In my research, I focused on the digital simulation and analysis of gear cutting for spiroid face gears. Gears are fundamental transmission elements in mechanical equipment, and their performance strongly affects operating efficiency, load capacity, vibration behavior, and service life. A spiroid face gear is a representative crossed-axis transmission component. Its tooth surface is characterized by variable module and variable helix angle, and during meshing it can form a relatively large contact area. For this reason, spiroid face gears offer advantages such as high load-carrying capacity, smooth transmission, compact layout, and low noise. These features make them attractive for high-end transmission systems, robotic joints, vehicle drivetrains, and sports and rehabilitation equipment. However, the geometric complexity of the tooth surface makes manufacturing difficult. Traditional casting, shaping, hobbing, grinding, and generic milling methods often struggle to achieve stable accuracy, especially when the helix angle is large and the module is small. In my work, I therefore studied gear cutting as a continuous generating method for spiroid face gears and combined experimental verification with digital simulation.
The main problem I addressed is that gear cutting of spiroid face gears still suffers from burrs, ragged edges, local interference, unstable cutting force, and tool wear. These problems are not only geometric issues. They arise from the coupled effects of tool geometry, spatial motion, material removal, heat generation, and contact friction. In conventional trial cutting, it is expensive and time-consuming to evaluate many tool structures and process parameters. Therefore, I developed a digital simulation framework based on finite element analysis and numerical machining verification. I used ABAQUS for cutting force simulation, DEFORM-3D for temperature and tool wear analysis, and VERICUT for tooth surface accuracy verification. The purpose of my study was to reveal how process parameters and cutter structures influence gear cutting behavior and to propose a practical way to improve tooth surface accuracy.
1. Background and Research Gap
Face gear manufacturing has been studied for many years. Shaping is based on the meshing relationship between a shaper cutter and the workpiece, and it has a mature theoretical foundation. However, shaping is an intermittent cutting process, and its efficiency is relatively low. Hobbing can provide continuous cutting and higher productivity, but the meshing relationship for complex face gear flanks is difficult to control, and accuracy may deteriorate. Grinding is often used as a finishing process and can improve surface quality, but the equipment cost is high and the process is sensitive to setup errors. CNC milling is flexible, but it is difficult to achieve high efficiency and high accuracy simultaneously for small-module spiroid face gears. Gear cutting, also called power skiving or skiving, has emerged as a promising continuous generating method. In gear cutting, the cutter rotates at high speed, the workpiece rotates in synchronization, and the cutter feeds along the tooth width. The relative motion creates a spatial envelope that removes material continuously. This makes gear cutting suitable for complex tooth surfaces, but the kinematics and cutting mechanics are more complicated than those of conventional shaping.
Previous studies have mainly concentrated on geometric modeling, tool design, and basic feasibility. Some researchers established envelope equations for face gear generation. Others analyzed cutter edge profiles and interference conditions. Digital simulation has also been used to study chip formation and tooth surface generation. However, for spiroid face gears with large helix angles, the relationship among cutting force, temperature, tool wear, interference, and final tooth accuracy has not been fully integrated. In particular, the influence of cutter structure on side flank contact and interference in gear cutting requires more detailed investigation. My research fills this gap by building a complete digital simulation chain from theoretical tooth surface modeling to finite element cutting analysis and numerical machining verification.
2. Experimental Feasibility of Gear Cutting
I first performed a feasibility experiment of gear cutting on a five-axis CNC machine. The cutter and workpiece parameters were selected to match the theoretical model. The cutter was designed as a gear-like cutting tool with a specified helix angle, normal module, pressure angle, rake angle, and relief angle. The workpiece was a spiroid face gear with a defined tooth number, modification coefficient, inner radius, outer radius, and offset distance. The main parameters are summarized in Table 1.
