The pursuit of higher power density and operational speeds in vehicle transmission systems, particularly for high-speed heavy-duty applications, has intensified the challenges associated with gear dynamics. Among these, spur gears remain fundamental components due to their simplicity and manufacturing ease. However, the inherent characteristics of spur gears, such as parametric excitation from time-varying mesh stiffness and nonlinearities from backlash, can induce significant dynamic loads under high-speed and high-torque conditions. These dynamic loads are primary sources of vibration and noise, leading to reduced reliability, fatigue life, and overall system performance. To address these issues, two critical design strategies are often employed: gear body lightweighting to enhance power density and tooth profile modification to mitigate vibration excitation.
Lightweight designs, such as spur gears with thin-rimmed bodies and web structures, effectively reduce rotational inertia. However, this reduction in structural mass comes at the cost of decreased gear body stiffness. This compliance can significantly alter the overall mesh stiffness, a key excitation source, potentially exacerbating dynamic behavior rather than improving it. Concurrently, tooth profile modification, specifically tip relief, is a well-established technique to soften the abrupt changes in mesh stiffness that occur during the transition from single-tooth to double-tooth contact in spur gears. While the individual effects of body flexibility and profile modification have been studied, their coupled influence on the complex, multi-directional dynamic response of spur gears is not fully understood, especially within a comprehensive nonlinear dynamic framework.
This investigation aims to bridge this gap by developing an integrated analytical-computational model to study the dynamic load characteristics of spur gears. The core of this study involves two coupled models: a refined mesh stiffness model that incorporates gear body compliance, linear tip relief, and the actual operating kinematics, and a multi-degree-of-freedom nonlinear dynamic model that captures the lateral-torsional-rocking coupled vibrations of the gear-rotor-bearing system. The primary objective is to quantitatively and qualitatively analyze how the dynamic load factor of spur gears is influenced by operational parameters (input speed and torque) and design parameters (amount and length of tip relief), with a direct comparison between solid and lightweight (thin-rimmed) gear bodies.
Methodology: Integrated Modeling Framework
The analysis is built upon a synergistic coupling of a detailed mesh stiffness model and a nonlinear dynamic model. This integrated framework allows for the accurate representation of the excitation mechanism and the subsequent system response.
1. Mesh Stiffness Model for Spur Gears with Flexible Bodies and Tip Relief
The time-varying mesh stiffness, $k_m(t)$, is the principal internal excitation in spur gear dynamics. The proposed model calculates this stiffness by considering the combined deflection of the mating teeth and the flexible gear bodies, adjusted for the loss of contact due to profile modification.
1.1 Single Tooth Pair Mesh Stiffness
The stiffness of a single tooth pair in contact, $k_s(\alpha_m(t))$, is derived from the composite deformation along the line of action. It is expressed as the inverse of the sum of deflections from the pinion, gear, and contact zone:
$$ k_s(\alpha_m(t)) = \frac{1}{\delta_{B1}(\alpha_m(t)) + \delta_{M1}(\alpha_m(t)) + \delta_{B2}(\alpha_m(t)) + \delta_{M2}(\alpha_m(t)) + \delta_{C}} $$
where $\delta_{Bj}$ is the bending and shear deflection of tooth $j$ calculated using an enhanced analytical method (e.g., Weber/Cornell approach), $\delta_{C}$ is the Hertzian contact deflection, and $\delta_{Mj}$ is the additional deflection due to the gear body flexibility of gear $j$, which is the novel contribution for lightweight spur gears.
1.2 Gear Body Compliance for Thin-Rimmed Spur Gears
For a thin-rimmed spur gear body with a web structure containing lightening holes, the body is modeled as three segments: the rim, the web, and the hub. The additional angular deflection at the pitch circle due to applied tooth load is calculated by integrating the shear strain across these segments. The total circumferential displacement, $s_{total}$, is found by summing contributions from each segment, considering the reduced effective load-carrying path around lightening holes in the web. The gear body deflection, $\delta_{Mi}$, at the base radius $R_i$ is then:
$$ \delta_{Mi} = \theta_t R_i = \frac{2(s_1 + s_2 + s_3)}{d_i} R_i $$
where $\theta_t$ is the body wind-up angle and $d_i$ is a reference diameter. This model provides a closed-form solution for the compliance of complex gear body geometries.
