[CIG-MC] CitcomCU side BC question

[Ethan Fahy]

Hello all,

In either of the functions velocity_side_bc or velocity_refl_vert_bc, the
velocity BCs are set.  In the following section:

if(E->parallel.me_loc[1] == 0)
                {
                    E->VB[1][node1] = 0.0;
                    if((ii != 1) && (ii != E->mesh.noz))
                        E->VB[3][node1] = 0.0;
                }
                if(E->parallel.me_loc[1] == E->parallel.nprocx - 1)
                {
                    E->VB[1][node2] = 0.0;
                    if((ii != 1) && (ii != E->mesh.noz))
                        E->VB[3][node2] = 0.0;
                }

The normal component to the x-faces is set by assigning E->VB[1][node1,2].
The z-component at the x-faces is also set using E->VB[3][node1,2].  My
question is why is the y-component not set here as well?  The very same
thing happens when assigning the y faces:

if(E->parallel.me_loc[2] == 0)
                {
                    E->VB[2][node1] = 0.0;
                    if((ii != 1) && (ii != E->mesh.noz))
                        E->VB[3][node1] = 0.0;
                }
                if(E->parallel.me_loc[2] == E->parallel.nprocy - 1)
                {
                    E->VB[2][node2] = 0.0;
                    if((ii != 1) && (ii != E->mesh.noz))
                        E->VB[3][node2] = 0.0;
                }

In this case the normal component to the y-faces is set as well as the
z-component, but this time the x-component on the y-faces is not set.

A little further on in these functions we see that the values for these
components set in the code above are designated to be either velocities or
stresses and that all 3 components are treated , but without the x or
y-component value assigned, what physical meaning does this have?  Any
guidance is most appreciated.

Thanks,

-Ethan Fahy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geodynamics.org/pipermail/cig-mc/attachments/20091104/9cbcab8d/attachment.htm