| Item | Parameter | Value |
|---|---|---|
| Cutter | Helix angle | 30.741° |
| Cutter | Number of teeth | 21 |
| Cutter | Normal module | 0.85 mm |
| Cutter | Normal pressure angle | 20° |
| Cutter | Addendum coefficient | 1 |
| Cutter | Clearance coefficient | 0.25 |
| Cutter | Rake angle | 5° |
| Cutter | Relief angle | 10° |
| Workpiece | Number of teeth | 34 |
| Workpiece | Modification coefficient | 0.2 |
| Workpiece | Inner radius | 16.8 mm |
| Workpiece | Outer radius | 19.8 mm |
| Workpiece | Offset distance | 7.5 mm |
During the experiment, the workpiece was mounted on the machine spindle and rotated about its own axis. The cutter was mounted on the tool spindle and rotated at high speed. The X, Y, and Z axes were used for positioning and feed. The motion relationship can be expressed by the transmission ratio:
$$ i_{12}=\frac{\omega_1}{\omega_2}=\frac{N_2}{N_1} $$
where \(\omega_1\) and \(\omega_2\) are the angular velocities of the workpiece and cutter, and \(N_1\) and \(N_2\) are their tooth numbers. The gear cutting experiment showed that the tooth profile was basically complete and continuous. The teeth were evenly distributed, and no missing teeth or severe discontinuities were observed. This verified that gear cutting is feasible for spiroid face gears. However, I also observed burrs at the tooth tip. This indicated that the material removal process was not uniformly stable. In the tooth tip region, the uncut chip thickness becomes small, and the cutting mechanism changes from shearing to extrusion and sliding. Plastic flow can accumulate at the free boundary and form burrs. The side flank of the cutter may also participate in contact, which can increase friction and interference. These observations motivated my subsequent simulation work on gear cutting.
3. Theoretical Modeling of the Spiroid Face Gear
To provide a reliable geometric reference, I established a theoretical model of the spiroid face gear based on the shaping principle. A virtual involute cylindrical pinion was used as the generating tool. The workpiece and cutter rotate about their respective axes, and their motion satisfies the meshing relationship. The coordinate systems include a fixed coordinate system for the cutter, a fixed coordinate system for the workpiece, a machine reference coordinate system, and moving coordinate systems attached to the cutter and workpiece. The transformation from the cutter coordinate system to the workpiece coordinate system can be written as:
$$ M_{pt}=M_{pm}M_{mn}M_{nq}M_{qt} $$
For example, the rotation matrix from the cutter moving frame \(S_t\) to the cutter fixed frame \(S_q\) is:
$$ M_{qt}=
\begin{bmatrix}
\cos\varphi_q & \sin\varphi_q & 0 & 0\\
-\sin\varphi_q & \cos\varphi_q & 0 & 0\\
0 & 0 & 1 & 0\\
0 & 0 & 0 & 1
\end{bmatrix}
$$
The transformation from the machine reference frame to the workpiece fixed frame includes the offset distance and axial position. After combining these transformations, the position of any point on the cutter can be mapped to the workpiece coordinate system. The involute helical tooth surface of the virtual pinion can be expressed in parametric form. For one tooth side, the surface vector is:
$$ \mathbf r_s(\theta_s,\mu_s)=
\begin{bmatrix}
r_{bs}\cos\theta_s+\mu_s\cos\lambda_{bs}\sin\theta_s\\
r_{bs}\sin\theta_s-\mu_s\cos\lambda_{bs}\cos\theta_s\\
p_s\theta_s-\mu_s\sin\lambda_{bs}
\end{bmatrix}
$$
where \(r_{bs}\) is the base radius, \(\lambda_{bs}\) is the base helix angle, \(p_s\) is the helix parameter, and \(\theta_s\) and \(\mu_s\) are surface parameters. The unit normal vector is obtained by:
$$ \mathbf n_s=
\frac{
\frac{\partial \mathbf r_s}{\partial \theta_s}
\times
\frac{\partial \mathbf r_s}{\partial \mu_s}
}{
\left|
\frac{\partial \mathbf r_s}{\partial \theta_s}
\times
\frac{\partial \mathbf r_s}{\partial \mu_s}
\right|
}
$$
For conjugate meshing, the relative velocity at the contact point must be perpendicular to the common normal. This gives the meshing equation:
$$ f(\theta_s,\mu_s,\varphi_p)=\mathbf n_t\cdot \mathbf v_t^{(tp)}=0 $$
The generated tooth surface of the spiroid face gear is then obtained by combining the family of cutter surfaces with the meshing equation:
$$ \mathbf r_p(\theta_s,\mu_s)=M_{pt}(\varphi_p)\mathbf r_t(\theta_s,\mu_s),
\quad
f(\theta_s,\mu_s,\varphi_p)=0
$$
I solved these nonlinear equations numerically. The tooth surface was divided into a grid, and the projection relationship between the three-dimensional surface point and the two-dimensional section was established. I used an iterative solver to obtain the point coordinates. In my calculation, the tooth height direction was divided into five intervals, and the tooth width direction was divided into nine intervals, giving forty-five points for each working flank. The transition surface was not discretized in detail because it does not directly participate in the main meshing contact. The resulting point cloud was imported into a three-dimensional modeling environment to build the theoretical model of the spiroid face gear. This model served as the reference for later gear cutting simulation and accuracy comparison.
