unphysical initial strain in axisymmetric elements

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view

unphysical initial strain in axisymmetric elements

Abaqus Users mailing list

Dear Community,

I observe strain in my model even before the load arrives and kindly ask your advice.

My simulations model the contact of two bars with different cross-sectional areas when exposed to pulse loading.

I have modelled this process in Abaqus/Explicit as a two-dimensional simulation. The investigation was conveniently done as a parametric study, defining the lateral dimension of the second bar as parameter. Both bars are slim (length 2 m, diameter 2 cm) and elastic. The same incident pulse is fed into all simulations. Abaqus 6.13 was run in double precision.

Five different simulations are run, with different elements:

CAX4R: unphysical strain in both bars for all bar aspect ratios
CAX3:  unphysical strain in the first bar for all bar aspect ratios, in the second bar for some bar aspect ratios
CAX6M: unphysical strain in the second bar for some bar aspect ratios

plane strain CPE4R: no unphysical strain

plane stress CPS4R: no unphysical strain

The unphysical initial strain of up to 0.00005 is already observed at an element representing a strain gage on the first bar, i.e., the contact of the bars is uncritical in this regard. The sign of this unphysical strain may be positive or negative.

I downsized my model:
both bars, but no load imposed: strain = 3.4*(10^(-5)) in the first bar
only first bar, no load imposed: strain = 2.0*(10^(-6))

The elements do not distort excessively (no friction, Poisson's ratio = zero, second bar is free to move axially). Hence, hourglassing should not be the cause.

Fringe information: the above simulations are unable to reproduce the reflection and transmission coefficients predicted by theory. The only advice was "Elastic waves are difficult to model explicitly". More on

Kindly advise me on that unphysical strain and, if possible, also on a tweak to reproduce the reflection and transmission coefficients.

Thank you for generously sharing your expertise,

Frank Richter