|
[Sponsors] |
September 10, 2007, 18:16 |
FSI and additional variable
|
#1 |
Guest
Posts: n/a
|
Hello, In ANSYS/CFX 11 I'm simulating flow as a result of solid deformation. The deformation repeats itself every second. I then want to simulate the behavior of an additional variable (say a dye in the fluid) for e.g. 30 sec. I now do this by recalculating the same mechanical deformation and fluid velocity every 'cycle' of 1 sec over and over again, the only difference is the additional variable. Is there a way to do this more efficient?
Thanks, Jorn |
|
September 10, 2007, 19:34 |
Re: FSI and additional variable
|
#2 |
Guest
Posts: n/a
|
Hi,
I can't think of a way of doing it more efficiently than just running it. If the flow is steady state you can turn the flow solver off and just run the additional variable solver but that won't work with a periodic flow like this. Glenn Horrocks |
|
September 27, 2007, 13:20 |
Re: FSI and additional variable
|
#3 |
Guest
Posts: n/a
|
Hi Jorn and Glenn,
While you won't be able to get around solving the flow field at each time step, you can get around the FSI and mechanical deformation repetition. The Ball Valve tutorial (21, I think) is essentially an FSI simulation that is setup entirely within CFX. It is the second part of this tutorial that is relevant. In particular, you can tell CFX to replace the mesh coordinates for all domain nodes using values from data files. This is done using junction box routines, as described in the tutorial. In your case, I'd recommend the following: * run one cycle with FSI and no dye AV * during that run, use your own junction box routine (i.e. not provided with the noted tutorial) to write a sequence of mesh files. - If you are running in serial, then the JB routine will be pretty simple. - Model your output based upon the format of the mesh files provided with the tutorial; you'll be able to re-use the JB routines provided with the tutorial when reading the meshes if you do this - Look at the JB routines provided with the tutorial to find out where the mesh coordinates are stored within CFX; once you have these data, you just write them to your mesh files * Now setup your run with the AV, and use the JB routines that are provided with the tutorial to read the meshes in If, by chance, you're running this in Linux, you'll need to use the Portland Group compiler. There is a note in the CFX installation documentation that highlights how the gnu compiler can not be used with JB routines that involve file I/O. Hope it helps. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Structural force in FSI mapping Fluent 12.0 | benjamon | FLUENT | 1 | October 24, 2011 11:32 |
FSI - Specified Mesh Displacement | Vinzent | CFX | 2 | September 17, 2010 08:09 |
CFX Additional Variable Transport Equation | Scott Nordsen | CFX | 3 | January 30, 2010 06:36 |
Fluid Structure Interaction (FSI) | Harendra | Siemens | 17 | February 20, 2005 14:38 |
How to do FSI using FLUENT? | Harendra | FLUENT | 0 | February 5, 2005 02:56 |