When I define a SHELL element with different thickness values at the nodes (parameters T1-T4 in *SECTION_SHELL card) Ls Prepost always shows that element as uniform (Ls Prepost uses the average of the 4 thickness values given in *element_shell_thickness) no matter which ELFORMs I choose (I found somewhere that ELFORMs 1 and 6 can act as tapered shells and retain unique nodal thicknesses-doesn't work)
Am I doing something wrong or maybe it is only a graphical issue?
Hello Paweł,
I think that it is a graphical issue. Maybe, however, you can try to run a simple benchmark test with two shells which are in contact, one of them is fixed and the second is movable. You should be able to see if the node with a largest thickness is in contact in the first place. Remember to turn on thickness consideration in the contact.
Hope it will help
Regards,
Paweł
Hello Pawel Baranowski ,
I made a simple test, like you advised. I modeled the contact between two shell elements, one of them (uniform) was fixed and the second one (with four different thickness values at nodes) had an initial velocity. I plotted the displacements of four nodes of the non uniform shell. Unfortunately the displacements were the same (co-linear) so probably it is not a graphical issue but my mistake or software bug.
Additionally I downloaded the .k file from the following tutorial (in the last 10 minutes there is shown how the non uniform shells should work):
https://www.youtube.com/watch?v=llNL8Kb1pQg&feature=youtu.be&fbclid=IwAR0EdrY9kQOxlxTWj7me8Z47V22J2NuTA2_OzwjVqon7J1DBQJObJVmMrCQ
I've reproduced everything step by step as in the tutorial and in my case it still looks like in the picture below:
Is there any option or parameter that should be checked for the non uniform shells to work properly?
Best Regards
I've found that in Ls-Prepost 4.6 and 4.7 the non uniform thickness is not visible. However, when I've opened the *.key file in Ls-Prepost 4.5 it works like a charm:)
If this is a graphical issue, you can also try to run this benchmark tension test and analyze the stress distribution within the sample. It should be non uniform, I think, when the variable thickness of shells is added.
Maybe, In your simple test ls-dyna sees an avarage thickness of shell?
You're right. I confirm that the way of non uniform shells displaying depends on the Ls-Prepost version. Everything works fine in the ver. 4.3 also.
Thanks for your help.
What about the alternate method of defining node thickness using NODE_THICKNESS or NODE_THICKNESS_SET to override thickness defined by SECTION_SHELL? This removes the necessity of matching T1 to T4 thickness specification to the element node order.
- Badges
- Science topic