In my research on the ZC1 worm gear pair used in escalator drives, I focus on the meshing performance characterized by the instantaneous contact lines on the worm wheel tooth surface. The ZC1 worm gear drive, also known as an arc-profile cylindrical worm gear, offers significant advantages over conventional worm gear designs, including better load capacity, smoother operation, and superior lubrication due to convex-concave tooth profiles. However, to fully exploit these benefits, it is essential to understand how design parameters influence the contact conditions. In this paper, I develop a mathematical model for the double-enveloping ZC1 worm gear pair based on the theory of spatial gearing. I derive the instantaneous contact line equations for the worm wheel tooth surface, visualize these lines using MATLAB numerical simulations, and systematically analyze the effects of key parameters such as center distance a, grinding wheel arc radius ρ, and lead angle γ on the distribution of contact lines. My results indicate that improper selection of these parameters can lead to undesirable contact line patterns—such as intersections or shortened total length—that degrade lubrication and heat dissipation, ultimately affecting the worm gear’s service life and performance. This analysis provides a foundation for future multi-objective optimization of ZC1 worm gear drives.
The ZC1 worm gear is widely adopted in escalator transmission systems due to its high reduction ratio, compact structure, and self-locking capability. Unlike conventional Archimedes or involute worm gears, the ZC1 type employs a circular arc tooth profile in the axial section of the worm, which is generated by a grinding wheel with a circular arc profile. This design promotes a concave-convex meshing pattern that enhances the formation of an oil film between the worm and worm wheel teeth, leading to improved lubrication and reduced wear. Nevertheless, the meshing performance of the ZC1 worm gear is highly sensitive to geometric parameters. The instantaneous contact line—the locus of points where the worm and worm wheel surfaces are in contact at a given instant—directly influences the load distribution, heat generation, and lubrication conditions. Poorly distributed contact lines, such as those crossing each other or concentrated in a narrow region, can cause localized overheating, oil film rupture, and accelerated pitting. Therefore, a detailed parametric study of these contact lines is necessary for optimizing worm gear design.
In the following sections, I first establish the mathematical framework for the single-enveloping process that generates the worm tooth surface from the grinding wheel. Then, I develop the double-enveloping model to obtain the worm wheel tooth surface and its instantaneous contact lines. Using numerical methods, I evaluate the influence of three critical design parameters—center distance a, grinding wheel arc radius ρ, and lead angle γ—on the contact line patterns. Finally, I summarize the findings and discuss their implications for worm gear optimization.
Mathematical Modeling of ZC1 Worm Gear Pair
Coordinate Systems for Single-Enveloping Process
To describe the generation of the worm tooth surface by a grinding wheel, I define the coordinate systems shown in Figure 1 (conceptually). The stationary coordinate system S(O, i, j, k) is fixed in space, with the worm axis coincident with the k-axis. The moving coordinate system S₁(O₁, i₁, j₁, k₁) is attached to the worm and rotates with it. The worm performs a helical motion with a spiral parameter p. The grinding wheel coordinate system Sσ(Oσ, iσ, jσ, kσ) is fixed to the grinding wheel, whose axis is oriented at an angle γ (lead angle) with respect to the worm axis. The distance between the origins O₁ and Oσ is the center distance Aσ.
The transformation matrix from Sσ to S is:
$$
M_{0σ} = \begin{bmatrix}
1 & 0 & 0 & A_σ \\
0 & \cosγ & -\sinγ & 0 \\
0 & \sinγ & \cosγ & 0 \\
0 & 0 & 0 & 1
\end{bmatrix}
$$
The transformation from Sσ to S₁ is given by:
$$
M_{1σ} = \begin{bmatrix}
\cosδ & \sinδ\cosγ & -\sinδ\sinγ & A_σ\cosδ \\
-\sinδ & \cosδ\cosγ & -\cosδ\sinγ & -A_σ\sinδ \\
0 & \sinγ & \cosγ & -pδ \\
0 & 0 & 0 & 1
\end{bmatrix}
$$
where δ is the rotation angle of the worm and p = H/(2π) is the helical parameter with H being the lead of the worm.
