Oszillating solution while solving

Log Out | Topics | Search Moderators | Register | Edit Profile

Author Message   matthiasbiese New member Username: matthiasbiese Post Number: 1 Registered: 05-2010 Posted on Thursday, May 13, 2010 - 10:33 am:    Hello, i try to simulate the breake-throug in a chromatography, and by that i get a problem with the concentration profile along the x-axis. I start with a smooth "injection" to avoid a hard pulse. But while solving the equation, the slope of the curve along the X-axis does a very hard change. And with that, i think the solution starts to "vibrate". Is it just because i use the student version, and there are not enough nodes? The Handbook told me how to avoid rectangular boundary conditions, but i don't know enougth to handle hard changes in the solution itself... breakthroug-curve in a chromatography 1d Equilibrium Model.pde (2.1 k)   matthiasbiese New member Username: matthiasbiese Post Number: 2 Registered: 05-2010 Posted on Thursday, May 13, 2010 - 11:39 am:    Me again. I found the FRONT statement. It seems to be a good way to avoid the oszillations. I think that statement forces FlexPDE to make a very dense mesh at "interesting" areas. And with that, FlexPDE does not make to big steps while walking along the slope of my function. Hope, this is the right way to solve my problem. When there is a better way, please tell me.   rgnelson Moderator Username: rgnelson Post Number: 1362 Registered: 06-2003 Posted on Thursday, May 13, 2010 - 03:40 pm:    I don't see any oscillation in this problem except for a little ringing at the foot of the steep front. And for this, FRONT is exactly the right thing.