4. Kinematics of Gear Cutting
For digital simulation, I built a kinematic model of gear cutting. The cutter and workpiece rotate about their own axes, while the cutter also feeds along the radial direction of the face gear. The ratio between the cutter rotation and the workpiece rotation must follow the tooth number relationship:
$$ \frac{\omega_m}{\omega_p}=\frac{Z_2}{Z_1} $$
Here \(\omega_m\) is the cutter angular velocity, \(\omega_p\) is the workpiece angular velocity, \(Z_1\) is the cutter tooth number, and \(Z_2\) is the workpiece tooth number. The feed motion can be decomposed into three linear axes. If \(\Delta L\) is the radial feed distance, \(h\) is the tooth height, and \(\gamma\) is the angle between the cutter spindle and the pinion axis, then the increments are:
$$ \Delta X=\Delta L\sin\gamma,\qquad
\Delta Y=h,\qquad
\Delta Z=\Delta L\cos\gamma $$
The rotational speed relationship for gear cutting can also be written as:
$$ \omega_s=\omega_p\pm \frac{v\sin\beta_m}{mZ} $$
where \(v\) is the radial feed speed, \(\beta_m\) is the nominal helix angle, \(m\) is the module, and \(Z\) is the tooth number. This expression shows that the cutter speed is not arbitrary. It must be synchronized with the workpiece motion and the feed motion. In my simulation, I ensured that the cutter and workpiece maintained strict meshing during gear cutting. The X and Y directions mainly contributed to material removal, while the Z direction mainly produced axial compression. This distinction is important when interpreting cutting force curves.
5. Cutter Structure Design for Gear Cutting
I designed and compared three cutter structures: a bowl-shaped cutter, a petal-shaped cutter, and a modified petal-shaped cutter. The cutter body can be regarded as a helical cylindrical gear, and the cutting edges are formed by the intersection of the rake face and the relief faces. The main geometric features include the top cutting edge, side cutting edge, rake face, top relief face, and side relief face. The rake face determines chip evacuation and cutting angle. The side relief face determines whether the side flank will interfere with the workpiece. In gear cutting of spiroid face gears, side flank contact is a critical factor because the tooth surface is curved in space.
For the bowl-shaped cutter, the rake face is approximately a conical surface. Its cutting edges are distributed on this conical surface. The rake angle and relief angle control the cone geometry. For the petal-shaped cutter, the rake face is formed by a plane with a certain inclination. This structure usually provides more cutting edges and better stability. However, if the tooth profile is not modified, the sharp edge and blunt edge may have different cutting conditions, leading to uneven load and wear. I therefore modified the petal-shaped cutter by adjusting the profile angles and helix parameters. The modified parameters are listed in Table 2.
| Parameter | Value |
|---|---|
| Normal module | 0.85 mm |
| Number of teeth | 21 |
| Normal pressure angle | 20° |
| Helix angle | 30.741° |
| Top relief angle | 5° |
| Axial relief angle | 8.62° |
| Sharp-edge pitch helix angle | 33.941° |
| Blunt-edge pitch helix angle | 27.541° |
| Sharp-edge pitch pressure angle | 23.938° |
| Blunt-edge pitch pressure angle | 22.556° |
| Sharp-edge base diameter | 18.982 mm |
| Blunt-edge base diameter | 19.179 mm |
The modified petal-shaped cutter was designed to make the load on the two side cutting edges more balanced. The modification reduces the difference between the sharp edge and the blunt edge and improves the contact condition during gear cutting. I built three-dimensional models of all cutter structures and used them in the finite element simulation. The cutter models were also used later in numerical machining verification. In this way, I could evaluate how cutter geometry affects cutting force, temperature, wear, interference, and tooth surface accuracy in gear cutting.
