Introduction. In my research, I focus on the dynamic behavior of hypoid bevel gears because these gears are widely used in vehicle drive axles, aviation transmissions, marine propulsion systems, and other staggered-axis transmission systems. Hypoid bevel gears offer high load capacity, smooth meshing, low noise, and a compact layout. However, their tooth surface geometry is complex, their assembly requirements are strict, and their sensitivity to manufacturing and installation errors is significant. For this reason, I treat hypoid bevel gears as a representative high-performance gear pair and investigate their intrinsic characteristics, dynamic response, dynamic stability, and parameter sensitivity under both healthy and faulty conditions.
I consider four major problems in gear dynamics: natural characteristics, dynamic response, dynamic stability, and the influence of system parameters on dynamic behavior. Natural characteristics include natural frequencies and mode shapes. Dynamic response mainly involves time-varying mesh stiffness. Dynamic stability is represented by transmission error. Parameter influence is studied through crack faults and installation errors. I use a finite element approach because the curved tooth surfaces of hypoid bevel gears make purely analytical solutions difficult and often inaccurate. My goal is to build a reliable numerical framework that can describe how hypoid bevel gears behave under real operating and fault conditions.
Geometric and machining fundamentals. The tooth surfaces of hypoid bevel gears are generated by a complex machine-tool motion. The gear pair is not a fully conjugate system in the traditional sense. Instead, it follows the local conjugate principle. In this principle, the theoretical conjugate pinion is first obtained, and then the tooth surface is modified near a chosen contact point. The modification introduces a controlled contact ellipse and avoids severe edge contact. For hypoid bevel gears, the local conjugate condition is especially important because the pinion axis is offset from the gear axis.
I describe the local conjugate condition through curvature differences. Let the normal curvature difference along a certain direction be \(\Delta k_n\). The separation between two tangent surfaces can be written as
$$ \Delta \delta = \frac{\Delta k_n \, l^2}{8}, $$
where \(l\) is the contact length along that direction. For the tooth length direction, I introduce a contact length coefficient \(f\), tooth width \(b\), and spiral angle \(\beta\). The correction of normal curvature along the tooth length direction can be expressed as
$$ \Delta A = \frac{\Delta k_n \cos^2 \beta}{f b}. $$
For the tooth height direction, the correction is related to the pitch radius \(r_1\), tooth number \(z_1\), and spiral angle \(\beta\):
$$ \Delta B = \frac{z_1 \Delta \delta \cos \beta}{2 r_1^2}. $$
To prevent curvature interference between the pinion and the gear, the corrections must satisfy
$$ \Delta A \, \Delta B – \Delta C^2 \ge 0, $$
where \(\Delta C\) is the correction of the geodesic torsion along the tooth length direction. This condition ensures that the modified pinion surface remains below the theoretical conjugate surface in all directions. If this condition is violated, the hypoid bevel gears may contact at an unintended location, causing noise, vibration, and premature failure.
I also analyze the boundary distances of the pinion and gear tooth surfaces. For a point on the pinion, the distance to the toe and heel can be written as
$$ \delta_{pto} = C_{pto} – z_p, \qquad \delta_{ph} = z_p – C_{ph}, $$
where \(C_{pto}\) and \(C_{ph}\) are reference positions along the pinion axis. The distance to the top line and root line is
$$ \delta_{pt} = C_{pt} – \left( z_p + y_p \tan \gamma_{pt} \right), \qquad \delta_{pr} = S_p. $$
For the gear, similar expressions are obtained:
$$ \delta_{gt} = C_{gt} – \left( z_g + y_g \tan \gamma_{gt} \right), \qquad \delta_{gr} = S_g. $$
The toe and heel distances of the gear are
$$ \delta_{gto} = z_g – C_{gp} + x_g \tan \gamma_{gp} + \frac{a_g}{\cos \gamma_{gp}}, $$
$$ \delta_{gh} = z_g – C_{gp} – x_g \tan \gamma_{gp} – \frac{a_g}{\cos \gamma_{gp}}. $$
These distances define the tooth blank boundaries and help me determine whether the generated hypoid bevel gears have sufficient tooth depth and proper clearance.
