General analysis procedures for piezoelectric materials

This problem contains basic test cases for one or more Abaqus elements and features.

In this section the general analysis procedures for elements that include piezoelectric coupling are discussed.

This page discusses:

ProductsAbaqus/Standard

Transient dynamic analysis for piezoelectric materials

Elements tested

C3D8E

Features tested

The transient dynamic analysis capability for elements that include piezoelectric coupling is illustrated. Both concentrated nodal electrical charges and potentials are applied in separate analyses.

Problem description

A piezoelectric bar [1 × 1 × 10] is subjected to an electrical potential. The potentials on the longitudinal top surface are prescribed to 1, and the potentials on the longitudinal bottom surface are prescribed to 0. The electrodes are simulated by using equations that set all the potentials on a face to the same prescribed value. The material is polarized in the local 3-direction.

The block is modeled using five C3D8E elements along the length. The material properties for the PZT-5H material that is used in the tests are as follows:

Elasticity properties:
Engineering constants
E1 60.61 GPa
E2 60.61 GPa
E3 48.31 GPa
ν12 0.289
ν13 0.512
ν23 0.512
G12 23.5 GPa
G13 23.0 GPa
G23 23.0 GPa
Piezoelectric coupling matrix (strain coefficients):
[0000741000000741-274-274593000]10-12    m/volt
Dielectric matrix for fully constrained material:
[1.5050001.5050001.301]10-8     farad/meter

The piezoelectric coefficient matrix gφ and the dielectric matrix Dφ(σ) for an unconstrained material, which are commonly used electrical properties in the piezoelectric literature, can be expressed in terms of the piezoelectrical properties mentioned above. These relationships are given in Piezoelectric analysis. These properties are commonly provided by the manufacturer. For the PZT-5H material the properties are as follows:

Piezoelectric coefficient matrix gφ:
[000026.800000026.8-9.11-9.1119.7000]10-3    volt m/N
Dielectric matrix for an unconstrained material Dφ(σ):
[2.770002.770003.01]10-8     farad/meter

The tests involve a transient dynamic step in which the potential on the top surface is ramped up to 1 volt in 0.014 seconds and then held constant for the remainder of the step. The results at the end of the step correspond to the static solution.

Results and discussion

The applied electrical potential of 1 volt results in a potential gradient of 1 volt/m. The piezoelectric constants d3 33φ and g3 33φ can be used to estimate the electrical charge per unit area. In the case of an unconstrained material

d3 33φ=ε33E3

and

g3 33φ=ε33q3,

where E3 is the potential gradient and q3 is the charge density in the local 3-direction. So the charge density q3 is equal to d3 33φ/g3 33φ= 3.01 × 10−8. The area to which the voltage is applied is 10; therefore, the static reaction charge should be about 3.01 × 10−7. The results of ppzodyn1.inp confirm this reaction charge. In the input file ppzodyn2.inp a concentrated nodal electrical charge of 3.01 × 10−7 is applied instead of a potential value of 1 at the top surface. This results in a potential of 1 volt on the top surface.

Input files

ppzodyn1.inp

Dynamic analysis with prescribed potentials.

ppzodyn2.inp

Dynamic analysis with concentrated nodal electrical charges.

Geometrically nonlinear static analysis for piezoelectric materials

Elements tested

C3D20E

Features tested

The geometrically nonlinear static analysis capability for a piezoelectric material is illustrated. A beam clamped at both ends is subjected to a potential that results in a loading that reaches the critical buckling load.

Problem description

A beam of piezoelectric material is clamped at both sides and is subjected to an electrical potential. The beam is 0.4 m long with a width of 0.006 m and a thickness of 0.005 m. The potentials at one end of the beam are prescribed to 500 kvolts, and the potentials at the other end are prescribed to 0 kvolt. The electrodes are simulated by using equations that set the potential of all the nodes at each end of the beam to the same prescribed value. In the first step a small load is applied to the center to induce a small geometric imperfection.

The block is modeled using 20 C3D20E elements. The material properties for PZT-5H, which is used for the simulation, are given in the previous section.

