1--- 2title: Release notes of the 3.2.3 version of `TFEL`, `MFront` and `MTest` 3author: Thomas Helfer 4date: 2020 5lang: en-EN 6numbersections: true 7documentclass: article 8from: markdown+tex_math_single_backslash 9geometry: 10 - margin=2cm 11papersize: a4 12link-citations: true 13colorlinks: true 14figPrefixTemplate: "$$i$$" 15tabPrefixTemplate: "$$i$$" 16secPrefixTemplate: "$$i$$" 17eqnPrefixTemplate: "($$i$$)" 18--- 19 20Version `3.2.3` is mainly a bug fix version of the `3.2` series. 21 22This version was released along with Version `3.1.5` and inherits from 23all the fixes of this version. 24 25# Noticeable fixed issues that may affect the results obtained with previous versions 26 27Ticket #256 reported that the scalar product of two unsymmetric tensors 28was not properly computed. 29 30This may affect single crystal finite strain computations to a limited 31extent, as the Mandel stress tensor is almost symmetric. 32 33# Issues solved 34 35## Ticket #249: Implementation of the Norton-Hoff in the StandardElastoViscoPlasticity brick may lead to negative equivalent plastic strain increment 36 37The implementation of the Norton-Hoff flow rule in the 38`StandardElastoViscoPlasticity` brick may in some exceptional cases lead 39to negative equivalent plastic strain increment. This was checked by the 40brick and led to a divergence of the Newton algorithm, so this flaw 41could not induce wrong results. 42 43For more details, see: <https://sourceforge.net/p/tfel/tickets/249/> 44 45## Ticket #242: The factor in front of the hyperbolic sine is discarded when using the `HyperbolicSine` inelastic flow 46 47The optional factor in in front of the hyperbolic sine was not generated when using the `HyperbolicSine` inelastic flow. 48 49For more details, see: <https://sourceforge.net/p/tfel/tickets/242/> 50 51## Ticket #228: Many stress criteria of the StandardElastoViscoplasticity brick fail when compiling in C++-17 52 53For more details, see: <https://sourceforge.net/p/tfel/tickets/228/> 54 55## Ticket #222: Handling stored and dissated energies is broken in `MTest` for `Abaqus/Standard` behaviours 56 57There were a bug in `MTest` in the `Abaqus/Standard` behaviours which 58didn't reinitialize the values of the stored and dissipated energies at 59the end of the time step to their values at the beginning of the time 60step before the behaviour integration. 61 62Thus, the dissipated and stored energies, when computed incrementally, 63are the sums of their increments at each iterations. 64 65For more details, see: <https://sourceforge.net/p/tfel/tickets/222/> 66 67## Ticket #203: Cyrano` interface: compilation process hangs when treating behaviours based on the Hencky strain measure in `AxisymmetricalGeneralizedPlaneStress` 68 69The trouble came from an infinite loop when looking for the external 70state variable holding the axial stress. 71 72For more details, see: <https://sourceforge.net/p/tfel/tickets/203/> 73