Hi

I modeled three straight pipes as beam elements (ELFORM=1). Please see the attached docx.

I applied 0.305 m upward displacement. I have some confusion. If anyone can help it would be really great.

Firstly, the stress results seem low when compared with shell element. More specifically, the time at which the pipe (beam elements) yielded does not match with matrix based approach and shell element. For instance, analysis with the shell element show that yielding occurs at t = 0.06 sec (total time = 1sec) which matched with matrix method. However, from beam element, the first yielding occurs at t = 0.12 sec, and maximum stress was about 20% less than what was found from shell element analysis.

In another note, if I took 20 elements per pipe, then I got little bit close result (4% less), however, the stress vs time curve looks very weird (attahced fig 2 in docx). If I took 10 elements or less tha 10 elements, then the curve looks good (attached fig 3).

Does anyone have some suggestions for me? Thanks.

Hi @mmia1

Where are you measuring the stress from?

And just to clarify, you have another simulation where you modeled the pipe with shell elements and it correlated well with your hand calculations?

You probably have read this already, but there's quite a bit that goes into Dyna calculations. Modeling a pipe as beams is a much more simplified approach with lots of assumptions. There will inherently be way more error depending on your parameters and model setup.

I increased the number of gaussian integration points QR=5 in the section_beam keycard and received better results.