|
[Sponsors] |
April 30, 2002, 18:11 |
Ignore cells on partition boundary
|
#1 |
Guest
Posts: n/a
|
When looping over cells in a udf:
begin_cell_loop(c,t,...){ program statements } end_cell_loop(....) I am processing data from all cells in a given zone. However, I am running udf in parallel. It seems that for parallel, the loop runs over all cells in each partition plus cells that it groups as being on the partition boundary. Ultimately, this means that, if my partition is 1,000,000 cells total, and 20,000 cells lie on a partition boundary, the loop runs over 1,200,000 cells...I assume it is counting those cells that lie on the partition boundary twice. Therefore I want to ignore cells that lie on the partition boundary. Does anyone know how to do this? |
|
May 2, 2002, 22:57 |
Re: Ignore cells on partition boundary
|
#2 |
Guest
Posts: n/a
|
Try
begin_cell_loop_int(c,t) { } end_cell_loop_int(c,t) this only loops over the interior cells! Greg |
|
May 8, 2002, 07:10 |
existing for non-interior cells, too?
|
#3 |
Guest
Posts: n/a
|
hi Greg!
is there a similar looping macro for non-interior cells, too? Christian |
|
May 8, 2002, 09:11 |
Re: existing for non-interior cells, too?
|
#4 |
Guest
Posts: n/a
|
Good question - don't know. Ask Fluent. Have a look in the udf header files in your Fluent src folder. Maybe there are some there too.
Greg |
|
May 8, 2002, 11:50 |
Re: existing for non-interior cells, too?
|
#5 |
Guest
Posts: n/a
|
Thank you Greg!
The question if they know this at FLUENT.Inc is in deed a good one. I have asked myself very often why there are so many things missing in the UDF-manual. I think they just WANT to be asked....;-) Thanks anyway, Christian |
|
May 8, 2002, 21:37 |
Re: existing for non-interior cells, too?
|
#6 |
Guest
Posts: n/a
|
Well actually, the Fluent 6 manual is now very good. I was doing this stuff two years ago with the Fluent 5 udf manual and that was really difficult!
There's no way they will document everything, because in fact these macros etc are actaully the code of Fluent so I think its fair enough that they don't describe it all, but rather describe the most commonly required code pieces. You can often find a lot out from the .h header files and then work it out from there. They have pretty good technical support if you can get a good link to them through your distributor. Admittedly, the code interface in Fluent 4 was easier to work out yourself because it used Fortran and couldn't use complex data structures, which Fluent 5/6 do with C. Greg |
|
May 9, 2002, 06:02 |
Re: existing for non-interior cells, too?
|
#7 |
Guest
Posts: n/a
|
hi greg!
it seems we are using different Fluent 6.0 manuals. Mine has got written 6.0 at the top, but it is almost the same as the 5.0 manual (at least the udf-section). they even failed to get the errors out. I think i cannot access the udf.h to have a look into it since our administrator is back, but thank you for the hint. christian |
|
May 11, 2002, 23:12 |
Re: existing for non-interior cells, too?
|
#8 |
Guest
Posts: n/a
|
Have you looked at the separate Fluent 6 udf manual - this is very different from the version 5 manual??
Have you installed all the documentation from the documentation CD?? If you follow the setup it should install all the Fluent documentation and enable you to view the manuals using your web browser. If you haven't done this, then I strongly recommend trying to get this setup. Greg |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Water subcooled boiling | Attesz | CFX | 7 | January 5, 2013 04:32 |
inlet velocity boundary condition | murali | CFX | 5 | August 3, 2012 09:56 |
[snappyHexMesh] snappyHexMesh won't work - zeros everywhere! | sc298 | OpenFOAM Meshing & Mesh Conversion | 2 | March 27, 2011 22:11 |
[Gmsh] Import problem | ARC | OpenFOAM Meshing & Mesh Conversion | 0 | February 27, 2010 11:56 |
Convective Heat Transfer - Heat Exchanger | Mark | CFX | 6 | November 15, 2004 16:55 |