In the field of aviation transmission systems, the demand for compact, high power-to-weight ratio components has led to the widespread use of small module spiral bevel gears. These bevel gears typically require high precision, often at levels 4 to 5, which is generally one to two grades higher than automotive bevel gears. Traditional manufacturing methods, such as the five-cut method, involve separate finishing of convex and concave surfaces, leading to inconsistent tooth flank clearance and reduced efficiency. To address these limitations, my research focuses on the application of the full process milling method for small module aviation bevel gears. This method enables simultaneous finishing of both convex and concave surfaces, improving surface roughness, ensuring uniform tooth clearance, and enhancing processing efficiency. In this study, I utilize KIMOS software for simulating and optimizing tooth contact patterns, generate adjustment cards, and validate the approach using domestic machine tools. The goal is to replace the five-cut method with the full process milling method, thereby achieving higher precision and efficiency in bevel gear manufacturing.
The full process milling method for bevel gears involves complex calculations to control contact characteristics. Unlike the five-cut method, which processes convex and concave surfaces separately, the full process method requires simultaneous consideration of five key elements on both the drive and coast sides: spiral angle, pressure angle, lengthwise curvature, profile curvature, and lengthwise geodesic torsion. These elements are optimized through a system of nonlinear equations to ensure optimal contact patterns. Specifically, for a bevel gear pair, the mathematical model involves solving for the tooth surface parameters based on the generating gear geometry. The tooth surface of the bevel gear can be represented using vector equations derived from the machine tool settings. For instance, the position vector of a point on the tooth surface is given by:
$$ \mathbf{r}(u, \theta) = \mathbf{r}_0(u) + \mathbf{R}(\theta) \cdot \mathbf{p}(u) $$
where \( u \) is the profile parameter, \( \theta \) is the rotation angle, \( \mathbf{r}_0 \) is the initial position, \( \mathbf{R} \) is the rotation matrix, and \( \mathbf{p} \) is the tool profile vector. The normal vector and curvature parameters are then derived through partial derivatives and differential geometry. To design the adjustment cards for the full process method, I start with the gear pair parameters. The following table summarizes the key parameters of the bevel gear pair used in this study:
| Parameter | Pinion (Small Gear) | Gear (Large Gear) |
|---|---|---|
| Module (mm) | 1.5 | |
| Normal Pressure Angle (°) | 20 | |
| Midpoint Spiral Angle (°) | 35 | |
| Total Tooth Height (mm) | 3 | |
| Hand of Spiral | Right | Left |
| Number of Teeth | 23 | 41 |
| Pitch Cone Angle (°) | 29 | 61 |
| Pitch Circular Tooth Thickness (mm) | 2.75 | 1.96 |
For the full process method, the calculation of adjustment cards involves determining the machine tool settings for both the pinion and gear. The gear’s tooth surface is first generated using given parameters or Gleason SB calculation cards. The pinion’s tooth surface is then derived based on meshing equations and curvature matching. The induced normal curvature at the reference point is crucial for contact analysis. The formula for induced normal curvature is:
$$ \kappa_{12} = \kappa_1 – \kappa_2 + \frac{\sin^2 \psi}{r} $$
where \( \kappa_1 \) and \( \kappa_2 \) are the normal curvatures of the pinion and gear surfaces, \( \psi \) is the angle between principal directions, and \( r \) is the relative curvature radius. By solving the nonlinear equations, I obtain the initial adjustment cards for the pinion, which include parameters such as radial distance, horizontal offset, vertical offset, and machine root angle. These parameters are optimized through iterative simulation to achieve the desired contact pattern.
To analyze the tooth contact pattern, I employ KIMOS software for simulation. The software allows for the evaluation of gear geometry, including tooth shape, backlash, and ease-off topography. For bevel gears, tooth shape analysis is essential to prevent undercutting and ensure proper tooth taper. The normal chordal tooth thickness at the tooth top is calculated along the tooth line to assess the tooth contraction. Additionally, the backlash variation during meshing is analyzed to ensure uniform clearance. The backlash as a function of gear rotation angle can be expressed as:
$$ B(\phi) = B_0 + \Delta B \cdot \sin(\phi + \delta) $$
where \( B_0 \) is the nominal backlash, \( \Delta B \) is the variation amplitude, \( \phi \) is the rotation angle, and \( \delta \) is the phase shift. In this study, the backlash variation is minimal, ranging from 0.1279 mm to 0.1294 mm, indicating good meshing characteristics.
