Hertz Contact Theory (Sphere-Plane)
Theory and Physics
Overview
Professor, I've heard that Hertz contact is the most commonly used problem for verifying FEA contact analysis. Why is that?
Because for the contact of two elastic bodies, there exist closed-form analytical solutions for all aspects: contact radius, maximum contact pressure, and contact displacement. Hertz derived them in 1882. The contact between a sphere and a plane is the most fundamental and serves as the entry point for Code Verification of nonlinear contact analysis.
Even though it's nonlinear, there's an analytical solution?
The cleverness of Hertz theory lies exactly there. It includes geometric nonlinearity where the contact area changes with load, but by simultaneously solving the superposition of Boussinesq solutions for an elastic half-space and the geometric compatibility condition, a closed-form solution is obtained. However, there are prerequisites: the contact area must be sufficiently small compared to the plane, it must be within the elastic limit, and there must be no friction.
Governing Equations
Please show me the specific equations.
Case of an elastic sphere of radius $R$ pressed against an elastic plane with load $P$. With equivalent elastic modulus $E^* = [(1-\nu_1^2)/E_1 + (1-\nu_2^2)/E_2]^{-1}$ and equivalent radius $R^* = R$ (plane has $R_2 = \infty$):
Contact radius: $a = \left(\frac{3PR^*}{4E^*}\right)^{1/3}$
Maximum contact pressure: $p_0 = \frac{2aE^*}{\pi R^*} = \frac{1}{\pi}\left(\frac{6PE^{*2}}{R^{*2}}\right)^{1/3}$
Approach: $\delta = \frac{a^2}{R^*} = \left(\frac{9P^2}{16R^*E^{*2}}\right)^{1/3}$
What about the pressure distribution?
The pressure within the contact area follows a semi-elliptical distribution.
Maximum $p_0$ at the center, zero at the contact edge $r = a$. This distribution can be reproduced by superposition of Boussinesq solutions, and the stress field beneath the contact surface can also be calculated exactly. The maximum von Mises stress occurs slightly below the surface at $z \approx 0.48a$ and becomes the origin of rolling contact fatigue.
Benchmark Numerical Example
Please show me a specific numerical verification example.
Steel sphere ($R = 10$ mm, $E = 200$ GPa, $\nu = 0.3$) vs. steel plane (same material). $P = 100$ N.
$E^* = 200/(2 \times (1-0.09)) = 109.9$ GPa
$a = (3 \times 100 \times 0.01 / (4 \times 109.9 \times 10^9))^{1/3} = 0.0727$ mm
$p_0 = 3P/(2\pi a^2) = 9013$ MPa
$\delta = a^2/R = 0.000529$ mm
| Physical Quantity | Theoretical Value | Abaqus (C3D20R) | Error |
|---|---|---|---|
| Contact radius $a$ [mm] | 0.0727 | 0.0731 | 0.55% |
| Max. contact pressure $p_0$ [MPa] | 9013 | 8940 | 0.81% |
| Approach $\delta$ [mm] | 0.000529 | 0.000527 | 0.38% |
Accuracy within 1%. How fine was the mesh?
This is the result with more than 20 elements placed within the contact area. Since the contact radius $a$ is very small, the element size in the contact region needs to be about $a/10 \approx 7$ μm. Areas far away can be coarser, so a biased mesh is essential.
Physical Meaning of Each Term
- Time Variation Term of Conserved Quantity: Represents the rate of change over time of the physical quantity in question. Becomes zero for steady-state problems. 【Image】When filling a bathtub with hot water, the water level rises over time—this "rate of change per time" is the time variation term. The state where the valve is closed and the water level is constant is "steady-state," and the time variation term is zero.
- Flux Term: Describes the spatial transport/diffusion of a physical quantity. Broadly classified into convection and diffusion. 【Image】Convection is like "a river's flow carrying a boat," where things are carried by the flow. Diffusion is like "ink naturally spreading in still water," where things move due to concentration differences. The competition between these two transport mechanisms governs many physical phenomena.
- Source/Sink Term: Represents the local generation or destruction of a physical quantity (external force/reaction term). 【Image】Turning on a heater in a room "generates" thermal energy at that location. When fuel is consumed in a chemical reaction, mass is "destroyed." A term representing physical quantities injected into the system from the outside.
Assumptions and Applicability Limits
- The continuum assumption holds for the spatial scale.
- The constitutive laws of the material/fluid (stress-strain relation, Newtonian fluid law, etc.) are within their applicable range.
- Boundary conditions are physically reasonable and mathematically well-defined.