1.3 Incorporating Linear Tip Relief
Linear tip relief is defined from a start point $s_b$ to the tip of the tooth $s_e$ along the line of action. The relief amount $C$ at any point $s$ is:
$$ C(s) = C_{max} \frac{s – s_b}{s_e – s_b} $$
The effective mesh stiffness when the contact point lies within a relief zone must account for the initial separation caused by the modified profile. The calculation logic depends on the contact state:
- Case 1 & 2: Contact in the non-relieved single or double tooth pair region. Stiffness is simply the sum of the relevant tooth pair stiffnesses.
- Case 3 & 4: Contact within the relief zone of one gear. The stiffness is modified based on whether the relief amount exceeds the elastic deflection of that tooth pair under the dynamic load $F_m(t)$. For example, if gear 1 is in its relief zone:
$$ \tilde{k}_{m3}(t) = \begin{cases}
\frac{k_{s1}(\alpha_m(t))}{1 + k_{s2}(\alpha_m(t)) C_1(t) / F_m(t)}, & \text{if } E_1(t) > 0 \\
k_{s1}(\alpha_m(t)), & \text{if } E_1(t) \le 0
\end{cases} $$
where $E_1(t) = \delta_1(t) – C_1(t)$, and $\delta_1(t)=F_m(t)/k_{s1}$. - Case 5: Contact within the relief zones of both mating spur gears. The logic determines which tooth pair remains in contact based on the relative relief amounts and dynamic deflections.
This formulation ensures the mesh stiffness $k_m(t)$ dynamically reflects the loss of contact due to profile modification under the operating load.
2. Nonlinear Dynamic Model of the Spur Gear System
A 10-degree-of-freedom (DOF) lateral-torsional-rocking coupled model is established for a single-stage spur gear transmission system supported by flexible bearings. Each gear (pinion $i=1$, gear $i=2$) has five DOFs: two translational motions $(x_i, y_i)$ in the plane perpendicular to the shaft axis, two rotational (rocking) motions $(\theta_{xi}, \theta_{yi})$ about the $x$ and $y$ axes, and one torsional motion $(\theta_{zi})$ about the $z$-axis.
The equations of motion are derived using Lagrange’s equation, incorporating several key dynamic effects:
- Geometric eccentricity ($e_i$): Introduces translational inertia coupling with torsional motion.
- Gyroscopic moments: Arising from the coupling between rocking and torsional velocities ($I_{zi}\dot{\theta}_{zi}\dot{\theta}_{xi}$, etc.).
- Moment from axial load variation: A moment arm ($k_h$) couples the mesh force to the rocking equations to simulate load distribution effects.
The governing equations for gear $i$ are of the form:
$$ m_i \ddot{x}_i – m_i e_i \dot{\theta}_{zi}^2 \cos(\theta_{zi}+\psi_i) – m_i e_i \ddot{\theta}_{zi} \sin(\theta_{zi}+\psi_i) + \mathbf{K}_{i1}\mathbf{q}_i + \mathbf{C}_{i1}\dot{\mathbf{q}}_i = \pm F_m(t) \sin(\alpha(t)+\gamma(t)) $$
$$ I_{xi}\ddot{\theta}_{xi} + \mathbf{K}_{i3}\mathbf{q}_i + \mathbf{C}_{i3}\dot{\mathbf{q}}_i + I_{zi}\dot{\theta}_{zi}\dot{\theta}_{xi} = \mp F_m(t) k_h \cos(\alpha(t)+\gamma(t)) $$
$$ (I_{zi} + m_i e_i^2)\ddot{\theta}_{zi} – m_i \ddot{x}_i e_i \sin(\theta_{zi}+\psi_i) + m_i \ddot{y}_i e_i \cos(\theta_{zi}+\psi_i) + \mathbf{K}_{i5}\mathbf{q}_i + \mathbf{C}_{i5}\dot{\mathbf{q}}_i = \mp F_m(t) R_i + T_i $$
where $\mathbf{q}_i = [x_i, y_i, \theta_{xi}, \theta_{yi}, \theta_{zi}]^T$, $\mathbf{K}_{ij}$ and $\mathbf{C}_{ij}$ are the $j$-th row of the support stiffness and damping matrices for gear $i$, $R_i$ is the base radius, $T_i$ is the external torque, $\alpha(t)$ is the dynamic pressure angle, and $\gamma(t)$ is the position angle. The upper/lower signs correspond to the pinion/gear, respectively.
