[petsc-dev] Jacobian lag persist with ROSW method

[Mani Chandra]

It doesn't converge at all if I use a first order method. The time step
required now is incredibly small. Is that an indication that something
might be wrong? In general should convergence be better with lower order?


On Sat, Mar 8, 2014 at 10:17 PM, Jed Brown <jed at jedbrown.org> wrote:

> Mani Chandra <mc0710 at gmail.com> writes:
>
> > Does linearly implicit mean that you can solve any nonlinear system with
> > the ROSW methods? I set my entire system using TSSetIFunction, without
> any
> > TSSetRHSFunction.
>
> Yes, that is fine.  Linearly implicit means that it only solves linear
> problems.  RosW methods use an (approximate) Jacobian from the beginning
> of the step for all subsequent stages.
>
> > I wanted to try the ROSW methods because the THETA methods seems to want
> a
> > time step much much smaller than the courant time step for the nonlinear
> > solver to converge.
>
> The solver convergence is a separate problem, for which error estimators
> and higher order will not help.  Work out the solver issues with beuler
> before doing anything else.
>
> > I checked the validity of the solution and it is indeed right, so the
> > residual evaluation should be correct. I'm solving for the general
> > relativistic mhd equations in a fixed background spacetime (kerr
> > blackhole). There are large variations in the density and pressure and
> > I use slope limiters.
>
> You can try assembling a Jacobian for a first order method (no limiters)
> and using TSROSW, but be very careful about stability.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20140309/be22c3ff/attachment.html>