Parameter selection and three-dimensional modeling. I determine the basic parameters of the hypoid bevel gears from the required transmission ratio, torque, shaft angle, offset distance, and mounting space. The outer module is calculated from the pinion pitch diameter \(d_1\) and pinion tooth number \(z_1\):
$$ m_d = \frac{d_1}{z_1}. $$
The gear tooth number is obtained from the transmission ratio \(i\):
$$ z_2 = i z_1. $$
The pitch cone angles are
$$ \delta_1 = \arctan \frac{z_1}{z_2}, \qquad \delta_2 = \Sigma – \delta_1, $$
where \(\Sigma\) is the shaft angle. The outer cone distance is
$$ R_e = \frac{d_2}{2 \sin \delta_2}. $$
The working tooth height and full tooth height are
$$ h_k = 1.70 m, \qquad h_t = 1.888 m. $$
The addendum and dedendum are calculated from standard hypoid bevel gear formulas. The face cone angle and root cone angle are obtained from the addendum angle and dedendum angle. I use these parameters to build the tooth blank and cutter geometry.
For the gear, I use a generating method. For the pinion, I use a cutter tilt method. The machine settings include radial cutter position \(S_q\), angular cutter position \(q\), machine root angle \(G_m\), horizontal wheel position \(X_p\), vertical wheel position \(E_m\), roll ratio \(i_G\), and bed position \(X_b\). The initial position of the blank and cutter is obtained through coordinate transformations. The blank position vector is
$$ \mathbf{R}_{mg} = \left[ X_p – X_b \cos G_m – X_b \sin G_m \pm E_m \right]^T. $$
The cutter position vector for the gear is
$$ \mathbf{R}_{mc} = \left[ S_q \cos q \quad 0 \quad S_q \sin q \quad 1 \right]^T. $$
For the pinion, the cutter tilt and cutter swivel introduce additional rotations. The final cutter position vector can be written as
$$ \mathbf{R}_{mc2} = \mathbf{R}_{mc0} \mathbf{M}_z(d) \mathbf{M}_{y1}(I) \mathbf{M}_{y2}(J), $$
where \(I\) is the cutter tilt angle, \(J\) is the cutter swivel angle, and \(d\) defines the tilt axis in the machine plane. The envelope curves of the cutter generate the tooth profile. I then use these curves to cut the tooth slot, rotate the slot by the angular pitch, and obtain the complete gear and pinion models. The assembled hypoid bevel gears are used for all subsequent finite element simulations.
Finite element preprocessing. I use a professional preprocessor to generate hexahedral meshes for the hypoid bevel gears. Hexahedral elements are preferred because they improve contact convergence and reduce numerical error. The meshing procedure includes importing the solid model, partitioning a single tooth, creating two-dimensional quadrilateral meshes, projecting them into three-dimensional meshes, merging coincident nodes, and rotating the single-tooth mesh to form the full gear. I check the mesh for free edges and T-shaped edges. To reduce computational cost, I use six tooth pairs in the final dynamic simulation. The material properties are summarized in Table 1.
| Property | Value |
|---|---|
| Density | \(7.8 \times 10^{-9} \, \mathrm{t/mm^3}\) |
| Young’s modulus | \(2.1 \times 10^5 \, \mathrm{MPa}\) |
| Poisson’s ratio | 0.3 |
| Contact type | Surface-to-surface |
| Tangential behavior | Penalty friction, \(\mu = 0.1\) |
| Normal behavior | Hard contact, no penetration |
For the contact definition, I select the pinion concave surface as the master surface because the two hypoid bevel gears have the same material and similar mesh density. I remove the non-meshing tooth root surfaces from the contact set to reduce unnecessary contact searches. I couple the inner ring of each gear to a reference point at the rotation center. This coupling transfers all nodal degrees of freedom to the reference point, allowing me to apply rotational velocity and torque. The pinion reference point receives a driving speed, and the gear reference point receives a resisting torque. I also introduce a short ramp time to eliminate rigid-body motion and improve convergence.
