Modeling and Machining Parameter Adjustment of Hypoid Bevel Gears

Hypoid bevel gears are widely used in automotive, mining, and heavy machinery because of their high load capacity, high contact ratio, and smooth transmission. As digital design and manufacturing technologies advance, the demand for accurate models and reliable machining parameters for hypoid bevel gears continues to grow. In my research, I focused on three main aspects: solid modeling of hypoid bevel gears using the HFM method, development of a tooth contact analysis (TCA) program, and adjustment of pinion control parameters to achieve desired contact patterns. This article summarizes my theoretical derivations, numerical procedures, and practical findings.

Hypoid bevel gears are a type of spiral bevel gear with offset axes. Their tooth surfaces are generated by complex cutting processes. The geometry and meshing performance depend strongly on machine settings and cutter parameters. Therefore, a mathematical model that reflects the actual cutting process is essential for simulation and optimization.

Conjugate Surface Principles

I began by reviewing the fundamental theory of conjugate surfaces, which forms the basis for modeling hypoid bevel gears. Two surfaces \(S^{(1)}\) and \(S^{(2)}\) are in conjugate contact if they satisfy the meshing equation. Let \( \mathbf{r}_1 \) and \( \mathbf{r}_2 \) be the position vectors of a contact point on each surface, and \( \mathbf{n}_1 \) and \( \mathbf{n}_2 \) be the unit normal vectors. The basic equations of gearing are:

$$ \mathbf{r}_2 = \mathbf{m} + \mathbf{r}_1, \quad \mathbf{n}_2 = \mathbf{n}_1 $$

where \( \mathbf{m} \) is the vector from the origin of coordinate system 1 to that of system 2. The relative velocity \( \mathbf{v}_{12} \) at the contact point must satisfy the meshing equation:

$$ \mathbf{n} \cdot \mathbf{v}_{12} = 0 $$

This condition ensures continuous engagement. The relative angular velocity and relative velocity are:

$$ \boldsymbol{\omega}_{12} = \boldsymbol{\omega}_1 – \boldsymbol{\omega}_2 $$

$$ \mathbf{v}_{12} = \boldsymbol{\omega}_{12} \times \mathbf{r}_1 – \boldsymbol{\omega}_2 \times \mathbf{m} + \dot{\mathbf{m}} $$

For conjugate surfaces, the induced curvature along a direction \( \mathbf{v} \) in the tangent plane is given by:

$$ k_{nv} = \frac{a_v}{q n + a v} $$

where \( a = \mathbf{v} \cdot \mathbf{n} \), \( q n + a v \) is a scalar function, and \( v \) is the relative velocity direction. The induced geodesic torsion is:

$$ \tau_{gv} = \frac{a_v a_t}{q n + a v} $$

These formulas allow me to compute the curvature of the generated tooth surfaces.

Equidistant Conjugate Surfaces

I also used the principle of equidistant conjugate surfaces to simplify the analysis. If \(S^{(1)}\) and \(S^{(2)}\) are conjugate, then their offset surfaces \(S_h^{(1)}\) and \(S_h^{(2)}\) are also conjugate under the same relative motion. The relationship between the induced curvatures is:

$$ \tan \omega_h = \frac{\rho_1 – h}{\rho_2 – h} \tan \omega $$

$$ \kappa_h = \kappa \left( \frac{\rho_1 – h}{\rho_1} \right)^2 \left( \frac{\rho_2 – h}{\rho_2} \right)^2 \left( \frac{\rho_1 – h}{\rho_1} \right)^2 \left[ 1 + \tan^2 \omega \right] – \left[ 1 + \tan^2 \omega_h \right] $$

This principle is useful for transforming complex contact problems into simpler ones at the pitch point.

Mathematical Modeling of Hypoid Bevel Gears

I established mathematical models for both the large gear and the pinion based on the HFM cutting method. The HFM method uses a forming method for the large gear and a modified roll method for the pinion. The cutter head parameters and machine settings determine the tooth surface geometry.

Large Gear Tooth Surface

The large gear is cut by a forming method. The cutter head rotates, but there is no relative motion between the imaginary generating gear and the workpiece. The position vector of a point on the cutter surface in the cutter coordinate system \(S_c\) is:

$$ \mathbf{r}_c(s, \theta) = \begin{bmatrix} (r_{02} \pm s \sin \alpha_{02}) \cos \theta \\ (r_{02} \pm s \sin \alpha_{02}) \sin \theta \\ -s \cos \alpha_{02} \\ 1 \end{bmatrix} $$

where \(r_{02}\) is the nominal cutter radius, \(\alpha_{02}\) is the profile angle, \(s\) is the distance along the blade, and \(\theta\) is the phase angle. The unit normal and tangent vectors are:

$$ \mathbf{n}_c = [\cos \alpha_{02} \cos \theta, \cos \alpha_{02} \sin \theta, \sin \alpha_{02}]^T $$

$$ \mathbf{t}_c = [\sin \alpha_{02} \cos \theta, \sin \alpha_{02} \sin \theta, \cos \alpha_{02}]^T $$

