[CIG-SHORT] Pylith's error including `dtStable > 0.0' failed

[hkj at mail.ustc.edu.cn]

Dear Professor Aagaard,

Thanks for your reply!
I checked my material properties and found it doesn't constrain the model's whole depth range. So I modified it and the error was gone. Now everything works fine.

Thanks a lot!
Kejing Huang
> -----Original Messages-----
> From: "Brad Aagaard" <baagaard at usgs.gov>
> Sent Time: 2018-08-17 23:23:11 (Friday)
> To: cig-short at geodynamics.org
> Cc: 
> Subject: Re: [CIG-SHORT] Pylith's error including `dtStable > 0.0' failed
> 
> Kejing Huang,
> 
> The error you are getting means that the stable time step computed for a 
> material is zero or negative. This is a very unusual error. You did not 
> include the .cfg files with the material info, so I cannot tell what 
> types of materials you are using. If you are using a linear Maxwell 
> viscoelastic material, be sure to check your viscosity to make sure it 
> gives reasonable Maxwell relaxation times. If you have a powerlaw 
> viscoelastic material, then check its parameters as well to make sure 
> they are reasonable.
> 
> I recommend first trying all linearly elastic materials to rule out 
> setup and mesh errors. Once that works and the results look okay, then 
> add a viscoelastic material one at a time. I would also work with a 
> coarse resolution version of the mesh and then switch to a finer 
> resolution once you have these errors fixed.
> 
> Regards,
> Brad
> 
> On 08/17/2018 02:45 AM, hkj at mail.ustc.edu.cn wrote:
> > Dear Sir,
> > 
> > When I running pylith using my own FEM model, the error including" 
> > `dtStable > 0.0' failed"  appears very frequently, and I found that this 
> > can be avoided by using fewer nodes like
> > 
> > from
> > /pylith postseis.cfg mat_viscoelastic.cfg solver_fieldsplit.cfg --nodes=20/
> > to
> > /pylith postseis.cfg mat_viscoelastic.cfg solver_fieldsplit.cfg --nodes=10/
> > /
> > /
> > But after I rebuilt my FEM model by using a larger area, this error 
> > still occurs even if I use only 1 node,
> > I also chose dt from dt = 1*year to dt = 0.001*year , but it didn't work.
> > 
> > 
> > *The .cfg file is attached, and the time formulation part is*
> > 
> > /```/
> > /# ----------------------------------------------------------------------/
> > /# problem/
> > /# ----------------------------------------------------------------------/
> > /[pylithapp.problem.formulation.time_step]/
> > /# Define the total time for the simulation and the time step size./
> > /total_time = 2*year/
> > /dt = 0.001*year/
> > /```/
> > /
> > /
> > *And the output error is*
> > *
> > *
> > /``` [OUTPUT]/
> > /......./
> > /......./
> > /......./
> > /-- Creating Jacobian matrix./
> > / >> 
> > /home/khuang/Pylith/lib/python2.7/site-packages/pylith/problems/Implicit.py:163:initialize/
> > / -- implicit(info)/
> > / -- Initializing solver./
> > / >> 
> > /home/khuang/Pylith/lib/python2.7/site-packages/pylith/problems/TimeDependent.py:132:run/
> > / -- timedependent(info)/
> > / -- Solving problem./
> > / >> 
> > /home/khuang/Pylith/lib/python2.7/site-packages/pylith/problems/TimeDependent.py:138:run/
> > / -- timedependent(info)/
> > / -- Preparing for prestep with elastic behavior./
> > /mpinemesis: 
> > ../../../pylith-2.2.1/libsrc/pylith/materials/ElasticMaterial.cc:405: 
> > virtual PylithScalar 
> > pylith::materials::ElasticMaterial::stableTimeStepImplicit(const 
> > pylith::topology::Mesh&, pylith::topology::Field*): Assertion `dtStable 
> >  > 0.0' failed./
> > /[0]0:Return code = 0, signaled with Aborted/
> > /```/
> > 
> > Can you tell me how to fix this error?
> > 
> > Sincerely
> > Kejing Huang
> 
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://lists.geodynamics.org/cgi-bin/mailman/listinfo/cig-short