[Paraview] Tetrahedral Mesh Quality

[Fastl, Thomas]

Hello Community,


I need to evaluate the quality of my tetrahedral mesh for simulations in structural mechanics and want to apply the filter 'Mesh Quality' using the 'Scaled Jacobian' metric. Paraview (4.0.1) calculates the values providing me with a range of [0.1,1.0] and the 'Histrogram' looks like a Gauss distribution with a mean around 0.6. I wanted to find out more about how this metric is calculated for tetrahedral elements and looked up the 'Verdict Library Reference Manual', which states the following:


acceptable range:    [0.500 , 0.707]

normal range:          [-0.707 , 0.707]

full range:                [-dbl_max , dbl_max]


It also states that the 'Scaled Jacobian' for a unit equilateral tetrahedron, in my humble opinion a perfect tetrahedron, is 1. As far as I understand these ranges the full range provides values if some error/problem occurs, the normal range should be the range of mathematically possible values and the acceptable range is where somebody wants his elements to be in. The 'Scaled Jacobian' of a unit equilateral tetrahedron, however, doesn't fall in the acceptable range or even worse, in the normal range. Do I misunderstand the provided ranges? I also tested a equilateral tetrahedron in Paraview and by hand with the 'Scaled Jacobian' metric and I always obtained 1. Thank you for your help!


Best, Thomas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/paraview/attachments/20140806/43717f74/attachment-0001.html>