After coordinate transformations, the large gear tooth surface in the assembly coordinate system \(S_2\) is:

$$ \mathbf{r}_2 = \mathbf{r}_o + \mathbf{m}_2, \quad \mathbf{n}_2 = \mathbf{n}_o, \quad \mathbf{t}_2 = \mathbf{t}_o $$

The curvature parameters \(A_2, B_2, C_2\) along the tooth length and tooth height directions are obtained from the cutter surface and the relative motion.

Pinion Tooth Surface

The pinion is cut by the modified roll method. The generating gear rotates, and the pinion rotates with a varying ratio. The cutter surface is similar to that of the large gear, but with different parameters. The pinion tooth surface in the assembly coordinate system \(S_1\) is:

$$ \mathbf{r}_1 = \mathbf{r}_o’ + \mathbf{m}_1, \quad \mathbf{n}_1 = \mathbf{n}_o’, \quad \mathbf{t}_1 = \mathbf{t}_o’ $$

The rotation angle of the pinion during cutting is:

$$ \phi_1 = i_{01} \Delta q_1 – c \Delta q_1^2 – d \Delta q_1^3 $$

where \(i_{01}\) is the roll ratio, \(\Delta q_1\) is the cradle angle increment, and \(c, d\) are the second and third order modification coefficients. The relative angular velocity and velocity are:

$$ \boldsymbol{\omega}_{12}’ = \mathbf{k}_1 – \frac{d\phi_1}{dt} \mathbf{p}_1 $$

$$ \mathbf{v}_{12}’ = \boldsymbol{\omega}_{12}’ \times \mathbf{r}_o’ – \frac{d\phi_1}{dt} \mathbf{p}_1 \times \mathbf{m}_1 $$

The meshing equation \(\mathbf{v}_{12}’ \cdot \mathbf{n}_1 = 0\) is solved to find the contact point.

Discrete Point Solving and Solid Modeling

I projected the tooth surface onto its axial section and divided the working area into a grid of \(5 \times 9 = 45\) points. For each grid point, I solved a system of nonlinear equations to find the corresponding cutter parameters. The equations are:

$$ \mathbf{r}_{ij} = (r_{02} + \Delta s_{ij}) \mathbf{r}_{02} + \Delta s_{ij} \mathbf{p}_2 $$

$$ \mathbf{r}_{ij} = (r_{02} + \Delta s_{ij}) \mathbf{r}_{02} + \Delta s_{ij} \mathbf{p}_2 $$

Using numerical methods, I obtained the 3D coordinates of the discrete points. Then I imported these points into SolidWorks to construct the solid model of the hypoid bevel gear pair. The process involves creating curves, surfaces, and then cutting the gear blank. Table 1 shows an example of discrete point coordinates for the large gear concave surface.

Table 1: Coordinates of points on the large gear tooth surface (sample)
Index X (mm) Y (mm) Z (mm)
(1,1) 141.0700 -23.6155 36.4076
(1,2) 147.0073 -20.8517 37.8178
(1,3) 152.8923 -17.7957 39.2281
(1,4) 158.7147 -14.4407 40.6383
(1,5) 164.4631 -10.7796 42.0485
(1,6) 170.1258 -6.8046 43.4588
(1,7) 175.6897 -2.5077 44.8690
(1,8) 181.1409 2.1200 46.2792
(1,9) 186.4644 7.0882 47.6895

Table 2 lists the basic design parameters and machine settings for an example hypoid bevel gear pair.

Table 2: Basic parameters of the hypoid bevel gear pair
Parameter Pinion (convex) Gear (concave)
Number of teeth 8 37
Mean pressure angle (deg) 22.5 22.5
Offset (mm) 38 38
Outer pitch diameter (mm) 147.68 372
Face width (mm) 50.89 45
Shaft angle (deg) 90 90
Spiral angle (deg) 49.99 37.1
Face angle (deg) 19.27 75.48
Pitch angle (deg) 14.83 74.80
Root angle (deg) 14.17 70.27
Hand Left Right
Table 3: Machine adjustment parameters
Parameter Pinion (convex) Gear (concave)
Cutter profile angle (deg) -28.5 -22.5
Cutter radius (mm) 151.73 149.78
Vertical offset (mm) 38 0
Radial distance (mm) 140.21 143.96
Angular position (deg) 72.88 56.88
Mounting angle (deg) 14.1671 70.4260
Roll ratio 4.3654 0
Slide base (mm) 2.1638 0

Using these parameters, I built the solid models of the large gear and pinion. The large gear model is shown in Figure 1 (image inserted above). The pinion model is built similarly, but with separate convex and concave flanks. The assembled model is then used for tooth contact analysis.

