So if anyone would be ready to provide some information, I would be very interested too! Thank you in advance. Virginie

March 20, 2009, 08:28		#3
elorriaux Member Etienne Lorriaux Join Date: Mar 2009 Location: Compiegne, France Posts: 45 Rep Power: 17	Hello, I'm also getting strong problems with patch interpolators around corners. For the moment, i've found a solution by removing all the interpolators and by using coincident meshes with vertex based motion solvers. I just copy the needed values from one patch to the other. If you can use coincident meshes, it's surely the easiest way. I'll take a look at motionSmoother, i've never used it at the moment. Etienne.

March 27, 2009, 05:24		#4
virginie_e Member Virginie Ehrlacher Join Date: Mar 2009 Posts: 52 Rep Power: 17	Hello, I am implementing an algorithm for mesh smoothing based on the article of S. Giuliani (1982) "An algorithm for continuous rezoning of the hydrodynamic grid in arbitrary lagrangian-eulerian computer codes". I have implemented it so far in 2D in interTrackFoam but I think it can easily be transferred to other dynamic mesh solvers. It is not very "clean" programming in the sense that I did not check my code to make it faster and so on, but it seems quite robust. So if some people are interested, you can e-mail me. There might be some bugs left that I would not have seen but if you are interested, I would gladly handle you the sources. I think I will release the sources on the forum when I have the 3D version ready too.