2.1 Gear Mesh Interface Model
The gears are coupled through the dynamic mesh force $F_m(t)$ acting along the line of action. This force is a function of the time-varying mesh stiffness $k_m(t)$ from the previous section, a mesh damping $c_m(t)$, and a nonlinear displacement function $f$ accounting for backlash $\tilde{b}$:
$$ F_m(t) = k_m(t) f(\tilde{b}, \Delta(t)) + c_m(t) \dot{f}_1(\tilde{b}, \Delta(t)) $$
$$ c_m(t) = 2\xi \sqrt{k_m(t) \frac{m_1 m_2}{m_1+m_2}} $$
The nonlinear displacement function is defined as:
$$ f(\tilde{b}, \Delta(t)) = \begin{cases}
\Delta(t) – \tilde{b}, & \text{if } \Delta(t) > \tilde{b} \\
0, & \text{if } |\Delta(t)| \le \tilde{b} \\
\Delta(t) + \tilde{b}, & \text{if } \Delta(t) < -\tilde{b}
\end{cases} $$
The dynamic transmission error (DTE), $\Delta(t)$, which is the relative displacement along the line of action, is the fundamental source of excitation:
$$ \Delta(t) = (x_1 – x_2)\sin(\alpha+\gamma) – (y_1 – y_2)\cos(\alpha+\gamma) + R_1\theta_{z1} – R_2\theta_{z2} + e(t) $$
where $e(t)$ represents the geometric deviation introduced by the tip relief profile.
2.2 Dynamic Load Factor (DLF)
To quantify the dynamic severity, the Dynamic Load Factor is defined as the ratio of the peak dynamic mesh force to the nominal static force:
$$ \text{DLF} = \frac{F_{max}}{F_s} = \frac{\max[F_m(t)]}{F_s} $$
The parameters for the two spur gear body designs studied are summarized below. Gear body 1 is a solid design, while Gear body 2 is a lightweight, thin-rimmed design with a web structure.
| Parameter | Gear Body 1 (Solid) | Gear Body 2 (Lightweight) | ||
|---|---|---|---|---|
| Pinion | Gear | Pinion | Gear | |
| Module (mm) | 6 | 6 | ||
| Number of Teeth | 45 | 35 | 45 | 35 |
| Mass, $m_i$ (kg) | 7.2 | 13.9 | 6.5 | 10.2 |
| Polar Moment, $I_{zi}$ (kg·mm²) | 0.049 | 0.135 | 0.043 | 0.110 |
| Backlash (mm) | 0.2 | 0.2 | ||
| Geometric Eccentricity (mm) | 0.05 | 0.05 | ||
Results and Discussion
The coupled model is solved numerically to investigate the dynamic response. The analysis focuses on the impact of operational conditions and profile modification on the DLF, with a consistent comparison between the solid and lightweight spur gear bodies.
1. Influence of Input Speed and Torque (Unmodified Spur Gears)
1.1 Speed Sensitivity: Under a constant nominal torque, the DLF for both spur gear designs exhibits a “hump-shaped” trend with increasing input speed, characteristic of systems passing through resonance. The lightweight spur gears (Gear body 2) consistently show a higher DLF across most of the speed range, with a maximum increase of approximately 6% compared to the solid gears. The peak DLF for both designs occurs around the same speed, but the lightweight gears exhibit a more pronounced response. A frequency domain analysis of the dynamic mesh force reveals that the amplitudes at the 2nd and 6th harmonics of the mesh frequency are significantly larger for the lightweight spur gears, indicating a stronger excitation of these dynamic modes due to the increased body compliance.
1.2 Torque Sensitivity: Under a constant high input speed, the DLF for both spur gear designs generally decreases in a parabolic fashion with increasing torque. This is because higher torque increases the mean load, which has a damping effect on the relative vibration within the backlash nonlinearity. However, the lightweight spur gears display localized peaks at specific torque values, a behavior not observed in the solid gears. Spectral analysis shows that while the solid gears’ response is dominated by the 2nd and 4th mesh harmonics, the lightweight gears show a prominent 5th harmonic component at certain torques, which is responsible for these local DLF maxima. This demonstrates that the dynamic response of compliant spur gears is more sensitive to load variations.
