Accurate prediction and control of the actual contact ratio are paramount for ensuring the reliability, efficiency, and noise-vibration-harshness (NVH) performance of locomotive traction gears. This critical parameter, defined as the average number of tooth pairs in simultaneous contact during meshing, is significantly influenced by tip relief modifications employed to mitigate undesired impact loads at the start and end of engagement. Conventional formulas, such as the classical G. Niemann method, often fail to adequately model the complex interplay between the applied tip relief amount and the resulting actual contact ratio, particularly under the diverse operating conditions encountered by railway traction systems – ranging from high-torque start-ups to high-speed cruising. This study delves into the fundamental mechanics of gear tip relief, establishes methodologies for determining optimal relief amounts on driver and driven gears, and proposes a refined analytical framework for calculating the actual contact ratio based on shifts in the direction of the normal load and variations in the effective length of the path of contact post-relief. This advancement in gear technology provides a practical and valuable index for assessing the quality of tip relief modifications.
1 Introduction
Locomotive bogie transmission systems predominantly utilize single-stage gear reduction. Given the wide operational speed range, designing tip relief requires careful consideration of multiple factors: the actual contact ratio, load distribution between tooth pairs, dynamic excitation forces, and the distinct requirements of start-up, continuous duty, and high-speed conditions. Among these, the actual contact ratio serves as a vital indicator of transmission smoothness and load-carrying capacity. Locomotive traction gears, subjected to harsh operating environments, typically have relief amounts designed based on tooth deformation under start-up or continuous torque. However, during high-speed operation, the reduced output torque from the traction motor leads to smaller elastic deformations. If the relief amount exceeds this reduced deformation, the effective path of contact shortens, diminishing the actual contact ratio. This can potentially lead to loss of contact and degraded performance. The prevalent classical G. Niemann formula falls short in this regard, as it does not incorporate the influence of the specific tip relief amount into its calculation model. This limitation hinders its precision for demanding applications like locomotive traction. Addressing this gap in gear technology, this paper analyzes the core principles of tip relief and introduces a novel computational method for the actual contact ratio. This method explicitly accounts for changes in the normal load vector direction and the true path of contact length induced by tip relief, offering enhanced accuracy over traditional approaches and contributing significantly to improved gear design and manufacturing quality.
2 Principles of Gear Tip Relief
Gear meshing dynamics reveal that a tooth pair entering mesh initially carries no load; the entire load is borne by the preceding pair. Factors like elastic deformation under load, thermal distortion, and manufacturing errors cause the driven gear’s effective base pitch to increase slightly and the driver gear’s effective base pitch to decrease. Consequently, the driven gear’s actual base pitch becomes larger than the driver’s (Figure 1). This discrepancy causes the incoming tooth pair to reach the theoretical mesh point prematurely. The tip of the driven tooth impacts and scrapes along the root fillet of the driver tooth until it reaches the correct contact position. Conversely, during mesh exit, the driver’s actual base pitch becomes larger than the driven’s (Figure 2). The presence of clearance delays the disengagement of the outgoing pair, causing the driver tooth tip to scrape along the driven tooth root fillet, leading to impact loading on the subsequent entering pair.
Applying tip relief to both driver and driven gears is a highly effective solution to these impact problems. Extensive theoretical and practical research confirms that this gear technology significantly enhances transmission performance, improves meshing smoothness, increases load capacity, and extends service life, often without substantial increases in production cost or development time. Designing the optimal relief amount primarily considers two sources of transmission error: (1) elastic deformation occurring during operation under load, and (2) deviations in tooth profile geometry arising from design and manufacturing tolerances.