One critical aspect of bevel gear design is the ease-off topography, which represents the deviation between the conjugate surface and the actual manufactured surface. This topography directly influences the contact pattern and transmission error. The ease-off value at any point on the tooth surface is given by:
$$ E(x, y) = z_{\text{conj}}(x, y) – z_{\text{act}}(x, y) $$
where \( z_{\text{conj}} \) is the conjugate surface height, and \( z_{\text{act}} \) is the actual surface height. By adjusting the machine tool settings, I optimize the ease-off to achieve a point contact pattern that reduces sensitivity to assembly errors. The contact ellipse dimensions are derived from the principal curvatures and the elastic deformation. The length of the contact ellipse’s major axis is calculated as:
$$ a = \sqrt{\frac{8 w R}{\pi E’ \kappa}} $$
where \( w \) is the load per unit length, \( R \) is the effective radius, \( E’ \) is the equivalent elastic modulus, and \( \kappa \) is the induced curvature. For the bevel gear pair in this study, the initial contact pattern adjustments are summarized in the table below:
| Surface Side | Adjustment Parameter | Before Adjustment | After Adjustment |
|---|---|---|---|
| Drive Side (Concave) | Spiral Angle Correction (°) | -0.0083 | -0.0100 |
| Pressure Angle Correction (°) | -0.1130 | 0.0000 | |
| Lengthwise Curvature Correction (μm) | 135.5785 | 137.5755 | |
| Profile Curvature Correction (μm) | 15.1747 | 15.1747 | |
| Diagonal Correction (°) | -1.7907 | -1.3000 | |
| Coast Side (Convex) | Spiral Angle Correction (°) | 0.0843 | 0.0100 |
| Pressure Angle Correction (°) | -0.0696 | -0.0600 | |
| Lengthwise Curvature Correction (μm) | 315.8355 | 315.8355 | |
| Profile Curvature Correction (μm) | 33.4225 | 33.4225 | |
| Diagonal Correction (°) | 3.6677 | 3.6677 |
After optimization, the contact pattern is centered on the tooth flank, and the transmission error is within acceptable limits. The transmission error function, which represents the deviation from ideal motion transfer, is approximated as a parabolic curve:
$$ TE(\phi) = C_0 + C_1 \phi + C_2 \phi^2 $$
where \( C_0 \), \( C_1 \), and \( C_2 \) are coefficients determined from the meshing simulation. For this bevel gear pair, the transmission error is minimal, ensuring smooth operation under light loads.
For experimental validation, I conducted cutting tests on a domestic CNC spiral bevel gear milling machine, specifically the YKF2260 model. The accuracy of tool grinding and setup is critical for achieving precise tooth profiles. The radial errors for the cutter blades are summarized as follows:
| Cutter Type | Radial Error (μm) |
|---|---|
| Gear Outer Blade | 2.2 |
| Gear Inner Blade | 2.0 |
| Pinion Outer Blade | 1.2 |
| Pinion Inner Blade | 1.8 |
The gear cutting process involves several steps. First, the gear is processed using a single indexing method from the toe to the heel. The initial cutting parameters include a cutter speed of 60 m/min, a feed rate corresponding to a cutter rotation of 280.60 rpm, and a swing angle of -35.6121°. The cutting sequence is divided into four positions with a cutting depth of 0.05 mm and a generating speed of 1.403°/s. The generating angles for these positions are 36.4176°, 45.0000°, 55.0000°, and 66.7045°. To ensure accuracy, the fixture alignment is maintained within 0.002 mm, and the workpiece is machined without additional benchmarking. After implementing thermal stabilization and variable-tooth processing, the gear accuracy improves to DIN4 level. The total processing time for the gear is 33.68 minutes. Following cutting, a reverse adjustment is performed based on measurement feedback. The reverse adjustment parameters are:
| Adjustment Parameter | Value |
|---|---|
| Radial Distance (mm) | 0.1478 |
| Horizontal Offset (mm) | -0.9375 |
| Vertical Offset (mm) | 0.9375 |
| Machine Center Distance (mm) | 0.7629 |
| Workpiece Mounting Angle (°) | 0.1221 |
| Ratio of Roll | 0.0057 |
After reverse adjustment, the gear tooth profile error is reduced from 0.06 mm to less than 0.01 mm, confirming the effectiveness of the compensation process.
The pinion is processed using similar parameters. The initial cutting involves a cutter speed of 60 m/min, a feed rate of 280.60 rpm, and a swing angle of -36.4176°. The generating angles for the four positions are 35.6121°, -40.0000°, -60.0000°, and -72.8051°. With thermal stabilization and variable-tooth processing, the pinion accuracy reaches DIN6 level, and the processing time is 18.80 minutes. The cumulative error and tooth morphology measurements meet the required standards.
To evaluate the meshing performance, I conducted contact pattern tests under standard mounting conditions. For the gear concave side, the setup includes a lateral clearance of 0.2 mm, a gear horizontal displacement (G) of 0.75 mm, a pinion horizontal displacement (H) of 0 mm, and a relative vertical displacement (V) of 0 mm. For the gear convex side, G is 0.74 mm, with H and V at 0 mm. The actual contact patterns align closely with the simulated patterns, demonstrating the accuracy of the full process milling method. The contact patterns are centered and exhibit minimal drift, indicating good load distribution and low transmission error.
The success of this study highlights the potential of the full process milling method for small module aviation bevel gears. Compared to the traditional five-cut method, the full process method reduces processing time by approximately 50% while maintaining or improving precision. This is particularly beneficial for aviation applications, where bevel gears must withstand high loads and operate with minimal noise and vibration. The ability to simulate and optimize contact patterns using KIMOS software enables precise control over tooth geometry, reducing the need for trial-and-error adjustments. Moreover, the use of domestic machine tools demonstrates the feasibility of implementing this method in local manufacturing environments.
In conclusion, my research demonstrates that the full process milling method is a viable alternative to the five-cut method for small module aviation bevel gears. Through simulation-based optimization and careful process control, I achieved high-precision bevel gears with improved efficiency. Future work could explore the application of this method to other types of bevel gears, such as hypoid or zerol bevel gears, and investigate the effects of different lubrication conditions on contact patterns. Overall, this study contributes to the advancement of bevel gear manufacturing technology, supporting the development of more efficient and reliable aviation transmission systems.