2. Influence of Tip Relief on Spur Gear Dynamic Loads
2.1 Optimal Relief Amount: For a fixed, short relief length, the DLF for both spur gear body types follows a distinct “U-shaped” curve as the relief amount $C_{max}$ increases. There exists an optimal relief amount (found to be approximately 0.019 mm for both cases) that minimizes the DLF. At this optimum, the DLF is reduced by about 17% for solid gears and 16% for lightweight gears compared to their unmodified versions. The optimal value is slightly less than the static deflection at the highest point of single tooth contact under nominal load. Beyond this optimum, over-relief causes the DLF to rise again. A critical finding is the existence of a critical relief amount beyond which the modified gear has a higher DLF than the unmodified gear. This critical value is larger for lightweight spur gears (0.073 mm) than for solid gears (0.053 mm), indicating a broader, more forgiving “effective relief window” for compliant gear bodies.
2.2 Influence of Relief Length: Holding the relief amount at a near-optimal value, the effect of varying the relief length $L$ was studied. The results reveal two distinct regimes based on whether the relief is shorter or longer than the theoretical length of the single tooth contact region ($L_h$).
- Short Relief ($L < L_h$): The DLF for both spur gear types shows a “U-shaped” variation, with an optimal short relief length (2.5 mm for solid, 3.0 mm for lightweight).
- Long Relief ($L \ge L_h$): When the relief extends through and beyond the single tooth contact zone, the DLF drops sharply and reaches a global minimum at a specific long relief length (5.4 mm for both). This minimum DLF under long relief is significantly lower than the minimum achieved with short relief, demonstrating that a sufficiently long relief is more effective in smoothing the stiffness transition and suppressing vibration in spur gears.
The time-domain mesh stiffness and dynamic mesh force plots confirm these findings. Optimal relief effectively smoothens the abrupt stiffness change at the tooth engagement/disengagement points. Under long relief, the dynamic force fluctuations are markedly smaller than under short relief, leading to the lowest overall dynamic loads. The frequency spectra show that optimal profile modification significantly reduces the higher harmonic content (2nd, 4th, 5th mesh harmonics) that is prominent in unmodified spur gears, shifting the energy more towards the fundamental mesh frequency.
| Parameter Varied | General Trend (Both Gear Bodies) | Notable Difference for Lightweight Spur Gears |
|---|---|---|
| Input Speed | DLF shows hump-shaped curve; peaks near resonance. | Higher DLF at most speeds; significantly larger 2nd & 6th mesh harmonic amplitudes. |
| Input Torque | DLF decreases parabolically with increasing torque. | Presence of local DLF peaks at specific torques; strong 5th mesh harmonic component. |
| Tip Relief Amount | DLF follows U-shaped curve; optimal amount exists. | Broader optimal region; larger critical relief amount before performance degrades. |
| Tip Relief Length | Short relief: U-shaped curve. Long relief (> single contact zone): DLF drops sharply. | Similar optimal trends, but dynamic force reduction is crucial for mitigating their higher inherent vibration. |
Conclusion
This study presented a comprehensive nonlinear dynamic analysis of spur gears, integrating the effects of gear body flexibility and tooth profile modification. The developed model couples a refined analytical mesh stiffness calculation with a 10-DOF lateral-torsional-rocking dynamic model, providing a robust tool for investigating the complex dynamic behavior of modern, high-power-density spur gear designs.
The key conclusions are as follows:
- Gear body compliance in lightweight spur gears significantly alters their dynamic signature. While reducing inertia, it generally increases the dynamic load factor across a wide speed range and introduces a more complex, torque-sensitive response with local peaks, primarily by amplifying specific mesh frequency harmonics (notably the 2nd, 5th, and 6th).
- Tooth profile modification is a highly effective strategy for reducing dynamic loads in both solid and lightweight spur gears. The relationship between relief amount and dynamic load factor is fundamentally U-shaped, underscoring the necessity for precise optimization. Under-relief is ineffective, while over-relief can be detrimental.
- Lightweight spur gears exhibit a greater tolerance to profile modification errors, characterized by a larger critical relief amount. However, they also require careful tuning to achieve minimal dynamic loads.
- The length of the relief profile is a critical design parameter. Extending the relief through the entire single-tooth contact region and slightly beyond (long relief) proves to be a more effective strategy than short relief for minimizing vibration, yielding a sharper reduction in the dynamic load factor for both gear body types.
These findings provide valuable guidelines for the design of high-performance spur gear transmissions. For applications requiring lightweight gears, a coupled analysis of body structure and profile modification is essential. The optimal design should target a specific combination of relief amount and length to mitigate the inherent dynamic amplification caused by body flexibility, thereby enabling the realization of high power density without compromising dynamic reliability and noise performance.