MICRESS Forum

Dear Bernd,

I am really happy to receive your letter, and I am very grateful to you for helping me many times. It is your help that gives me more confidence to persevere.
You said that the critical supercooling degree should be less than 0, but I did not have the option of critical undercooling degree here. I have set the others according to what you said. The following is the driving force file I have modified.but there may still be some problems. If you have time, you still hope Could you please help me to see if there is still a problem with the driver file? Or what problems still exist?
Thank you very much for your support and help.

Lv

Hi lv,

It seems you forgot to attach the driving file...

Bernd

Dear Berrnd,

I'm sorry, I'll attach my file and look forward to hearing from you.

Lv

Dear lv,

I am still not sure about how you want to design your recrystallisation model. By specifying "all_interfaces" in the phase input section for bcc, and by specifying the maximum value of reX-energies (1,0 J/cm3) as energy threshold for recrystallisation you have decided that deformed grains can grow into each other without lowering the reX-energy (each of both switches alone would be sufficient for that).

This is the less usual notion of the recrystallisation model. The "normal" approach is to assume that interfaces between deformed grains do not move, and that recrystallisation is modelled by nucleation of new grains without reX-energy (or at least below the threshold). If you want to go for the usual model, you need to remove "all_interfaces" and set the energy threshold to a small value which is smaller than the usual values of deformed grains. Furthermore, you need a third nucleation type for recrystallisation. The process would be

1.) γ-α-transformation below 1200K (seed type 1)
2.) deformation in a short period below 1000K (i.e. introduction of reX-energy into bcc by seed type 2)
3.) recrystallisation (seed type 3)

For the second step to work, you need to apply a negative critical undercooling for seed type 2 and limit the deformation period (e.g. by specifying a minimum temperature).

For the third step, the reX-energy range must be 0 or at least below the energy threshold.

Bernd

Dear Bernd,
First of all, I am very happy to receive your reply. After receiving your reply, I began to understand and simulate. After a few days of thinking, I still have some questions that I would like to ask you. According to your reply this time, I think this should be It is the first method you said, which is to give the ferrite deformation amount. Here are my questions:
1. In the short-time deformation in the second step (introducing recrystallization energy to BCC through seed 2), can I set the second seed here to any type? If yes, is the matrix phase here FCC? Or BCC? Also, you said that a negative critical subcooling degree needs to be applied to seed 2. What is the negative critical subcooling degree here represented by? I didn't find the negative critical subcooling you mentioned. I don't particularly understand the step 2 you mentioned here. I hope you can explain it further, or can you explain it to me in the form of a driver file? I think this will be easier to understand.
2. According to what you said, the recrystallization energy needs to be introduced, so can I directly introduce the recrystallization energy into the seed type 1? Is it possible to omit seed type 2 directly?
The following is my driver file. I think I understand the two methods you mentioned, but there are still some problems with the driver file. Thank you very much for your technical support.
Lv

Dear Lv,

Yes, you can easily omit step 2 in your approach and directly introduce recrystallisation energy using seed type 1.

But using seed type 2 instead is more straightforward and also more flexible: Imagine that after recrystallisation you would like to apply another deformation and recrystallisation step, then you would need this technique using "add_to_grain". The negative undercooling applied in that case has no physical meaning, it also has nothing to do with real "nucleation", we just use the existing input specification and mechanisms and apply nucleation as a "trick" to introduce recrystallisation energy into an already existing microstructure...

Bernd

Dear Bernd,

Yes, you are right. If you have time or you are free, can you read my driver file? Could you point out my error through my driver file? Thank you for your support.

lv

Dear Lv,

Your driving file looks ok. The only detail which I do not understand is why you have two different seed types for ferrite on austenite, where one creates reX-energy and the other doesn't...

Bernd