Skip to main content

Elastic failure criterion

Submitted by HussamNasreddin on

Dear all,

I would like to define an elastic material with a
failure value of maximum tensile strain only. Any element reaches a max.
tensile strain value will crack and suffer complete loss of load
carrying capacity(possibly delete that element). Thus its a change in
geometry that is only considered when redistributing stresses.

 Could this be done in ABAQUS or any other Finite Element Software?...I have been working with ABAQUS for some time now but I could not reach any where

 

Thanks for the help

Hussam 

Forums
Fracture Mechanics Forum
Jayadeep U. B.

Hi Hussam,

I am not much familiar with scripting in Abaqus, but it can be done easily with user subroutines (APDL) in Ansys, as we have done in this paper:  http://www.doaj.org/doaj?func=abstract&id=492835.  The details of the program are not given in the paper, but if you are interested, I can send you the code (please give me few days time to locate the code, and provide some comments for better readability).

I assume that it should be possible in Abaqus also.  May be some expert users of Abaqus can give suggestions.

Regards,

Jayadeep

Mon, 09/26/2011 - 04:38 Permalink
Jayadeep U. B.

Hi,

We used element "kill" option in ANSYS, which effectively makes "elasticity value nearly zero"; we didn't do it manually. Probably you are right; if Young's modulus is made exactly zero (or even extremely small), we will have convergence issues due to ill-conditioning.

The stress-strain curve of concrete given in the paper is the actual material model from a reference.  As far as I remember, ANSYS does not allow a "descending branch" for stress-strain curve.  So the curve was truncated at the maximum stress point.  It does not make any difference as the compressive strains never become that high.  Anyway, it is a very good observation from your side.

Regards,

Jayadeep

P.S.: I will send the script to your mail ID (within a few days).

Wed, 09/28/2011 - 04:20 Permalink