Page 1 of 1

Reg. order of elements in Ges file and Micress input file

Posted: Sun Feb 28, 2016 3:36 am
by deepumaj1
Hi. I had a very basic doubt in running a Micress TQ coupled simulation. I'm a little confised while giving the order of elements under the following section of the input file:
# Reading GES5 workspace ...
# Index relations between TC and MICRESS
# --------------------------------------
# The database contains the following components:
# 1: C
# 2: FE
# 3: MN
# Specify relation between component indices Micress -> TC!
# The main component has in MICRESS the index 0
# Thermo-Calc index of (MICRESS) component 0?
2
# Thermo-Calc index of (MICRESS) component 1?
1
# Thermo-Calc index of (MICRESS) component 2?
3
My system is a 5 component system. How should the order be given. Should we give the same order, as given in the *.TCM file used to generate the *.GES file?

Thanks
Deepu

Re: Reg. order of elements in Ges file and Micress input fil

Posted: Mon Feb 29, 2016 10:41 am
by Bernd
Dear Deepu,

In the .TCM file (this is the macro which you may use with ThermoCalc to create the .ges5 file) the order in which you specify the elements and phases is completely arbitrary and has no consequences for the MICRESS simulation. When you open the .ges5 file with MICRESS, the order in which elements are displayed will always be alphabetic, while the order of phases depends on the order in the database (not in the .TCM macro).
The order how you define the elements inside MICRESS can be freely chosen (without any numerical consequences) apart from the matrix component: This component (defined in MICRESS as component 0) should be major component in the most important phases (e.g. liquid, primary solid) and have a broad solubility range in most phases (otherwise stoichiometric conditions should be applied to all other elements in those phases.
The order of phases in MICRESS in principle also is arbitrary. However, (in contrary to the elements) the their order may sometimes have some minor numerical consequences. The reason is that initialisation of theses phases is always done based on the already initialized phases. It therefore makes sense to define them in the order in which they are expected to appear during the MICRESS simulation. And there are some restrictions for phase 0 which cannot be anisotropic.

Bernd