DiekA is an in-house finite element code for forming simulations of the University of Twente, the Netherlands.
The code was founded in the early 1980's by Han Huétink and since then continuously developed. The code is designed to solve non-linear plasticity problems and therefore well suited to model forming processes like sheet forming, bulk forming and welding.
Currently DiekA is adopted as 'community software' for the Netherlands Institute for Metals Research (NIMR). DiekA is freely available for all official partners of the M2i.
DiekA 10.1 is the latest version. It was released in September 2011.
1. Importing constraints via a neutral file.
Bug: The filename is erroneously converted to upper case.
Workaround: Make a copy of the neutral file with the filename in capitals.
- Updated Lagrangian formulation and Arbitrary Lagrangian Eulerian (ALE-) formulation (coupled and de-coupled)
- Static and dynamic implicit code
- Double sided contact. Thickness changes are accounted when using shell element
- Available element types: solid and shell elements
- Thermo-mechanical coupled calculations
- Direct and iterative solvers
- Incremental solution procedure, comprising the Newton Raphson algorithm, Line search and Arc length Method
- Advanced material models (both yield functions and hardening models)
- Pre- and post processing is suited for Patran format
- Special purpose pre-processors
- Operational on Linux, Solaris and Windows and MacOSX.
Q. I use keywords as described in the manual, but DiekA responds that the keyword does not exist.
A. The keywords are read in blocks. *DIMENSION can only be used as first command (after title), followed by the definition of nodes and then the elements. Before *ELEMENTS also bandwidth optimization and output switches may be used, no other keywords. After *ELEMENTS optionally *CONNECTS and *SUPPRESS can be used in that order. Thereafter all other keywords can be used, but all keywords up to *SUPPRESS may not be repeated.
Q. DiekA fails to read an input file that worked for someone else. It gives:
*** EXECUTION IS NOT TERMINATED BUT CALCULATION MAY BE INCORRECT***
HINT: Maybe forgotten the closing EOG or EOD ????? OTHER TYPE FOUND WHILE EXPECTING INTEGERS
What is wrong here?
A. The input file should be consistent with the file system. E.g. files that were made on Windows will not run correctly on Unix/Linux machines because of different line break conventions. Sometimes mail-systems will automatically convert a text file to the Windows/DOS convention.
Q. How can I plot the localization data that is used in the *FAILURE tests?
A. Add the keyword "FAIL" in the *OUTPUT block. It will add the principal strain increments, the principal strain ratios and the element status (failed or not) to the nodal files.