6. Finite Element Simulation of Cutting Force in Gear Cutting
I used ABAQUS to simulate the gear cutting process. The simulation model included the cutter and the workpiece. The workpiece material was 20CrMnTi, and the cutter material was S390 high-speed steel. The workpiece was modeled as a deformable body, while the cutter was treated as a rigid body because its hardness and strength are much higher than those of the workpiece. The Johnson-Cook constitutive model was used to describe the workpiece material behavior:
$$ \sigma=(A+B\varepsilon^n)\left(1+C\ln\dot\varepsilon^*\right)\left(1-T^{*m}\right) $$
The fracture criterion was also based on the Johnson-Cook damage model:
$$ \varepsilon_f=\left(d_1+d_2e^{-d_3\eta}\right)\left(1+d_4\ln\dot\varepsilon^*\right)\left(1+d_5T^*\right) $$
The material parameters are summarized in Table 3.
| Model | Parameter | Value |
|---|---|---|
| Johnson-Cook constitutive | A | 303 MPa |
| Johnson-Cook constitutive | B | 192 MPa |
| Johnson-Cook constitutive | n | 0.5 |
| Johnson-Cook constitutive | C | 0.1 |
| Johnson-Cook constitutive | m | 0.9 |
| Johnson-Cook damage | d1 | 1.508 |
| Johnson-Cook damage | d2 | 1.094 |
| Johnson-Cook damage | d3 | -1.4 |
| Johnson-Cook damage | d4 | 0.005 |
| Johnson-Cook damage | d5 | 0.94 |
The mesh was generated with three-dimensional hexahedral elements for the workpiece. The cutter had a coarser mesh because it was rigid. Contact was defined between the cutter and the workpiece using surface-to-surface contact and general contact. Boundary conditions included rotational velocity and translational feed. The total cutting force was calculated from the three components:
$$ F_{\mathrm{total}}=\sqrt{F_x^2+F_y^2+F_z^2} $$
In my simulation, the X direction corresponded to the radial feed direction, the Y direction corresponded to the cutting tangential direction, and the Z direction corresponded to the axial direction of the face gear. The X and Y components mainly contributed to material removal, while the Z component mainly represented axial compression. The cutting force curves showed periodic fluctuations because the cutter and workpiece maintained meshing while the cutter fed radially. These fluctuations are a characteristic feature of gear cutting of spiroid face gears.
6.1 Influence of Feed Speed on Cutting Force
I studied three feed speeds: 0.016 m/min, 0.032 m/min, and 0.047 m/min. As the feed speed increased, the average total cutting force increased from 308.021 N to 498.175 N. The peaks and fluctuations also increased. The reason is that a higher feed speed increases the material volume removed per unit time and increases the instantaneous uncut chip thickness. The radial force followed the same trend. The axial force was negative in most of the process, indicating that the cutter pressed the workpiece. At higher feed speeds, the absolute axial force increased because the cutting edge pressed more strongly into the material. The results are summarized in Table 4.
| Feed speed (m/min) | Total force trend | Observed average total force (N) | Fluctuation |
|---|---|---|---|
| 0.016 | Lowest | 308.021 | Small |
| 0.032 | Medium | Approximately 395 | Moderate |
| 0.047 | Highest | 498.175 | Large |
These results indicate that feed speed must be selected carefully in gear cutting. A higher feed speed improves productivity, but it also increases cutting force and vibration. Excessive force can accelerate tool wear, worsen tooth surface accuracy, and increase the risk of burr formation. Therefore, I concluded that feed speed optimization is necessary for stable gear cutting.
