UD = Unidirectional (continuous fiber)
MFH = Mean Field Homogenization
The purpose of this post is to:
1) Explain and demonstrate how non-isotropic material test data is imported into the material calibration app.
a) The images above define the axes nomenclature we use in the app for test data import.
b) Explain how stiffness property data can be converted to stress-strain data for purpose of import.
2) Show a worked example of the creation of a MFH material model for a transversely isotropic UD lamina.
In characterizing UD composite lamina, it is common to use handbook data for the fiber properties, for the matrix properties and to test one or more coupons (specimens) cut from the lamina. Of these three pieces of information, it is typical to trust the fiber properties and the coupon testing data more than the matrix handbook data. For this example, we will assume that the fiber and matrix properties are isotropic linear elastic.
Because I do not have test data, I am going to generate some synthetic test data using the Abaqus/CAE Micromechanics plugin (FE-RVE plugin). We will use the default properties in the plugin, the default type of UD_HexPack, and an Interface ratio of 0.0
Volume Fraction of the fiber = 0.40 | ||
Fiber Properties | E = 8.0e+10 Pa | Poisson's ratio = 0.2 |
Matrix Properties | E = 3.35e+9 Pa | Poisson's ratio = 0.35 |
After running the FE-RVE plugin, the following properties can be found in the property module. While there are 9 numbers in the table below, only 5 are independent (transversely isotropic).
We will imagine that there are/were 3 tests:
Test #1: The E1 modulus and nu12 value would come from a simple pull test on a specimen cut from the lamina such that the fibers are aligned with the direction of the pull. In the app, we show this with a picture. The actual test data would be stress versus strain, but since it is linear, it is common to take the slope of this line (stiffness or modulus) and record it in handbooks.
For the current calibration app (R2022x, FD01), the coupon level test data must be imported as stress and strain, so we reconstitute that stress-strain test data as shown in the table below (and contained in the attached Excel file). Notice that there is a color coding, the E1 value and the nu12 value are used to generate the table below. See the attached Excel file for more details.
In the calibration app, we speak of the "primary" direction / loading direction. And the transverse direction is in the plane of the lamina at 90 degrees to the loading direction. This is seen in the tool tip in the app by hovering your mouse over the text "In-plane rotation angle".
Test #2 is a uniaxial pull test on a specimen cut from the lamina such that the loading direction is at 90 degrees to the fibers. This test would have been used to determine E2 and nu23 (orange colored in the output from the FE-RVE).
The E2 and nu23 property values are used to reconstitute stress-strain information like this:
As per the diagram and definition at the top of this post, we define "Lateral" to mean the "through-the-thickness" direction.
Test #3 is a shear test (Google "rail shear test image"). The G12 shear modulus is used to reconstitute a shear stress-shear strain data set like this.
In the narrated video below, we will demonstrate how the test data sets are imported into the calibration app and how the creation of a MFH material model is performed.This narrated video was created using 3DX R2022x HotFix 1.21. It was created on April 2, 2022.
On listening to this video, I noticed a small mistake. At the 4 minute mark, I said "...any of the Poisson ratio or transverse / longitudinal things..." It should have been "... transverse / lateral things"
Note: You may see the axes referred to as X,Y,Z or 1,2,3 or L,T,T'
Excel file:
Back to: Sharing Material Test Data
Back to: Material Modeling and Calibration - An Overview and Curriculum