The transition points between single and double tooth contact regions are critical locations for dynamic load fluctuations. The optimal relief zone is typically defined from the tooth tip down to the nearest transition point. The primary goals are to maximize the proportion of the meshing cycle under double tooth contact and to maintain or enhance the actual contact ratio. Neglecting secondary error sources, the maximum relief for the driver tooth tip is determined by the total elastic deformation of the tooth pair located at the end of the path of contact when the subsequent pair reaches the start of contact (Point D). Similarly, the maximum relief for the driven tooth tip is determined by the total elastic deformation of the tooth pair at the start of the path of contact when the preceding pair reaches the end of contact (Point B). In practice, manufacturing errors, base pitch deviations, and thermal deformations must also be considered to define the tolerance band for the relief profile. The elastic deformation governing relief amount calculation is expressed as:
$$ \begin{cases} \Delta_{\text{in}} = \delta_{D1} + \delta_{D2} \pm \Delta f_b \\ \Delta_{\text{out}} = \delta_{B1} + \delta_{B2} \pm \Delta f_b \end{cases} $$
Where:
- $\delta_{D1}, \delta_{D2}$: Elastic deformation of driver and driven gears at transition point D.
- $\delta_{B1}, \delta_{B2}$: Elastic deformation of driver and driven gears at transition point B.
- $\Delta f_b$: Base pitch error due to manufacturing tolerances.
- $\Delta_{\text{in}}$: Combined deformation and error affecting driven tip relief (mesh-in impact mitigation).
- $\Delta_{\text{out}}$: Combined deformation and error affecting driver tip relief (mesh-out impact mitigation).
The boundaries of the single and double tooth contact regions (Points B and D) are found using geometric relationships on the base circle. For modified gears (common in traction applications), the operating pitch radii ($r’_1$, $r’_2$) replace the standard pitch radii:
$$ r’_1 = \frac{m \cdot z_1 \cdot \cos \alpha}{2 \cos \alpha’}, \quad r’_2 = \frac{m \cdot z_2 \cdot \cos \alpha}{2 \cos \alpha’} $$
$$ r_B = \sqrt{CB^2 + r_1’^2 – 2 \cdot CB \cdot r_1′ \cdot \cos \angle OCA} $$
$$ CB = CA – BA = CA – (\varepsilon_\alpha – 1) \cdot p_b $$
$$ r_D = \sqrt{CD^2 + r_2’^2 – 2 \cdot CD \cdot r_2′ \cdot \cos \angle OCE} $$
$$ CD = CE – DE = CE – (\varepsilon_\alpha – 1) \cdot p_b $$
Where $r_{B}, r_{D}$ are the radial distances to transition points B and D, $\varepsilon_\alpha$ is the theoretical transverse contact ratio, and $p_b$ is the base pitch.
3 Analysis of Actual Contact Ratio
3.1 Path of Contact Length Calculation
3.1.1 Length Before Tip Relief
The actual path of contact ($\overline{AE}$) defines the line segment along which tooth contact occurs. The contact ratio ($\varepsilon_\alpha$) is the ratio of this length to the base pitch ($p_b$). Analyzing variations in $\overline{AE}$ under different loads is therefore essential for determining the actual contact ratio. The length is calculated as:
$$ \overline{AE} = \overline{CA} + \overline{CE} $$
$$ \overline{CE} = r_{b1} (\tan \alpha_{a1} – \tan \alpha_{0}), \quad \overline{CA} = r_{b2} (\tan \alpha_{a2} – \tan \alpha_{0}) $$
Where $r_{b1}, r_{b2}$ are the base circle radii, $\alpha_{a1}, \alpha_{a2}$ are the tip circle pressure angles, and $\alpha_{0}$ is the operating pressure angle.
3.1.2 Length After Tip Relief
Focusing on linear tip relief, commonly used in industrial practice, the modification alters the tooth profile near the tip. Post-relief, the direction of the normal contact force vector ($\vec{F’}_n$) deviates from the theoretical path of contact direction ($\vec{F}_n$) due to the modified profile slope (Figure 4). Establishing a local coordinate system at the relief start point allows derivation of the modified profile’s slope. The direction of $\vec{F’}_n$ is determined by this slope, while $\vec{F}_n$’s direction depends on the local pressure angle. The actual radius of curvature ($r’_x$) at a point on the relieved tip profile, crucial for contact mechanics, is given by:
$$ r’_x = \sqrt{(s_{a1} – \delta)^2 + h^2} $$
Where $s_{a1}$ is the half tooth thickness at the tip before relief, $\delta$ is the relief amount at the point, and $h$ is a profile height parameter. The angle $\omega$ between the actual and theoretical lines of action varies along the path of contact due to the changing pressure angle on the involute and the relief profile geometry. This deviation $\omega$ and the local radius $r’_x$ define the actual trajectory of contact points under load.
