Error in reading diffusivity from database

technical aspects of .dri file generation (e.g. debug mode ) etc...
Post Reply
kamalnath
Posts: 13
Joined: Mon Dec 11, 2017 9:17 pm
anti_bot: 333
Location: Columbus, Ohio

Error in reading diffusivity from database

Post by kamalnath » Thu Apr 05, 2018 12:59 am

Hi Brend,

I am relatively new to MICRESS, so I might be doing some silly mistake. I am having a trouble reading diffusivity value from database. I am trying to simulate directional solidfication of Ni-5Nb (wt%)

When running the driving file, I getting the following error message.( The image is from the log file)
Error_Msg.PNG
Snapsnot of the log file
Error_Msg.PNG (82.42 KiB) Viewed 7894 times
But, If I use diff command and give user-defined values, the driving file is working fine.

I created the database file using "make_ges5.tcm" (attached) and used the driving file "NbNi_in.txt" (attached)

I wonder whether it is because my MICRESS version is very old ( v5.4 ) or problem in input file. Kindly have a look when you get time.

Regards,
Kamal
Attachments
NbNi_in.txt
Driving file
(14.14 KiB) Downloaded 339 times
NbNi_log.txt
Log file from MICRESS
(7.03 KiB) Downloaded 310 times
make_ges5.tcm
Thermo-Calc script to generate database file
(262 Bytes) Downloaded 311 times

deepumaj1
Posts: 104
Joined: Fri Feb 13, 2015 11:00 am
anti_bot: 333
Contact:

Re: Error in reading diffusivity from database

Post by deepumaj1 » Thu Apr 05, 2018 9:56 am

Hi Kamal,
No problem, everyone starts with silly mistakes. The way you create GES file and everything looks fine. When I ran your driving file with the new micress version, it started fine, though the thermodynamics didn't go well, and I had trying hard errors. But, I'm not getting the same error as yours (invalid diffusion coefficient). So, I think this error might be because you use a very old version of micress, which may not be compatible with the new thermocalc version and the TQ interface. Maybe Bernd can confirm on this. If this is the case, then I would suggest you to use manual diffusion coefficient for time being and see if you are able to simulate your problem. And once you have a stable simulation with manual thermodynamic and diffusion data, you can think of the TQ coupling, which can increase the accuracy of your simulation. At this stage, you may need a new micress version.. 8-)

Best Regards,
Deepu

ralph
Posts: 167
Joined: Wed Apr 27, 2011 4:42 pm
anti_bot: 333

Re: Error in reading diffusivity from database

Post by ralph » Thu Apr 05, 2018 10:12 am

Welcome, Kamal!

next quesion ...
Which Thermo-Calc version did you use to create the GES5 file?

MICRESS 5.4 use a Thermo-Calc library from TC version R (32 bit) and requires an appropriate GES5 file format. For sure, a file generated with a 32 bit version of Thermo-Calc. But there were other internal format change, esp. regarding diffusion data, which broke backwards compatibility of GES5 files.

Ralph

kamalnath
Posts: 13
Joined: Mon Dec 11, 2017 9:17 pm
anti_bot: 333
Location: Columbus, Ohio

Re: Error in reading diffusivity from database

Post by kamalnath » Thu Apr 05, 2018 2:50 pm

Dear Ralph and Deepu,
Thank you very much for your quick responses. I greatly appreciate that.

Deepu: Probably you are right. I am also planning to use manual diffusivity instead of getting diffusivity from the database. Interestingly the reading the information from the thermodynamic database (TCNI8) seems to have no problem. Only when reading mobility database (MOBNI4), I have this issue.

Ralph: Since, I want to use the recent databases TCNI8 and MOBNI4, I am using the thermocalc version R2017b(32bit). As you have suggested, mobility database may have lost compatibility but the thermodynamic database works fine. So, I think I will use user-defined values for now.

Regards,
Kamal

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

Re: Error in reading diffusivity from database

Post by Bernd » Thu Apr 05, 2018 5:12 pm

Hi Kamal,

I am pretty sure that the problem is the combination of a .ges5 file created with a recent version of Thermo-Calc and the old TQ library of MICRESS version 5.4. However, I believe the problem could be only the mobility data, and if you are lucky, perhaps only those for the liquid phase.

In this case you can simply use the following diffusion data which you would have got with a newer MICRESS version at T=1694.98K for manual setting:

Diffusion coefficients in LIQUID (global):
Time: 0.0000 s
Temperature: 1694.98000000000 K
Composition NB: 5.0000E+00 wt%
Components Diff. Coeff. Frequency Term Activation energy
Flux Gradient [cm**2/s] [cm**2/s] [J/mol]
NB NB +5.183E-05 +1.608E-04 +1.596E+04


Diffusion coefficients in FCC_A1 (global):
Time: 0.0000 s
Temperature: 1683.46000000000 K
Composition NB: 3.2224E+00 wt%
Components Diff. Coeff. Frequency Term Activation energy
Flux Gradient [cm**2/s] [cm**2/s] [J/mol]
NB NB +1.090E-08 +0.879 +2.548E+05



Bernd

kamalnath
Posts: 13
Joined: Mon Dec 11, 2017 9:17 pm
anti_bot: 333
Location: Columbus, Ohio

Re: Error in reading diffusivity from database

Post by kamalnath » Fri Apr 06, 2018 5:29 pm

Thank you very much, Brend.

I plan to use these data. But in the manual entry, I need only "Diff.Coeff" and "Activation energy" . What is meant by "Frequency term". I tried looking it up in the manual(v6.3) but could not find it.

For example, for the LIQUID phase:
I understand the numbers 5.183E-05 and 1.596E+04 but what does 1.608E-04 mean ?

-Kamal

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

Re: Error in reading diffusivity from database

Post by Bernd » Sun Apr 08, 2018 6:14 pm

Hi Kamal,

When you input diffusion data manually by using "diagonal d" or the old-style "diff", you are requested for the frequency term and the activation energy (Arrhenius description).
Alternatively, you can also specify the diffusion coefficient and input zero for the activation energy. In the latter case you will have no temperature dependency, of course.

Bernd

Post Reply