The straight bevel gear is one of the most important machine elements used to transmit motion and power between intersecting shafts. It is widely found in machine tools, vehicle differentials, agricultural machinery, aerospace actuators, marine transmissions and many other industrial applications. Compared with spiral bevel gears, the tooth trace of the straight bevel gear is simpler, and therefore its design, manufacturing and assembly are usually easier. The production cost of a straight bevel gear is also lower, and for this reason it remains an indispensable solution in many non-high-speed transmission systems.
Although the straight bevel gear is structurally simpler than curved-tooth bevel gears, its gear tooth geometry is still based on a spherical involute surface. This makes the generation of the tooth flank more complex than that of a cylindrical involute gear. In current manufacturing practice, the dominant processes for producing a straight bevel gear are planing, milling and broaching. These methods are usually intermittent because the blank must be indexed after each tooth is cut. They contain a considerable amount of idle stroke, require special cutters, and do not realize continuous generating motion. With the increasing demand for high efficiency and high precision, these traditional methods are becoming difficult to satisfy modern production requirements.
In this research, I propose a continuous generating hobbing method for the straight bevel gear. The method uses a conical hob whose basic geometric form is similar to a mating straight bevel gear. During cutting, the conical hob rotates continuously while the workpiece rotates at a fixed ratio. The tooth flank of the workpiece is generated by the enveloping motion of the hob cutting edges. This idea brings the well-known advantages of cylindrical gear hobbing into the field of bevel gear manufacturing. I studied the tooth surface mathematics, designed the conical hob, performed finite-element verification and carried out a virtual machining simulation in VERICUT.
1. Tooth Surface Equation and Parametric Design of the Straight Bevel Gear
Before developing the hobbing process, I first established a reliable geometric model of the straight bevel gear. The theoretical tooth flank of a straight bevel gear is a spherical involute surface. This surface can be generated by a plane rolling without slipping on a base cone. A point on the circumference of the rolling plane traces a spherical involute on a sphere whose radius is equal to the cone distance. All spherical involutes from the large end to the small end together form the complete tooth flank of the straight bevel gear.
Let the base cone angle be \(\theta_b\), and let \(R\) be the cone distance. I set a fixed coordinate system \(S(O;x,y,z)\) whose \(z\)-axis coincides with the cone axis. A moving coordinate system \(S_1(O;x_1,y_1,z_1)\) is attached to the tangent plane. After rolling through an angular parameter \(\varphi\), the no-slip condition gives
\[
\psi=\varphi\sin\theta_b
\]
where \(\psi\) describes the angular position of the generating point on the tangent plane. The coordinates of a point on the large-end spherical involute of the straight bevel gear can then be expressed as
\[
\begin{cases}
x_e = R\left[\sin\theta_b\cos\varphi\cos\psi+\sin\varphi\sin\psi\right]\\[2mm]
y_e = R\left[\sin\theta_b\sin\varphi\cos\psi-\cos\varphi\sin\psi\right]\\[2mm]
z_e = R\cos\theta_b\cos\psi
\end{cases}
\]
For the small end, the cone distance is reduced to \(R-B\), where \(B\) is the face width of the straight bevel gear. Therefore the small-end spherical involute is
\[
\vec r_i(\varphi)=(R-B)
\begin{bmatrix}
\sin\theta_b\cos\varphi\cos\psi+\sin\varphi\sin\psi\\
\sin\theta_b\sin\varphi\cos\psi-\cos\varphi\sin\psi\\
\cos\theta_b\cos\psi
\end{bmatrix}
\]
Combining the two ends, the entire convex or concave flank of the straight bevel gear can be represented by
\[
\vec r(l,\varphi)=l
\begin{bmatrix}
\sin\theta_b\cos\varphi\cos\psi+\sin\varphi\sin\psi\\
\sin\theta_b\sin\varphi\cos\psi-\cos\varphi\sin\psi\\
\cos\theta_b\cos\psi
\end{bmatrix},\qquad
l\in[R-B,\;R]
\]
This surface equation was the starting point for the parametric modeling of the straight bevel gear in SolidWorks. I used an equation-driven modeling approach in which the characteristic parameters of the straight bevel gear are treated as independent variables. The characteristic parameters include the number of teeth, module \(m\), pressure angle \(\alpha\), addendum coefficient \(h_a^*\) and tip clearance coefficient \(c^*\). All dependent dimensions are calculated through the equations listed in Table 1.