Modal analysis. The equation of motion for the hypoid bevel gear system is
$$ \mathbf{M} \ddot{\mathbf{x}} + \mathbf{C} \dot{\mathbf{x}} + \mathbf{K} \mathbf{x} = \mathbf{F}(t), $$
where \(\mathbf{M}\), \(\mathbf{C}\), and \(\mathbf{K}\) are the mass, damping, and stiffness matrices, respectively. For free vibration without damping, the equation reduces to
$$ \mathbf{M} \ddot{\mathbf{x}} + \mathbf{K} \mathbf{x} = \mathbf{0}. $$
Assuming a harmonic solution \(\mathbf{x} = \boldsymbol{\phi} e^{i \omega t}\), I obtain the eigenvalue problem
$$ \left( \mathbf{K} – \omega_i^2 \mathbf{M} \right) \boldsymbol{\phi}_i = \mathbf{0}, $$
where \(\omega_i\) is the \(i\)-th natural frequency and \(\boldsymbol{\phi}_i\) is the corresponding mode shape. I extract the first six modes of the hypoid bevel gears. The natural frequencies are listed in Table 2.
| Mode order | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Natural frequency (Hz) | 100.22 | 101.56 | 104.28 | 118.97 | 150.61 | 207.10 |
The first mode is torsional vibration. The teeth twist around the gear axis, and the stress distribution is approximately uniform in the circumferential direction. The second and third modes are bending modes. The gear bends along an axial direction, and the stress distribution is symmetric. The symmetry axis of the second mode is different from that of the third mode. The fourth and fifth modes are higher-order bending modes. The deformation becomes more distributed, and the maximum stress region is no longer concentrated at a single tooth. The sixth mode combines bending and torsion. The teeth experience increased stress, and the gear shows a centrifugal tendency. These results are important because I can use them to avoid resonance when the excitation frequency of the surrounding transmission system is close to one of the natural frequencies of the hypoid bevel gears.
Time-varying mesh stiffness. Mesh stiffness is one of the most important internal excitations in gear dynamics. For hypoid bevel gears, the contact force and contact displacement vary along the curved tooth surface. I define the instantaneous mesh stiffness as
$$ K_n(t) = \frac{F_n(t)}{u_n(t)}, $$
where \(F_n(t)\) is the normal contact force and \(u_n(t)\) is the comprehensive elastic displacement. Because an analytical solution for \(F_n(t)\) and \(u_n(t)\) is difficult for hypoid bevel gears, I extract these quantities from the finite element solution. The contact force is obtained from the contact history output. The contact displacement is obtained from the average elastic deformation of nodes with positive contact pressure.
The single-tooth mesh stiffness is calculated from the contact force and contact displacement. I then couple single-tooth stiffnesses to obtain the multi-tooth mesh stiffness. The meshing time of a single tooth is \(\Delta T\), and the time difference between adjacent teeth entering mesh is \(\Delta t\). The contact ratio is
$$ \varepsilon = \frac{\Delta T}{\Delta t}. $$
For the hypoid bevel gears studied here, I obtain \(\Delta T = 0.00053 \, \mathrm{s}\) and \(\Delta t = 0.00039 \, \mathrm{s}\), giving \(\varepsilon = 1.76\). The angular shift between adjacent tooth pairs is
$$ \Delta \alpha = \frac{\varphi}{\varepsilon}, $$
where \(\varphi\) is the rotation angle of the gear during one complete meshing cycle. By shifting and superimposing the single-tooth stiffness, I obtain the multi-tooth time-varying mesh stiffness. The mesh stiffness increases as a new tooth pair enters contact and decreases as a tooth pair leaves contact. This periodic variation is the main source of internal dynamic excitation in hypoid bevel gears.
I also study the effect of load on mesh stiffness. I apply the original load, 1.5 times the original load, and 2 times the original load. The results are summarized in Table 3. As the load increases, the contact ratio increases, the time difference between adjacent teeth decreases, and the average mesh stiffness increases. The increase in mesh stiffness is more pronounced at higher loads. This behavior occurs because a larger load expands the contact area and delays the separation of the tooth pair, allowing more teeth to share the load.
| Load case | Relative torque | Contact ratio | Average mesh stiffness \(K_{\mathrm{avg}}\) \((10^4 \, \mathrm{N/m})\) | Observed trend |
|---|---|---|---|---|
| Case I | 1.0 | 1.76 | 11.22 | Reference |
| Case II | 1.5 | 1.88 | 12.05 | Increase |
| Case III | 2.0 | 2.03 | 13.16 | Larger increase |
I define the average mesh stiffness over one meshing period as
$$ K_{\mathrm{avg}} = \frac{1}{T} \int_{0}^{T} K_n(t) \, dt. $$
The numerical results show that the average mesh stiffness of hypoid bevel gears is sensitive to load. This is different from the ideal constant-stiffness assumption often used in simple gear models. For accurate dynamic analysis, the time-varying and load-dependent nature of mesh stiffness must be included.
