Orthotropic Elastic Material Calibration

 

Let’s imagine that testing has been performed on 3 samples taken from an orthotropic material. Each sample is subjected to a uniaxial deformation and the following elastic responses recorded:
 

The purpose of this post is to explain and show how one would use this test data in the calibration of an elastic material model, using the 3DExperience material model calibration app.  As one imports test data that comes from non-isotropic materials, one must tell the software something about the orientation of the sample.  Using a series of images below, I show the import of the test data for the leftmost sample above (we will call it "Uni_X", loaded in the X-direction).  I begin the test data import as shown here, highlighting the test data of interest.  The Excel file used is attached in the zip file at the bottom of this post.
 

Clicking "Next", we get a dialog where the app wants us to tell it what kind of data is in each column.
 

The software uses the terms "Transverse" strain and "Lateral" strain, and one must refer to the documentation to better understand the definition for these two terms.   In the docs:

Doc section:  About Built-in Time Domain Simulation Models

The convention used in the Material Calibration app is that the 1-direction is the direction of applied load. The transverse and lateral directions are respectively aligned with the global 2- and 3-directions.

As we click "Next", we get to the dialog where we must click on the "Apply orientation" option. 
 

We have a 2D option and a 3D option for specifying the orientation of the sample.  The 2D option is pretty well explained by the diagram.
 

The 3D option is a little more complicated.

Note:  There was a bug in this feature prior to version R2024x FD01.  I am using a development version to create this post.  This version was released to the public cloud on February 10, 2024.
 

Comparing the above image to the test image below.

For this "Uni-X" test data, the correct definition of the 1-direction is that it is aligned with X.  By definition the "Transverse" strain is in the 2-direction.  So, the 2-direction below is aligned with Y.   The "Lateral" strain is in the 3-direction, which is aligned with Z.
 

For the "Uni-Y" test data, the correct definition of the 1-direction is that it is aligned with Y.  By definition the "Transverse" strain is in the 2-direction.  So, the 2-direction below is aligned with X.   The "Lateral" strain is in the 3-direction, which is aligned with Z.
 

For the "Uni-Z" test data, the correct definition of the 1-direction is that it is aligned with Z.  By definition the "Transverse" strain is in the 2-direction.  So, the 2-direction below is aligned with X.   The "Lateral" strain is in the 3-direction, which is aligned with Y.
 

By the way, this synthetic test data was produced by running Abaqus models with this material model (units are SI, stress is Pa):
 

Here is a narrated video of setting up this calibration. It is performed in numerical mode. For FE mode this whole orientation issue is moot, because in FE model, the user is responsible for providing run-ready FE models that are correctly representative of the tests performed.  This video was created on December 8, 2023 using a development version of the 3DExperience platform that will look like the public version to be released in February of 2024.

The final answer we get from the calibration agrees with our known answer:
 

The zip file below contains the Excel file of the test data, the final 3dxml file and the 3 Abaqus .inp files that were used to generate the synthetic test data.

 

Back to:  Sharing Material Test Data

Back to:  Material Modeling and Calibration - An Overview and Curriculum