My research focuses on the tooth modification design and tooth profile optimization of involute straight bevel gear, which is a crucial issue in the field of mechanical transmission. This work systematically investigates the modification theory, dynamic contact simulation, orthogonal optimization of isometric modification parameters, and variable curvature crowning technology for involute straight bevel gear, aiming at improving transmission stability and homogenizing tooth load distribution under specific working conditions.
The involute straight bevel gear is widely employed in intersecting-axis transmission components such as automotive differential mechanisms or industrial gear reducers. Its meshing quality directly determines the overall performance of the entire gear system. However, during actual service, the gear pair inevitably suffers from manufacturing errors, installation errors, elastic deformations and thermal deformations, which can lead to severe meshing interference impact, unstable load transition and end stress concentration phenomena. To solve these problems, tooth modification technology has been proven to be an effective approach without significantly increasing manufacturing cost. In this paper, I present a comprehensive study on the modification design and tooth profile optimization of involute straight bevel gear, combining theoretical analysis, three-dimensional solid modeling, dynamic contact finite element simulation, orthogonal experimental design, and reverse engineering-based variable curvature crowning technique.
This thesis is organized as follows. After introducing the background and significance of my research, I discuss the theoretical basis of gear modification, including classical mechanics solutions and finite element theory. Then I demonstrate the digital modeling technology for involute straight bevel gear based on SolidWorks and ANSYS/LS-DYNA. Subsequently, I present the orthogonal optimization design of isometric modification parameters for involute straight bevel gear and compare the comprehensive modification effect with symmetric circular arc crowning. Finally, I propose a novel method for accurate variable curvature crowning of involute straight bevel gear considering the elastic deformation of gear shafts, which is verified through dynamic contact finite element simulation.
1. Introduction
Gear mechanisms transmit motion and power between arbitrary axes through direct contact between tooth flanks. Due to their advantages such as high transmission efficiency, constant transmission ratio, wide power range, and convenient manufacturing and installation, involute gears are extensively used in industrial applications. Among them, the involute straight bevel gear is especially suitable for intersecting shaft transmission with the shaft angle commonly being 90 degrees. Typical applications include the differential gears in automobiles and various right-angle power transmission units.
With the rapid development of modern industry, higher requirements are imposed on gear transmission accuracy, smoothness and load-carrying uniformity. Under high-speed and heavy-duty conditions, the elastic deformations of gears and their supporting systems become more significant. Combined with manufacturing and installation errors, these deformations seriously disturb the theoretical meshing state of the gear pair, resulting in base pitch mismatch, meshing interference, line-of-action discontinuities, and severe edge stress concentration. Consequently, the gear pair generates excessive vibration, noise and impact, which may lead to premature failure.
In my research, I recognize that simply improving manufacturing accuracy cannot fundamentally eliminate these adverse phenomena, and may increase cost dramatically. Tooth surface coating treatments also have limitations. Tooth modification, however, provides a cost-effective solution. By removing a small amount of material from specific tooth flank regions, modification can compensate for elastic deformations and errors, optimize the meshing process, and improve the overall transmission quality. It is important to emphasize that modification does not alter the basic gear design parameters, but only performs micro-adjustment on the tooth flank geometry.
Many researchers have investigated gear modification technology. Most studies focus on cylindrical gears, whereas research on bevel gears is still limited. The profile modification and lead modification are often treated separately, but a comprehensive modification combining both aspects usually achieves better results. Moreover, to obtain an ideal modification effect, the specific working conditions must be taken into account. In this regard, the objective of my research is to develop a reliable methodology for designing and optimizing the tooth modification of involute straight bevel gear under given working conditions.
2. Theoretical Basis for Modification Design of Involute Straight Bevel Gear
Accurate analysis of gear deformation and stress after loading is the fundamental basis for gear modification design. In this section, I present the classical mechanics solutions for gear contact problems and the finite element method that serves as a powerful numerical tool.
