2010/7/6 Daniel Wheeler <daniel.wheel...@gmail.com> > > Hi Benny, The oscillations may be a result of pressure velocity > decoupling. To avoid this on a collocated grid one needs to use an > interpolation method such as Rhie-Chow interpolation or a more modern > variant: > > < > http://apps.isiknowledge.com/full_record.do?product=WOS&search_mode=GeneralSearch&qid=6&sid=1c...@1f69fl81hkpdm&page=1&doc=1#output_options > > > > In FiPy this generally required getting the matrix diagonal and using > it as part of the interpolation. I can discuss more if you want to > pursue it further. > > I am not sure if it is possible to set up a staggered grid in FiPy in > a practical way. That may be another option. One grid would probably > have to be created in Gmsh in order to have the cell centers at the > same place as the face centers on the other grid. I have never tried > this, but it is certainly possible. > > Hope this helps. Cheers. >
I read this article and some other references, and updated the stokes cavity example to avoid the oscillations. I hope the example is updated like this, because although the flow lines don't change much most of the time, seeing the oscillations does not give confidence. All it attached to http://matforge.org/fipy/ticket/305 I attached a before and after of my experiment to show clearly how the oscillations are removed nicely. Greetings, Benny > On Tue, Jul 6, 2010 at 5:48 AM, Benny Malengier > <benny.maleng...@gmail.com> wrote: > > Hi, > > > > I am doing a flow field calculation like > > > http://www.ctcms.nist.gov/fipy/examples/flow/generated/examples.flow.stokesCavity.html > > > > However, I obtain spurious oscillations for the pressure solution. > Patankar > > in his works is very clear that a staggered grid should be used for the > > SIMPLE algorithm, in the example averaging is applied so as to work on > one > > grid, which I believe leads to these oscillations. > > > > Does anybody has an idea how best to alleviate this problem? I am > thinking > > about eg a CornerCellVariable for vx/vy, which would be the staggered > grid. > > In essense that means a mesh holds two grids then. It becomes complicated > > fast however. > > Perhaps other ideas? Or better to calculate the flow field with another > > tool? > > > > Greetings, > > Benny > > > > > > -- > Daniel Wheeler > >
