In the field of precision machinery, especially for robotic joints, the rotary vector reducer plays a critical role due to its compact size, high reduction ratio, and exceptional motion accuracy. As a key component of the rotary vector reducer, the cycloidal-pin gear transmission mechanism significantly impacts overall performance, particularly transmission precision. During operation, friction heat generated at the meshing interfaces between the cycloidal gear and pins can lead to thermal deformation and even scuffing, thereby affecting the longevity and efficiency of the rotary vector reducer. Therefore, conducting a comprehensive thermal analysis is essential for enhancing the performance and reliability of rotary vector reducers. This article delves into the thermal behavior of cycloidal gears through theoretical calculations and finite element simulations, focusing on friction heat flux, convective boundary conditions, and temperature field distributions. The insights gained will lay the groundwork for improving transmission accuracy, preventing tooth failures, and guiding tooth profile modifications in rotary vector reducers.
The transmission mechanism within a rotary vector reducer involves a cycloidal gear engaging with multiple pins arranged in a circular pattern. This setup allows for high torque transmission with minimal backlash. The cycloidal gear, mounted on an eccentric input shaft, undergoes both rotation and revolution relative to the pin gear center. This planetary motion is analogous to that in epicyclic gear trains, where the cycloidal gear acts as the planet and the pin gear as the sun. The tooth profile of the cycloidal gear is derived from a curtate epicycloid curve, which ensures smooth and efficient power transfer in rotary vector reducers. Understanding this geometry is fundamental for subsequent thermal analysis.
The tooth profile of the cycloidal gear is mathematically defined based on the transmission principles of the rotary vector reducer. Using a coordinate system centered at the geometric center of the cycloidal gear, with the x-axis aligned along the symmetry axis of a tooth gap, the standard tooth profile equation for the curtate epicycloid is given by:
$$x = r_p \left(1 – \frac{r_{rp}}{s}\right)^{-\frac{1}{2}} \cos[(1-i)\phi] – \frac{a}{r_p} \left(r_p – z_p r_{rp} s^{-\frac{1}{2}}\right) \cos(i\phi)$$
$$y = r_p \left(1 – \frac{r_{rp}}{s}\right)^{-\frac{1}{2}} \sin[(1-i)\phi] + \frac{a}{r_p} \left(r_p – z_p r_{rp} s^{-\frac{1}{2}}\right) \sin(i\phi)$$
where:
- $i = \frac{z_p}{z_c}$ is the transmission ratio between the cycloidal gear and pin gear,
- $\phi$ is the meshing phase angle (in radians),
- $r_p$ is the radius of the pin circle,
- $r_{rp}$ is the radius of the pin sleeve (or pin),
- $a$ is the eccentric distance,
- $z_p$ is the number of pins,
- $z_c$ is the number of cycloidal gear teeth,
- $s = 1 + K_1^2 – 2K_1 \cos \phi$, and
- $K_1 = \frac{a z_p}{r_p}$ is the shortening coefficient.
This equation generates the unique tooth shape that enables the high-performance characteristics of rotary vector reducers. The curvature of this profile varies along the tooth flank, influencing contact stresses and heat generation.
To analyze the thermal effects in a rotary vector reducer, we first compute the friction heat flux generated at the meshing interfaces. The friction heat flux $q$ is a function of contact stress $\sigma_p$, relative sliding velocity $v_r$, friction coefficient $f$, and energy conversion coefficient $\gamma$, expressed as:
$$q = \sigma_p v_r f \gamma$$
For accurate thermal simulation in rotary vector reducers, each parameter must be determined precisely. The model parameters used in this analysis are summarized in Table 1, which are typical for a medium-sized rotary vector reducer.
