instruction

solid-solid phase transformations, influence of stresses and strains
Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: instruction

Post by Bernd » Wed May 04, 2022 12:53 pm

Dear Jane,

In a first place, your input seems not to make much sense:

# Initial Microstructure
# ======================
# Type of grain positioning?
# Options: deterministic random [deterministic_infile] from_file
deterministic
# NB: the origin of coordinate system is the bottom left-hand corner,
# all points within the simulation domain having positive coordinates.
# Number of grains at the beginning?
9
#
# Grain number 1

# --------------
# Geometry?
# Options: round rectangular elliptic round_inverse
round
...

...
# Structure from restart file
# ---------------------------
# Shall grain(s) be replaced by initial structure(s) from a restart file(s) ?
# Options: restart_file | no_restart_file
restart_file
# How many restart files shall be read?
1
# For each restart file a grain number and (optionally)
# shift (in grid cells) and zoom factor for all 3 dimensions
# as well as a character for rotation options must be specified:
# grain number [shift X (int) shift Y (int) shift Z (int)
# zoom X (int) zoom Y (int) zoom Z (int) rot(string)] ?
# Rotation options: "xz+90" "xz-90" "xz180" "xy+90" ... "yz180"
2 0 0 -4000
# Name of restart file?
FeCMn
...



The idea of reading initial microstructures from restart file is as follows: With the deterministic grain input you define a grain as a "mask" for filling in the microstructure which is stored in a .rest-file. In the most simple case, you just do not define any grain (grain number is 0) and replace default grain 0 afterwards with the .rest-file microstructure.
What you essentially do is define 9 grains, and replace grain 2 with the microstructure from the .rest-file. On top of that, you apply a shift of -4000 cells in z-direction on this microstructure! I guess that you just copied that from another place without change...

The iosError error may also come from other reasons. Did you change something related to the definition of grain orientations? You e.g. cannot change a phase from isotropic to anisotropic upon restart, because the data is missing then..

Bernd

Jane
Posts: 50
Joined: Tue Mar 29, 2022 1:06 pm
anti_bot: 333

Re: instruction

Post by Jane » Wed May 04, 2022 5:08 pm

Dear Bernd,

Thank you very much, I have been able to read the initial file successfully, but I still have a problem, the initial file I read here has two phases, austenite and ferrite, I want to add another phase , that is, taking the file I read as the initial structure, I added another phase to make it nucleate and grow inside the crystal. How do you think it should be implemented? Very much looking forward to your reply.

Jane

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: instruction

Post by Bernd » Wed May 04, 2022 7:02 pm

Dear Jane,

Exactly as you say...

There should be no problem with reading the .rest-file if you add a further phase, as long as you do not change the phase numbers for the phases already existing in the microstructure of the .rest-file.

Bernd

Jane
Posts: 50
Joined: Tue Mar 29, 2022 1:06 pm
anti_bot: 333

Re: instruction

Post by Jane » Thu May 05, 2022 3:03 am

Dear Bernd,

I'm glad to hear that you can go further and add a phase, but unfortunately I added phase3 and its interactions without any change in the initial organization, but the following error occurs, when I add phase and its The reaction can proceed normally after the relevant parameters are deleted. As you said, the third phase can be added, so I can't find the specific reason for the error. The following is my driver file.

# Reading restart file #1: FeCMn.rest
# Restart file version number: 30
# Concentration field read
# Grain number field read

Unable to read from file FeCMn.rest
STOP in routine initRestart
while reading grain data, iosError: 5


Jane
Attachments
FeCMn _restart_add.dri
(24.33 KiB) Downloaded 94 times

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: instruction

Post by Bernd » Thu May 05, 2022 6:47 pm

Dear Jane,

As I already indicated in my latest post, the iosError = 5 could be the result of any other change which you did in the driving file between the one which created the .rest-file and the one you attached. As I don't know all differences, I can only speculate that you changed something related to the orientations of the already existing grains. Can you tell me the changes, or attach the corresponding driving file?

