[CIG-SHORT] Linear Convergence Using SNES

[Charles Williams]

In addition to Brad's comment, once you're happy with your mesh you could define a number of different nodesets representing the crack, each with a different length.  This would allow you to have a single mesh, and then choose which nodeset you want for a particular crack length (and thus minimize the number of vertices on the crack for each problem).

Another thing I just thought of.  In this sort of problem you need to make sure you address the issue of ambiguity in the splitting of the fault at the ends of the fault.  You will be able to tell if this is happening by looking at the fault mesh produced by running PyLith.

Cheers,
Charles


On 20/09/2012, at 6:27 AM, Brad Aagaard wrote:

> Jeff,
> 
> Have you done the same setup with prescribed slip where the slip 
> distribution has the same shape as the traction distribution? I *highly* 
> recommend you do that problem first to optimize the size of the domain 
> and mesh discretization. You should use the nonlinear solver and verify 
> that it requires only 1 SNES iteration.
> 
> I will look at the files you posted some more and see if I spot any issues.
> 
> Brad
> 
> _______________________________________________
> CIG-SHORT mailing list
> CIG-SHORT at geodynamics.org
> http://geodynamics.org/cgi-bin/mailman/listinfo/cig-short

Charles A. Williams
Scientist
GNS Science
1 Fairway Drive, Avalon
PO Box 30368
Lower Hutt  5040
New Zealand
ph (office): 0064-4570-4566
fax (office): 0064-4570-4600
C.Williams at gns.cri.nz

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://geodynamics.org/pipermail/cig-short/attachments/20120920/fc33cd5c/attachment.htm