Dear all,
I have formulated a user finite element (2D 9-node, plane stress conditions) based on a damage-plastic isotropic constitutive law, to performe dynamic analysis.
Then the tangent matrix is been defined in a element routine as:
S=ctan(1)K+ctan(2)C+ctan(3)M
and the residual R is been consistently adapted.
At the beginning i made linear elastic analysis (adopting Newmark implicit method with beta=0.25, gamma=0.5) and i compared my results with those obtained with FEAP element (using 2d solid element in plane state stress).
I got the same results in terms of displacements, accelerations, reactions.., but if i compare the 'L'file i can observe that for the user element there are only 2 iterations for each time-step, while for the FEAP element there are more.
I can't understand the reason of these differences. am I forgetting somethings on my code? I know that for linear system it is possible to modify Newmark's original formulation to permite solution without iteration, but i didn't make any change to the transient method.
I attached my input files and my 'L' files.
Thanks all