2.1 Classical Mechanics Analysis of Gear Contact
The contact deformation of gear tooth flanks can be treated using Hertz contact theory. For the involute straight bevel gear, the tooth flank contact can be modeled as the contact between two frusta of cones, with the equivalent contact radius at each point being the arc length of the spherical involute from its starting point to that point. The contact deformation can be expressed as:
$$ \delta_c = \frac{2(1-\nu^2)P_n}{\pi E b}\left(1.27 + 0.781\ln\frac{m b}{a}\right) $$
where $\nu$ is Poisson’s ratio, $E$ is Young’s modulus, $P_n$ is the normal load per unit tooth width, $b$ is the tooth width, $a$ is the contact half-band width, and $m$ is the gear module.
For the calculation of contact stress in involute straight bevel gear, the equivalent spur gear at the mean pitch cone is adopted. The contact stress is:
$$ \sigma_H = \sqrt{\frac{4.439 K T_1}{\Phi_R (1-0.5\Phi_R)^2 d_1^3 u}} \cdot Z_H Z_E $$
where $K$ is the load factor, $T_1$ is the torque on the driving gear, $\Phi_R$ is the tooth width factor, $d_1$ is the pitch diameter at the large end, $u$ is the gear ratio, $Z_H$ is the zone factor, and $Z_E$ is the elasticity influence coefficient.
For the bending strength of involute straight bevel gear, the equivalent spur gear at the mean pitch cone is used. The bending stress is calculated as:
$$ \sigma_F = \frac{K F_t Y_{Fa} Y_{Sa}}{b m (1-0.5\Phi_R)} $$
where $F_t$ is the tangential force, $Y_{Fa}$ is the tooth form factor, and $Y_{Sa}$ is the stress correction factor.
For the tooth profile modification, the modification amount mainly originates from the comprehensive elastic deformation of the tooth, including contact deformation, bending deformation, shear deformation, and the deformation of the gear body and shaft. The recommended profile modification amount can be determined by:
$$ \delta_\alpha = W_t / c_r $$
where $W_t = F_t / b$ is the unit load on the tooth width and $c_r$ is the meshing stiffness per unit width.
For the lead modification, the modification amount should compensate the comprehensive deformation including gear body bending, torsion, tooth bending and contact deformation. The total deformation is expressed as:
$$ \delta = \delta_u + \delta_t $$
These classical formulas provide approximate results under simplified conditions. However, when the boundary conditions and load cases are complex, the analytical solutions deviate notably from the real situation. Therefore, numerical methods such as the finite element method are necessary.
2.2 Finite Element Method Fundamentals
The finite element method is a discretized numerical analysis technique. The basic principle is to divide the continuous solution domain into a finite number of elements connected at nodes. By constructing shape functions, the displacement within each element is interpolated from nodal values. The element stiffness matrix is formulated in the local coordinate system and then transformed into the global coordinate system. By assembling all element equations and applying boundary conditions, the global equation system is solved to obtain nodal displacements. Subsequently, strains and stresses in each element are derived from the displacement results.
For gear contact problems, which involve material nonlinearity, geometric nonlinearity and contact nonlinearity, the finite element method offers significant advantages over classical analytical methods. It can handle complex geometries, boundary conditions and loading conditions realistically, and provides more accurate results. The finite element analysis procedure includes the following steps: discretization of the continuum, construction of shape functions, formulation of local element equations, coordinate transformation, assembly of the global equation system, application of boundary conditions, and solution of the global equation system. The computational results converge to the exact solution as the element density is increased or the interpolation function order is elevated.
3. Digital Modeling Technology of Involute Straight Bevel Gear
3.1 Accurate 3D Solid Modeling Based on Spherical Involute
Accurate three-dimensional solid modeling of involute straight bevel gear is the prerequisite for finite element simulation and subsequent modification design. The conventional method of using planar involute on the back cone to approximate the tooth profile causes unavoidable model errors, especially when the ratio of cone distance to large-end module is small. To improve accuracy, I adopt the spatial spherical involute to generate the exact tooth flank of involute straight bevel gear.