| Dependent Parameter | Equation |
|---|---|
| Transmission ratio | \(i_{12}=z_g/z_p\) |
| Pinion pitch cone angle | \(\delta_p=\arctan(z_p/z_g)\) |
| Gear pitch cone angle | \(\delta_g=90^\circ-\delta_p\) |
| Pitch diameter | \(d=mz\) |
| Addendum | \(h_a=h_a^*m\) |
| Dedendum | \(h_f=(h_a^*+c^*)m\) |
| Cone distance | \(R=d_p/(2\sin\delta_p)\) |
| Tip circle diameter | \(d_a=d+2h_a\cos\delta\) |
| Base circle diameter | \(d_b=d\cos\alpha\) |
| Root circle diameter | \(d_f=d-2h_f\cos\delta\) |
| Face width | \(B\le R/3\) (selected) |
| Addendum angle | \(\theta_a=\arctan(h_a/R)\) |
| Dedendum angle | \(\theta_f=\arctan(h_f/R)\) |
| Tip cone angle | \(\delta_a=\delta+\theta_a\) |
| Root cone angle | \(\delta_f=\delta-\theta_f\) |
Every dimension in the three-dimensional solid model was linked to these equations or to the characteristic parameters. When I changed the characteristic parameters in the embedded Excel worksheet, the model of the straight bevel gear was automatically reconstructed. This parametric feature is important because the same model is later used both as the workpiece blank and as the reference geometry for the comparison measurement in VERICUT.
2. Hobbing Principle of the Straight Bevel Gear
The hobbing process for a straight bevel gear is based on the meshing of two bevel gears. In a pair of straight bevel gear drives, the two pitch cones roll without sliding. The common generator of the two pitch cones is the instantaneous axis of relative rotation. For a shaft angle \(\Sigma=\delta_1+\delta_2\), the angular velocity ratio is
\[
\frac{\omega_1}{\omega_2}=\frac{\sin\delta_2}{\sin\delta_1}
\]
where \(\omega_1\) and \(\omega_2\) are the angular velocities of the two gears. The relative angular velocity vector \(\omega_{12}=\omega_1-\omega_2\) lies along the instantaneous axis. When I apply this idea to machining, the conical hob is treated as one member of the meshing pair and the workpiece is the other member. The cutting edges on the hob correspond to the tooth surfaces of a mating straight bevel gear. As both parts rotate, the workpiece flank is generated by the conjugate motion.
The required cutting motion is a simple relation
\[
\frac{n_w}{n_h}=\frac{z_h}{z_w}
\]
where \(n_w\) is the workpiece rotational speed, \(n_h\) is the hob rotational speed, \(z_h\) is the number of teeth on the conical hob and \(z_w\) is the number of teeth on the workpiece straight bevel gear. This equation is the fundamental synchronizing equation of the hobbing process.
In the hobbing machine model, the machining motion includes four basic axes, as listed in Table 2.
| Axis | Function |
|---|---|
| \(C_1\) | Rotation of the conical hob spindle; principal cutting motion |
| \(C\) | Rotation of the workpiece; generating motion synchronized with the hob |
| \(X\) | Radial infeed motion that determines the depth of cut |
| \(Z\) | Axial feed motion along the workpiece axis |
To describe the position and motion of the cutting edge, I established four coordinate systems: the tool frame \(S_1\), the tool-orientation frame \(S_2\), the machine reference frame \(S_3\), and the workpiece frame \(S_4\). A point \(\mathbf{p}_t\) in the tool frame is transformed to the workpiece frame by
\[
\mathbf{p}_w=\mathbf{T}_{41}\mathbf{p}_t=\mathbf{T}_{43}\mathbf{T}_{32}\mathbf{T}_{21}\mathbf{p}_t
\]
where \(\mathbf{T}_{21}\) represents the rotation of the conical hob about its own axis, \(\mathbf{T}_{32}\) represents the installation angle of the hob axis relative to the workpiece axis, and \(\mathbf{T}_{43}\) represents the workpiece rotation together with the translational feed axes. This transformation is the basis for generating the CNC program used in the VERICUT simulation.
3. Structural Design of the Conical Hob
The conical hob is the most critical tool in this new hobbing method. Because the workpiece tooth flank is generated by the envelope of the hob teeth, the conical hob must have the same fundamental geometry as a mating straight bevel gear. I therefore designed the hob by starting from a straight bevel gear and adding cutting edges, flutes and clearance angles.