Transmission error. Transmission error is a key indicator of dynamic stability and motion accuracy. It represents the deviation between the actual output rotation and the ideal output rotation. For a hypoid bevel gear pair with pinion tooth number \(z_1\) and gear tooth number \(z_2\), I define the transmission error as
$$ TE = \left( \theta_2 – \theta_{2,0} \right) – \frac{z_1}{z_2} \left( \theta_1 – \theta_{1,0} \right), $$
where \(\theta_1\) and \(\theta_2\) are the instantaneous rotation angles of the pinion and gear, and \(\theta_{1,0}\) and \(\theta_{2,0}\) are their initial angles. I extract the angular displacement histories from the finite element solution and calculate the transmission error using the above equation.
The transmission error curve has several notable features. At the beginning of meshing, the fluctuation amplitude is large because the system is accelerating from rest and the contact conditions are not yet stable. As meshing continues, the amplitude decreases and the curve gradually becomes harmonic. This indicates that the hypoid bevel gears enter a stable meshing state. I also study the effect of load on transmission error. The results are shown in Table 4. As the load increases, the fluctuation amplitude of the transmission error decreases. The reduction from 1.0 to 1.5 times the original load is larger than the reduction from 1.5 to 2.0 times the original load. This suggests that increasing the load can improve the contact stability of hypoid bevel gears, but the benefit diminishes at very high loads.
| Load case | Relative torque | Maximum TE amplitude \((10^{-3} \, \mathrm{rad})\) | Change relative to Case I |
|---|---|---|---|
| Case I | 1.0 | 3.01 | Reference |
| Case II | 1.5 | 2.44 | Reduced |
| Case III | 2.0 | 2.15 | Further reduced |
Crack fault analysis. The failure modes of hypoid bevel gears include pitting, scuffing, wear, plastic deformation, and tooth breakage. Among these, tooth breakage is one of the most dangerous because it can cause sudden loss of transmission. Tooth breakage usually begins with a crack. I use fracture mechanics to describe crack behavior. For a mixed-mode crack, the crack-tip stress field can be expressed as
$$ \sigma_x = \frac{K_I}{\sqrt{2\pi r}} \cos\frac{\theta}{2} \left( 1 – \sin\frac{\theta}{2} \sin\frac{3\theta}{2} \right) – \frac{K_{II}}{\sqrt{2\pi r}} \sin\frac{\theta}{2} \left( 2 + \cos\frac{\theta}{2} \cos\frac{3\theta}{2} \right), $$
$$ \sigma_y = \frac{K_I}{\sqrt{2\pi r}} \cos\frac{\theta}{2} \left( 1 + \sin\frac{\theta}{2} \sin\frac{3\theta}{2} \right) + \frac{K_{II}}{\sqrt{2\pi r}} \sin\frac{\theta}{2} \cos\frac{\theta}{2} \cos\frac{3\theta}{2}, $$
$$ \tau_{xy} = \frac{K_I}{\sqrt{2\pi r}} \cos\frac{\theta}{2} \sin\frac{\theta}{2} \cos\frac{3\theta}{2} + \frac{K_{II}}{\sqrt{2\pi r}} \cos\frac{\theta}{2} \left( 1 – \sin\frac{\theta}{2} \sin\frac{3\theta}{2} \right), $$
where \(K_I\), \(K_{II}\), and \(K_{III}\) are the stress intensity factors for opening, sliding, and tearing modes, respectively. I calculate the stress intensity factors for the hypoid bevel gears and find that the opening mode dominates. The average stress intensity factors are listed in Table 5.
| Crack type | Opening mode \(K_I\) | Sliding mode \(K_{II}\) | Tearing mode \(K_{III}\) |
|---|---|---|---|
| Average value \((\mathrm{MPa \cdot mm^{1/2}})\) | 168.47 | 4.32 | 3.98 |
Because the opening mode is much larger than the sliding and tearing modes, the crack propagates primarily in the direction perpendicular to the maximum tensile stress. I choose a semi-elliptical crack for the fault simulation. The crack is inserted at the location of maximum contact stress. The crack gradients are given in Table 6.