6.2 Influence of Cutting Depth on Cutting Force
Cutting depth had the most significant effect on cutting force. I simulated depths of 0.5 mm, 1.0 mm, and 1.5 mm. The average total cutting force increased from 192.663 N to 665.527 N. The peaks also increased significantly. This is because cutting depth directly changes the cross-sectional area of the removed material. A larger cutting depth means more material is engaged by the cutting edge, which increases the cutting resistance. The radial and axial forces also increased with cutting depth. The results are shown in Table 5.
| Cutting depth (mm) | Total force trend | Observed average total force (N) | Radial force trend | Axial force trend |
|---|---|---|---|---|
| 0.5 | Lowest | 192.663 | Low | Small compression |
| 1.0 | Medium | Approximately 395 | Medium | Medium compression |
| 1.5 | Highest | 665.527 | High | Strong compression |
Because cutting depth strongly affects force, it also affects tool wear, vibration, and tooth surface quality. A large cutting depth may cause excessive load on the cutter, leading to chipping and accelerated wear. It can also cause unstable material removal and local overcut or residual material. Therefore, in gear cutting of spiroid face gears, cutting depth should be controlled to balance efficiency and accuracy.
6.3 Influence of Top Relief Angle on Cutting Force
I compared top relief angles of 9°, 10°, and 11°. The average total cutting forces were 409.068 N, 395.659 N, and 398.098 N, respectively. When the relief angle was too small, friction between the flank face and the workpiece increased, and the cutting force increased. When the relief angle was increased to 10°, the force decreased slightly. However, when the relief angle was too large, the cutting edge strength decreased, and the force fluctuation increased. Table 6 summarizes the results.
| Top relief angle | Average total force (N) | Observed effect |
|---|---|---|
| 9° | 409.068 | Higher friction and larger force |
| 10° | 395.659 | Lower friction and more stable force |
| 11° | 398.098 | Slightly larger fluctuation |
This result shows that the relief angle should not be minimized or maximized without consideration. A suitable relief angle reduces friction and maintains edge strength. This is important for stable gear cutting and for reducing tool wear.
6.4 Influence of Cutter Structure on Cutting Force
I compared the bowl-shaped cutter, petal-shaped cutter, and modified petal-shaped cutter. The average total cutting forces were 367.735 N, 395.659 N, and 507.705 N, respectively. The bowl-shaped cutter produced the lowest average force, but its force peak at the initial engagement was relatively high. The petal-shaped cutter produced a moderate force with relatively smooth fluctuations. The modified petal-shaped cutter produced a higher average force and larger peaks, but it removed material more effectively and had a stronger cutting action. The results are presented in Table 7.
| Cutter structure | Average total force (N) | Force fluctuation | Observed cutting behavior |
|---|---|---|---|
| Bowl-shaped | 367.735 | Moderate | Lower average force, higher initial peak |
| Petal-shaped | 395.659 | Small to moderate | Smoother cutting, more stable |
| Modified petal-shaped | 507.705 | Large | Stronger cutting action, higher local load |
The simulation also showed that the side flank of the cutter participated in the cutting process for all three structures. This means the side flank is not merely a guide surface. It contacts the workpiece and influences the tooth surface formation. In gear cutting, if the side flank contacts the tooth surface improperly, it can cause interference, local extrusion, and uneven material removal. ABAQUS can show the stress distribution and force on the side flank, but it cannot fully describe wear and interference evolution. Therefore, I used DEFORM-3D for supplemental simulation.
7. Temperature Distribution and Tool Wear in Gear Cutting
I used DEFORM-3D to simulate the thermal and wear behavior of gear cutting. Unlike ABAQUS, DEFORM-3D does not allow arbitrary coordinate system creation in the same way. I therefore rebuilt the machining coordinate system and adjusted the model positions so that the cutter rotated about its own central axis. The workpiece blank was modeled as a plastic body, and a lower die was added as a rigid body to constrain the blank and provide rotation. The cutter was treated as a rigid body. The mesh was generated with tetrahedral elements, and automatic remeshing was enabled to handle large plastic deformation. Contact between the cutter and workpiece used Coulomb friction with a coefficient of 0.6. Contact between the workpiece and the lower die used shear friction to ensure proper constraint. The thermal boundary conditions included an ambient temperature of 20°C, a convection coefficient of 0.02 N/(s·mm·°C), radiation emissivity of 0.7, and a contact heat transfer coefficient of 50 W/(m·K).
