XANSYS Message: 93026 [Go back to message list]
[bookmark on del.icio.us]
No rating yet
Subject: Re: Ansys FSI nightmare
Author: Anup Paul
Date: 2008-04-25 21:01:55Tan,
What is the end goal of this exercise? i.e. what information or data are you
trying to extract from this simulation. Do you really need a FSI analysis to
answer your questions? Can break up the problem into smaller practical
pieces
Simulating a buckling problem by itself is very challenging, so I am not
surprised that you are having problems with fluid coupled buckling problem.
Sequential is always better than simultaneous solving. The simultaneous
option should only be used if you have weakly coupled fields and it only
serves to save wall clock time for a very large FE & CFD model. Sequential
should be used for strongly coupled field [RFTM please]
I am not sure if you are gaining anything by using the CFX/MFX approach.
Since it is a laminar flow, I would think using FLOTRAN would be sufficient.
This might make life easier, since it is all within a single code and I
assume FLOTRAN will handle large deformations better than CFX.
Where are you having problems - in the structural solution or the CFD
solution? Usually large deformations cause problems with mesh morpher in
CFX.
I am no expert on bucking analysis, but have you thought of using an
explicit procedure for this problem. I have heard explicit solvers can
handle buckling problems better (probably with less accuracy and you have to
careful about dynamic effects). Maybe codes like LS-Dyna have enough fluid
mechanics to compute pressure drop due to area changes.
Regards,
Anup Paul
www.stress.com
This message is not provided as a Stress Engineering employee.
-----Original Message-----
From: xansys-bounces_at_xansys.org [mailto:xansys-bounces_at_xansys.org] On Behalf
Of Y Y Tan
Sent: Friday, April 25, 2008 4:30 PM
To: xansys_at_xansys.org
Subject: [Xansys] Ansys FSI nightmare
Hi guys,
Its now 1 week away from my dissertation deadline, I have posted up my
problem
several times and people here have helped me clarify alot of things. Yet, I
still can not get the results that I want, I really feel that my FE skills
are
letting me down or maybe its just how i approach things. I have ran my FE
model
for almost 300 times now for the past few months and I have really ran out
of
ideas what I can do.
Let me just explain my model again, I have a circular pipe, with a steady
state
laminar flow through it. The pipe walls are flexible and an external
pressure
is applied onto the pipe walls. Because the external pressure is greater
than
the internal pressure of the fluid, the pipe will buckle inwards and because
of
that, the flow will have to squeeze through the constriction and the local
pressure will drop because of Bernoulli, and as a result the pipe will
further
deform. This is a highly non-linear and highly coupled problem.
My question is, can Ansys actually solve this type of problem? I am
currently
using MFX and doing the sequential analysis where I run CFX first before I
run
Ansys. Would a simultaneous analysis be a better option?
I'm about to burst out in tears as I write this post, I really really hope
that
someone here could understand the situation which I'm going through and help
me
go through this modelling problem and consequently help save me from failing
my
degree. :(
Regards,
Tan Yi Yong
The University of Sheffield
E: mea05yyt_at_sheffield.ac.uk
^--------------------------------------------------------
| XANSYS - www.xansys.org |
| The Discussion List for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
^--------------------------------------------------------
^--------------------------------------------------------
| XANSYS - www.xansys.org |
| The Discussion List for users of ANSYS, Inc. Software |
| Hosted by PADT - www.padtinc.com |
^--------------------------------------------------------