Warning message during simulation

Discussion of numerical issues which are not directly linked to specific applications
Post Reply
betleenkim
Posts: 41
Joined: Tue May 07, 2013 1:01 pm
anti_bot: 333

Warning message during simulation

Post by betleenkim » Mon May 13, 2013 6:19 am

Hi!

I have a question about the simulation.
During simulation i saw the message as below.

Warning in routine solveCGetDiffMatTQ!
Invalid Arrhenius description of diffusion coefficient of
MN (flux) and AL (gradient) in FCC_A1
Diffusion coefficient [m**2/s] : +2.04147E-18
Frequency term [m**2/s] : +7.09599E-21
Activation energy [J/mol] : -59598.

I do not know what the problem is.
Does this effect much on the result of my simulation?

Thank you!!

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

Re: Warning message during simulation

Post by Bernd » Mon May 13, 2013 1:00 pm

Hi betleenkim,

welcome to the MICRESS forum!

The warning which you got is a very common one, and there is no reason to worry about!

The reason for the warning is the following: When reading diffusion data from database, MICRESS is converting the data to an Arrhenius-type description, i.e. extracting a frequency factor and an activation energy. This is done numerically by requesting the diffusion coefficient at two slightly different temperatures.
Sometimes it happens that a negative value for the activation energy is obtained. This seems unreasonable from a physical point of view, but may happpen nevertheless. In those cases, MICRESS is creating this warning and "flattens" the Arrhenius description, i.e. sets the activation energy to a value of 0 and the frequency factor to the value of the diffusion coefficient. This is done in order to be on the safe side in case of numerical problems. Strictly speaking,this "flattening" may reduce accuracy because no extrapolation for temperature is done for the diffusion term for which the warning has been issued. But, in most cases, those diffusion coefficients are anyway not relevant (like in your case very small), or the temperature dependency is minimal.

Bernd

betleenkim
Posts: 41
Joined: Tue May 07, 2013 1:01 pm
anti_bot: 333

Re: Warning message during simulation

Post by betleenkim » Tue May 14, 2013 2:43 am

Dear Bernd

Thank you for your reply.
I think that i will keep bothering you asking so many other questions :D

Thank you again!!

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

Re: Warning message during simulation

Post by Bernd » Tue May 14, 2013 3:19 pm

I am looking forward to your questions! ;)

Bernd

Post Reply