The pursuit of compact, high-load-capacity power transmission solutions in demanding fields such as aerospace and maritime engineering has driven the development of novel gear drives. Among these, the internal gear enveloping crown worm gear drive presents a significant advancement. This configuration, consisting of a crown worm generated by an enveloping process using a planar internal gear as the master, inherits favorable characteristics from traditional worm drives—like multi-tooth contact, high load capacity, and excellent lubrication potential—while offering a more compact structure due to its internal meshing form. However, the theoretical performance of this worm gear drive, often analyzed under ideal, error-free conditions, does not accurately reflect its real-world operational behavior. Manufacturing inaccuracies and assembly errors are inevitable, potentially leading to deviations in tooth contact patterns, which can severely impact transmission performance, causing issues like uneven load distribution, increased noise, and even jamming. Therefore, a comprehensive analysis of how various errors affect the real meshing state is crucial for understanding its true capabilities, guiding manufacturing tolerance specifications, and improving the robustness of the design. This work delves into the error analysis of a planar internal gear single-enveloping crown worm gear drive, establishing its meshing geometry under the influence of errors, constructing precise three-dimensional models, and systematically investigating the impact of key error sources on the actual contact conditions between the worm and the gear.
The foundational step for any error analysis in gearing is the establishment of a rigorous mathematical model that incorporates potential deviations from the ideal geometry. For the planar internal gear enveloping crown worm gear drive, we begin by defining a set of coordinate systems that account for the error sources. A fixed coordinate system is used as a global reference. The crown worm is rigidly connected to a moving coordinate frame that rotates about its axis with an angular velocity $\omega_1$. Similarly, the planar internal gear is attached to its own moving frame, rotating with angular velocity $\omega_2$. The transmission ratio is defined as $i_{12} = \omega_1 / \omega_2 = Z_2 / Z_1$, where $Z_1$ is the number of worm threads and $Z_2$ is the number of gear teeth. The generating plane, which is the tooth surface of the internal gear, is defined within another moving frame. The key design parameters include the center distance $a$, the shaft crossing angle $\delta$, the inclination angle of the generating plane $\beta$, and the base circle radius $r_b$. The corresponding error sources are introduced as $\Delta a$ (center distance error), $\Delta \delta$ (shaft angle error), $\Delta \beta$ (inclination angle error), $\Delta r_b$ (base circle radius error), and $\Delta l$ (worm axial displacement error).
The mathematical derivation starts with the equation of the generating plane (the internal gear tooth surface) in its coordinate system. It can be represented parametrically using parameters $u$ and $v$:
$$ \mathbf{r}_2(u, v) = (r_b + \Delta r_b – v \sin(\beta + \Delta\beta)) \mathbf{i}_p – u \mathbf{j}_p + v \cos(\beta + \Delta\beta) \mathbf{k}_p $$
The unit normal vector to this plane is constant: $\mathbf{n}_2 = \mathbf{k}_p$. The core of the enveloping process is the relative velocity between the tool (gear) and the generated workpiece (worm). The relative velocity vector $\mathbf{v}_p^{(12)}$ at a point on the generating plane is derived based on the kinematic relationship between the moving coordinate systems, incorporating all relevant error terms $\Delta a, \Delta \delta, \Delta \beta, \Delta l$. According to the theory of gearing, the necessary condition for the generating plane to envelope a conjugate worm surface is given by the meshing equation: $\Phi(u, v, \varphi_2) = \mathbf{n}_2 \cdot \mathbf{v}_p^{(12)} = 0$, where $\varphi_2$ is the rotation angle of the internal gear. This equation establishes a functional relationship between the surface parameters and the motion parameter. Solving this meshing equation yields the condition that must be satisfied for a point on the generating plane to be a contact point at a given instant. The resulting meshing function $\Phi$ is:
$$ \Phi = M_1 \sin \varphi_2 + M_2 \cos \varphi_2 + M_3 = 0 $$
$$ \text{where: } M_1 = -v \cos(\delta + \Delta\delta) + (r_b + \Delta r_b)\cos(\delta + \Delta\delta)\sin(\beta + \Delta\beta) $$
$$ M_2 = (a + \Delta a)\sin(\delta + \Delta\delta)\cos(\beta + \Delta\beta) + u \cos(\delta + \Delta\delta)\sin(\beta + \Delta\beta) $$
$$ M_3 = -u \sin(\delta + \Delta\delta)\cos(\beta + \Delta\beta) – i_{12} u \cos(\beta + \Delta\beta) – (a + \Delta a)\cos(\delta + \Delta\delta)\sin(\beta + \Delta\beta) $$
The surface of the crown worm in its own coordinate system is then obtained by applying the coordinate transformation from the generating plane system to the worm system, subject to the constraint $\Phi=0$:
$$ \mathbf{r}_1(u, v, \varphi_2) = x_1 \mathbf{i}_1 + y_1 \mathbf{j}_1 + z_1 \mathbf{k}_1 $$
