Re: [Gfs-users] Size of the box and embedded objects...
Brought to you by:
popinet
From: Geordie M. <gdm...@fr...> - 2008-10-26 22:18:05
|
On Mon, Oct 27, 2008 at 8:58 AM, pascal CASTRO <pas...@gm...> wrote: > Hello ! > > I notice that using PhysicalParams { L = 2.0 } do not change the relative > size of the embedded objects. > > For example, If I insert a object of length 0.1 in the GfsBox of length 1 > (default value) the object is one tenth of the length of the GfsBox. But now > if I use PhysicalParams { L = 2.0 }, the same object is still one tenth of > the length of the GfsBox !!!! So in the simulation is own length was double. > > I don't know your opinion about this behaviour but personally I would prefer > that the object was scaled at the it's relative size when using > PhysicalParams { L = X.X } I don't think this is right. Could you recheck? In the example I discussed on this list in July I had a sphere or cylinder of unit diameter in a box with PhysicalParams { L = 8.0 }, so the obstacle wouldn't have fitted inside the domain if it had also been scaled with L. The same unit obstacle diameter approach is taken in the axisymmetric test http://gfs.sourceforge.net/tests/tests/axi.html |