search

Dhondtguido / CalculiX

This repository contains the source files of CalculiX, a three-dimensional Finite Element Program (www.calculix.de).
GNU General Public License v2.0
75 stars 19 forks source link

Posible error on Manual formula 363 and 364 ccx manual 2.20 #23

Closed disla2 closed 1 year ago

disla2 commented 1 year ago

screenshot 104

Subindex is recommended on Manual formula 363 to avoid confusion between the second order Taylor approximation of the logarithmic Strain and the Eulerian Strain. One is computed using initial length Lo and Eulerian using final length L.

FEA-eng commented 1 year ago

There are quite a few typos and mistakes in the ccx manual. @Dhondtguido said that he will create a separate repository with the documentation where we could create issues (or possible PRs) to correct such mistakes. Unfortunately, this repository hasn't been created yet.

Dhondtguido commented 1 year ago

Hi, your remark is correct, my expansion of the Eulerian strain was incorrect (the coefficient of the second term should be -3/2 instead of -1/2). Therefore, the Eulerian and logarithmic strain can be quite different. I decided to define the *DEFORMATION PLASTICITY routine as a Abaqus umat routine using logarithmic strain (using the umat_abaqusnl.f routine in CalculiX). Now the results are more in line with Abaqus. This new approach will be available in the next CalculiX version. Thanks for pointing this out. Guido