Page 1 of 2

Ferrite / Pearlite Formation

Posted: Mon Nov 12, 2018 1:35 pm
by R.Hess
Dear Bernd,
i want to model the formation of an ferrite / pearlite structure with cementite as an unresolved phase - like in the examples.
my approach was to use an ferrite / pearlite structure as initial, but to declare the pearlite grains as austenite.
by cooling down this intial strucutre i want to achieve a matrix of ferrite and unresolved cementite.

the only thing i did, was to edit the GammaAlphaPearlite by changing the initial structure and matching the initial concentrations
GammaAlphaPearlite_initial_dri.txt
Inputfile
(29.03 KiB) Downloaded 179 times
But there are some problems during running the simulation. it is extremly slow. i started the simulation on friday and today (on monday) i cancelled. furthermore, the results were very confusing, for example : the grain boundarys are vanishing. BUT by considering other output files (like fraction of phase 3 (Cementite)) the structure is visible and it seems to work.
recently i started the simulation again. for that reason i cannot upload the results - maybe tomorrow. but maybe you can already help me by the problem of simulation duration?

kind regards

raphael

Re: Ferrite / Pearlite Formation

Posted: Tue Nov 13, 2018 2:34 pm
by Bernd
Hi Rafael,

I don't see anything which is obviously wrong, apart from some "," instead of "." in the initial concentration input which could lead to a wrong interpretation.
However, you changed strongly the initial composition and initial microstructure, so that numerical parameters may be completely wrong in this context. I would need all the input files (.dri, .ges5, initial microstructure) in order to find out more!

Bernd

Re: Ferrite / Pearlite Formation

Posted: Wed Nov 14, 2018 10:14 am
by R.Hess
Hi Bernd,
as attached you can find the ges file, dri.-file and images of the phase-output and fraction of cementite (frac3) output.
The Initial microstructure is too lagre, so i am not able to upload it. Is there another way to upload such large files ?

Re: Ferrite / Pearlite Formation

Posted: Wed Nov 14, 2018 10:39 pm
by Bernd
Dear Rafael,

At a first glimpse your results do not look so bad...

If the initial microstructure is too big for uploading, you could send it to me by email. Much better, however, would be to reduce the domain size of your simulation which also makes testing much faster. Then, you could easily upload the initial microstructure to the forum. It is always better to have small simulations for testing and calibrating, and make the big simulation once this has been achieved!

Bernd

Re: Ferrite / Pearlite Formation

Posted: Thu Feb 07, 2019 3:32 pm
by Mehnoush
Dear Bernd,
I made the initial structure much smaller (Matrix size). I ran the simulation. An improvement is that we do not have the austenite phase any more at the end of the transformation time. However, the result is still not satisfying. The main problem i think is thick interfaces. also the small pearlite phases doses not tranform. I wanted to ask if you have any idea how we cam improve the model?
Best regards,
Mehnoush

Re: Ferrite / Pearlite Formation

Posted: Thu Feb 07, 2019 11:44 pm
by Bernd
Dear Mehnoush,

is the input file different from what Rafael showed before? Could you please attach it?

Bernd

Re: Ferrite / Pearlite Formation

Posted: Fri Feb 08, 2019 9:06 am
by Mehnoush
a little bit.i have attached it. :)

Re: Ferrite / Pearlite Formation

Posted: Fri Feb 08, 2019 1:02 pm
by Bernd
Dear Mehnoush,

I have seen that you changed the nucleation conditions. Now, you additionally nucleate phase 3 (cementite) without using the "unresolved" model - what are the reasons for that?
The reason for using the "unresolved" model for cementite was that it is not possible to resolve this phase correctly at the given resolution. Therefore it is allowed spreading as diffuse eutectoid mixture (see .frac3 distribution which Rafael showed above). At least for the cementite which nucleates at triple points you do not allow that anymore. Maybe, this is the problem.

Bernd

Re: Ferrite / Pearlite Formation

Posted: Wed Feb 13, 2019 5:19 pm
by Mehnoush
HI Bernd,
we finaly could obtain the Ferrite/pearlite structure by cooling down austenite/ferrite structure.
kow we want to simulation the heating of the obtained structure with 100000 K/s but the results are not looking good. the austenite phase is formed as big massive grain also grain interfaces are fuzzy and thick. do you have any hint that can help us i attached the results.
Many thanks,
Mehnoush

Re: Ferrite / Pearlite Formation

Posted: Wed Feb 13, 2019 7:15 pm
by Bernd
Hi Mehnoush,

I cannot see anything from the results. Did you change the length scale? May I see the driving file?

Bernd