[petsc-users] Problems with nonlinear convergence when density or pressure is too low

[Jed Brown]

Mani Chandra <mc0710 at gmail.com> writes:

> Hi,
>
> I'm trying to solve for accretion flows around blackholes using the theta
> method in TS. The accretion disk is surrounded by a very low density and
> pressure "atmosphere". The lower the values of the density and pressure in
> the atmosphere, the harder it is becoming for the nonlinear convergence.
> Eventually, it leads to diverging linesearches and the time step becomes
> very small.
>
> Is there a way to deal with very low regions of density and pressure and
> avoid killing the nonlinear convergence?

Large-amplitude dynamics with near-vacuum states/rarefied gasses are
hard for lots of reasons, including nonlinear convergence if you choose
an implicit method.  You might be able to pull most of the nonlinearity
out of the solve using an IMEX method, but there's no free lunch.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 835 bytes
Desc: not available
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20140313/f4d97370/attachment.pgp>