3.2 Mathematical Model for Actual Contact Ratio
In the theoretical double contact zone, two pairs should share the load. However, inappropriate or excessive tip relief can cause one pair to lose contact, effectively shortening the load-bearing path of contact and reducing the actual contact ratio (Figure 5). Define the actual base pitches of driver and driven gears as $p_{b1}$ and $p_{b2}$, and the relief amounts on the leading and trailing pairs as $\lambda_1$ and $\lambda_2$. The loaded normal deformations affecting base pitch equivalence are $\Delta p_{b1}, \Delta p_{b2}$ (at mesh entry) and $\Delta’ p_{b1}, \Delta’ p_{b2}$ (at mesh exit).
Deformation compatibility dictates the contact state:
- If $p_{b2} + \Delta p_{b2} – \lambda_2 > p_{b1} + \Delta p_{b1} – \lambda_1$, the leading pair loses contact prematurely, shortening the double contact zone.
- If $p_{b1} + \Delta’ p_{b1} – \lambda_1 > p_{b2} + \Delta’ p_{b2} – \lambda_2$, entry into the double contact zone is delayed.
Consequently, if the relief amount ($\delta$) exceeds the elastic deformation ($\Delta$) at a given load, the double tooth contact region shrinks, lowering the actual contact ratio. By determining the contact state (engaged or separated) at critical points along the meshing cycle using these inequalities and the calculated path of contact trajectory considering $\omega$ and $r’_x$, the true length of the load-bearing path of contact and thus the actual contact ratio can be derived.
The trajectory of the actual path of contact depends on the operating condition due to load-dependent deformations:
- Continuous Duty (High Torque): Deformation $\Delta$ is significant. The actual path (e.g., A1-B-D-E1 in Figure 6) may deviate from the theoretical line (A-E), but its length within the double contact zone can be longer than the theoretical path, potentially increasing the actual contact ratio compared to the unrelieved case.
- High Speed (Low Torque): Deformation $\Delta$ is small. If $\delta > \Delta$, the relief prevents full contact near the theoretical start/end points. The actual path (e.g., A1-B-D-E1 in Figure 7) is shorter than the theoretical path within the double contact zone, leading to an actual contact ratio lower than the theoretical value.
4 Case Study
The proposed methodology is applied to analyze a specific locomotive traction gear set. Key parameters are listed in Table 1.
| Parameter | Driver Gear | Driven Gear | Unit |
|---|---|---|---|
| Module ($m$) | 12 | mm | |
| Pressure Angle ($\alpha$) | 25° | deg | |
| Helix Angle ($\beta$) | 0° (Spur) | deg | |
| Tip Clearance Coeff. ($c^*$) | 0.35 | – | |
| Number of Teeth ($z$) | 16 | 77 | – |
| Center Distance ($a$) | 565 | mm | |
| Profile Shift Coeff. ($x$) | 0.228 | 0.37165 | – |
| Base Pitch Error ($f_{pb}$) | 0.016 | 0.018 | mm |
| Rated Power per Axle | 1130 | kW | |
| Continuous Speed / Motor RPM | 48.6 km/h / 1034 | – | |
| High Speed / Motor RPM | 116 km/h / 2553 | – | |
The classical G. Niemann formula for actual contact ratio $\varepsilon’_\alpha$ is:
$$ \varepsilon’_\alpha = \varepsilon_\alpha + \Delta\varepsilon_\alpha – \Delta p $$
Where $\varepsilon_\alpha$ is the theoretical contact ratio, $\Delta\varepsilon_\alpha$ is an empirical increment (0.15-0.25) accounting for meshing deformation, and $\Delta p$ is the root mean square of the base pitch errors ($\Delta p = \sqrt{f_{pb1}^2 + f_{pb2}^2}$). Using $\varepsilon_\alpha = 1.38625$, $\Delta\varepsilon_\alpha = 0.15$, and $\Delta p = \sqrt{0.016^2 + 0.018^2} = 0.0241 \approx 0.024$, the Niemann calculation yields:
$$ \varepsilon’_\alpha = 1.38625 + 0.15 – 0.024 = 1.51225 $$
This value remains constant regardless of load or speed condition using the Niemann approach.
