CFD Online Logo CFD Online URL
www.cfd-online.com
[Sponsors]
Home > Forums > Software User Forums > ANSYS > FLUENT

UDS and Periodic BC

Register Blogs Members List Search Today's Posts Mark Forums Read

Reply
 
LinkBack Thread Tools Search this Thread Display Modes
Old   July 19, 2002, 03:12
Default UDS and Periodic BC
  #1
iker
Guest
 
Posts: n/a
Hello.. everybody....

I have a diffusio convection problem, and I'm trying to solve it with UDS. In fact, I'd like to set PEriodic Boundary conditions for the Scalar transport..

I wonder if someone has already tried it before,,and any suggestions to manage it are welcome.

thans from Switzerland...

iker

  Reply With Quote

Old   July 23, 2002, 09:58
Default Re: UDS and Periodic BC
  #2
iker
Guest
 
Posts: n/a
hello.. hello..I've changed the mail adresse..

once again .. Does anybody try to set Periodic BC for a User DEfined Scalar variable... in fact I'm working with a Diffusion/Convection stationary problem with a corrugated geometry...

thanks.. I need someone's suggestions..

iker

  Reply With Quote

Old   July 23, 2002, 20:58
Default Re: UDS and Periodic BC
  #3
Matt
Guest
 
Posts: n/a
Are you asking how to set a periodic boundary condition? Link the faces/edges in Gambit and then select both faces/edges in the boundary conditions panel and select "periodic."

Is this what you are looking for?

Matt
  Reply With Quote

Old   July 24, 2002, 03:34
Default Re: UDS and Periodic BC
  #4
iker
Guest
 
Posts: n/a
Bonjour, Hi Matt..

thanks a lot for the response...

My questions is not concerning how to do the mesh. I've already done it.. it's more concerning a solver problem (I'm working with fluent 5.6)

In fact, my geometry is a corrugated channel(like a period of a cosinus signal). This channel lies inside a straight tube. So I have a Navier Stokes with transport diffusion problem. I'm working with UDS for solving the tranport/Diffusion.

As boundary conditions I'd like to set Periodic COnditions trough the streamlines (like a cosinus signal with periodic geoemtry).

I run the case (laminar and stationary case) but afer a hundred of iterations the residual error starts oscilating... without geting the convergence...

So I do not know if I have to run a Turbulent case, or if I made a mistake somewhere, or if it's possible to set Periodic Boundary conditions working with UDS...

thanks once again Matt,

from lausanne..

iker
  Reply With Quote

Old   July 25, 2002, 13:28
Default Re: UDS and Periodic BC
  #5
Matt
Guest
 
Posts: n/a
iker, The oscillating residuals is not uncommon. There are a number of issues surrounding why they do that. I went to the FLUENT Users Group Meeting and they informed me that there are often problems in the continuity equation that causes the oscillations. This problem happens if you have a large gradient in a value. (ie, if you have an inlet condition on your pipe and the you set a no slip condition on a wall.) My suggestion is to monitor other things in your model. I often monitor the mass-weighted average of the outflow velocity, or the average temperature on the outflow. These should level off even though your residuals don't.

Matt
  Reply With Quote

Old   July 31, 2002, 07:56
Default Re: UDS and Periodic BC
  #6
iker
Guest
 
Posts: n/a
Hello Matt,

I'm not sure I've understood what you proposed me..when you talk about displaying or monitoring other variables you're proposing to monitor them during solving processus???

Anyway, even if I achieve to display them, continuity residual curve will continue to oscillate??

could you give some more details concerning your proposition please?? t

thanks once again..

iker . .from geneva-lausanne
  Reply With Quote

Old   August 1, 2002, 01:11
Default Re: UDS and Periodic BC
  #7
Matt
Guest
 
Posts: n/a
Yes, the continuity variable will oscillate even though you are monitoring other variables. You see, residuals are only one way to monitor convergence. If you go to Solve --> Monitors --> Surface/stastistic/force/etc, you can define other things to monitor as a function of iteration number. (ie-wall shear stress or a mass-weighted average exit velocity). Usually when you define one of these monitors, you will see that they will converge to a solution. If this is converged, the rest of the model is likely converged even though your residuals oscillate. I know what you are experiencing. It has happened to me on a few occasions. However, I felt that my solution was correct by simply monitoring other parameters.

Matt-from Provo, UT--USA BYU Graduate Student
  Reply With Quote

Reply

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are On
Refbacks are On



All times are GMT -4. The time now is 05:42.