|
[Sponsors] |
August 31, 2005, 18:37 |
Boundaries for pipeflow
|
#1 |
New Member
Chris Somerlot
Join Date: Mar 2009
Posts: 2
Rep Power: 0 |
Which boundaries should I use to model velocity distribution in a pipe?
I believe I need 'fixedValue' for pand U in the inlet, and 'calculated' at the outlet, but thats not working. Is there any further documentation for the 'calculated' boundary? [Moderator note: was originally located in the blockMesh forum] Last edited by wyldckat; April 27, 2019 at 22:09. Reason: see "Moderator note:" |
|
August 31, 2005, 18:45 |
Not a particularly good locati
|
#2 |
Senior Member
Hrvoje Jasak
Join Date: Mar 2009
Location: London, England
Posts: 1,907
Rep Power: 33 |
Not a particularly good location for this thread, but never mind :-)
The "calculated" boundary condition is just a place-holder for the situations where the code does not know what to do - for example when you are calculating a gradient of the field and you're not sure what the boundary conditions for that field should be. Also, you cannot specify both p and U at the same location for incompressible flows. In your case, the appropriate set of b.c.-s are fixedValue U and zeroGradient p at the inlet and zeroGradient U and fixedValue p at the outlet. Have a look at the tutorials, for example tutorials/simpleFoam/pitzDaily Enjoy, Hrv
__________________
Hrvoje Jasak Providing commercial FOAM/OpenFOAM and CFD Consulting: http://wikki.co.uk |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
UDF and boundaries | deneb | FLUENT | 1 | October 24, 2008 04:32 |
Pro-am boundaries | Nalleli | Siemens | 1 | July 13, 2005 02:08 |
boundaries | Peter | Siemens | 3 | June 17, 2004 17:14 |
Moving Boundaries | Olga | Siemens | 0 | September 5, 2002 11:20 |
DNS of fully developed turbulent pipeflow | Emad khalifa | Main CFD Forum | 0 | October 6, 2001 17:57 |