<span style="font-family: arial,helvetica,sans-serif; font-size: 12px;">‌Hi,<br> <br>  I'm trying to run a Sfepy example : poisson_short_syntax.py with a mesh coming from tetgen (see bar.node and bar.ele files).<br> I have two groups of vertices: 4 and 5 on which I want to impose the temperature.<br> When I use the following instruction:</span><br> <br> regions = {<br>     'Omega' : 'all', # or 'cells of group 6'<br>     'Gamma_Left' : ('vertices of group 4', 'facet'),<br>     'Gamma_Right' : ('vertices of group 5', 'facet'),<br> }<br> <br> The command failed because, I think,  groups are defined on nodes in my mesh and not on facets.<br> So I change the previous instruction to:<br> <br> regions = {<br>     'Omega' : 'all', # or 'cells of group 6'<br>     'Gamma_Left' : ('vertices of group 4', 'vertex'),<br>     'Gamma_Right' : ('vertices of group 5', 'vertex'),<br> }<br> <br> In that case Sfepy runs but the solution is a T = 0 in all the domain:<br> <br> sfepy: left over: ['absolute_import', '__builtins__', '__spec__', 'verbose', '__doc__', '__cached__', '__file__', '__name__', 'data_dir', '_<br> _loader__', '__package__', '_filename']<br> sfepy: reading mesh [tetra4] (C:\Users\msaudreau\AnacondaProjects\bar.node)...<br> sfepy: ...done in 0.01 s<br> sfepy: creating regions...<br> sfepy:     Omega<br> sfepy:     Gamma_Left<br> sfepy:     Gamma_Right<br> sfepy: ...done in 0.01 s<br> sfepy: equation "Temperature":<br> sfepy: dw_laplace.i.Omega( coef.val, s, t ) = 0<br> sfepy: using solvers:<br>                 ts: no ts<br>                nls: newton<br>                 ls: ls<br> sfepy: updating variables...<br> sfepy: ...done<br> sfepy: setting up dof connectivities...<br> sfepy: ...done in 0.00 s<br> sfepy: matrix shape: (232, 232)<br> sfepy: assembling matrix graph...<br> sfepy: ...done in 0.00 s<br> sfepy: matrix structural nonzeros: 2596 (4.82e-02% fill)<br> sfepy: updating materials...<br> sfepy:     coef<br> sfepy: ...done in 0.00 s<br> sfepy: nls: iter: 0, residual: 0.000000e+00 (rel: 0.000000e+00)<br> <br> <br> Does someone understand what's going on ??<br> <br> Thanks a lot for any help.<br> <br> All the very best<br> <br>    Marc