The heat generation in gear cutting can be described by the coupled energy equation:
$$ \rho c_p \frac{\partial T}{\partial t}
=
\nabla\cdot(k\nabla T)
+
\dot q_{pl}
+
\dot q_{fr} $$
where \(\dot q_{pl}\) is the heat generated by plastic deformation and \(\dot q_{fr}\) is the heat generated by friction. The simulation showed that the high-temperature region was concentrated in the tooth slot and in the contact zone between the cutter and the workpiece. The temperature distribution was non-uniform. The highest temperature occurred near the contact between the cutting edge and the workpiece, where plastic deformation and friction were both intense. The chip carried away a large portion of the heat, but the remaining heat diffused into the cutter and the workpiece. This caused local thermal softening and influenced tool wear.
I used the Usui wear model to predict tool wear:
$$ \dot w=C_1\sigma_n v_s\exp\left(-\frac{C_2}{T}\right) $$
where \(\dot w\) is the wear rate, \(\sigma_n\) is the normal stress, \(v_s\) is the sliding velocity, \(T\) is the temperature, and \(C_1\) and \(C_2\) are model constants. In my simulation, the wear constants were set as \(C_1=1\times10^{-7}\) and \(C_2=855\). The results showed that the bowl-shaped cutter and the unmodified petal-shaped cutter had obvious wear on the flank face. The wear band was concentrated near the cutting edge, which indicated interference or severe contact between the flank face and the workpiece. In contrast, the modified petal-shaped cutter showed a smaller wear region and a narrower wear band. This means the modification reduced interference and improved the contact condition during gear cutting.
The temperature results also showed that the modified petal-shaped cutter produced a higher cutting temperature than the bowl-shaped cutter, approximately 25% higher in some regions. This indicates that the modified cutter had a stronger cutting action and higher material removal efficiency. Although higher temperature can increase tool wear, the reduced interference in the modified cutter compensated for this effect and produced a better wear distribution. Therefore, the modified petal-shaped cutter is more suitable for gear cutting of spiroid face gears.
8. Numerical Machining Verification of Gear Cutting Accuracy
I used VERICUT to verify the gear cutting process. A five-axis CNC machine model was established, and the Siemens 840D control system was selected. The cutter models were exported as STP files and imported into VERICUT. To evaluate interference, the entire cutter body was defined as the cutting surface rather than only the rake face. This allowed all possible contact areas between the cutter and the workpiece to be considered. The NC program was generated according to the kinematic model of gear cutting. The workpiece rotated about its own axis, the cutter rotated about its own axis, and the cutter fed along the radial direction of the face gear. The synchronization between the cutter rotation and workpiece rotation was maintained.
After simulation, I used the automatic comparison module in VERICUT to compare the machined tooth surface with the theoretical tooth surface. The software calculated the geometric deviation and identified overcut and residual regions. The comparison results are summarized in Table 8.
| Cutter structure | Overcut points | Maximum overcut (mm) | Residual points | Maximum residual (mm) |
|---|---|---|---|---|
| Bowl-shaped | 661 | 0.08692913 | 2655 | 0.098290015 |
| Petal-shaped | 455 | 0.03795907 | 1438 | 0.047242282 |
| Modified petal-shaped | 105 | 0.011689839 | 418 | 0.029986124 |
The results show a clear improvement as the cutter structure was optimized. For the bowl-shaped cutter, the overcut and residual errors were the largest. This indicates that the bowl-shaped cutter caused more interference and left more uncut material during gear cutting. For the petal-shaped cutter, the number of overcut points decreased to 455, and the maximum overcut decreased to about 0.038 mm. The residual points decreased to 1438, and the maximum residual decreased to about 0.047 mm. For the modified petal-shaped cutter, the overcut points decreased further to 105, and the maximum overcut decreased to about 0.0117 mm. The residual points decreased to 418, and the maximum residual decreased to about 0.030 mm. These results confirm that the modified petal-shaped cutter provides the best tooth surface accuracy among the three structures.
The improvement can be explained by the contact condition. The modified cutter reduces side flank interference and distributes the cutting load more reasonably. As a result, the material removal process becomes more uniform, and the machined surface is closer to the theoretical surface. This is important for gear cutting of spiroid face gears because the tooth surface is curved and the contact area changes continuously. If the cutter interferes with the workpiece, the generated tooth surface will deviate from the theoretical one. The VERICUT comparison provided a quantitative evaluation of this deviation.