The tooth width of the conical hob should be slightly larger than the face width of the workpiece straight bevel gear. This ensures that the complete tooth flank from the large end to the small end can be generated. The conical flute helix follows the tapered surface of the hob. The conical angle of the flute is selected between the root cone angle and the tip cone angle of the mating straight bevel gear.
The cutting face of each flute is inclined with respect to the axial plane of the hob. This inclination improves chip flow and reduces cutting force. If the inclination angle is too large, however, the cutting edge may become weak and chipping may occur. I therefore set the inclination angle in the range from \(5^\circ\) to \(15^\circ\). The tooth space width of the hob was selected as \(5\%\) to \(15\%\) of the tool face width. The top relief angle was chosen between \(10^\circ\) and \(12^\circ\), and the side relief angle was kept greater than \(3^\circ\). These values follow conventional gear-hob design practice and guarantee sufficient edge strength. The main design parameters are listed in Table 3.
| Design Parameter | Recommended Value |
|---|---|
| Flute face inclination angle | \(5^\circ\)–\(15^\circ\) |
| Top relief angle | \(10^\circ\)–\(12^\circ\) |
| Side relief angle | \(>3^\circ\) |
| Tooth space width ratio | \(5\%\)–\(15\%\) of tool face width |
| Tool face width | Slightly larger than workpiece face width |
| Flute cone angle | Between root angle and tip angle |
I created the three-dimensional model of the conical hob in SolidWorks. The hob body is made of high-speed steel. The material properties used in the finite-element analysis are given in Table 4.
| Property | Value |
|---|---|
| Density | \(8300\ \mathrm{kg/m^3}\) |
| Poisson ratio | \(0.27\) |
| Elastic modulus | \(200\ \mathrm{GPa}\) |
4. Finite-Element Verification of the Conical Hob
4.1 Static Analysis
I imported the conical hob into ANSYS Workbench and performed a static structural analysis to examine whether the tooth root and the cutting edges can withstand the cutting load. The three-dimensional model was meshed using a swept meshing method, and the contact region on the tooth flank was locally refined so that the stress gradient could be captured accurately.
Two loading cases were considered. In the first case, the load was applied to simulate the initial cutting state, where only a small part of the cutting edge is involved. In the second case, the load was applied over a larger contact area to simulate a deeper cutting state. In both cases, the total deformation of the conical hob was very small. The stress distribution was concentrated mainly near the cutting edge and the tooth root, but the maximum equivalent stress remained below the yield strength of the high-speed steel. The strain values were also small. This indicates that the conical hob structure is sufficiently rigid for the hobbing of a straight bevel gear.
4.2 Modal Analysis
Resonance is a major concern in a continuous generating process because the tool is always rotating at high speed. I performed a modal analysis of the conical hob with a fixed constraint applied at the shank. The first six natural frequencies are listed in Table 5.
| Mode | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Frequency / Hz | 4153.5 | 4155.3 | 5868.1 | 17399 | 18571 | 18636 |
The lowest natural frequency of the conical hob is \(4153.5\ \mathrm{Hz}\), which is far above the normal working frequency of the machine spindle and the tooth-meshing frequency. Therefore the hobbing process will not excite a destructive resonance in the conical hob. The first two mode shapes are bending modes in the \(x\)-\(y\) plane, the third mode is mainly torsion about the hob axis, the fourth and fifth modes are bending modes in the \(x\)-\(z\) plane, and the sixth mode is an axial stretching mode. The mode shape results provide useful guidance for optimizing the tool holder and clamping system.
4.3 Dynamic Contact Analysis
I also studied the dynamic contact behavior between the conical hob and the workpiece straight bevel gear. The same characteristic parameters used in the hobbing simulation are listed in Table 6.
| Parameter | Pinion | Gear |
|---|---|---|
| Number of teeth | 20 | 21 |
| Module / mm | 1 | 1 |
| Pressure angle / ° | 20 | 20 |
| Addendum coefficient | 1 | 1 |
| Face width / mm | 5 | 5 |
| Tip clearance coefficient | 0.25 | 0.25 |
The conical hob flank was set as the target surface and the workpiece tooth flank was set as the contact surface. The friction coefficient was set to \(0.15\). I selected the augmented Lagrange method because it provides good balance between accuracy and convergence. The workpiece was fixed, while the conical hob was allowed to rotate under a small torque. The simulation time was set to \(5\ \mathrm{ms}\) with 500 substeps.