Grinding Wheel Profile and Surface Equation
The axial section of the grinding wheel is a circular arc with radius ρ. The center of this arc is located at a distance b from the worm axis and c from the grinding wheel axis, as shown in Figure 2 (conceptually). The parameters satisfy:
$$
c = ρ \cosα,\quad b = r_1 + ρ \sinα,\quad A_σ = b + d
$$
where α is the pressure angle, r₁ is the worm pitch circle radius, and d is the distance from the arc center to the grinding wheel axis. In the local coordinate system Sσ’ attached to the grinding wheel profile, the generating line is expressed as:
$$
r^{(σ’)} = \begin{bmatrix} x_{σ’} \\ y_{σ’} \\ z_{σ’} \end{bmatrix} = \begin{bmatrix} -ρ\sinθ \\ 0 \\ ρ\cosθ \end{bmatrix}
$$
where θ is the angular parameter along the arc. By introducing a rotation parameter β around the grinding wheel axis, the surface of the grinding wheel in Sσ becomes:
$$
r^{(σ)} = \begin{bmatrix} x_σ \\ y_σ \\ z_σ \end{bmatrix} = \begin{bmatrix} -ρ\sinθ\cosβ – d\cosβ \\ ρ\sinθ\sinβ + d\sinβ \\ ρ\cosθ – c \end{bmatrix}
$$
The unit normal vector at any point on the grinding wheel surface is derived from the first and second fundamental forms:
$$
n^{(σ)} = \begin{bmatrix} n_x^{(σ)} \\ n_y^{(σ)} \\ n_z^{(σ)} \end{bmatrix} = \begin{bmatrix} \sinθ\cosβ \\ -\sinθ\sinβ \\ -\cosθ \end{bmatrix}
$$
Meshing Condition for Single Enveloping
The relative velocity between the grinding wheel and the worm during the generating process is computed based on their motions. With ω₁ = ωσ = 1 (a normalization that simplifies equations without loss of generality), the relative velocity vector in Sσ is:
$$
v_{σ1} = \begin{bmatrix} v_{σ1x} \\ v_{σ1y} \\ v_{σ1z} \end{bmatrix} = \begin{bmatrix} (\rho\sinθ\sinβ + d\sinβ)\cosγ – (\rho\cosθ – c)\sinγ \\ – (-\rho\sinθ\cosβ – d\cosβ + A_σ)\cosγ – p\sinγ \\ (-\rho\sinθ\cosβ – d\cosβ + A_σ)\sinγ – p\cosγ \end{bmatrix}
$$
The meshing condition for conjugate surfaces requires that the relative velocity be perpendicular to the common normal:
$$
φ_{σ1} = n^{(σ)} \cdot v_{σ1} = 0
$$
Substituting the expressions yields the simplified equation:
$$
\tanθ – \frac{A_σ – d\cosβ – p\cotγ}{c\cosβ + A_σ\sinβ\cotγ + p\sinβ} = 0
$$
This equation defines the instantaneous contact line on the grinding wheel surface. By transforming these points back to the worm coordinate system using M₁σ, I obtain the worm tooth surface equation:
$$
\begin{aligned}
x_1 &= x_σ\cosδ + y_σ\sinδ\cosγ – z_σ\sinδ\sinγ + A_σ\cosδ \\
y_1 &= -x_σ\sinδ + y_σ\cosδ\cosγ – z_σ\cosδ\sinγ – A_σ\sinδ \\
z_1 &= y_σ\sinγ + z_σ\cosγ – pδ
\end{aligned}
$$
together with the meshing condition φσ1 = 0. These equations represent the worm tooth surface generated by the single-enveloping process.