Results and discussion

The critical buckling load for the beam is

P=4π2EIL2=745N,

where E is the Young's modulus in the longitudinal direction and I is the appropriate moment of inertia for the beam section. The analysis shows a critical compressive force of 773 N. The compressive force converges to the analytical buckling load with mesh refinement.

Large rotations in piezoelectric analysis

Elements tested

C3D4E

C3D6E

C3D8E

C3D10E

C3D20RE

Features tested

Large rotations for different piezoelectric element types.

Problem description

Five blocks modeled with different piezoelectric element types are subjected to an electrical potential. The potentials at one side are prescribed to 1 volt, and the potentials at the opposite side are prescribed to 0 volt. The blocks are tied to three orthogonal surfaces to prevent unconstrained rigid body motions but are free to move tangentially with respect to the surfaces. The surfaces are also used to prescribe the rigid body rotations.

Results and discussion

The applied potential gradient remains constant in magnitude but rotates appropriately with the element.

Input files

ppzolarrot.inp

Large rotations with piezoelectric elements.

Validation of piezoelectric material behavior

Elements tested

C3D8E

Features tested

Validation of piezoelectric material properties using a general static analysis.

Problem description

A block of PZT-5H is subjected to different loadings from which the piezoelectric material properties can be validated.

Results and discussion

In the first step the potentials at the two opposite sides in the local 3-direction of the material are prescribed. Sufficient boundary conditions are applied to prevent rigid body motions, but the model is otherwise unconstrained. The piezoelectric constants d3 33φ= 593 × 10−12 and g3 33φ = 19.7 × 10−3 can be expressed in terms of the strain ε33, the potential gradient E3, and the charge density q3 as

d3 33φ=ε33E3

and

g3 33φ=ε33q3.

The piezoelectric constants d3 11φ, d3 22φ, d3 33φ, g3 11φ, g3 22φ, and g3 33φ are verified by using the numerically obtained values of the strains ε11, ε22, and ε33. The dielectric constant D33φ(σ)in the local 3-direction for an unconstrained material is given by

D33φ(σ)=q3E3.

The numerical results for q3 and E3 confirm the above relationships. In Steps 2–4 the model is charged in different ways verifying the same piezoelectrical material parameters as in Step 1. In Step 2 the potentials of the bottom and the top surface are switched. In Step 3 a nodal concentrated electrical charge is applied, and in Step 4 a distributed electrical charge is applied instead of prescribing the potentials. In Step 5 a potential gradient is applied in the local 1-direction to verify the piezoelectric properties d1 13φ, g1 13φ, and D11φ(σ).

In Steps 6–7 an open circuit condition is applied (the potential gradient is not prescribed by voltage boundary conditions), which results in reaction charges that are equal to zero. The piezoelectric constitutive equations can be written in different forms. In particular, the strain can be expressed in terms of either the potential gradient E or the charge density q. If the constitutive relation is expressed in terms of the potential gradient, the compliance data (typically denoted as SE in the piezoelectric literature) define the mechanical behavior at zero potential gradient. In Abaqus the stiffness data at zero potential gradient are used to specify the mechanical behavior. If the constitutive relation is expressed using the charge density, the compliance matrix (typically denoted as SD in the piezoelectric literature) defines the mechanical behavior at zero charge density. The compliance SD can be obtained from the compliance SE and the electrical properties. For the PZT-5H material, S11D = 14.05 × 10−12, S12D = −7.27 × 10−12, S13D = −3.05 × 10−12, S31D = −3.05 × 10−12, and S33D = 8.99 × 10−12. By loading the model at zero charge (in open circuit condition), these elastic compliances are verified.

Input files

ppzovallin.inp

Geometrically linear static analysis used to validate piezoelectric material properties.

ppzovalnlg.inp

Geometrically nonlinear static analysis used to validate piezoelectric material properties.

ppzovalnlg_tfv.inp

Geometrically nonlinear static analysis used to validate temperature- and field-variable-dependent piezoelectric material properties.