<html><p style="margin:0cm 0cm 8pt"><span style="font-size:11pt"><span style="line-height:107%"><span style="font-family:Calibri, sans-serif">Hello,<br /><br />I'm a phd student working currently on seismology in Strabourg (France) and we use GMSH to create and mesh a simple plate geometry in which holes are drilled. The plate and the holes must be meshed uniformly because we propagate wave using the code specfem3D in the holes too. We have a problem with some elements at the boundaries of our plate. The code only takes in account hexahedral elements. The 2D surfacic elements at the borders should be defined by 4 nodes. Moreover, the volumic elements at the boundaries (defined by 8 nodes) should be such that the intersection of its nodes with the list of nodes of the border is of length =4. Thus, the type of elements highlighted in the screenshot joined are problematic. I must precise that the geometry of my plate is fixed and cannot be changed. Even if I changed the characteristic length, the meshing algorithm, the only way I found to have an accurate mesh is to enlarge my plate (with an unchanged localization of holes). I also joined the geometry file I use. <br /><br />Is there a parameter I didn't tested or a simple way to suppress this kind of elements in my mesh? </span></span></span></p><br /><br />--<p style="font-family: Arial, sans-serif; color: black;"><span style="font-weight: bold">Jérôme AZZOLA</span><br />Doctorant en géosciences<br />Tél : +33 (0)7 68 54 92 57<br /><a title="Contacter azzola@unistra.fr" href="mailto:azzola@unistra.fr">azzola@unistra.fr</a><br /><a title="Contacter jerome.azzola@gmail.com" href="mailto:jerome.azzola@gmail.com">jerome.azzola@gmail.com</a><br /><a title="Université de Strasbourg" href="http://www.unistra.fr"><img style="border: 0;" alt="www.unistra.fr" src="http://www.unistra.fr/images/logo-uds-signature.gif" /></a><br />EOST - Ecole et Observatoire des Sciences de la Terre<br />5, rue René Descartes<br />F - 67000 Strasbourg</p></html>