$$ \begin{aligned}
x_1 = & -x_2(\sin(\delta+\Delta\delta)\sin\varphi_1\cos\varphi_2 + \cos\varphi_1\sin\varphi_2) \\
& +y_2(\cos\varphi_1\cos\varphi_2 – \sin(\delta+\Delta\delta)\sin\varphi_1\sin\varphi_2) \\
& +z_2\sin(\delta+\Delta\delta)\sin\varphi_1 + (a+\Delta a)\cos\varphi_1 \\
y_1 = & x_2(-\sin(\delta+\Delta\delta)\cos\varphi_1\cos\varphi_2 + \sin\varphi_1\sin\varphi_2) \\
& -y_2(\sin\varphi_1\cos\varphi_2 + \sin(\delta+\Delta\delta)\cos\varphi_1\sin\varphi_2) \\
& +z_2\sin(\delta+\Delta\delta)\cos\varphi_1 – (a+\Delta a)\sin\varphi_1 \\
z_1 = & x_2\cos(\delta+\Delta\delta)\cos\varphi_2 + y_2\cos(\delta+\Delta\delta)\sin\varphi_2 \\
& + z_2\cos(\delta+\Delta\delta) – \Delta l \\
\end{aligned} $$
$$ \text{Subject to: } \Phi(u, v, \varphi_2) = 0 $$
Here, $\varphi_1 = i_{12} \varphi_2$ is the rotation angle of the worm. This set of equations, $\mathbf{r}_1(u, v, \varphi_2)$ with $\Phi=0$, defines the crown worm tooth surface locus in the presence of specified errors. The instantaneous contact line on the generating plane for a fixed $\varphi_2$ is simply the set of points $(u, v)$ that satisfy the meshing equation $\Phi=0$, which typically describes a straight line on that plane.
Before analyzing errors, it is essential to establish the baseline performance of the ideal worm gear drive. Using the derived equations with all error terms $(\Delta a, \Delta \delta, \Delta \beta, \Delta r_b, \Delta l)$ set to zero, the theoretical tooth surface of the crown worm is generated. For this study, a specific set of geometric parameters is chosen, as summarized in Table 1.
| Parameter | Symbol | Value |
|---|---|---|
| Center Distance | $a$ | 100 mm |
| Transmission Ratio | $i_{12}$ | 63 |
| Number of Worm Threads | $Z_1$ | 1 |
| Number of Gear Teeth | $Z_2$ | 63 |
| Shaft Crossing Angle | $\delta$ | 25° |
| Inclination Angle | $\beta$ | 28° |
| Base Circle Radius | $r_b$ | 45 mm |
A highly accurate 3D solid model of the worm is constructed using a point-mapping technique. For discrete values of the gear rotation angle $\varphi_2$, the corresponding parameters $(u, v)$ satisfying $\Phi=0$ are calculated to define points on the instantaneous contact lines. These lines are then swept along the worm’s kinematic motion path to generate the worm tooth surface. The accuracy of this digital model is critical for subsequent contact analysis. The deviation between the mathematical surface and the CAD model is verified by checking distances from calculated theoretical points to the modeled surface. As shown in Table 2, the maximum deviation is on the order of $4.5 \times 10^{-3}$ mm, indicating sufficient accuracy for contact simulation studies.
| Checking Location (Helix) | Average Deviation (×10³ mm) | Maximum Deviation (×10³ mm) |
|---|---|---|
| Root Arc | 1.4 | 2.0 |
| Pitch Arc | 1.9 | 2.9 |
| Tip Arc | 2.7 | 4.5 |
Under ideal conditions, the contact pattern of this worm gear drive exhibits highly favorable characteristics. The instantaneous contact lines on the worm tooth surface are straight lines that are parallel to each other along the tooth height direction. When assembled with its conjugate internal gear, the drive achieves simultaneous contact across multiple tooth pairs. For the parameters in Table 1, at least five pairs of teeth are in contact at any given time. This multi-tooth engagement is a key feature that contributes to the high load-carrying capacity of this type of worm gear drive. The contact lines are evenly distributed across the active tooth flanks, suggesting a theoretically uniform load distribution.
The real-world performance deviates from this ideal due to various errors. Each error source perturbs the meshing condition defined by $\Phi=0$ and alters the generated worm surface geometry $\mathbf{r}_1$, leading to changes in the contact pattern. We analyze the effects of individual errors by introducing them one at a time into the mathematical model, generating the corresponding “as-manufactured” worm and internal gear models, and simulating their assembly and contact.
1. Center Distance Error ($\Delta a$): This error changes the nominal distance between the worm and gear axes. A positive error ($\Delta a = +0.5$ mm) increases the center distance, while a negative error ($\Delta a = -0.5$ mm) decreases it. The contact analysis reveals asymmetric effects. With a positive error, contact occurs primarily near the exit region of the mesh. Only about two tooth pairs are in simultaneous contact, and the second pair only contacts over approximately the upper two-thirds of the worm tooth flank. This reduction in contact area and shift in load zone can lead to stress concentration. With a negative error, contact shifts towards the entry region, with up to four pairs in contact. However, the trailing pairs contact only over the upper half to two-thirds of the tooth depth. Crucially, the overall contact pattern and number of engaging teeth under negative error are generally better than under an equivalent positive error, indicating that a slightly tighter center distance is more forgiving in this worm gear drive design.
