CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   ANSYS Meshing & Geometry (http://www.cfd-online.com/Forums/ansys-meshing/)
-   -   Structured Meshing Query (http://www.cfd-online.com/Forums/ansys-meshing/82960-structured-meshing-query.html)

Bennp2000 December 9, 2010 21:47

Structured Meshing Query
 
Hi,
First post here.

I'm in my final year and just tidying up a lot of simulation work that I've done over the last few years regarding solute transport through urban drainage networks.

A lot of my work is concerning manholes.

So far I've only had around a handful of geometries to try, all of which had associated lab studies that allowed me to verify my results (the representation of the jet strength being key).
For each manhole (effectively two cylinders [one thick one thin] intersecting perpendicular to each other, I've generated a structured grid.
I'm assuming the free surface in the manhole is fixed and using a non slip wall to represent this.
First I've generated the manhole itself (to start with the same height as the pipe), then a box the same height as the pipe but wider. I've then used this box to split the manhole into two hemispheres and a curved box (deleting the excess volumes). I then generate the pipe, and use this to split the curved box.

Boundary conditions are:
Velocity inlet
Pressure outlet (either end of the pipe)
Wall (named lid) to represent the free surface

I then specify tri-elements in the vertices where the pipe intersects with the curved box (two faces, one entering, one leaving the manhole).

Next I use quad pave to place approximately 120 cells on the pipe face on the curved box, I then quad/tri pave the surrounding face and then use this to cooper along the pipe, box and then the extremeties.

Up until now this has worked fine but all of a sudden in a new geometry I can't get it to work. I can follow the above but I end up with negative volume errors and genuinely I can't seem to get a good mesh in place.

I'm rapidly running out of simulation time and have about 50 cases to do quicky so any help would be much appreciated.

I'm currently looking into tet meshing but it seems that brings me back to my initial problem of the solver (Fluent 12) over representing the jet.

Thanks for reading this long (and probably confusing) first post.
:confused:


All times are GMT -4. The time now is 07:55.