| Crack parameter | Crack 1 | Crack 2 | Crack 3 |
|---|---|---|---|
| Major semi-axis \(a\) (mm) | 10 | 10 | 10 |
| Minor semi-axis \(b\) (mm) | 1.0 | 1.5 | 2.0 |
I then perform finite element simulations for the cracked hypoid bevel gears. The mesh stiffness and transmission error are compared with the healthy case. The results are shown in Tables 7 and 8. The average mesh stiffness decreases as the crack depth increases. Crack 3 causes the largest reduction, reaching 48.8% of the original mesh stiffness. The transmission error amplitude increases with crack depth. For Crack 3, the amplitude is 210.82% of the healthy value. The crack mainly affects the mesh stiffness, while the transmission error fluctuation occurs in the same meshing period as the stiffness reduction. Once the cracked tooth leaves mesh, the transmission error tends to return to the healthy level.
| Crack level | Average mesh stiffness \((10^4 \, \mathrm{N/m})\) | Reduction percentage (%) | Additional reduction relative to previous crack (%) |
|---|---|---|---|
| Healthy | 11.22 | — | — |
| Crack 1 | 7.26 | 13.9 | — |
| Crack 2 | 7.06 | 24.5 | 76.5 |
| Crack 3 | 4.99 | 48.8 | 98.9 |
| Crack level | TE amplitude \((10^{-3} \, \mathrm{rad})\) | Amplitude increase (%) | Additional increase relative to previous crack (%) |
|---|---|---|---|
| Healthy | 3.01 | — | — |
| Crack 1 | 5.31 | 76.39 | — |
| Crack 2 | 6.49 | 115.48 | 39.09 |
| Crack 3 | 9.36 | 210.82 | 95.35 |
Installation error analysis. Installation errors are another major source of faults in hypoid bevel gears. Because hypoid bevel gears operate on staggered axes, even a small installation error can shift the contact pattern, change the contact pressure, and alter the transmission error. I consider three installation errors: pinion axial error \(\Delta Z\), gear axial error \(\Delta S\), and offset distance error \(\Delta E\). The positive direction is chosen away from the meshing zone, and the negative direction is chosen toward the meshing zone. The error range is from \(-0.3 \, \mathrm{mm}\) to \(0.3 \, \mathrm{mm}\).
For the pinion axial error, I find that the transmission error curve shifts in time. A positive pinion axial error delays the transmission error, while a negative error advances it. The larger the error magnitude, the larger the shift. The negative error has a stronger effect than the positive error. The pinion axial error also changes the transmission error amplitude. The mesh line extends toward the tooth tip. The analysis step time changes with the error direction, which is consistent with the transmission error shift. The mesh stiffness increases slightly, but the increase is not as large as the load-induced increase.
For the gear axial error, the transmission error shift follows a similar trend. A positive gear axial error delays the transmission error, and a negative gear axial error advances it. The negative effect is stronger than the positive effect. However, the influence of the gear axial error is weaker than that of the pinion axial error. The mesh line also extends toward the tooth tip. The mesh stiffness increases, and the negative error produces a larger increase than the positive error.
For the offset distance error, the transmission error curve does not show a clear time shift. Instead, the offset error mainly changes the transmission error amplitude. The positive offset error has a stronger effect than the negative offset error. The mesh line position does not change significantly. The mesh stiffness increases slightly, but the change is small and almost symmetric for positive and negative errors.
The transmission error amplitudes under different installation errors are listed in Table 9. The values show that the pinion axial error has the strongest effect on the transmission error shift and fluctuation. The offset distance error has the largest effect on the transmission error amplitude. The gear axial error has the weakest effect among the three error types. The average mesh stiffness under different installation errors is listed in Table 10. The pinion axial error produces the largest change in average mesh stiffness. The offset distance error produces the smallest change. The maximum increase in average mesh stiffness is only about 2.51% of the healthy value, which indicates that installation errors mainly affect the contact pattern and transmission error rather than the average stiffness magnitude.
