Page 1 of 1

phase transformation

Posted: Fri Jul 29, 2022 1:38 pm
by Jane
In the process of phase transition, the problem shown in the figure began to appear with time, but I did not find the reason for this phenomenon, and I look forward to your answer.

Jane

Re: phase transformation

Posted: Fri Jul 29, 2022 5:32 pm
by Bernd
Dear Jane,

What is clear from the picture you show is that the interfaces between phase 1 and 2 are completely destroyed. I cannot tell you more without having more information about the system and the used input parameters...

Bernd

Re: phase transformation

Posted: Mon Aug 01, 2022 5:53 pm
by Jane
Dear Bernd,

First of all, thank you for your reply, I will attach my driver file, please check it, I hope you can answer the reasons for some errors for me.

Jane

Re: phase transformation

Posted: Mon Aug 01, 2022 7:13 pm
by Bernd
Dear Jane,

Do you use BCC_A2 or BCC_B2 phase? Please send a clean .in file, otherwise the comments are wrong, and I cannot see what you are doing...

Bernd

Re: phase transformation

Posted: Tue Aug 02, 2022 4:26 pm
by Jane
Dear Bernd,

Sorry, the following is the clean file I created, please check it out.

Jane

Re: phase transformation

Posted: Tue Aug 09, 2022 11:25 pm
by Bernd
Dear Jane,

Sorry for my late reply, I am currently on holidays. I have checked you simulations and noted that nucleation is not working correctly with version 7.000, while everything is fine with 7.003 and 7.1. Do you have access to 7.003? Otherwise you should ask Ralph how you can get one.

Bernd

Re: phase transformation

Posted: Thu Aug 11, 2022 5:25 pm
by Jane
Dear Bernd,

First of all, thank you for your reply. It should be noted that I use 7.003, but the interface problem I mentioned will appear, but you said that you can run normally, I am very strange.

Jane

Re: phase transformation

Posted: Tue Aug 16, 2022 7:15 pm
by Bernd
Dear Jane,

You can find the exact version number and compilation date at the top of the .in output file which is created when you run your simulation.

However, I focussed on the wrong error: With version 7.000 I did not get nucleation while I got it with newer versions. But your error occurs later...

Meanwhile I identified two other potential problems:

1. You use the full diffusion matrix for fcc and bcc ("multi ggg") together with "nple" redistribution behavior. This combination is prone to create numerical problems, because the nple-Model implies unrealistic concentration steps for the corresponding elements (see .cpha output). These steps lead to unrealistic cross diffusion of C which then may lead to numerical problems. I recommend using "diagonal g" for diffusion of all elements in both phases.

2. Given the total simulation time of 300s and the cooling rate of 2K/s, the simulation runs into temperature ranges where C-diffusion is very low and grid resolution is too low to resolve the C-pile-up. This will lead to wrong behaviour towards the end of the simulation.

Bernd