The simulation results show that the maximum contact stress first appears where the tooth enters the meshing zone. As the hob rotates, the contact zone moves gradually along the tooth flank of the straight bevel gear. The contact pattern is smooth and continuous, with the main contact area located near the middle of the tooth flank. No severe edge loading is observed. This dynamic behavior is consistent with the expected meshing of a real pair of straight bevel gear teeth, confirming that the geometric design of the conical hob is suitable for the continuous hobbing process.
5. VERICUT Simulation and Virtual Measurement
5.1 Construction of the Simulation Model
Since a physical prototype was not yet available, I used VERICUT software to verify the complete hobbing process. VERICUT can simulate the motion of CNC axes, the removal of material by the cutting tool, and the difference between the machined model and the design model. The simulation workflow includes several steps.
First, I created the three-dimensional models of the workpiece blank, the conical hob and the reference straight bevel gear in SolidWorks. The workpiece blank was exported as an STL file and loaded into the VERICUT project. The conical hob was created in the VERICUT tool manager by importing its CAD geometry. Then I built the machine configuration according to the hobbing principle. The topological structure consists of the machine base, the \(Z\)-axis, the \(X\)-axis, the workpiece \(C\)-axis, the tool spindle \(C_1\), the tool holder and the conical hob.
I selected the FANUC 16IM control system as the virtual CNC controller. Because the standard controller cannot directly handle two synchronized rotary axes, I added a custom macro named C2AxisMotion to realize the synchronous rotation of the workpiece axis and the hob spindle. The NC program was written manually according to the coordinate transformation and the process requirement. The main machining stages are listed in Table 7.
| Stage | Motion |
|---|---|
| Positioning | Rapid traverse to the starting point |
| Spindle start | Hob spindle set to cutting speed |
| Radial infeed | \(X\)-axis feeds to the required depth |
| Axial feed | \(Z\)-axis feeds along the gear axis |
| Synchronized rotation | Workpiece \(C\)-axis and hob \(C_1\)-axis rotate with a fixed ratio |
| Retraction | Rapid traverse back to the starting point |
5.2 Simulation Result and Comparison
The VERICUT simulation process generated the complete tooth flanks of the straight bevel gear. To evaluate the accuracy of the simulated machining result, I created a standard straight bevel gear model using the same parametric design procedure. The standard model was then imported into VERICUT and compared with the machined workpiece by using the automatic comparison function.
The comparison result showed that the machined workpiece model is in very good agreement with the standard model. Both the convex flanks and the concave flanks of the straight bevel gear are well generated from the large end to the small end. The amount of overcut and remaining material is very small and lies within a reasonable tolerance range. The tooth profile is a proper spherical involute, and no severe undercutting or interference is observed. This proves that the proposed continuous hobbing method and the designed conical hob are capable of producing a correct straight bevel gear tooth form.
6. Conclusion and Future Work
In this work, I presented a systematic study on the hobbing technology of the straight bevel gear. The main contributions are summarized as follows.
First, I derived the spherical involute equation of the straight bevel gear tooth flank. Based on this equation, I established a parametric three-dimensional model of the straight bevel gear in SolidWorks. The model can be updated by changing only the characteristic parameters, which greatly improves the efficiency of gear design and provides a solid basis for subsequent simulation.
Second, I proposed a continuous generating hobbing method for the straight bevel gear. The method is based on the meshing of a pair of straight bevel gear teeth. I designed a conical hob whose geometry is derived from the mating gear and which carries cutting edges along conical helically arranged flutes. The static analysis, modal analysis and dynamic contact analysis all confirm that the conical hob has enough stiffness, avoids resonant vibration, and produces a smooth and continuous contact pattern on the tooth flank of the straight bevel gear.
Third, I built a virtual hobbing process in VERICUT. The simulation of the straight bevel gear hobbing process showed that the complete tooth space is generated correctly. The comparison between the machined workpiece and the standard model demonstrated that the hobbing method can produce an accurate straight bevel gear profile. Therefore the proposed technology offers a practical and efficient alternative to the traditional intermittent manufacturing methods for the straight bevel gear.
In the future, I plan to manufacture a real conical hob and carry out cutting experiments on a five-axis machining center. I will also study the influence of cutting speed, feed rate and hob wear on the surface quality and accuracy of the straight bevel gear. Further development may include the optimization of the hob geometry and the design of a dedicated CNC hobbing machine for the straight bevel gear. With the continuous improvement of the proposed method, it is expected that the straight bevel gear can be produced with higher efficiency and better quality in the near future.