9. Integrated Discussion
My study shows that gear cutting of spiroid face gears is feasible but sensitive to process parameters and cutter structure. The experimental results confirmed that a complete tooth profile can be generated. However, burrs at the tooth tip and local extrusion indicate that the cutting process is not ideal. The finite element simulation explained these phenomena. The cutting force fluctuates periodically, and the side flank participates in cutting. When the cutter geometry is not properly designed, the side flank may contact the workpiece too strongly, causing interference and wear. The temperature simulation showed that the contact zone reaches high temperature, which accelerates wear. The wear simulation further showed that the modified petal-shaped cutter has a smaller wear band. The VERICUT simulation confirmed that the modified cutter produces the smallest overcut and residual errors.
The influence of process parameters can be summarized as follows. Feed speed increases the cutting force and its fluctuation. Cutting depth has the strongest influence on cutting force. Top relief angle affects friction and edge strength. Among these parameters, cutting depth should be controlled most carefully in gear cutting. Feed speed should be selected to balance productivity and stability. The relief angle should be chosen to reduce friction without weakening the cutting edge. Cutter structure is also critical. The bowl-shaped cutter has a lower average force but suffers from more interference and residual material. The petal-shaped cutter provides smoother cutting. The modified petal-shaped cutter gives the best tooth surface accuracy, although it may produce higher local force and temperature. This trade-off should be considered in practical gear cutting.
10. Conclusions
I conducted experimental, theoretical, and digital simulation research on gear cutting of spiroid face gears. The main conclusions are as follows.
First, gear cutting is a feasible method for manufacturing spiroid face gears. The five-axis CNC experiment produced a complete and continuous tooth profile. However, burrs appeared at the tooth tip, indicating that tool geometry and process parameters need further optimization.
Second, a theoretical model of the spiroid face gear was established based on the shaping principle. The tooth surface equations were derived, and a numerical solution was used to obtain the point cloud. The resulting three-dimensional model provided a reliable reference for gear cutting simulation and accuracy evaluation.
Third, ABAQUS simulation showed that feed speed, cutting depth, and cutter geometry all affect cutting force. Cutting depth had the greatest influence. The side flank of the cutter participates in cutting, which means interference analysis must include the side flank. The modified petal-shaped cutter produced a stronger cutting action and higher local load, but it also removed material more effectively.
Fourth, DEFORM-3D simulation showed that high temperature is concentrated in the contact zone and tooth slot. Tool wear mainly occurs near the cutting edge and flank face. The bowl-shaped and unmodified petal-shaped cutters showed severe flank wear and possible interference. The modified petal-shaped cutter showed reduced wear and improved contact conditions.
Fifth, VERICUT verification showed that the modified petal-shaped cutter significantly reduced overcut and residual errors. Compared with the bowl-shaped cutter, the number of overcut points decreased from 661 to 105, and the maximum overcut decreased from about 0.0869 mm to about 0.0117 mm. The number of residual points decreased from 2655 to 418, and the maximum residual decreased from about 0.0983 mm to about 0.0300 mm. This confirms that cutter modification improves tooth surface accuracy in gear cutting.
Overall, my research demonstrates that digital simulation is an effective tool for studying gear cutting of spiroid face gears. By combining theoretical modeling, finite element analysis, thermal and wear simulation, and numerical machining verification, I established a systematic method to evaluate cutting force, temperature, tool wear, interference, and tooth surface accuracy. The modified petal-shaped cutter is recommended for gear cutting of spiroid face gears because it provides better contact conditions and higher tooth surface accuracy.
11. Future Work
Further work can improve the accuracy of cutter modeling. In my study, the modified cutter was built by parametric modeling in a three-dimensional environment. This may introduce small geometric deviations. In future work, the cutter tooth surface can be calculated directly from mathematical equations, and the point cloud can be used to construct the cutter model. This would improve the geometric accuracy of gear cutting simulation. In addition, more experimental validation should be carried out. Different cutter structures can be manufactured and tested on a five-axis CNC machine. The measured tooth surface can be compared with the simulated surface. This would further verify the optimization method and provide more reliable guidance for industrial gear cutting of spiroid face gears.