Learning Forum

[mpirkle]

I'm having difficulty obtaining meaningful results from ANSYS when using PDA code. The code was taken from "Determination of Material Properties for ANSYS Progressive Damage Analysis of Laminated Composites" by Barbero and Shahbazi. I am trying to apply this code to a different layup sequence and geometry, with a slightly smaller lamina thickness (t=0.127 mm as opposed to t=0.144 mm). In effect, I am trying to perform a simple in-plane uniaxial tension test to a rectangular plate which contains a centrally located hole, in which enough external load is applied to cause damage. The element used is SHELL281, which is the same as what is used in the paper (see link below).

 

Copy paste this link into your browser: 

barbero.cadec-online.com/papers/2017/AnsysPDABarberoShahbazi_website.pdf 

 

The code runs and converges (nonlinear solution obtained), however ANSYS displays a lamina which is completely red, with no colored fiber damage legend (see attached). This is the case no matter what type of PLESOL, PDMG damage I ask it to display (FT, FC, MT, STAT, etc.). I have tried the MPDG method, only to obtain the same all red lamina results. It leads me to believe that there is either no damage happening, or the entire lamina is damaged. If I increase the applied load to some exceedingly large value, ANSYS will no longer converge. If I lower the applied load, I still get the same results. Can this be a sensitivity issue in the strength, energy, viscosity, etc. values required by the CDM or MPDG damage models in ANSYS? 

 

I notice in a presentation on ANSYS damage from caeai.com "Progressive Damage of Fiber-Reinforced Composites in ANSYS V15.0" that certain substeps are plotted in the damage contours. How does one know how many substeps to use in the code, and which substep is of interest to plot? I notice in many books and papers that :

NSUBST,100,200,100 !SUBSTEPS: DESIRED, MAX #, MIN #

seems to be common. I am currently using 125 substeps, but do not know if this is better or worse than 100 substeps. Do I need more substeps to capture the composite damage that I'm missing with too few substeps? I notice that as I increase the number of substeps, the maximum stress in the lamina increases.

Am I simply looking at the wrong layer or substep, or am I missing some vital line/s of code?

Any advice is very welcome. Thank you!