[Nek5000-users] "vnormal_desired = ?" for turbulent outflow boundary condition

[nek5000-users at lists.mcs.anl.gov]

Hello,

I am running a LES case for backward facing step with Re=28000. The
incoming mean velocity is 1.0. I am having trouble with setting the
turbulent outflow boundary condition.

I am using the fill_div(usrdiv) subroutine. The "vnmo_desired = 0.05" for
this case and for this value the simulation does not look right. As seen
in the two figures for instantaneous and average x-velocity countour
pictures.

Instantaneous x-vel:
http://www2.uic.edu/~hkanch1/x-vel-instantaneous.jpeg

Average x-vel:
http://www2.uic.edu/~hkanch1/x-vel-avg0000.jpeg

For a value of "vnmo_desired" less than 0.02, the simulation blows up. Is
there a way I can estimate the value of "vel_normal desired". The incoming
channel height is 4.0 and mean velocity is 1.0. The outflow channel height
is 5.0. The value of "dmax = 1.488", this is used in calculation "cdiv".

Attached is the *.usr file for this case.

Any help will be appreciated.

Thanks,

Harish.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: bfs3d.usr
Type: application/octet-stream
Size: 8406 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/nek5000-users/attachments/20100604/86d8821f/attachment.obj>