The spherical involute is generated by a plane tangent to the base cone rolling purely on the base cone. Any radial line on the rolling plane sweeps a conical involute surface. Intersecting this surface with a sphere centered at the cone apex produces the spherical involute curve. When the sphere radius approaches infinity, the spherical involute degenerates into the planar involute.
In the Cartesian coordinate system, the spherical involute equations are expressed as:
$$ \begin{cases} x = l(\sin\phi\cos\theta\cos\varphi + \cos\phi\sin\varphi) \\ y = l(\sin\phi\cos\theta\sin\varphi – \cos\phi\cos\varphi) \\ z = l\cos\phi\cos\theta \end{cases} $$
where $\theta$ is the base cone angle, $l = \sqrt{x^2 + y^2 + z^2}$ is the radius of the sphere on which the spherical involute lies, $\phi$ is the angle between the initial line on the rolling plane and the instantaneous axis of rotation, and $\varphi$ satisfies $\sin\varphi = \sin\phi / \sin\theta$.
Based on the basic structural parameters of the involute straight bevel gear pair, as listed in Table 1, I established the accurate three-dimensional solid models of the planetary gear and the side gear.
| Parameter | Planetary gear | Side gear |
|---|---|---|
| Number of teeth | 10 | 15 |
| Module / mm | 4.438 | |
| Pressure angle / ° | 22.5 | |
| Cone distance / mm | 40 ± 0.025 | |
| Shaft angle / ° | 90 | |
| Addendum / mm | 4.98 | 3.28 |
| Whole depth / mm | 9.17 | 9.169 |
| Pitch diameter / mm | 44.38 | 66.57 |
| Pitch cone angle / ° | 33.69 | 56.31 |
| Face cone angle / ° | 46 | 66 |
| Root cone angle / ° | 24 | 44 |
Using the equation-driven curve function in SolidWorks, I drew the spatial spherical involute and then constructed the tooth flank by the boundary surface feature. Through mirroring, rotating, patterning and extruding operations, I completed the accurate models of the standard involute straight bevel gear pair and assembled them without interference. A critical aspect is to ensure the absence of interference in the assembled model for subsequent numerical simulation.
3.2 Finite Element Modeling of Involute Straight Bevel Gear Based on ANSYS/LS-DYNA
Considering the powerful nonlinear dynamic analysis capability of ANSYS/LS-DYNA, I utilized it for the dynamic contact simulation of the involute straight bevel gear pair. The finite element model was established through the following steps.
Element type and material properties. The gear body was meshed with SOLID164 elements, which are eight-node hexahedral elements for explicit three-dimensional structural analysis. Since the SOLID164 element has no rotational degrees of freedom, I defined the inner cylindrical surfaces of the two gears as rigid bodies using SHELL163 elements. The rigid bodies share nodes with the solid elements to drive the gear rotation and speed up computation. The gear material is 20CrMnTi steel, treated as linear elastic with density $\rho = 7.8 \times 10^3 \text{ kg/m}^3$, elastic modulus $E = 2.07 \times 10^5 \text{ MPa}$, and Poisson’s ratio $\nu = 0.3$. For the SHELL163 elements, the thickness was set to 0.0001 m, and the S/R co-rotational multi-integration-point element formulation was selected to reduce hourglass modes.
Mesh generation. To obtain precise elastic deformation of the tooth flank, I generated hexahedral mesh models by partitioning the gear into sweepable regions. At the same time, to reduce computational expense without sacrificing accuracy in the contact stress analysis, I also created tetrahedral mesh models with appropriate mesh density in critical contact areas. The number of elements in the hexahedral model was 83796 for the planetary gear and 195299 for the side gear, while for the tetrahedral model the numbers were 39528 and 68290, respectively.