Double-Enveloping Model for Worm Wheel
In the second enveloping process, the worm tooth surface obtained above serves as the cutting tool to generate the worm wheel. Figure 3 (conceptually) illustrates the coordinate systems for worm–wheel meshing. The fixed coordinate system Sg(Og, ig, jg, kg) has its kg-axis aligned with the worm axis. The worm rotates by φ₁ about its axis, while the worm wheel rotates by φ₂ about its axis, which is perpendicular to the worm axis. The center distance between the worm and worm wheel is a. The transmission ratio is i₁₂ = z₂/z₁ = ω₁/ω₂, where z₁ and z₂ are the numbers of starts and teeth, respectively.
The transformation from the worm coordinate system S₁ to the worm wheel coordinate system S₂ is:
$$
M_{21} = \begin{bmatrix}
\cosφ_1\cosφ_2 & -\sinφ_1\cosφ_2 & -\sinφ_2 & a\cosφ_2 \\
-\cosφ_1\sinφ_2 & \sinφ_1\sinφ_2 & -\cosφ_2 & -a\sinφ_2 \\
\sinφ_1 & \cosφ_1 & 0 & 0 \\
0 & 0 & 0 & 1
\end{bmatrix}
$$
The unit normal vector of the worm tooth surface in S₁ is obtained by transforming the grinding wheel normal using L₁σ:
$$
n^{(1)} = \begin{bmatrix} n_x^{(1)} \\ n_y^{(1)} \\ n_z^{(1)} \end{bmatrix} = \begin{bmatrix}
\sinθ(\cosβ\cosδ – \sinβ\sinδ\cosγ) + \cosθ\sinδ\sinγ \\
-\sinθ(\cosβ\sinδ + \sinβ\cosδ\cosγ) + \cosθ\cosδ\sinγ \\
-\sinθ\sinβ\sinγ – \cosθ\cosγ
\end{bmatrix}
$$
The relative velocity between worm and worm wheel in S₁ is:
$$
v_{12} = \begin{bmatrix} v_{12x} \\ v_{12y} \\ v_{12z} \end{bmatrix} = \begin{bmatrix}
-y_1 – i_{21}z_1\cosφ_1 \\
x_1 + i_{21}z_1\sinφ_1 \\
i_{21}(x_1\cosφ_1 – y_1\sinφ_1 + a)
\end{bmatrix}
$$
The double-enveloping meshing condition is:
$$
φ_{12} = n^{(1)} \cdot v_{12} = 0
$$
Expanding this gives:
$$
φ_{12} = W_1\cosφ_1 – W_2\sinφ_1 – W_3 = 0
$$
where:
$$
\begin{aligned}
W_1 &= i_{21}(x_1 n_z^{(1)} – z_1 n_x^{(1)}) \\
W_2 &= -i_{21}(z_1 n_y^{(1)} – y_1 n_z^{(1)}) \\
W_3 &= -(i_{21}a n_z^{(1)} – y_1 n_x^{(1)} + x_1 n_y^{(1)})
\end{aligned}
$$
The instantaneous contact lines on the worm wheel tooth surface are obtained by simultaneously solving the worm tooth surface equations, the meshing condition φ₁₂ = 0, and the coordinate transformation M₂₁. The complete set of equations is:
$$
\begin{cases}
φ_{12} = W_1\cosφ_1 – W_2\sinφ_1 – W_3 = 0 \\
x_1 = x_σ\cosδ + y_σ\sinδ\cosγ – z_σ\sinδ\sinγ + A_σ\cosδ \\
y_1 = -x_σ\sinδ + y_σ\cosδ\cosγ – z_σ\cosδ\sinγ – A_σ\sinδ \\
z_1 = y_σ\sinγ + z_σ\cosγ – pδ \\
n^{(1)} = (n_x^{(1)}, n_y^{(1)}, n_z^{(1)})^T \\
r^{(σ)} = (x_σ, y_σ, z_σ)^T \\
φ_{σ1} = 0 \\
x_2 = x_1\cosφ_1\cosφ_2 – y_1\sinφ_1\cosφ_2 – z_1\sinφ_2 + a\cosφ_2 \\
y_2 = -x_1\cosφ_1\sinφ_2 + y_1\sinφ_1\sinφ_2 – z_1\cosφ_2 – a\sinφ_2 \\
z_2 = x_1\sinφ_1 + y_1\cosφ_1
\end{cases}
$$
These equations form the foundation for numerically computing the instantaneous contact lines on the worm wheel tooth surface.