2. Inclination Angle Error ($\Delta \beta$): This error modifies the angle of the generating plane of the internal gear. Both positive and negative errors of $\Delta \beta = \pm 0.25^\circ$ significantly degrade the contact. A positive error results in contact concentrated at the mesh entry, with three pairs engaged, but the last pair has minimal contact (only about a quarter of the tooth depth). A negative error causes contact to be limited to the exit region with only two pairs in full contact. The sensitivity of the meshing to this angle is high, meaning precise control during the manufacturing of the internal gear is essential for good performance of the worm gear drive.
3. Base Circle Radius Error ($\Delta r_b$): An error in the radius of the base circle associated with the generating process alters the fundamental geometry of the enveloping action. A positive error ($\Delta r_b = +0.5$ mm) produces a contact pattern similar to a positive center distance error, with two-pair contact at the exit. A negative error ($\Delta r_b = -0.5$ mm) leads to two-pair contact at the entry, but the contact on the second pair is very poor, covering only about one-third of the tooth depth. The negative error appears to be more detrimental, causing a more severe reduction in effective contact area.
4. Shaft Angle Error ($\Delta \delta$): This is one of the most critical errors for a worm gear drive with crossed axes. An error of $\Delta \delta = \pm 0.25^\circ$ drastically reduces the simultaneous contact to just a single tooth pair. A positive error forces contact to the exit side, while a negative error forces it to the entry side. The loss of multi-tooth contact is severe and would lead to a substantial increase in contact stress and a high risk of failure. This highlights the extreme importance of accurate alignment during the assembly of this type of worm gear drive.
5. Worm Axial Displacement Error ($\Delta l$): This error represents a shift of the worm along its own axis during assembly. Interestingly, its effect is less catastrophic than shaft angle or inclination errors. A positive error ($\Delta l = +0.5$ mm) maintains multi-tooth contact (all five pairs), but the nature of contact changes from the entry to the exit: from full-face contact to contact primarily on the worm tooth tip. A negative error ($\Delta l = -0.5$ mm) results in four-pair contact at the exit. The first pair contacts only on the lower half of the worm tooth root, while the others have full contact. This negative error configuration functionally creates a tip relief at the entry side of the worm, which can be beneficial for lubrication entry and reducing mesh impact, making it a potentially favorable condition compared to the positive error.
The systematic analysis of these errors is summarized in Table 3, which compares their impact on key meshing characteristics.
| Error Source | Magnitude | Primary Contact Zone | Approx. No. of Contacting Pairs | Contact Quality | Relative Severity |
|---|---|---|---|---|---|
| Center Distance ($\Delta a$) | +0.5 mm | Exit | 2 | Poor (partial contact) | High |
| -0.5 mm | Entry | 4 | Moderate (partial on trailing pairs) | Moderate | |
| Inclination Angle ($\Delta \beta$) | +0.25° | Entry | 3 | Poor (very partial on last pair) | Very High |
| -0.25° | Exit | 2 | Moderate | Very High | |
| Base Circle Radius ($\Delta r_b$) | +0.5 mm | Exit | 2 | Poor (partial contact) | High |
| -0.5 mm | Entry | 2 | Very Poor (severe partial contact) | Very High | |
| Shaft Angle ($\Delta \delta$) | +0.25° | Exit | 1 | Very Poor (single pair) | Critical |
| -0.25° | Entry | 1 | Very Poor (single pair) | Critical | |
| Axial Displacement ($\Delta l$) | +0.5 mm | Full length (graded) | 5 | Moderate (tip contact at exit) | Low |
| -0.5 mm | Exit | 4 | Good (root contact on first pair acts as relief) | Low / Beneficial |
In conclusion, the real meshing behavior of a planar internal gear single-enveloping crown worm gear drive is highly sensitive to manufacturing and assembly errors. While the ideal drive promises robust multi-tooth contact, practical imperfections can significantly degrade its performance. The analysis leads to several key findings. First, errors that alter the fundamental geometry of the enveloping process—specifically the shaft angle error $\Delta \delta$, the generating plane inclination error $\Delta \beta$, and the base circle radius error $\Delta r_b$—have the most severe impact. These errors often reduce the number of simultaneously contacting tooth pairs dramatically and cause poor, localized contact, making them critical tolerances to control during production and assembly of the worm gear drive. Second, the effect of center distance error $\Delta a$ is asymmetric; a negative error (tighter center distance) is generally less detrimental than a positive one. Finally, worm axial displacement error $\Delta l$ has a relatively mild effect. Notably, a negative axial displacement can create a favorable entry relief, potentially improving lubrication flow and meshing smoothness, which might be intentionally incorporated as a design modification. This comprehensive error analysis provides vital insights for designing, manufacturing, and assembling this compact and potentially high-performance worm gear drive, ensuring its reliable operation under real-world conditions.