Bernd

Jane
Posts: 50
Joined: Tue Mar 29, 2022 1:06 pm
anti_bot: 333

Re: instruction

Post by Jane » Fri May 06, 2022 8:01 am

Dear Bernd,

I am very happy to hear from you, I will attach my relevant drivers.
FeCMn _restart.dri, this file is the file that completes the initial organization reading.
FeCMn _restart_add.dri, this file adds the third phase on the basis of ①.
Very much looking forward to your reply.

Jane
Attachments
FeCMn _restart.dri
(34.06 KiB) Downloaded 85 times
FeCMn _restart_add.dri
(24.33 KiB) Downloaded 99 times

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: instruction

Post by Bernd » Fri May 06, 2022 9:22 pm

Dear Jane,

Your driving files are not consistent because the first one already reads the initial microstructure from a .rest file! Could you provide the unaltered .in file from the first run as clean reference driving file? Then I could try to reproduce the problem.

My guess still is that your first simulation has been using anisotropic phase definitions (in phase data), and you now try reading the already existing anisotropic grains as isotropic...

Unfortunately, I am in a train right now, and I don't have a sufficiently performant internet connection to proof my guess by checking again the MICRESS implementation with respect to iosErr =5...

Bernd

Jane
Posts: 50
Joined: Tue Mar 29, 2022 1:06 pm
anti_bot: 333

Re: instruction

Post by Jane » Sat May 07, 2022 5:12 am

Dear Bernd,

First of all thank you for writing back to me on the road, I'll attach the .in and .dri files that were unmodified from my first run here for your convenience.
Finally, I wish you a happy trip.

Jane
Last edited by Jane on Tue Oct 11, 2022 3:31 am, edited 1 time in total.

Bernd
Posts: 1504
Joined: Mon Jun 23, 2008 9:29 pm

Re: instruction

Post by Bernd » Mon May 09, 2022 10:56 pm

Dear Jane,

Unfortunately, your case is more complicated than expected, and I found a problem which impedes including anisotropy of a phase via restart if all phases have been isotropic in the first run which created the .rest-file. This is a problem which we interestingly never have seen before and which leads to the iosError. For this specific case it can be corrected in the code, but not backward in version 7.0 of MICRESS. Therefore, you need to work around: By setting at least one of the phases (e.g. phase 2) to anisotropic in the phase data section for the first run (you can still use "isotropic" in the 1/2-interaction data though), or set phase 3 isotropic in the second run. I am sorry for that!

Further, there are two things you definitively need to change in the driving file of the second run to allow for correct initialisation of thermodynamic phase interactions and successfully nucleating the cementite phase:


1.) You should define initial phase concentrations for all phases (using "input" instead of "equilibrium") in order to allow for a smooth initialisation. For that it is best to use the existing average values from the last row of the .TabC-output (you need to activate "tab_conc" in the output selection to get this output), and use a reasonable estimation for the phases which are not yet existent (cementite). These values do not define the initial concentration if you read the initial microstructure plus compositions from the .rest-file, but serve as start values for initialisation and in case of later trouble. You should further set the "temperature at which the initial equilibrium will be calculated" to the temperature at the end of the first run (as well given in the last line of the .TabC output).

2.) Phase interaction data are needed for the 1/3- interaction (because phase 1 is reference phase for nucleation of phase 3)

Bernd

Jane
Posts: 50
Joined: Tue Mar 29, 2022 1:06 pm
anti_bot: 333

Re: instruction

Post by Jane » Tue May 10, 2022 7:23 am

Dear Bernd,

Thank you very much. Your reply is very helpful to me.
In addition, I would like to ask you a question about composition distribution. Can I use MICRESS to simulate the problem of element distribution in the process of austenite to bainite or martensite transformation? If so, will the elements in it segregate ? Or will the elements present an even distribution?

I wish you a pleasant life.
Jane

Post Reply