Numerical Simulation and Parametric Analysis
Using the derived equations, I implemented a numerical program in MATLAB to compute and plot the instantaneous contact lines on the worm wheel tooth surface for various design parameters. The baseline parameters for the ZC1 worm gear are: center distance a = 180 mm, axial module m = 9.5 mm, diameter coefficient q = 7.684, pressure angle α = 23°, grinding wheel arc radius ρ = 55 mm, worm starts z₁ = 5, worm wheel teeth z₂ = 29, and lead angle γ = 33° (derived from the helix geometry). I systematically varied each parameter while keeping others constant to observe the effects on contact line distribution.
The quality of contact line distribution is assessed based on two criteria: (1) the lines should be evenly spread across the tooth surface to promote heat dissipation and avoid localized overheating; (2) intersections of contact lines should be minimized because crossing lines double the meshing cycles at those points, leading to accelerated pitting and poor lubrication. The total length of contact lines also matters—shorter lines reduce the load-carrying area and worsen lubrication conditions.
Effect of Center Distance a
I varied the center distance from 160 mm to 190 mm in steps of 10 mm while keeping ρ = 55 mm and γ = 33°. The resulting contact line patterns are summarized in Table 1.
| Center distance a (mm) | Contact line distribution | Intersections observed | Total line length |
|---|---|---|---|
| 160 | Sparse, concentrated near tooth root | Minor crossings near tip | Short |
| 170 | Moderately spread | Some crossings | Moderate |
| 180 | Uniform across tooth surface | None | Long |
| 190 | Well spread but slightly shifted | None | Long |
Figure 4 (conceptually) shows that as the center distance decreases, the contact lines become sparser and shorter, with a tendency to intersect near the tooth tip. For a = 180 mm and 190 mm, the lines are uniformly distributed with no intersections, indicating better lubrication and heat transfer. Therefore, a larger center distance is generally favorable for meshing performance, although practical constraints such as overall gearbox size must be considered.
Effect of Grinding Wheel Arc Radius ρ
I tested four values of ρ: 50, 55, 60, and 65 mm, with a = 180 mm and γ = 33°. The results appear in Table 2.
| ρ (mm) | Contact line distribution | Intersections observed | Total line length |
|---|---|---|---|
| 50 | Non-uniform, dense near tooth root | Yes, near tip | Moderate |
| 55 | Uniform across tooth surface | None | Long |
| 60 | Slight crowding near tip | Yes, near tip | Moderate |
| 65 | Non-uniform, more crossings | Yes, multiple | Short |
From Figure 5 (conceptually), the optimal value is ρ = 55 mm, which yields evenly distributed contact lines without intersections. Both smaller and larger radii cause crossings near the tooth tip, which create localized high-temperature zones and accelerate oil film breakdown. The total contact line length is also maximized at ρ = 55 mm, improving load capacity and lubrication.
Effect of Lead Angle γ
I examined lead angles of 28°, 33°, 38°, and 40° while fixing a = 180 mm and ρ = 55 mm. The findings are summarized in Table 3.
| Lead angle γ (°) | Contact line distribution | Intersections observed | Total line length |
|---|---|---|---|
| 28 | Uniform but shortened | None | Short |
| 33 | Uniform, well-spread | None | Long |
| 38 | Dense near tip | Yes, near tip | Moderate |
| 40 | Crowded, many crossings | Yes, multiple | Short |
Figure 6 (conceptually) reveals that γ = 33° produces the best distribution: contact lines are evenly spaced across the entire tooth surface with no intersections. For γ = 28°, although no crossings occur, the total line length is reduced, which diminishes the effective load-bearing area and worsens lubrication. For γ = 38° and 40°, crossings appear near the tooth tip, leading to poor heat dissipation and accelerated wear.
