Modifying Grain Growth example Mesh

[Jonathon Gardner]

I am trying to calculate the grain growth using the example, but I am having trouble when changing the Mesh. When I change the Mesh size it shows that there are no grains, when I have an initial condition of 20. Here is my input file. 

Thanks

Jonathon

[Cody Permann]

On Wed, Apr 15, 2015 at 2:35 PM Jonathon Gardner <jonathon...@gmail.com> wrote:

I am trying to calculate the grain growth using the example,

This is vague, please be specific here. Are you trying to calculate the average grain size? The change in grain size?

 

but I am having trouble when changing the Mesh.

How are you changing the mesh? Bigger, smaller, number of elements?

 

When I change the Mesh size it shows that there are no grains, when I have an initial condition of 20. Here is my input file. 

What field? You are only showing order parameter fields "PolycrystalVariables" Do you want to see the "unique_grain" field? If so, try the "*_graintracker.i" file instead.

Cody

 

Thanks

Jonathon

--
You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users...@googlegroups.com.
Visit this group at http://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/a841b1f6-7b2d-4022-8c41-f64d6125a1a6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Tonks, Michael R]

Jonathon,

There are several problems with your edited input file. 

First, in the phase field method, order parameters are used to represent grains. To reduce computational cost, more grains can be used then order parameters, such that order parameters are used to model more than one grain. In your input file you are using 5 order parameters to represent 20 grains, which isn't really enough. You are going to have multiple grains merge together because you aren't using enough variables to represent them (only 5). You should use at least 17 variables in 2D, and you will need to turn on the grain tracker user object to remap grains when necessary (see the grain tracker example). The op_num is set in the global params block in the example. 

Second, in the phase field model interfaces have a finite width that is set as an input in the material. The grain boundary width is set to 14 nm in the material, but your domain size is only 10 nm so no grains will ever form because the grain boundary is wider than the domain. You decreased the domain size by 100 so you should decrease the grain boundary width by 100 as well, i.e. wGB = 0.14.

Finally, you are using a domain size of 10 nm (see the length scale set in the material). That is really small. You are basically running a mesoscale method at the atomic scale which isn't a good idea. If you really need to model 10 nm of material, you might want to use MD.

--
You received this message because you are subscribed to the Google Groups "moose-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to moose-users...@googlegroups.com.
Visit this group at http://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/a841b1f6-7b2d-4022-8c41-f64d6125a1a6%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

*************************

Michael R. Tonks

Fuel Modeling and Simulation Department

Idaho National Laboratory

208-526-6319

michae...@inl.gov

[Cody Permann]

Jonathon,

I'll add to Mike's explanation and make one more suggestion, try starting from the working example and changing just one parameter at a time. Rerun your simulation to make sure that the change you are making makes sense and has the desired effect. As Mike pointed out, there are several problems with your input file so try to keep your input file working at each stage. Peacock (our GUI tool for building input files, executing and visualizing) is great for this kind of workflow.

Cody

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/CA%2BEq2494XeJzBff9OURcYOCAS7yVyKLcqpr4Zrq5C73gcNSs_A%40mail.gmail.com.

[Jonathon Gardner]

Okay thank you that does help. I started with the "grain_growth_2D_voronoi" input file and everything I had changed up until the Mesh was working. Changing the grain boundary width fixed the problem.

Thanks again

Jonathon