CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   FSI-time dependent load in cfx10.0 (https://www.cfd-online.com/Forums/cfx/24522-fsi-time-dependent-load-cfx10-0-a.html)

siavashghassemi September 11, 2007 11:46

FSI-time dependent load in cfx10.0
 
hello, i am working on a fsi simulation in cfx10.0.load is a pressure like "sin 5t".(t is time) but cfx in post processing doesn't accept t.how can i apply this load?should i write a fortran program?can anyone guide me? thanks.

V. Kumar September 12, 2007 06:12

Re: FSI-time dependent load in cfx10.0
 
How can you define loads in CFX-POST? In order to define load or a boundary condition (B.C) you have to go either to ANSYS or CFX-PRE. If it is load on the structure you can edit your ANSYS input like following under /SOLU or /PREP7:

************

nsel,s,..... !!with nsel select you nodes where you want to apply time-dependent load

*dim, press,table,100,1 !define a variable array e.g. press of length 100

delt=1e-4 !define size of time-step

FREQ=10 !define freqency e.g. 10 Hz

pi = 4*atan(1) ! define your Pi or pi = 3.14157

omega = 2*pi*FREQ

p_max = 10 ! amplitude of the load

*do,tt,0,100,1

press(tt,0)=tt*delt

press(tt,1)=p_max*sin(omega*tt*delt)

*enddo

SF,all,PRES,%press% ! apply pressure load vector on selected nodes

*********

If you want to apply a B.C on the fluid side, this can be done in CFX-PRE. In CFX-PRE a time-dependent B.C can be set by CEL which is quite easy using variable "t". Refer to CFX manuals for this.


siavash ghassemi September 15, 2007 11:28

Re: FSI-time dependent load in cfx10.0
 
as you wrote i could apply a time dependent b.c in cfx-pre with no error.("sin t" as a mesh displacement in y direction).but when i want to obtain a solution in cfx-solver(start run) an error massage displays that "t is not an angle"! how can i solve this problem? thanks, s.ghassemi

V. Kumar September 16, 2007 04:28

Re: FSI-time dependent load in cfx10.0
 
ofcourse "t" is not an angle.

Here is an example:

pi = 4*atan(1)

freq = 100 [s^-1]

theta = 2 * pi * freq * t

disp = 0.01 [m] sin(theta)

this should work.



All times are GMT -4. The time now is 10:57.