[MOAB-dev] Possible bug in 2-D quadratic elements

Wollaber, Allan B. awollaber at anl.gov
Thu Jan 15 13:24:08 CST 2009


Thanks again, Jason. I added the following lines to my code:
      ! Determine the topology of the side
      call iMesh_getEntTopo(%VAL(iMesh), %VAL(sides(j)), sideTopology,
err)
      ! Get MBCN type equivalent to the iMesh topology 
      call iMesh_MBCNType(%VAL(sideTopology), sideMBCNtype)
      ! Get the number of corner vertices in this side
      call  MBCN_VerticesPerEntity(%VAL(sideMBCNtype),
sideCornerVerticesSize)
And changed the call to:
      call MBCN_SideNumberULong(adjElemConn,   %VAL(elementMBCNtype),
sideVertices, &
           %VAL(sideCornerVerticesSize), %VAL(numDimensions-1), tmpInt,
idum, idum)
 
And this seems to have fixed the problem I was having (checked HEX27,
TRI6, and QUAD9 so far).

- Allan


-----Original Message-----
From: Jason Kraftcheck [mailto:kraftche at cae.wisc.edu] 
Sent: Thursday, January 15, 2009 12:54 PM
To: Wollaber, Allan B.
Cc: moab-dev at mcs.anl.gov; Smith, Micheal A.; Tautges, Timothy J.
Subject: Re: [MOAB-dev] Possible bug in 2-D quadratic elements

Wollaber, Allan B. wrote:
> Jason (Tim),
> 
> The connectivity indices are now the correct length for all the 
> elements that I've tried (HEX8/20/27, QUAD4/8/9, TRI3/6).  This is
great!
> 
> I have a new question now, however: At one point in the mesh 
> processing I must determine the local reference number of a facet (in 
> a sideset with the NEUMANN_SET tag) with respect to the element it is
bounding.
> To do this, I use the MBCN_SideNumber routine.  This has worked well 
> for every 3-D element that I've tried (hexes, tets) and for the linear

> 2-D elements (tri3, quad4), but not for the quadratic 2-D elements 
> (tri6, quad8/9), even before the change that was just made to the
repository.
> For the quadratic 2-D elements, the local facet number is always 
> returned as "1".  The following code snippet should help illuminate 
> how I am doing this:
> 

The 'child_num_verts' argument to SideNumber should be the number of
corners, not the number of nodes.  The function works the same
regardless of the number of nodes in the element.

- jason
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.mcs.anl.gov/mailman/private/moab-dev/attachments/20090115/a5b6bdae/attachment.htm>


More information about the moab-dev mailing list