How to input isotropic thermal expansions

[mcteagu]

So i'm sure its already out there but I'm not seeing it. How would i go about inputting different thermal expansion coefficients in different directions? I.e. i'm modeling alumina bi-crystals, which are trigonal. So alpha 11 and alpha 33 are different, do I put it in as a tensor material of some sort?

[Daniel Schwen]

You could use a ComputeVariableEigenstrain material for this (as described http://mooseframework.org/wiki/PhysicsModules/TensorMechanics/PlugAndPlayMechanicsApproach/#stress-free-strains)

The eigen_base parameter would contain the anisotropy, the var_dep property would be a function of T.

Daniel

On Thu, Jul 14, 2016 at 3:19 PM mcteagu <mct...@sandia.gov> wrote:

So i'm sure its already out there but I'm not seeing it. How would i go about inputting different thermal expansion coefficients in different directions? I.e. i'm modeling alumina bi-crystals, which are trigonal. So alpha 11 and alpha 33 are different, do I put it in as a tensor material of some sort?

--
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 https://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/36862f44-1f3f-4e49-81fb-aec6116554cb%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Teague, Melissa C]

Daniel, 

  i'm still struggling with how to input this. So for eigen_base parameter I would input the thermal expansion tensor? For reference what i'm trying to do, but failing miserably at is to model my experiments 

I have two single crystals of alumina bonded together and cooled to room temperature. Due to the anisotropic thermal expansion there is residual stress around the interface, which I can map/measure. I'm trying to figure out how to represent this in a model. Any help would be greatly greatly appreciated

From: moose...@googlegroups.com [mailto:moose...@googlegroups.com] On Behalf Of Daniel Schwen.

Sent: Thursday, July 14, 2016 3:25 PM
To: moose-users
Subject: [EXTERNAL] Re: How to input isotropic thermal expansions

 

You could use a c material for this (as described http://mooseframework.org/wiki/PhysicsModules/TensorMechanics/PlugAndPlayMechanicsApproach/#stress-free-strains)

--
You received this message because you are subscribed to a topic in the Google Groups "moose-users" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/moose-users/gblPCBc34k8/unsubscribe.
To unsubscribe from this group and all its topics, send an email to moose-users...@googlegroups.com.
Visit this group at https://groups.google.com/group/moose-users.
To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/CAPxoKqdyE4s9d%3D9C5xQ3xnOgyJUTS8sZW7J5PWuEGASD6nqPsA%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

[Daniel Schwen]

Hello Melissa,

if you are using EBSD reader you are probably working with order parameters, right? Let's talk about eigenstrain first. Your setup is a bit complicated. I'm assuming that the anisotropic thermal expansion rotates with the crystal orientation. I.e. you need two different thermal expansions for the two crystals. Assuming that one crystal is defined through the order parameter eta1 and the other through the order parameter eta2 you'd need a CompositeEigenStrainMaterial. 

Step one: figure out what your thermal expansion tensor is

0.1 0.0 0.0

0.0 0.2 0.0

0.0 0.0 0.3

for example would be one with the major axis along z, the minor axis along x and intermediate expansion along y. You asked the list about rotating tensors using EBSD data once. I don't know if you made any headway in implementing this. If so you can use your class to create a rotated Eigenstrain, otherwise compute a rotation matrix and use it to rotate the tensor.

I'm attaching a simple input. Check it out and get back to me with your followup questions.

Daniel

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/1468606753806.80359%40sandia.gov.

[Teague, Melissa C]

yes the expansion depends on the plane of crystals, one thing I don't quite get is order parameter. Are those just way of defining the two crystals. I have mesh with two blocks (one block for each crystal). I'm not planning on doing any phase field or grow grains or such. "Just" mechanics and thermal modelling

---

From: moose...@googlegroups.com <moose...@googlegroups.com> on behalf of Daniel Schwen <dan...@schwen.de>
Sent: Friday, July 15, 2016 2:09 PM
To: moose...@googlegroups.com
Subject: Re: [EXTERNAL] Re: How to input isotropic thermal expansions

 

To view this discussion on the web visit https://groups.google.com/d/msgid/moose-users/CAPxoKqe%2BBbBhCLESzVP-nhohswdqgi_j0c_MSj20WdLFHEQL%2Bg%40mail.gmail.com.