Thank you.<br>[ I was on a wrong trail. I did as Birendra but for the SCEC-TP11 dynamic rupture and had a convergence issue when resolution was 25m on the fault. I didn't think about the iteration parameters since i asked Pylith for ExplicitLumped modelisation]<br>
Sorry<br><div class="gmail_quote">2010/10/17 Brad Aagaard <span dir="ltr"><<a href="mailto:baagaard@usgs.gov">baagaard@usgs.gov</a>></span><br><blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
<div class="im">On 10/17/10 12:10 PM, Niz wrote:<br>
<blockquote class="gmail_quote" style="margin: 0pt 0pt 0pt 0.8ex; border-left: 1px solid rgb(204, 204, 204); padding-left: 1ex;">
I am only guessing here, but if the displacement induced by your<br>
simulation is of the order of 200m on the fault, model of cohesive cell<br>
would fail as you use a smaller grid of 500m. I don't think that the<br>
Implicit.LgDeform in Pylith is useful.<br>
</blockquote>
<br></div>
As long as you are using the infinitesimal strain formulation, the amount of deformation does not affect the rate of convergence. Additionally, the cohesive cell formulation assumes the slip direction is independent of the amount of slip even in the small strain formulation. In other words, the slip direction is relative to the original, undeformed configuration.<br>
<font color="#888888">
<br>
Brad<br>
</font></blockquote></div><br><br clear="all"><br>-- <br>==========================================<br>Nizar MOUSSATAT<br>LGIT - Grenoble <br>tel: 06.43.59.39.44<br>==========================================<br>