| Error value (mm) | Pinion axial error TE \((10^{-3} \, \mathrm{rad})\) | Gear axial error TE \((10^{-3} \, \mathrm{rad})\) | Offset distance error TE \((10^{-3} \, \mathrm{rad})\) |
|---|---|---|---|
| -0.3 | 3.57229 | 3.40076 | 4.11998 |
| -0.2 | 3.38693 | 3.35063 | 3.70475 |
| -0.1 | 3.30363 | 3.21979 | 3.30414 |
| 0 | 3.08586 | 3.08586 | 3.08586 |
| 0.1 | 3.27718 | 3.12212 | 3.27772 |
| 0.2 | 3.27553 | 3.19167 | 3.22119 |
| 0.3 | 3.38966 | 3.27887 | 3.39007 |
| Error gradient (mm) | Pinion axial error \(K_{\mathrm{avg}}\) \((10^4 \, \mathrm{N/m})\) | Gear axial error \(K_{\mathrm{avg}}\) \((10^4 \, \mathrm{N/m})\) | Offset distance error \(K_{\mathrm{avg}}\) \((10^4 \, \mathrm{N/m})\) |
|---|---|---|---|
| -0.3 | 11.64455593 | 11.407792 | 11.221419 |
| -0.1 | 11.538922 | 11.460914 | 11.429027 |
| 0 | 11.358989 | 11.358989 | 11.358989 |
| 0.1 | 11.460913 | 11.538922 | 11.303954 |
| 0.3 | 11.407791 | 11.644556 | 11.265779 |
To quantify sensitivity, I define a sensitivity index for average mesh stiffness as
$$ S_K = \frac{\Delta K_{\mathrm{avg}} / K_{\mathrm{avg},0}}{\Delta e / e_0}, $$
where \(\Delta K_{\mathrm{avg}}\) is the change in average mesh stiffness, \(K_{\mathrm{avg},0}\) is the healthy average mesh stiffness, \(\Delta e\) is the installation error increment, and \(e_0\) is the reference error. In my results, the pinion axial error gives the highest sensitivity index, followed by the gear axial error and then the offset distance error. For transmission error amplitude, the offset distance error gives the highest sensitivity, followed by the pinion axial error and then the gear axial error. This means that different installation errors excite different dynamic features of hypoid bevel gears.
Discussion. My results show that hypoid bevel gears have rich dynamic behavior. Their natural frequencies increase with mode order, and their mode shapes transition from torsion to bending and then to combined bending-torsion. The time-varying mesh stiffness is periodic and load-dependent. As load increases, the contact ratio increases and the average mesh stiffness rises. The transmission error initially fluctuates strongly, then stabilizes into a harmonic-like curve. Higher load reduces the transmission error amplitude. These results confirm that load is an important parameter for controlling the dynamic stability of hypoid bevel gears.
Cracks change the local stiffness of the tooth. The opening-mode stress intensity factor dominates, and the crack reduces the average mesh stiffness while increasing the transmission error amplitude. The effect is localized in the meshing period of the cracked tooth. Once the cracked tooth leaves contact, the transmission error returns to a level close to the healthy case. This periodicity is useful for fault detection because the vibration signature of a cracked hypoid bevel gear contains a once-per-revolution component related to the cracked tooth.
Installation errors mainly change the contact pattern. The pinion axial error shifts the transmission error in time and has a strong effect on mesh stiffness. The gear axial error has a similar but weaker effect. The offset distance error does not shift the transmission error in time, but it strongly changes the transmission error amplitude. These differences are important for assembly tolerance design. If the main concern is motion accuracy, the pinion axial error should be controlled tightly. If the main concern is vibration amplitude, the offset distance error should be controlled tightly.
Conclusions. I studied the dynamic characteristics of hypoid bevel gears under healthy, cracked, and misaligned conditions. The main conclusions are as follows. First, the natural frequencies of hypoid bevel gears increase with mode order, and the first six modes include torsion, bending, higher-order bending, and combined bending-torsion. Second, the multi-tooth mesh stiffness rises when a new tooth pair enters contact and falls when a tooth pair leaves contact. Increasing load increases the contact ratio and average mesh stiffness. Third, the transmission error fluctuates strongly at the beginning of meshing and then stabilizes. Increasing load reduces the transmission error amplitude. Fourth, cracks reduce the average mesh stiffness and increase the transmission error amplitude. The opening mode dominates crack propagation. Fifth, installation errors affect the contact pattern and dynamic response. The pinion axial error has the strongest influence on transmission error shift and mesh stiffness, while the offset distance error has the strongest influence on transmission error amplitude. These findings provide a useful basis for the design, assembly, and fault diagnosis of hypoid bevel gears in reducer transmission mechanisms.
In future work, I plan to include shaft and bearing flexibility, thermal effects, and lubrication conditions in the dynamic model. I also intend to combine the finite element results with experimental measurements to validate the predicted mesh stiffness, transmission error, and fault signatures of hypoid bevel gears. Such extensions will make the dynamic model more realistic and more useful for engineering applications involving hypoid bevel gears.