We have several verification problems labeled "te0.1feb" to "te05.feb" in the verification suite, but these all use the nodally integrated tetrahedrons. We don't have any verification problems for the quadratic tetrahedrons. Presumably we could incorporate some of the problems from the quadratic tet paper.
The existing te*.feb problems are not labeled as using the nodally integrated tetrahedrons. This is confusing so if we decide to keep them, we should label their descriptions in the .xlsx file and the header sections as using the nodally integrated tets.
The User's Manual section on the nodally integrated tet elements is quite terse. I suggest that we expand the section on tetrahedral elements and explain that although the nodally integrated tet elements are available, they were determined to be inferior to other formulations and are not recommended, and here's why. We wrote a SCI Technical Note describing the their performance and issues:
http://www.sci.utah.edu/publications...I-2011-007.pdf
We have several verification problems labeled "te0.1feb" to "te05.feb" in the verification suite, but these all use the nodally integrated tetrahedrons. We don't have any verification problems for the quadratic tetrahedrons. Presumably we could incorporate some of the problems from the quadratic tet paper.
The existing te*.feb problems are not labeled as using the nodally integrated tetrahedrons. This is confusing so if we decide to keep them, we should label their descriptions in the .xlsx file and the header sections as using the nodally integrated tets.
The User's Manual section on the nodally integrated tet elements is quite terse. I suggest that we expand the section on tetrahedral elements and explain that although the nodally integrated tet elements are available, they were determined to be inferior to other formulations and are not recommended, and here's why. We wrote a SCI Technical Note describing the their performance and issues:
http://www.sci.utah.edu/publications...I-2011-007.pdf