Jonathan and Daniel,
We've managed to figure out how to generate quads using gmsh and the cell
numbering isn't as much of an issue, but we're still running into cell
volume problems. I've attached a simple example, in which I extrude a gmsh
file once, making one layer. The vertical spacing between vertices is 5,
except of one point were the spacing is 10. All cells have a basal area of
1 (1 by 1 in x-y plane), but the cell volumes I get are not correct, with
one cell having a volume of 7.5 and one having a volume of 2.5. These two
cells should have the same volume.
Does the top surface have to be defined by a plane?
Andy
----------------------
Andrew Reeve
Dept. of Earth Science
University of Maine
Orono, ME 04469
On Fri, 21 Mar 2008, Jonathan Guyer wrote:
Date: Fri, 21 Mar 2008 17:34:12 -0400
From: Jonathan Guyer <[EMAIL PROTECTED]>
Reply-To: [email protected]
To: Multiple recipients of list <[email protected]>
Subject: Re: FiPy vertex numbering convention
On Mar 21, 2008, at 5:06 PM, asreeve wrote:
cells in input file to be extruded labeled:
cell vertices
--- --------
1 1,5,6,2
2 2,6,7,3
3 3,7,8,4
Is this correct?
Depends. How are your faces defined? In order to extrude, your cells need to
be 2D polygons and your faces need to be 1D line segments. Is that what you
have?
Is there a place in the manual that indicates how
this should be done?
Probably the closest we have is in the class documentation in
fipy/meshes/pyMesh/grid2D.py
No, this isn't the most useful place for this, but it made some historical
sense.
I've also tried counterclockwise ordering but get
negative cell volumes...
What happens when you do clockwise ordering?