Contact definition. The automatic surface-to-surface contact algorithm was employed to simulate the complex tooth contact. The static friction coefficient was 0.5 and the dynamic friction coefficient was 0.3. For the hexahedral mesh model, four parts were defined, and the planetary gear and side gear were set as the contact and target surfaces, respectively. To prevent hourglass deformation from undermining the simulation accuracy, I used the stiffness-based hourglass control type 4 with a coefficient of 0.145, which was only applied to the part with the most severe hourglass energy.
Boundary conditions and loading. The planetary gear (driving gear) was constrained against all translational degrees of freedom and the rotations around x and y axes of the inner bore. An angular velocity of $\omega = 157 \text{ rad/s}$ was applied to the driving gear, and a resistance torque of $T = 100 \text{ Nm}$ was applied to the driven gear. To avoid initial impact, both the angular velocity and torque were applied gradually as functions of time. The arrays of time, angular velocity and torque are defined as follows:
$$ \text{Time: } [0, 0.0005, 0.001, \ldots, 0.005] \text{ s} $$
$$ \Omega: [0, 15.7, 31.4, \ldots, 157] \text{ rad/s} $$
$$ \text{Torque: } [0, -10, -20, \ldots, -100] \text{ Nm} $$
With this loading method, the gear pair enters a steady working state smoothly without severe transient impact. The total simulation time was set to 0.004 s, corresponding to one complete tooth meshing cycle of the planetary gear.
Solution and post-processing. The minimum time step was set to $-1 \times 10^{-7}$ s with a time step scale factor of 0.6. The output files were written in LS-DYNA format for post-processing with LS-Prepost. From the dynamic contact simulation of the involute straight bevel gear pair, I extracted the equivalent stress contours at various meshing times. The contact stress at the pitch cone of the planetary gear reached a maximum value of 1208 MPa, while the theoretical value is 1338.45 MPa. The difference is about 8.2 percent. The tooth root bending stress of the side gear reached 603 MPa, compared with the theoretical value of 667.63 MPa. The difference is about 9.68 percent. These deviations are reasonable and mainly result from the fact that the theoretical method is based on the equivalent spur gear at the mean cone with higher safety margins, whereas the finite element method accounts for more realistic dynamic and friction effects. The comparison validates the correctness and rationality of my finite element simulation.
4. Orthogonal Optimization Design of Isometric Modification Parameters for Involute Straight Bevel Gear
4.1 Isometric Modification and Orthogonal Design Scheme
To avoid the end stress concentration and reduce meshing impact, I proposed an isometric modification method for the involute straight bevel gear pair. In this method, only the planetary gear is modified, while the side gear remains unmodified. A new tooth flank is generated by offsetting the original tooth flank along its normal direction within a designated region. This region still retains the involute surface characteristics. The boundary of the isometric modification region is formed by two straight lines parallel to the gear axis at the pitch cone angle and two circular arcs smoothly connecting them. Proper fillets at the boundary prevent excessive stress concentration during both the forging process and the finite element dynamic contact analysis.
The isometric modification can be conveniently realized through the chemical milling process for finishing the electrode gear used in precision forging. The forged involute straight bevel gear can meet the modification requirements and exhibit improved surface quality and internal material flow. Since the modification amount is negligible compared with the basic gear dimensions, the influence of isometric modification on the meshing point position and rotation angle is negligible, meaning that the transmission ratio remains essentially unchanged.
For the orthogonal optimization design of isometric modification, I selected three factors that significantly affect the modification effect: the modification amount (factor A), the modification height (factor B), and the modification position (factor C). Each factor has three levels, as shown in Table 2.
| Level | A: Modification amount / μm | B: Modification height | C: Modification position |
|---|---|---|---|
| 1 | 25 | 50% of mean tooth height | 1/3 tooth width from small end |
| 2 | 35 | 60% of mean tooth height | Middle of tooth width |
| 3 | 45 | 70% of mean tooth height | 1/3 tooth width from large end |
Based on the gear accuracy grade 8 and the effective meshing trace requirements (35% to 65% along tooth length and 40% to 70% along tooth height), I set the meshing trace percentage along the tooth length to 60%. The modification height levels of 50%, 60%, and 70% of the tooth height at the middle of the tooth width were chosen to ensure geometrically similar modification shapes. The modification position is determined by the center of the modified profile: along the tooth height it is always at the middle, and along the tooth width it varies among three positions. Since the interactions among these three factors are not considered, the $L_9(3^4)$ orthogonal array is appropriate. The three factors are assigned to the first three columns of the orthogonal table, and the fourth column is left blank as the error column.
