[Nek5000-users] Split/Stress Formulations

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

Shriram,

In general, if you have variable viscosity or free surface
you should likely use the stress fomulation, unless the
extra terms in the stress tensor are treated explicitly
in the .usr file.

For marginally resolved turbulent flows we've found the PnPn formulation
to be more accurate, in part because of the more accurate pressure
treatment.

hth,

Paul



On Wed, 14 Apr 2010, nek5000-users at lists.mcs.anl.gov wrote:

> Hello,
>
> I am quite interested in knowing in what instances one would employ Stress /
> Non-stress / Split formulation.
> I noticed in turbchannel.usr that one must use the stress formulation if
> PN-PN-2 basis is chosen.
> Should a different formulation be used for say PN-PN basis or are they
> conditions under which one would
> judiciously choose these formulations ?
>
> Regards
> Shriram
>