| Parameter | Symbol | Value |
|---|---|---|
| Number of cycloidal gear teeth | $z_c$ | 17 |
| Number of pin teeth | $z_p$ | 18 |
| Eccentric distance | $a$ | 0.004 m |
| Shortening coefficient | $K_1$ | 0.5625 |
| Pin circle radius | $r_p$ | 0.128 m |
| Pin sleeve radius | $r_{rp}$ | 0.0085 m |
| Cycloidal gear tooth width | $B$ | 0.016 m |
| Input shaft speed | $n_H$ | 1500 rpm |
The contact between the cycloidal gear and pins in a rotary vector reducer can be approximated as Hertzian contact between two cylinders. The maximum contact stress $\sigma_H$ and contact width $L$ are calculated using Hertz formulas:
$$\sigma_H = 0.418 \sqrt{\frac{E_c P_i}{B \rho_{ei}}}$$
$$L = \sqrt{\frac{8 P_i \rho_{ei} (1 – \nu^2)}{\pi B E_c}}$$
where $P_i$ is the normal contact force at meshing position $i$, $E_c$ is the equivalent elastic modulus, $\nu$ is Poisson’s ratio (taken as 0.3 for steel), and $\rho_{ei}$ is the equivalent curvature radius. The average contact stress $\sigma_p$ is then:
$$\sigma_p = \frac{\pi}{4} \sigma_H = \frac{\pi}{4} \times 0.418 \sqrt{\frac{E_c P_i}{B \rho_{ei}}}$$
The equivalent curvature radius $\rho_{ei}$ varies with the meshing phase angle $\phi$ and is derived from the curvature radius of the cycloidal tooth profile $\rho_i$ and the pin radius $r_{rp}$. The curvature radius of the theoretical tooth profile $\rho_{0i}$ is:
$$\rho_{0i} = \frac{(1 + K_1^2 – 2K_1 \cos \phi_i)^{\frac{3}{2}} r_p}{K_1 (1 + z_p) \cos \phi_i – (1 + z_p K_1^2)}$$
Thus, the actual curvature radius $\rho_i$ and equivalent curvature radius $\rho_{ei}$ are:
$$\rho_i = \rho_{0i} + r_{rp}$$
$$\frac{1}{\rho_{ei}} = \frac{1}{r_{rp}} – \frac{1}{\rho_i} = \frac{\rho_i – r_{rp}}{r_{rp} \rho_i} = \frac{\rho_{0i}}{r_{rp} (\rho_{0i} + r_{rp})}$$
The equivalent elastic modulus $E_c$ for steel components in rotary vector reducers is:
$$E_c = \frac{2 E_1 E_2}{E_1 + E_2} = E_1 = E_2 = 2.06 \times 10^5 \text{ MPa}$$
The normal contact force $P_i$ at each meshing position in a rotary vector reducer is given by:
$$P_i = \frac{2.2 M_v}{K_1 z_c r_p} \frac{\sin \phi_i}{s^{\frac{1}{2}}}$$
where $M_v$ is the output torque resistance. The relative sliding velocity $v_r$ between the cycloidal gear and pins, which drives frictional heating, is:
$$v_r = \left( r_p s^{\frac{1}{2}} – r_{rp} \right) \frac{\omega_H}{z_c}$$
$$\omega_H = \frac{2 \pi n_H}{60}$$
Here, $\omega_H$ is the angular velocity of the input shaft. The friction coefficient $f$ depends on material properties, surface roughness, and lubrication conditions. For rotary vector reducers, it can be estimated using empirical formulas from gear scuffing studies:
$$f = 0.002 \left( \frac{P_{ti}}{B \times 0.001} \right)^{0.2} \cdot \left( \frac{2}{0.001 \rho_{ei} v_n \cos \alpha} \right)^{0.2} \eta^{-0.05} X$$
where $P_{ti}$ is the tangential load, $v_n$ is the tangential velocity, $\eta$ is the dynamic viscosity of lubricant, $\alpha$ is the pressure angle, and $X$ is the roughness factor:
$$X = 21.4 \left( \frac{S_1 + S_2}{2d} \right)^{0.25}$$
with $S_1$ and $S_2$ as surface roughness values, and $d$ as the pitch diameter of the cycloidal gear. The energy conversion coefficient $\gamma$ is typically 0.9 to 0.95. Combining these, the friction heat flux $q$ across the tooth surface can be computed, as illustrated in Table 2 for various meshing positions in a rotary vector reducer.
| Meshing Position Index (i) | Meshing Phase Angle $\phi_i$ (rad) | Friction Heat Flux $q$ (W/m²) |
|---|---|---|
| 0 | 0.0 | 1.2 × 10⁵ |
| 4 | 0.2 | 1.5 × 10⁵ |
| 8 | 0.4 | 1.8 × 10⁵ |
| 12 | 0.6 | 2.1 × 10⁵ |
| 16 | 0.8 | 2.4 × 10⁵ |
| 20 | 1.0 | 2.7 × 10⁵ |
| 24 | 1.2 | 3.0 × 10⁵ |
| 28 | 1.4 | 3.3 × 10⁵ |
| 32 | 1.6 | 3.6 × 10⁵ |
In addition to heat generation, convective heat dissipation plays a vital role in the thermal equilibrium of rotary vector reducers. The cycloidal gear tooth exchanges heat with lubricating oil and air. The convective heat transfer coefficient $h$ for the tooth surface can be approximated using formulas from gear thermal analysis:
$$h = 0.664 \lambda P^{\frac{1}{3}} \left( \frac{\omega}{\nu} \right)^{\frac{1}{2}}$$
where $\lambda$ is thermal conductivity, $P$ is Prandtl number, $\omega$ is the rotational speed of the cycloidal gear, and $\nu$ is kinematic viscosity. For typical lubricating oil used in rotary vector reducers, properties are listed in Table 3.
