Re: Simulation of delta-ferrite/austenite solidifcation sequence under SLM conditions
Posted: Mon Aug 22, 2022 12:45 am
Dear Ahmet,
Sorry for my late answer! What you are doing is correct. Using paratq was problematic due to the triple junctions: When using paratq this option must be used for all interfaces, otherwise there is a conflict of non-matching phase compositions. The reason is that paratq uses the Thermo-Calc para-equilibrium model, so that fcc and bcc have almost the same composition. This is in conflict with different equilibrium compositions with the liquid phase and typically leads to numerical issues!
This is different with "para", because this model is a MICRESS redistribution model which is compatible at triple junctions. In your case it is better to use the bottom temperature as reference for the transformation direction, the other option in some cases may lead to fluctuations.
I still had no time to analyze the "segmentation" problem...
Bernd
Sorry for my late answer! What you are doing is correct. Using paratq was problematic due to the triple junctions: When using paratq this option must be used for all interfaces, otherwise there is a conflict of non-matching phase compositions. The reason is that paratq uses the Thermo-Calc para-equilibrium model, so that fcc and bcc have almost the same composition. This is in conflict with different equilibrium compositions with the liquid phase and typically leads to numerical issues!
This is different with "para", because this model is a MICRESS redistribution model which is compatible at triple junctions. In your case it is better to use the bottom temperature as reference for the transformation direction, the other option in some cases may lead to fluctuations.
I still had no time to analyze the "segmentation" problem...
Bernd