4.2 Finite Element Results and Range Analysis
I established the isometric modified involute straight bevel gear models according to the nine parameter combinations listed in the orthogonal table, and carried out dynamic contact finite element simulations under the same working conditions as described in Section 3. The maximum contact stress at the tooth tip of the driven gear during meshing-in was chosen as the evaluation index for the meshing impact severity. The simulation results and range analysis are presented in Table 3.
| No. | A | B | C | Maximum contact stress / MPa |
|---|---|---|---|---|
| 1 | 1 | 1 | 1 | 3091.2 |
| 2 | 1 | 2 | 2 | 2521.8 |
| 3 | 1 | 3 | 3 | 3120.3 |
| 4 | 2 | 1 | 2 | 2642.8 |
| 5 | 2 | 2 | 3 | 3761.6 |
| 6 | 2 | 3 | 1 | 3502.9 |
| 7 | 3 | 1 | 3 | 2920.2 |
| 8 | 3 | 2 | 1 | 2290.8 |
| 9 | 3 | 3 | 2 | 3720.2 |
| $T_{1j}$ | 2916.5 | 2885.0 | 2961.6 | |
| $T_{2j}$ | 3302.4 | 2860.2 | 2963.7 | |
| $T_{3j}$ | 2977.2 | 3451.1 | 3270.8 | |
| $R_j$ | 385.9 | 590.9 | 309.2 |
From the range analysis in Table 3, it is evident that $R_B > R_A > R_C$, indicating that the modification height has the most significant influence on the meshing impact improvement, followed by the modification amount, and the modification position has the least influence. The optimal isometric modification parameter combination is $A_1B_2C_1$, which means the modification amount of 25 μm, the modification height of 60% of the mean tooth height, and the modification position at 1/3 tooth width from the small end. This combination was not included in the original nine test groups, so I established an additional model with the combination $A_1B_2C_1$ and performed dynamic contact simulation. The maximum contact stress at the driven gear tooth tip was 1928.8 MPa, which is significantly lower than the unmodified value of 4068.5 MPa.
4.3 Dynamic Response Analysis of the Optimal Isometric Modified Involute Straight Bevel Gear
To evaluate the improvement of the meshing impact and vibration, I extracted the angular acceleration and axial acceleration time histories of the same node on the side gear tooth tip before and after the optimal isometric modification. The results are summarized as follows.
| Condition | Angular acceleration extremum / 10^6 rad/s² | Axial acceleration extremum / 10^6 m/s² | Improvement of angular impact | Improvement of axial impact |
|---|---|---|---|---|
| Unmodified | 0.872 | 0.594 | — | — |
| Optimal isometric modification | 0.186 | 0.102 | 78.6% | 82.8% |
From Table 4, the angular acceleration extremum is reduced from $0.872 \times 10^6$ rad/s² to $0.186 \times 10^6$ rad/s², representing a 78.6 percent improvement in meshing impact. The axial acceleration extremum is reduced from $0.594 \times 10^6$ m/s² to $0.102 \times 10^6$ m/s², representing an 82.8 percent improvement in axial impact. These results show that the optimized isometric modification effectively reduces the meshing interference impact and vibration of the involute straight bevel gear pair.