Tooth Contact Analysis

Tooth contact analysis (TCA) simulates the meshing of hypoid bevel gears under light load. I formulated the TCA equations by assembling the two tooth surfaces in a common coordinate system. The contact condition requires that the position vectors and normal vectors coincide, and the meshing equation is satisfied. The system of equations is:

$$ \mathbf{r}_2^{(2)} = \mathbf{O}_2\mathbf{O}_1 + \mathbf{r}_1^{(1)} $$

$$ \mathbf{n}_2^{(2)} = \mathbf{n}_1^{(1)} $$

After eliminating the rotation angles, the system reduces to three scalar equations with four unknowns: \(\Delta q_1, \theta_1, \Delta s_2, \theta_2\). To solve this, I specify one parameter, typically the pinion rotation angle, and solve for the rest. The initial guess for the contact point is critical for convergence. I developed an automatic initial value algorithm that uses the properties of the nonlinear function \(f(\theta_1)\).

The function \(f(\theta_1)\) is periodic, and its roots can be found easily. I observed that for a given \(\Delta q_1’\), the root \(\theta_1\) is close to a constant value. Therefore, I can choose a reasonable \(\Delta q_1’\) and solve \(f(\theta_1)=0\) to get \(\theta_1\). This provides a good initial guess. Table 4 shows examples for different gear ratios.

Table 4: Initial values and roots for different gear parameters
\(z_1\) \(z_2\) \(d_2\) \(b_2\) \(E\) \(\Delta q_1’\) Initial \([\Delta q_{10}, \theta_{10}]\) Root \([\Delta q_1, \theta_1]\)
11 45 273.5 40.6 38 -0.05 [-0.0500, 0.4293] [-0.0319, 0.4212]
8 45 273.5 40.6 38 -0.05 [-0.0500, 0.4474] [-0.0328, 0.4422]
6 45 273.5 40.6 38 -0.05 [-0.0500, 0.4611] [-0.0337, 0.4570]
8 37 372 45 38 -0.05 [-0.0500, 0.4345] [-0.0577, 0.4825]
8 39 372 45 38 -0.05 [-0.0510, 0.3300] [-0.0439, 0.3264]

Once the contact point is found, I compute the contact ellipse. The relative curvature along any direction \(\theta\) in the tangent plane is:

$$ \Delta K = \Delta A \cos^2 \theta + \Delta B \sin^2 \theta + \Delta C \sin \theta \cos \theta $$

The principal curvatures are:

$$ \Delta K_{\max} = \frac{\Delta A + \Delta B}{2} + \sqrt{\left( \frac{\Delta A – \Delta B}{2} \right)^2 + \Delta C^2 } $$

$$ \Delta K_{\min} = \frac{\Delta A + \Delta B}{2} – \sqrt{\left( \frac{\Delta A – \Delta B}{2} \right)^2 + \Delta C^2 } $$

The lengths of the major and minor axes of the contact ellipse are:

$$ l_1′ = \sqrt{\frac{0.0127}{\Delta K_{\min}}}, \quad l_2′ = \sqrt{\frac{0.0127}{\Delta K_{\max}}} $$

I then project the contact ellipse onto the axial section to obtain the contact pattern. The transmission error is computed as:

$$ \Delta \varepsilon = \Delta \varepsilon_2 – \frac{z_1}{z_2} \Delta \varepsilon_1 $$

where \(\Delta \varepsilon_1\) and \(\Delta \varepsilon_2\) are the rotation angles of the pinion and gear. The transmission error curve is plotted against the pinion rotation angle.

Figure 2 shows a typical contact pattern and transmission error curve for the example hypoid bevel gear pair. The contact pattern should be centered and not too wide, and the transmission error should be small.

Pinion Control Parameter Adjustment

Hypoid bevel gears are point-contact conjugate gears. The contact pattern obtained from TCA may not meet design requirements. Therefore, I studied the adjustment of pinion control parameters to correct the contact pattern. The correction is usually applied to the pinion because it has fewer teeth and is cut by a single-flank method.

The distance between the theoretical tooth surface and the corrected tooth surface is approximated by:

$$ \Delta \delta = \frac{1}{2} (\Delta k_n) s^2 $$

where \(\Delta k_n\) is the difference in normal curvature along a direction. To achieve a contact pattern length \(L\) and a maximum gap of 0.00635 mm, the required curvature correction is:

$$ \Delta k_n = \frac{0.0127}{L^2} $$

For the tooth length direction, \(L = b / \cos \beta\), where \(b\) is the face width and \(\beta\) is the spiral angle. For the tooth height direction, \(L = \frac{2 \pi r}{z} \cos \beta\). These corrections are used to adjust the pinion control parameters.

