I am trying to model quasi static three point flexure of hollow circular composite beam in LS DYNA. I am using MAT 58 and B-T shell elements. The beam undergoes progressive loading in an expected manner until the failure strains are reached in some elements. Beyond this point there are weird vibrations in the beam. There is also a simultaneous increase in hourglass energy. I have tried hourglass control without any success. Has someone experienced this before? What could be the possible reasons behind this non-physical behaviour?

Any help would be deeply appreciated.

Thanks.