[Gmsh] Broken mesh for coarser mesh size at centre of circle

[Rohracker, Maurice]

Dear GMSH user and developer,

I have a small example where some weird things are going on. We have a 
quadric box with a circle inside and interphase around it. For the FE 
simulation, it's sufficient to have a coarser mesh at the centre point 
of the circle. We increase here the computed mesh size by a factor 
(centreMeshFactor in the .geo-file). For some reasons, the resulting 
mesh has for the factor = 10 and > 15 not a good shape anymore and the 
elements kind of stretch up.
Could you please tell me why this happens and how the mesh sizes are 
interpolated between two specified points. Our goal is here to 
understand the issue or interpolation.
I agree that for an increasing factor the mesh size gets unproportional 
in comparison to the other mesh sizes on the circle, but why does this 
occur, that at some point the mesh breaks. It seems that the mesh 
generator could not interpolate the mesh sizes in a correct way at some 
point anymore, or am I wrong?

Thank you very much in advance.

Best regards,
Maurice Rohracker
Master Student, FAU Erlange-Nürnberg
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: bubble_interphase_layers.geo
URL: <http://onelab.info/pipermail/gmsh/attachments/20200208/e3cf18ca/attachment.geo>