Discussion and Optimization Implications
The parametric analysis clearly demonstrates that the instantaneous contact line pattern on the worm wheel tooth surface is highly sensitive to the design parameters of the ZC1 worm gear. The center distance a, grinding wheel arc radius ρ, and lead angle γ all have significant influence. Inappropriate values lead to contact line intersections or reduced total length, both detrimental to meshing performance. Intersections cause double meshing at certain points, generating more heat and stress, while shortened contact lines concentrate load over a smaller area, increasing contact pressure and wear.
From a design optimization perspective, these three parameters should be treated as decision variables in a multi-objective optimization framework. The objectives could include maximizing the total contact line length, minimizing intersections, and ensuring uniform distribution. Constraints would include strength, manufacturing feasibility, and geometric limitations. The mathematical model developed in this study provides the necessary analytical expressions to compute contact line coordinates efficiently.
Moreover, the lubrication performance of a worm gear is closely related to the contact line pattern. A well-distributed set of contact lines facilitates the formation of a continuous oil film and prevents localized overheating. In escalator applications, where the transmission operates under varying loads and frequent start-stop cycles, maintaining a stable oil film is crucial for durability. Therefore, selecting parameters that yield long, non-intersecting contact lines is essential for reliable operation.
The findings also have practical implications for manufacturing. The grinding wheel radius ρ directly relates to the dressing tool profile; thus, choosing the optimal ρ reduces the need for specialized tooling adjustments. Similarly, the lead angle γ affects the helix geometry and the ease of hobbing or grinding the worm. By incorporating the contact line analysis early in the design phase, engineers can avoid iterative prototyping and achieve higher first-pass success.
Future work could extend this study to include the induced normal curvature and lubrication angle, which further quantify the meshing quality. Additionally, dynamic load analysis and thermal simulation could be coupled with the contact line distribution to predict the actual service life of the worm gear under operating conditions. The ultimate goal is to develop a comprehensive design methodology for ZC1 worm gears that balances performance, manufacturability, and cost.
Conclusions
In this paper, I have presented a comprehensive analysis of the meshing performance of the ZC1 worm gear pair used in escalator drives, focusing on the instantaneous contact lines on the worm wheel tooth surface. Based on the spatial meshing theory, I established the mathematical models for both the single-enveloping (worm generation) and double-enveloping (worm wheel generation) processes. The key equations for the worm tooth surface and the worm wheel contact lines were derived and implemented in MATLAB for numerical simulation.
Through systematic parametric studies, I found that:
(1) The center distance a significantly affects the distribution and length of contact lines. Larger values (e.g., 180–190 mm) produce uniform, long contact lines without intersections, enhancing heat dissipation and lubrication. Smaller values lead to sparse or intersecting lines.
(2) The grinding wheel arc radius ρ has an optimal value around 55 mm for the baseline case. Both smaller (50 mm) and larger (60–65 mm) radii cause contact line intersections near the tooth tip, degrading performance.
(3) The lead angle γ must be carefully selected. For the studied worm gear, γ = 33° yields the best contact line pattern. Lower values (28°) shorten the lines, while higher values (38°–40°) introduce crossings.
These results confirm that the instantaneous contact line is a critical indicator of worm gear meshing quality. The three parameters—center distance a, grinding wheel radius ρ, and lead angle γ—should be treated as primary design variables in any multi-objective optimization aimed at improving the lubrication, load capacity, and lifespan of ZC1 worm gears for escalator applications. The mathematical framework and numerical methodology developed here provide a solid foundation for such optimization efforts.