| Property | Value |
|---|---|
| Kinematic viscosity $\nu$ | 320 × 10⁻⁶ m²/s |
| Density $\rho$ | 900 kg/m³ |
| Thermal conductivity $\lambda$ | 0.1337 W/(m·K) |
| Specific heat capacity | 900 J/(kg·K) |
With these parameters, we proceed to finite element analysis (FEA) to simulate the steady-state temperature field of the cycloidal gear in a rotary vector reducer. A single-tooth model is sufficient due to symmetry and similar loading conditions across teeth. The 3D model is created based on the tooth profile equation and meshed appropriately for thermal analysis. The contact region on the tooth flank is divided into multiple bar-shaped areas along the tooth width direction, each corresponding to a specific meshing line. This division allows for accurate application of friction heat flux, which varies along the tooth profile but is constant along the tooth width for a given meshing position.
The friction heat flux applied in FEA must account for the intermittent nature of meshing in rotary vector reducers. Heat is generated only during the brief contact time $t_1$ when a point on the tooth surface engages with a pin. The contact time $t_1$ and the meshing cycle time $t_2$ are:
$$t_1 = \frac{L}{v_r}$$
$$t_2 = \frac{1}{\omega_H} \frac{2\pi (z_p – 1)}{z_p}$$
Thus, the average friction heat flux $\bar{q}$ applied to each bar-shaped area is:
$$\bar{q} = \frac{t_1}{t_2} q \Lambda$$
where $\Lambda$ is the heat partition coefficient, taken as 0.5, assuming equal heat distribution between the cycloidal gear and pin. Convective boundary conditions are applied to the tooth flank and sides to model heat dissipation. The FEA is conducted using ANSYS Workbench, with material properties set for steel.
The impact of input power on the temperature field in rotary vector reducers is significant. Simulations were run for input powers of 7.5 kW, 11.25 kW, and 15 kW. The results, summarized in Table 4, show that higher input power increases overall tooth temperature due to greater frictional heating, while the temperature distribution pattern remains consistent.
| Input Power (kW) | Maximum Temperature (°C) | Minimum Temperature (°C) | Temperature Gradient (°C/mm) |
|---|---|---|---|
| 7.5 | 85.3 | 45.2 | 12.5 |
| 11.25 | 102.7 | 52.8 | 15.6 |
| 15.0 | 120.1 | 60.4 | 18.7 |
The temperature field exhibits a gradient from the contact region to the non-contact region, with higher temperatures near the tooth surface and lower temperatures in the tooth core. This non-uniformity can induce thermal deformation, affecting the transmission accuracy of rotary vector reducers. Along the tooth width, temperatures are symmetric, with the highest values at the mid-width due to reduced convective cooling compared to the edges.
The number of bar-shaped divisions in the contact region influences the accuracy of FEA results. Simulations with 8, 16, 32, and 64 divisions were performed for an input power of 11.25 kW. As shown in Table 5, increasing the division number refines the temperature prediction, with results converging at 32 divisions, indicating an optimal balance between computational efficiency and accuracy for rotary vector reducers.
| Division Number | Maximum Temperature (°C) | Minimum Temperature (°C) | Computational Time (s) |
|---|---|---|---|
| 8 | 98.5 | 50.3 | 120 |
| 16 | 101.2 | 51.9 | 240 |
| 32 | 102.7 | 52.8 | 480 |
| 64 | 103.0 | 53.0 | 960 |
In conclusion, this thermal analysis of cycloidal gears in rotary vector reducers provides valuable insights into heat generation and dissipation mechanisms. The friction heat flux, dependent on contact stresses and sliding velocities, drives temperature rises that can impact performance. Convective cooling mitigates these effects but does not eliminate thermal gradients. Finite element simulations reveal that input power directly influences temperature levels, while the subdivision of contact regions affects result precision, with 32 divisions being optimal for this rotary vector reducer model. These findings underscore the importance of thermal management in rotary vector reducers to maintain transmission accuracy and prevent failures like scuffing. Future work could explore advanced lubrication techniques, material enhancements, and tooth profile modifications to further optimize the thermal behavior of rotary vector reducers, ensuring their reliability in demanding applications such as robotics and precision machinery.
The methodology presented here can be extended to other gear systems, but the unique geometry of cycloidal gears in rotary vector reducers necessitates tailored approaches. By integrating theoretical calculations with numerical simulations, engineers can better predict thermal effects and design more robust rotary vector reducers. As the demand for high-performance rotary vector reducers grows, continued research in thermal analysis will be crucial for advancing the state of the art in motion control and power transmission.