4.4 Tooth Stress Distribution Analysis of the Optimal Isometric Modified Gear
I further examined the equivalent stress distribution on the tooth flank along the tooth width direction. Before modification, severe end stress concentration occurs at both tooth ends, which is mainly caused by the edge effect of the nearly right-angle tooth edges. Moreover, the contact stress at the large end is higher than at the small end, reaching approximately 1880 MPa. After the optimal isometric modification, the maximum contact stress is significantly reduced to about 1100 MPa and the location of the maximum stress shifts to the isometric modification region, i.e., at the tooth width middle near the small end. This is the desirable contact area for involute straight bevel gear, since the contact area gradually extends toward the large end under load, leading to a more uniform and rational load distribution along the tooth width and improving the load-carrying capacity of the gear pair.
5. Comparison with Centrally Symmetric Constant-Radius Circular Arc Crowning
Conventionally, the centrally symmetric constant-radius circular arc crowning is widely adopted in engineering practice for lead modification. To compare the comprehensive modification effect with the optimized isometric modification, I designed a symmetric crowned involute straight bevel gear with the crown point located at the middle of the tooth width. The crowning amount was taken as 25 μm, which is the same as the optimal isometric modification amount obtained in my orthogonal design. The crowned tooth surface was generated by sweeping a circular arc guide curve along the tooth flank, and the modified gear model was established in SolidWorks.
The dynamic contact finite element simulation was carried out under the same working conditions. The angular acceleration extremum of the side gear tooth tip was $0.474 \times 10^6$ rad/s², representing a 45.6 percent improvement compared with the unmodified gear. The axial acceleration extremum was $0.245 \times 10^6$ m/s², representing a 58.8 percent improvement. The maximum contact stress on the tooth flank was about 1400 MPa, which is lower than the unmodified end stress but higher than that of the optimally isometric-modified gear. The stress concentration region was shifted to the middle of the tooth width, which is an improvement over the unmodified gear.
Comparing the two modification forms, the optimally isometric-modified involute straight bevel gear outperforms the symmetric crowned gear in the following aspects:
- More significant reduction of angular acceleration extremum: The improvement of the optimally isometric-modified gear surpasses the symmetric crowned gear by about 33 percent.
- More significant reduction of axial acceleration extremum: The improvement of the optimally isometric-modified gear surpasses the symmetric crowned gear by about 24 percent.
- Lower maximum contact stress: The optimally isometric-modified gear has a maximum contact stress of 1100 MPa, whereas the symmetric crowned gear has 1400 MPa under the same load.
- More rational stress distribution: The stress concentration of the optimally isometric-modified gear is located at the tooth width middle near the small end, which is the best contact area for involute straight bevel gear, while the symmetric crowned gear concentrates the stress at the middle of the tooth width.
It can be concluded that under the given working conditions, the optimized isometric modification achieves better comprehensive modification performance than the conventional centrally symmetric constant-radius circular arc crowning for the involute straight bevel gear.
6. Variable Curvature Surface Modification Technology of Involute Straight Bevel Gear
6.1 Motivation for Variable Curvature Crowning
In actual service, the involute straight bevel gear pair is affected by various installation errors, manufacturing errors, and complex tooth load distributions. As a result, the deformation along the tooth width direction is complex and does not generally follow a symmetric circular arc pattern. Therefore, the centrally symmetric crowning and the optimal isometric modification may not guarantee the ideal modification effect when the meshing misalignment is large or uncertain. In my research, I propose a target variable curvature crowning method. The basic idea is to obtain the actual tooth flank elastic deformation through dynamic contact finite element analysis under a specific working condition, and then establish the variable curvature crowned tooth surface that exactly compensates the deformation. In this way, the tooth load can be more uniformly distributed and the edge stress concentration can be avoided more effectively.
6.2 Accurate Modeling Method for Variable Curvature Crowned Involute Straight Bevel Gear
Since the variable curvature crowning involves complex curved surfaces, conventional modeling methods that remove material from the standard gear tooth flank by sweeping cannot accurately represent the variable curvature crowning shape and cannot realize parametric modeling. To solve this problem, I developed a novel method that comprehensively utilizes the strengths of ANSYS/LS-DYNA, SolidWorks, Matlab and Imageware.
