Hi everyone,
I am modeling a spherical TNT charge in air using a 2D axisymmetric MM-ALE model. The symmetry axis is x=0, and both the charge center and detonation point are located on the axis.
During shock-wave propagation, I consistently observe two abnormal pressure spikes along the axial direction, as shown in the attached figure. The remaining shock front is almost circular.
I have tested different *CONTROL_ALE settings (METH, AFAC, NADV, MMGPREF), changed the detonation-point location, and refined the mesh near the axis. The charge-region mesh is about 1 mm, but the spikes remain.
Interestingly, when I constrain the axial nodes with DOFY=1, the spikes disappear. I am unsure whether this boundary condition is appropriate or whether it is simply suppressing a numerical instability.
Has anyone seen this behavior in a 2D axisymmetric ALE blast model? Could it be related to the symmetry-axis boundary condition, ALE advection, or the treatment of cells next to the axis?
Any suggestions would be greatly appreciated.
Thanks!



Because axisymmetric element formulations for all type of elements, use a formulation involving 1/r, there is a potential singularity near the Y-axis as the integration point approached r=0.
This is why there are two ELFORM options on the *SECTION_ALE2D keyword: one for area and one for volume weighting.
You might experiment with these two options. –len
PS – I doubt setting DOFY=1 is a good thing, but I have never tried your idea.



--
You received this message because you are subscribed to the Google Groups "LS-DYNA2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ls-dyna2+u...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ls-dyna2/7ff8f537-d403-4efa-8570-30c26c8508d3n%40googlegroups.com.