CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Problem with unwanted oscillations (http://www.cfd-online.com/Forums/main/120076-problem-unwanted-oscillations.html)

 patovalades June 29, 2013 22:19

Problem with unwanted oscillations

Hello,
I am trying to write my own Finite Volume code for a very simple problem.
it is a mass transport problem on a pipe. 1D (axial) Unsteady state for the mass balance equation, only forced convection is considered. The Momentum equation on the pipe is considered plug flow (constant at all radius and all axial positions).

In the program I consider an impulse function (delta dirac) of compound A, I made an animation of a plot of Concentration vs Axial Position that changes with time and started and following the pulse (or the evolution of the concentration profile inside the tube).

The pulse moves in the expected way (forward) the width of the peak doesnt change (as expected since there is no axial dispersion), HOWEVER, on the trail behind the pulse I notice some "unwanted" oscillations...

I am using a 1st order in space approximation for the FV and 4th order Runge-Kutta Method. I don't know what to do to get rid of those oscillations, I already tried moving the segments length in the axial direction and the time step, but nothing happens...

I am new to this so all advices are welcome, however I would like to stress that I already checked and reckecked the logic behind the FV volume, and it is all right...maybe some problem with the numerical stability?

 patovalades June 29, 2013 22:33

1 Attachment(s)
Here I attach an image of the concentration profile inside the pipe. at time zero the pulse is at Z=0, this image however is taken at time= 22 seconds. The velocity in the pipe is about 0.7 m/s , so the pulse is at about 15-16 meters (about halfway through the pipe since the pipe is 27 m long).

 All times are GMT -4. The time now is 18:52.