In-file returns strange range of seed radii

technical aspects of .dri file generation (e.g. debug mode ) etc...
Post Reply
Hans_IKV
Posts: 8
Joined: Tue Oct 13, 2009 12:13 pm
anti_bot: 333

In-file returns strange range of seed radii

Post by Hans_IKV » Mon Dec 14, 2009 1:28 pm

Dear Bernd,

I have written a dri-file with one seed type, seed density model and two classes of radii.
I am wondering why micress returns this strange range of radii in the in-file.

# How many classes shall be chosen for the critical radius?
2
# Specify radius [micrometers] and seed density [cm**-3] for class1
0.04 50000000000
# Specify radius [micrometers] and seed density [cm**-3] for class2
0.25 1000000
# Class 1: 5429 seed(s), 1.4500E-01 < radii < ********** [micrometers]
# Class 2: 4 seed(s), 1.0000E-08 < radii < 1.4500E-01 [micrometers]

Do you have any explanation?

Thank you!

Hans

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

Re: In-file returns strange range of seed radii

Post by Bernd » Mon Dec 14, 2009 6:37 pm

Dear Hans,

the reason is that you have to input the radii in decreasing order - I admit that this is not sufficiently explained in the text and that it should also be checked during execution...

Bernd

Hans_IKV
Posts: 8
Joined: Tue Oct 13, 2009 12:13 pm
anti_bot: 333

Re: In-file returns strange range of seed radii

Post by Hans_IKV » Wed Dec 16, 2009 11:04 am

This information helped!
Thank you!

Hans_IKV
Posts: 8
Joined: Tue Oct 13, 2009 12:13 pm
anti_bot: 333

Re: In-file returns strange range of seed radii

Post by Hans_IKV » Mon Jan 04, 2010 2:13 pm

Dear Bernd,
wish you a happy new year!

Unfortunately I have problems with the seed radii again. I have started a simulation with three phases and with one seed type for each phase (seed_density model). And I received asterisks in the In-file again:

# Number of types of seeds?
3
#
# Input for seed type 1:
# ----------------------
# Type of 'position' of the seeds?
# Options: bulk region interface triple quadruple [restrictive]
bulk
# Phase of new grains?
1
# Reference phase?
0
# Which nucleation model shall be used?
# Options: seed_undercooling seed_density
seed_density
# Integer for randomization?
13
# How many classes shall be chosen for the critical radius?
1
# Specify radius [micrometers] and seed density [cm**-3] for class1
0.025 10000000
# Class 1: 18 seed(s), 3.0429+247 < radii < ********** [micrometers]
# Shield effect:
# Shield time [s] ?
1.00000E-04
# Time between checks for nucleation? [s]
1.00000E-04
# Shall random noise be applied?
# Options: nucleation_noise no_nucleation_noise
no_nucleation_noise
#
# Input for seed type 2:
# ----------------------
# Type of 'position' of the seeds?
# Options: bulk region interface triple quadruple [restrictive]
bulk
# Phase of new grains?
2
# Reference phase?
0
# Which nucleation model shall be used?
# Options: seed_undercooling seed_density
seed_density
# Integer for randomization?
13
# How many classes shall be chosen for the critical radius?
1
# Specify radius [micrometers] and seed density [cm**-3] for class1
0.019 1000000000
# Class 1: 400 seed(s), 9.5000E-03 < radii < 2.8500E-02 [micrometers]
# Shield effect:
# Shield time [s] ?
1.00000E-04
# Time between checks for nucleation? [s]
1.00000E-04
# Shall random noise be applied?
# Options: nucleation_noise no_nucleation_noise
no_nucleation_noise
#
# Input for seed type 3:
# ----------------------
# Type of 'position' of the seeds?
# Options: bulk region interface triple quadruple [restrictive]
bulk
# Phase of new grains?
3
# Reference phase?
0
# Which nucleation model shall be used?
# Options: seed_undercooling seed_density
seed_density
# Integer for randomization?
13
# How many classes shall be chosen for the critical radius?
1
# Specify radius [micrometers] and seed density [cm**-3] for class1
0.01 500000000000
# Class 1: **** seed(s), 3.0429+247 < radii < ********** [micrometers]
# Shield effect:
# Shield time [s] ?
1.00000E-04
# Time between checks for nucleation? [s]
1.00000E-04
# Shall random noise be applied?
# Options: nucleation_noise no_nucleation_noise
no_nucleation_noise
#

Are there any other rules I have to consider beside the one you already mentioned?

Besides this, is there a possibility to make the range smaller?

Hans

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

Re: In-file returns strange range of seed radii

Post by Bernd » Mon Jan 04, 2010 4:08 pm

Dear Hans_IKV,

Thank you very much for reporting this bug which is occurring if only one radius class is entered! I fixed the bug right know, so that with the coming release the problem will be solved.

The behaviour should be the following: the radius range for class i is determined as
(r(i+1)+r(i))/2 < radius < (r(i-1)+r(i))/2,
the first class is assumed to be symmetric around r(i),
the last class (smallest radius) goes down to nearly zero

Example:
r(1) = 0.5
r(2) = 0.4
r(3) = 0.2

==> 0.45 < radius1 < 0.55
==> 0.30 < radius1 < 0.45
==> 1E-8 < radius1 < 0.3

If there is only one class, of course :? , this scheme is not working. In the new release (with bug-fix), with only one class the width of the class will be zero.

To define a single class with a given width, a second dummy class has to be added which has a low density and thus does not occur. For examlple, to define a class single between 0.45 and 0.55 µm, the input would be as follows:


# Which nucleation model shall be used?
# Options: seed_undercooling seed_density
seed_density
# Integer for randomization?
13
# How many classes shall be chosen for the critical radius?
2
# Specify radius [micrometers] and seed density [cm**-3] for class1
0.5 10000000
0.4 0.00001

Note: In the present version, a plain 0 is not allowed for the density. In the fixed version, it will be allowed!

Note also: For random clustering of the seeds like it is observed for example for primary Silicon in aluminium piston alloys, a extra (cluster) radius and (cluster) density value may be given in each line.

Bernd

mogeritsch
Posts: 15
Joined: Fri Feb 07, 2014 10:27 am
anti_bot: 333

Re: In-file returns strange range of seed radii

Post by mogeritsch » Sat May 12, 2018 2:41 pm

Hello Bernd,

can you give a sample about seed density for the clustering? The first two numbers are clear, the radius and the seed density, but what exactly do I define with the cluster and the cluster density? An example would be helpful.
Many Thanks,

Johann

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

Re: In-file returns strange range of seed radii

Post by Bernd » Mon May 14, 2018 9:40 am

Dear Johann,

When you define seed classes with different radii in the seed density model, for each class a range is assigned in order to get a smooth distribution function. In the case you show, there seems to be a format problem when outputting the upper range of the first class. I need to look up the place in the code to get an idea what could be the reason for that. Did you use the newest MICRESS version?

Bernd

Post Reply