Effect of Second-Order Modification Coefficient

I investigated the effect of the second-order modification coefficient \(c_{2x}\) on the contact pattern. I varied \(c_{2x}\) from -0.04 to 0.04 while keeping other parameters constant. The results are summarized in Table 5.

Table 5: Effect of second-order modification coefficient \(c_{2x}\) on contact pattern
\(c_{2x}\) Pinion convex Pinion concave
-0.04 Narrow, outer diagonal Wide, inner diagonal
-0.02 Slightly narrow, slight outer diagonal Slightly wide, slight inner diagonal
0 Nominal Nominal
0.02 Slightly wide, slight inner diagonal Slightly narrow, slight outer diagonal
0.04 Wide, inner diagonal Narrow, outer diagonal

For the left-hand pinion convex flank, a negative \(c_{2x}\) makes the contact pattern narrower and produces an outer diagonal. A positive \(c_{2x}\) makes it wider and produces an inner diagonal. The opposite holds for the concave flank.

Adjustment for Different Gear Ratios

I also studied how the typical control parameters affect TCA results for different gear ratios. Table 6 lists the gear ratios I tested by varying the pinion tooth number while keeping the gear tooth number constant.

Table 6: Gear ratios tested
\(z_1\) \(z_2\) Gear ratio \(z_2/z_1\)
11 45 4.09
10 45 4.5
9 45 5
8 45 5.625
7 45 6.429
6 45 7.5
5 45 9

I found two main observations: (1) As the gear ratio increases, the nonlinear equations for TCA become more difficult to converge. (2) With other parameters unchanged, higher gear ratios tend to produce less ideal contact patterns and transmission error curves.

To address these issues, I developed a general adjustment procedure for pinion control parameters. The steps are:

  1. Run TCA and examine the contact pattern shape and transmission error curve.
  2. If the contact pattern appears diamond-shaped or close to it, adjust the cutter profile angle \(\alpha_{01}\) to correct it. Table 7 shows the effect of \(\alpha_{01}\).
  3. If diagonal contact or fish-tail contact remains, adjust the generating cone distance \(R_{01}\) to correct diagonal contact. Table 8 shows the effect of \(R_{01}\).
  4. If fish-tail contact persists, adjust the vertical offset \(E_{Mx}\) to correct it. Table 9 shows the effect of \(E_{Mx}\).
Table 7: Effect of cutter profile angle \(\alpha_{01}\) on contact pattern
\(\alpha_{01}\) (deg) Contact pattern
-29.0 Diamond, narrow
-27.0 Optimal
-26.0 Wide, slightly diagonal
Table 8: Effect of generating cone distance \(R_{01}\) on contact pattern
\(R_{01}\) (mm) Contact pattern
2.0 Diagonal, inward
2.5 Optimal
3.0 Diagonal, outward
Table 9: Effect of vertical offset \(E_{Mx}\) on contact pattern
\(E_{Mx}\) (mm) Contact pattern
-9 Fish-tail, toe
-3 Optimal
6 Fish-tail, heel

I summarize the correction order as: first third-order correction (shape), then second-order correction (size), and finally first-order correction (position). In most cases, the third-order and second-order corrections are sufficient to achieve a satisfactory contact pattern and transmission error curve.

Conclusion

In my research, I established a complete mathematical model for hypoid bevel gears based on the HFM cutting method. I solved for discrete tooth surface points and built solid models in SolidWorks. I developed a TCA program with an automatic initial value algorithm that significantly improves convergence and stability. I also investigated the effect of pinion control parameters on contact patterns and developed a systematic adjustment procedure. The main findings are:

  • The second-order modification coefficient \(c_{2x}\) strongly affects the width and direction of the contact pattern. For the pinion convex flank, negative \(c_{2x}\) narrows the pattern and causes outer diagonal contact, while positive \(c_{2x}\) widens it and causes inner diagonal contact. The opposite is true for the concave flank.
  • Higher gear ratios make TCA convergence more difficult and often produce less ideal contact patterns.
  • A general adjustment procedure is: correct diamond contact with cutter profile angle, correct diagonal contact with generating cone distance, and correct fish-tail contact with vertical offset. This order of correction is effective and efficient.

Future work should include modeling the fillet and root transition surfaces to make the solid model more realistic for loaded contact analysis. Additionally, the combined effect of multiple machine settings should be studied to better match actual manufacturing conditions. My work provides a foundation for the digital design and manufacturing of hypoid bevel gears.

Scroll to Top