[Tobi]

Hello Foamers,

I have a question to some schemes in OpenFOAM.


grad-Scheme
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

What is the differences between

grad(U)         Gauss linear;
grad(U)         cellLimited Gauss linear 1;

Its clear that I use a limiter for the grad schemes. But does that mean, that the change of U^{n-1} to U^n is limited? Is my interpretation correct?

div-Scheme
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

Whats the difference between

div(phi, Z)           Gauss limitedLinear 1 0 1
div(phi, Z)           Gauss limitedLimitedLinear 1 0 1

I think the first scalar is for the limiter. 1=full limiter 0=non limiter. The values after that are bouding parameters? so that the variable Z is bounded between [0-1] ? is that correct? And the differences between both?


Any suggestion would be very nice.
Regard
Tobi

[immortality]

hi
can anyone answer to these questions?
thanks.

[Tobi]

Hi all,

results of different schemes can be found on my homepage (its unfortunatelly german but with graphics)

http://www.holzmann-cfd.de/index.php...rische-schemen

[ngj]

Hi Tobias,

Nice graphics Just a quick question from a non-German speaking person. Is it a pure advection problem, which you have been solving?

At least with a quick view on a couple of the cases, it seems that you are only changing the divergence scheme.

Thanks,

Niels

[akidess]

Niels, indeed the site states pure convection ("reine Konvektion").

[Tobi]

Hi,

yes I am only changing the div schemes to get influence of konvection influence ... "there is no diffusivity". That s necessary to get see what influence the different schemes are make in the equations.

[dkokron]

I found the following paper that includes some other results along the same lines as Tobi. The paper includes the step function and two other shapes.

http://powerlab.fsb.hr/ped/kturbo/Op...GammaPaper.pdf

Dan