Abaqus Orthotropic material

[Harryboy]

Hi,

Working on the implementation of a orthotropic Kirshhoff material in ABAQUS EXPLICITE using finite strains.
When the result is compared to the result from ABAQUS *ELASTIC,type=ENGINEERING CONSTANTS the result differ significantly. I am aware of the shortcomings of the Kirshhoff material, but I cant realy figure out why the difference is that large.

So, my question to you all:
Do anyone know what type of model ABAQUS is using for the card *ELASTIC,type=ENGINEERING CONSTANTS??
My guess is that it is a hypo elastic model based on the Green-Naghdi rate, since the Jaumann rate wont do the trick.

Best regards
Harrysboy

[Jorgen]

Harrysboy,

:arrow: How do you define a Kirchhoff material?
:arrow: The *Elastic, type=engineering constants material is an anisotropic elastic material model. I am not sure of the details of how ABAQUS has implemented that model, but why do you think that a elastic material would need to be implemented using a rate equation?

- Jorgen

[Tompa]

Hi
I think that the model used by ABAQUS for *elastic is only valid for small strains, cf. *ELASTIC in the Keyword reference manual (v6.6) and 17.2.1 Linear elastic behavior in the Analysis User's Manual.
The rate equations used by ABAQUS is reported in 1.5.3 Stress rates in the theory manual (v6.6). Jaumann rate for built in constitutive models for continuum elements and Green-Naghdi for structual elements.

Take care
Tompa

[arajaneesh]

Hi, i have doubt as follows:

1) we use only D11,D12,D22 and D16 are used in calculation of deflection of the plate, from the theory or analytical solutions, when we doing for orthotropic material .

Let in Abaqus, i want to verify the result from my analytical formulation:
I wanted to use orthotropic material model in abaqus with 3D deformable element. but when i add only D11,D12,D22 and D16; and remaining as zeros then abaqus gives warning as "constitutive matrix is not diagonally dominant" and error as "trying to invert singular matrix"

I am able to undestand warning, caused because of zero values in the Dij matrix.

Now my question is : how to model 3d deformable orthotropic plate with transverse loading.

2) Or we can model only using shell element.

3) let I input all the values of Dijkl matrix using 3rd direction properties of material, then my Abaqus model is more stiff compared to analytical model because we considered 3rd direction properties. so how to aviod this and how to represent the orthotropic plate which contain stiffness of D11,D12,D22 and D16; so that i can represent represent as i assumed in analytical solution.

4) in calculation of Dij for the analytical formulation we use height of the plate. But when i am calculating the Dijkl properties required in abaqus input i dont have to use. what about this matter.

Thanks and regards

[VTN]

Why don't you use *ELASTIC,type=ENGINEERING CONSTANTS. Having Dij you can convert it back to E11, E12 ......The best way is to use shell element. If you use solid element, make sure that you have enough element on the thickness (3 to 5 if you have bending effect).

Regards

[arajaneesh]

Thank you VTN for your reply. :)