Dimensional Analysis and Unit Systems
| Variable | SI Unit | Notes / Conversion Memo |
|---|---|---|
| Characteristic length $L$ | m | Must match the unit system of the CAD model. |
| Characteristic time $t$ | s | For transient analysis, time step must consider CFL condition and physical time constants. |
Visualization of Verification Data
Quantitative comparison between theoretical and computed values. Error within 5% is the passing criterion.
| Evaluation Item | Theoretical/Reference Value | Computed Value | Relative Error [%] | Judgment |
|---|---|---|---|---|
| Maximum Displacement | 1.000 | 0.998 | 0.20 | PASS |
| Maximum Stress | 1.000 | 1.015 | 1.50 | PASS |
| Natural Frequency (1st) | 1.000 | 0.997 | 0.30 | PASS |
| Total Reaction Force | 1.000 | 1.001 | 0.10 | PASS |
| Energy Conservation | 1.000 | 0.999 | 0.10 | PASS |
Judgment Criteria: Relative error < 1%: ■ Excellent, 1–5%: ■ Acceptable, > 5%: ■ Needs Review
Numerical Methods and Implementation
Contact Algorithm Selection
What types of contact algorithms are there in FEA?
Mainly three types.
| Method | Principle | Accuracy | Computational Cost |
|---|---|---|---|
| Penalty Method | Reaction force proportional to penetration | Depends on penalty constant | Low |
| Lagrange Multiplier Method | Strictly enforces penetration = 0 | High accuracy | High (increased DOF) |
| Augmented Lagrangian Method | Hybrid of Penalty + Lagrange | High accuracy | Medium |
For Hertz verification, the Lagrange Multiplier or Augmented Lagrangian method is recommended. With the Penalty method, the contact pressure changes with the penalty coefficient value, so it's not suitable for verification.
What are the default settings in each solver?
Abaqus defaults to Augmented Lagrangian (*SURFACE INTERACTION + *CONTACT PAIR). Nastran uses penalty-based for SOL 101 GLUE/SLIDE contact. Ansys defaults to Augmented Lagrangian. For V&V of contact problems, it is mandatory to clearly state the algorithm used.
Mesh Design
What is the most important thing in mesh design for contact problems?
The element size on the contact surface must be 1/10 or less of the contact radius $a$. A coarse mesh makes the contact surface stepped, leading to inaccurate pressure distribution.
Recommended configuration:
- Near contact surface: $h_{elem} = a/10$. Structured mesh with HEX20.
- Areas away from contact surface: Coarsen with bias. Ratio around 2.0.
- Quarter-symmetry model of the sphere (symmetry conditions on two faces).
- Plane side model size with radius of at least $10a$.
Which surface should be master/slave for the contact pair?
In Abaqus convention, the convex surface (sphere) is slave, the plane is master. The slave side should have a finer mesh. Similar thinking applies in Nastran and Ansys; the stiffer surface is master. If the mesh density differs greatly between the two surfaces, it affects pressure accuracy, so it's desirable to match the mesh density on both sides in the contact region.
Convergence Points
Contact analysis often fails to converge. What are the countermeasures?
Smooth contact like Hertz problems usually converges without issue. If it doesn't converge:
1. Check the initial contact gap. If the gap is too large, contact is not detected in the initial step.
2. Split the load into multiple steps. Enable geometric nonlinearity with NLGEOM=ON.
3. Temporarily enable contact stabilization (Abaqus *CONTACT CONTROLS, STABILIZE).
4. Use smoothing options (Abaqus *SURFACE SMOOTHING) to reduce discontinuities on the contact surface.
Why is NLGEOM necessary?
In Hertz contact, the contact area changes as the sphere and plane approach. This is geometric nonlinearity, so NLGEOM=ON is mandatory. If not ON, the solver uses a linear approximation based on the initial gap state and cannot obtain the correct contact area.
Low-Order Elements
Low computational cost and easy to implement, but accuracy is limited. Large errors may occur with coarse meshes.
High-Order Elements
Achieve higher accuracy with the same mesh. Computational cost increases, but the required number of elements is often smaller.
Newton-Raphson Method
Standard method for nonlinear problems. Quadratic convergence within the convergence radius. Convergence judged by $||R|| < \epsilon$.
Time Integration
Explicit: Conditionally stable (CFL condition). Implicit: Unconditionally stable but requires solving simultaneous equations at each step.
Visualization of Verification Data
Quantitative comparison between theoretical and computed values. Error within 5% is the passing criterion.
| Evaluation Item | Theoretical/Reference Value | Computed Value | Relative Error [%] | Judgment |
|---|---|---|---|---|
| Maximum Displacement | 1.000 | 0.998 | 0.20 | PASS |
| Maximum Stress | 1.000 | 1.015 | 1.50 | PASS |
| Natural Frequency (1st) | 1.000 | 0.997 | 0.30 | PASS |
| Total Reaction Force | 1.000 | 1.001 | 0.10 | PASSRelated TopicsV&V・品質保証Boussinesq problem (point load on a semi-infinite elastic body)用語集Hertz接触 — CAE用語解説StructuralContact mechanics in MBDV&V・品質保証Cantilever beam bending (concentrated load)Structural3次元弾性体解析Open Source CAECalculiX非線形解析
この記事の評価 ご回答ありがとうございます! 参考に なった もっと 詳しく 誤りを 報告 |