CFD Online Discussion Forums

CFD Online Discussion Forums (
-   CFX (
-   -   Urgent..Needs for FSI modeling using ANSYS (

Eugene February 7, 2005 14:04

Urgent..Needs for FSI modeling using ANSYS
Hi!, everybody?

My name is Eugene and I am very interested in Fluid-structure interaction analysis. But it is really hard to start the study. There is a ANSYS8.0 and CFX5.7 available in school.

For example, I'd like to have a stress(or acceleration or pressure) change at some specific nodes by time (every second or less)of light pole by wind-induced vibration.

What do I have to start? Do I need to model structure using ANSYS8.0 and fluid model using CFX5.7 after importing the structure model from ANSYS8.0?

or opposite order?

I am not sure how these two programS work together?

or Do I need another program or higher version of programs? Some says I need MPCCI. I do not know where I can buy this or free sofeware..etc.

If anybody has a simple 3D (even 2D is ok) FSI model using ANSYS (&CFX)..could you send me the log file( or paste here (just log file). I also want to know the procedure of the modelling.

In addition, could you suggest any books websites that has Theory of FSI? I want to study the basic concept and theory like shape function in solid FEM.

Robin February 7, 2005 17:04

Re: Urgent..Needs for FSI modeling using ANSYS
It sounds like you want to do a transient FSI calculation. This will be available natively with the release of CFX-5.8 and ANSYS 10.0 this summer.

With the versions you currently have, you would require MPCCI and some manual effort. There is an example with the appropriate file in the CFX-5.7 installation in ...\examples\mpcci. Note that the native version will not require MPCCI.

Regards, Robin

Glenn Horrocks February 7, 2005 17:34

Re: Urgent..Needs for FSI modeling using ANSYS

If your light pole can be assumed to be a simple linear beam you might be able to do the motion calculations in CFX and not worry about an FEA code at all. Have a look at tutorial 20, Fluid structure interaction and mesh deformation. In this example the motion of the ball is a simple function of the forces acting on it, so the ball motion is calculated in CFX with no FEA code. You might be able to do a similar thing for your model.

Regards, Glenn Horrocks

antoine February 8, 2005 05:52

Re: Urgent..Needs for FSI modeling using ANSYS

Do you know how will work CFX and Ansys together if there is no MPCCI? I am interested in doing FSI myself. Will they become one product or will it still be two different products?


Robin February 8, 2005 09:59

Re: Urgent..Needs for FSI modeling using ANSYS
Hi Antoine,

The solvers remain as two separate solver codes, with communication done through sockets. At run time, the machine and TCP/IP port number of the Multiphisics solver is provided to the CFX-5 solver. Periodically, the solvers pass data through these sockets, but in between they can be running simultaneously on the same or separate computers (including different platforms, 32 or 64 bit, etc).

MPCCI was only necessary in the past because it provided a standard means for codes from different vendors to communicate. Since all the code is owned by a single company, ANSYS, there is no need for a third party anymore and the developers can implement the connection natively. The beta level MPCCI implementation that is available in CFX-5.7.x was created for consulting work and merged into that release as an interim solution.

Regards, Robin

antoine February 9, 2005 06:21

Re: Urgent..Needs for FSI modeling using ANSYS


seb March 23, 2005 11:11

FSI with Ansys
hye everybody,

I am trying to modelling a wind blade in 3d using a FSI method. Building my model, I have the error message saying "number of fluid/solid interfaces do not match for FSIN boundary condition" and actually I don't understand what is the problem. Does anybody have any idea to help me? Thank you very much. Sincerely. BERGER Sebastien

raghunath March 27, 2005 11:19

fw: Flotran CFD with Ansys
hi buddy,

well i am also engaged withthe work CFD Flotran of Ansys and during the FSI analysis I too have come across some doubts. well the help files written for FSI analysis are not ofthat help are of more confusing nature. well i am involved with axial piston pumps where there is a rotating barrel with equally spaced cylinders and butting a cylindrical plate,with a 40 micron gap between the two. My doubt is how to go about modelling the 40 micron distance between both the surfaces.Anybody please clarify.

thanks, With regards, Raghunath

All times are GMT -4. The time now is 11:56.