First, I established a three-dimensional model of the involute straight bevel gear pair with long shafts, and performed dynamic contact finite element analysis to obtain the elastic deformation of the tooth flank at the pitch cone under the given working condition. For a representative node, the elastic deformation was determined by extracting the circumferential displacement difference between the driving and driven sides within a time interval around the contact moment. The deformation values of the planetary gear and the side gear were superimposed to obtain the modification amount.
Second, I extracted the nodal coordinates of the undeformed tooth flank at the pitch cone from the finite element model and imported them into SolidWorks to construct the normal reference plane. On this plane, the modified nodal positions were determined according to the deformation data. Since a variable curvature crowning curve is the envelope of the deformed points, I also incorporated an additional 5 μm modification amount at the five nodes closest to each tooth end to prevent edge effects.
Third, the deformed nodal coordinates were imported into Matlab, where nonlinear curve fitting was conducted to obtain the variable curvature crowning curve equation. The fitted equation has the form:
$$ y’ = a z’^2 + b z’ + c $$
where, for the specific case in this study, $a = 7.971$, $b = -0.398$, and $c = 0.744$.
Based on the standard spherical involute equations and the crowning curve, I derived the tooth flank equation of the variable curvature crowned involute straight bevel gear:
$$ \begin{cases} x = l[\sin\phi\sin(\varphi + \eta_i) + \cos\phi\cos(\varphi + \eta_i)\sin\theta] \\ y = l[\sin\phi\cos(\varphi + \eta_i)\sin\theta – \cos\phi\sin(\varphi + \eta_i)] \\ z = l\cos(\varphi + \eta_i)\cos\theta \end{cases} $$
The angle $\eta_i$ at each tooth width position is determined by the crowning curve as:
$$ \eta_i = \frac{[a(z’)^2 + b z’ + c]B}{B r_{b1} + z'(r_{b2} – r_{b1})} $$
where $B$ is the tooth width, $r_{b1}$ and $r_{b2}$ are the base circle radii at the small end and the large end, respectively, and $z’$ is the coordinate along the tooth width direction measured from the small end.
With this derived equation, I drew the variable curvature crowned tooth surface in Matlab and verified its correctness by comparing with the standard tooth flank. The comparison shows that the crowned surface deviates slightly from the standard surface at both the small end and the large end, while coinciding with it at the middle of the tooth width, which matches the intended crowning behavior.
Fourth, I extracted the coordinate points of the fully parameterized crowned tooth surface and saved them in .pts format. The point cloud was then imported into Imageware for surface reconstruction. The reconstructed surface was checked for smoothness and precision using the high-light line method. Finally, the surface was exported in .IGES format and imported into SolidWorks, where mirroring, shearing and rotating operations were performed to complete the three-dimensional solid model of the variable curvature crowned involute straight bevel gear. The assembly model was subsequently established without interference.
6.3 Finite Element Verification of Variable Curvature Crowning
The established variable curvature crowned involute straight bevel gear pair with long shafts was imported into ANSYS/LS-DYNA for dynamic contact finite element simulation under the same boundary and loading conditions as the unmodified one. For comparison, I also simulated the centrally symmetric crowned model and the optimally isometric-modified model with long shafts under identical conditions. The maximum contact stress values on the tooth flank are listed in Table 5.
| Modification form | Maximum contact stress / MPa |
|---|---|
| Unmodified with long shaft | 1900 |
| Centrally symmetric crowned | 1760 |
| Optimally isometric modified | 1490 |
| Variable curvature crowned | 1250 |
From Table 5, it is clear that the variable curvature crowning achieves the lowest maximum contact stress among all modification forms. A detailed examination of the stress distribution along the tooth width direction shows that the unmodified gear suffers from severe stress concentration at the large end due to the elastic deformation of the shaft and gear body. The centrally symmetric crowning shifts the stress concentration slightly toward the middle and reduces the maximum stress, while the isometric modification further reduces the stress. However, the variable curvature crowning nearly eliminates the stress concentration at both ends and yields the most uniform load distribution along the tooth width. The maximum contact stress is located in the intended contact area and is significantly lower than that of the other modification forms. Therefore, the variable curvature crowning is necessary and superior for involute straight bevel gear under specific working conditions with pronounced deformation asymmetry.
