[CIG-LONG] Boundary conditions

[Walter Landry]

nicolas riel <nicolas.riel at free.fr> wrote:
> Hello everyone,
> 
> I was working on a mantelic plume with Gale 1.4 but in the last version of
> Gale the boundary conditions does not reacts in the same way.
> Here is the description of the old comportement:
> 
> In my 3D model, I have 2 materials: the mantle and a small and thin cylinder
> of hot and less dense plume like material laid on the bottom boundary.
> I set the "vy" bottom BC to 0 (same for other BC except for the top), and the
> initialParticleShape (In the splitting routine) to BottomBoundaryShape.

All of the machinery around initialParticleShape is gone.  However,
you should remember to specify inflow conditions when creating
particles.  Section 5.6, step 5 describes how to do this.  There is an
example in input/cookbook/viscous_inflow.xml.

> Because, my plume material at the bottomBoundary was lighter than the
> surrounding material, the plume particles were going up and new particles were
> automattically created to fill the gap (freely adding material to the model
> with direct dependency on the plume buyancy).It was a very efficient way to
> study plume ascent.
> 
> However this is not working anymore because there is no creation of new
> particle anymore.
> 
> I was wondering if there was a possibility to keep a "vy" = 0 and in the same
> timing allowing creation of new particles at boundaries ?

I do not quite understand what you mean by a vy=0 boundary.  Does that
mean that there is no inflow across the boundary?  In that case, you
should not need to worry about inflow at all.  If you are still having
problems, can you send your input file?

Cheers,
Walter Landry
walter at geodynamics.org