Applying the proposed methodology:
- Theoretical Contact Ratio: Calculated as $\varepsilon_\alpha = 1.38625$.
- Deformation Calculation: Using Ishikawa’s formula for mesh stiffness and load sharing principles, the deformation pattern across the path of contact is determined for continuous duty torque (Figure 8). Deformations at transition points are $\delta_B \approx 0.0641$ mm and $\delta_D \approx 0.0621$ mm.
- Relief Amount Design: Considering deformation and manufacturing errors, the relief tolerance bands are defined. Linear relief profiles are designed based on the optimal zone principle (Figure 9):
- Driver Relief: $y_1 = 0.062 \left( \frac{x}{6.54} \right)$ mm
- Driven Relief: $y_2 = 0.064 \left( \frac{x}{5.26} \right)$ mm
- Actual Contact Ratio Calculation:
- Continuous Duty (High Torque): The deformation ($\Delta$) is significant and generally larger than the relief amount ($\delta$) near the theoretical contact endpoints. The actual path of contact length increases slightly compared to the unrelieved case under load. Calculated $\varepsilon’_\alpha = 1.44235$.
- High Speed (Low Torque): The deformation ($\Delta$) is small. The relief amount ($\delta$) exceeds $\Delta$ near the tip, preventing contact initiation until further along the path. The actual path of contact length is shortened. Calculated $\varepsilon’_\alpha = 1.21355$.
| Contact Ratio Type | G. Niemann Method | Proposed Method |
|---|---|---|
| Theoretical ($\varepsilon_\alpha$) | 1.38625 | 1.38625 |
| Actual – Continuous Duty | 1.51225 | 1.44235 |
| Actual – High Speed | 1.51225 | 1.21355 |
The comparison (Table 2) highlights the key limitation of the Niemann formula: its inability to model the load/speed-dependent effect of tip relief on contact ratio. While both methods predict an increase over theoretical $\varepsilon_\alpha$ under continuous load, the proposed method shows a more moderate increase. Crucially, it reveals a significant decrease below $\varepsilon_\alpha$ under high-speed, low-torque conditions – a critical insight missed by the classical approach. This underscores the importance of the proposed gear technology for accurate performance prediction across the entire operational envelope.
5 Conclusion
This study presents a refined analytical framework for calculating the actual contact ratio of locomotive traction gears, explicitly incorporating the influence of tip relief modifications and their interaction with load-dependent elastic deformations. By analyzing the fundamental mechanism of tip relief and establishing the relationship between relief amount, normal load vector deviation ($\omega$), and the resulting actual path of contact trajectory, the proposed method overcomes significant limitations inherent in classical formulas like G. Niemann’s.
The case study on a typical locomotive gear set demonstrates the method’s practical value. Results confirm that under continuous high-torque operation, appropriately designed tip relief can slightly increase the actual contact ratio compared to the unrelieved case under load. However, under high-speed, low-torque conditions, the reduced elastic deformation means the relief amount exceeds the deformation required for contact near the tooth tips. This shortens the effective path of contact, leading to an actual contact ratio potentially significantly lower than the theoretical value – a phenomenon accurately captured by the proposed model but entirely missed by the constant-value Niemann approach.
This research provides a powerful and practical tool within modern gear technology for evaluating the effectiveness of tip relief designs. It enables engineers to accurately predict gear meshing performance across diverse operating conditions, ensuring sufficient actual contact ratio to maintain smooth transmission, adequate load sharing, and minimal dynamic excitation, particularly at high speeds. The method serves as a valuable quantitative index for assessing and optimizing tip relief quality in the demanding application of locomotive traction gears and other heavily loaded, variable-speed gear transmissions. Future work could extend the model to include helical gears and more complex relief profiles.