6.4 Parameterization Prospects
The method I proposed in this section can also be extended to realize the parametric modeling of variable curvature crowned involute straight bevel gear. Based on the derived variable curvature crowning tooth flank equations and the theory of uniform B-spline surface interpolation, a parametric modeling program can be developed. By simply replacing the crowning surface equation and parameters in the program, the variable curvature crowned tooth surface suitable for different specific working conditions can be conveniently generated. This provides a theoretical foundation for the future development of a fully parametric variable curvature crowning design system for involute straight bevel gear.
7. Conclusions and Future Work
In this thesis, I systematically studied the tooth modification design and tooth profile optimization of involute straight bevel gear. The main conclusions are summarized as follows.
(1) The accurate three-dimensional solid model of involute straight bevel gear was established based on the spatial spherical involute equation in SolidWorks, which overcomes the accuracy deficiency of the conventional back-cone planar involute approximation. The dynamic contact finite element model of the involute straight bevel gear pair was built in ANSYS/LS-DYNA. The comparison between the numerical results and the classical analytical results shows that the contact stress and bending stress errors are within 10 percent, confirming that the finite element simulation is accurate and capable of providing reliable deformation and stress information for modification design.
(2) The deformation and stress analysis of involute straight bevel gear based on Hertz theory and material mechanics is fundamental for modification design. However, the analytical solutions are essentially approximate under simplified assumptions. The finite element method is superior in handling complex geometry, boundary conditions and load cases, and is indispensable for high-quality gear modification design. Dynamic contact simulation is necessary to realistically capture the continuous dynamic behavior of the involute straight bevel gear pair during meshing.
(3) The orthogonal experimental design combined with dynamic contact finite element simulation was successfully applied to optimize the isometric modification parameters of involute straight bevel gear. The modification height is the most influential factor, followed by the modification amount, and the modification position has the least influence. The optimal isometric modification parameters under the given working condition are a modification amount of 25 μm, a modification height of 60 percent of the mean tooth height, and a modification position at one-third of the tooth width from the small end. The optimized isometric modification significantly reduces the angular acceleration extremum by 78.6 percent and the axial acceleration extremum by 82.8 percent compared with the unmodified gear, and effectively eliminates the end stress concentration.
(4) Compared with the centrally symmetric constant-radius circular arc crowning, the optimized isometric modification achieves better comprehensive performance for involute straight bevel gear under the given working condition, including a greater reduction of meshing impact and vibration, lower maximum contact stress, and more rational tooth load distribution.
(5) For specific working conditions where the deformation distribution is asymmetric or the meshing misalignment is large, the target variable curvature crowning is necessary. I proposed a novel method for the accurate three-dimensional solid modeling of variable curvature crowned involute straight bevel gear, which integrates dynamic contact finite element analysis, curve fitting in Matlab, surface reconstruction in Imageware, and solid modeling in SolidWorks. The derived variable curvature crowning tooth flank equations provide a theoretical basis for parametric modeling. The dynamic contact simulation results show that the variable curvature crowned involute straight bevel gear achieves the most uniform tooth load distribution and the lowest maximum contact stress among all modification forms, proving the necessity and superiority of variable curvature crowning.
Future research directions include: (1) establishing a complete gear transmission system model including the housing and bearings to analyze the coupled elastic deformation and stress distribution on the involute straight bevel gear in a more realistic environment; (2) incorporating thermal deformation into the modification design of involute straight bevel gear; and (3) conducting physical experiments to validate the proposed modification design methods under realistic operating conditions.
In summary, this thesis provides a systematic methodology for the tooth modification design and tooth profile optimization of involute straight bevel gear, which is valuable for reducing vibration and noise, improving load-carrying capacity, and broadening the application range of involute straight bevel gear in modern industry.