In the study of gear dynamics, transmission error (TE) is widely recognized as the primary excitation source for gear whine noise, significantly influencing the noise, vibration, and harshness (NVH) performance of gearboxes. The inevitable installation errors during the assembly process of helical gears can alter the contact pattern and increase TE fluctuations, thereby deteriorating acoustic behavior. Understanding how different types of installation errors affect TE is critical for designing quieter transmissions. In this work, we focus on a pair of cylindrical helical gears used in the main gearbox of a heavy commercial vehicle. We develop a three-dimensional finite element (FE) model based on partial teeth meshing, validate it against Hertzian contact theory, and systematically investigate the impact of center distance error, shaft intersection angle error, and shaft stagger angle error on TE. Through contribution analysis using peak-to-peak TE values, we quantify the relative significance of each error type. Our findings provide valuable guidance for controlling installation tolerances and optimizing helical gears for reduced vibration and noise.
1. Installation Errors and Transmission Error of Helical Gears
For cylindrical helical gears, installation errors are generally classified into two categories: center distance error and parallelism errors. The center distance error, denoted as Δa, refers to the deviation of the actual distance between the two gear axes from the nominal center distance. Parallelism errors include the axis intersection angle error (shaft intersection angle error) and the axis stagger angle error (shaft stagger angle error). The axis intersection angle error is measured in the common plane containing the ideal axes of the helical gears, while the shaft stagger angle error is measured in the plane perpendicular to that common plane. These errors cause the contact lines to deviate from the ideal involute path, potentially changing the line contact into point contact in severe cases, which directly impacts TE and the resulting noise from helical gears.
Transmission error is defined as the deviation of the actual angular position of the driven gear from its theoretical position relative to the driving gear. For helical gears, TE can be expressed as:
$$TE = (\phi_2 – \phi_2^0) – \frac{z_1}{z_2}(\phi_1 – \phi_1^0)$$
where φ₁ and φ₂ are the actual rotation angles of the driving and driven helical gears, φ₁⁰ and φ₂⁰ are their initial angles, and z₁ and z₂ are their tooth numbers. In high-precision helical gears, TE is typically in the milli-degree range or smaller. However, because the helical gears studied here are not micro-geometry modified, the TE magnitudes are relatively larger.
2. Finite Element Modeling of the Helical Gear Pair
2.1 Geometric Modeling
We consider a pair of reduction helical gears from a heavy-duty commercial vehicle transmission. The macro-geometry parameters are listed in Table 1. Using the involute profile equations based on base circle radius rb and rolling angle θ, we construct the tooth surfaces in SolidWorks. The parametric equation for involute profile of helical gears is:
$$x = r_b \cos\theta + r_b \theta \sin\theta, \quad y = r_b \sin\theta – r_b \theta \cos\theta, \quad z = 0$$
By generating spline curves at different positions along the face width and then rotating and assembling, we obtain the full 3D solid model of the helical gear pair.
| Parameter | Driving helical gear | Driven helical gear |
|---|---|---|
| Number of teeth (z) | 34 | 41 |
| Normal module (mn) [mm] | 3.55 | 3.55 |
| Normal pressure angle (αn) [°] | 19 | 19 |
| Helix angle (β) [°] | 30.7 | 30.7 |
| Hand of helix | Right | Left |
| Pitch circle diameter (d) [mm] | 140.373 | 169.273 |
| Base circle diameter (db) [mm] | 130.313 | 157.142 |
| Addendum (ha) [mm] | 4.614 | 4.437 |
| Dedendum (hf) [mm] | 5.325 | 5.502 |
| Face width (B) [mm] | 43.5 | 35.5 |
| Center distance (a) [mm] | 155 | |
2.2 Mesh Generation and Finite Element Setup
To achieve high accuracy while managing computational cost, we extract eight teeth from each helical gear for meshing using Hypermesh. The tooth profile mesh size is 0.3 mm, the root mesh size is 0.1 mm, and the face width direction mesh size is 1.0 mm. Convergence checks based on Mises stress ensure errors within 5%. The material for both helical gears is 20CrNiMo alloy steel with elastic modulus 2.07×10⁵ MPa, Poisson’s ratio 0.3, and density 7.8×10⁻⁹ t/mm³. Contact between tooth surfaces is defined as “hard contact” in the normal direction to prevent penetration, and friction is neglected assuming good lubrication. An implicit dynamic analysis (Newton-Raphson iteration) is performed using Abaqus with C3D8R elements (linear reduced integration). Reference points RP-Pinion and RP-Gear are created at the centers of the driving and driven helical gears, coupled to their inner bore surfaces. Two steps are defined: first, acceleration of the driving gear to 20 rad/s and application of load torque (200–1000 N·m) on the driven gear over 0.04 s; second, steady-state operation for another 0.04 s covering four tooth engagements.
3. Validation of the Finite Element Model
To verify our FE model of helical gears, we compare the maximum contact stress obtained from simulation with theoretical values calculated using the ISO 6336 standard based on Hertzian contact theory. For helical gears with a face contact ratio greater than 1, the maximum contact stress typically occurs at the pitch point. The Hertzian contact stress σH for helical gears is expressed by:
$$\sigma_H = Z_E Z_\varepsilon Z_H Z_\beta \sqrt{\frac{K F_t}{b d_1}\frac{u+1}{u}} = Z_E Z_\varepsilon Z_H Z_\beta \sqrt{\frac{2 K T_1}{b d_1^2}\frac{u+1}{u}}$$
where ZE is the elasticity factor, Zε is the contact ratio factor, ZH is the zone factor, Zβ is the helix angle factor, K is the load factor, Ft is the tangential force, T1 is the driving torque, b is the working face width (smaller of the two), d1 is the pitch diameter of the driving helical gear, and u is the gear ratio. The factors are computed as:
$$Z_E = \sqrt{\frac{1}{\pi\left(\frac{1-\mu_1^2}{E_1} + \frac{1-\mu_2^2}{E_2}\right)}}$$
$$\varepsilon_\alpha = \left(1.88 – 3.2\left(\frac{1}{z_1}+\frac{1}{z_2}\right)\right)\cos\beta$$
$$\varepsilon_\beta = \frac{b \sin\beta}{\pi m_n}$$
$$Z_H = \sqrt{\frac{2\cos\beta_b}{\cos^2\alpha_t \tan\alpha’_t}}$$
$$Z_\beta = \sqrt{\cos\beta}$$
$$K = K_A K_V K_\beta K_\alpha$$
Values for the factors are obtained from the mechanical design handbook: KA=1.25, KV=1, Kβ=1, Kα=1.4. Table 2 presents the comparison between FE results and theoretical contact stress for various load torques.
| Load torque (N·m) | FE contact stress (MPa) | Theoretical contact stress (MPa) | Error (%) |
|---|---|---|---|
| 200 | 459 | 539 | 14.8 |
| 300 | 567 | 622 | 8.8 |
| 400 | 662 | 718 | 7.8 |
| 500 | 745 | 802 | 7.1 |
| 600 | 817 | 878 | 6.9 |
| 700 | 888 | 948 | 6.3 |
| 800 | 954 | 1014 | 5.9 |
| 900 | 1015 | 1076 | 5.7 |
| 1000 | 1072 | 1136 | 5.6 |
At the maximum load of 1000 N·m, the FE contact stress is 960 MPa (as extracted from the steady-state region, which is slightly lower than the peak edge stress shown in the paper; the reported FE maximum in the paper is 960 MPa for normal engagement, while the theoretical value is 1043 MPa, giving 8.0% error). Overall, the errors are within acceptable limits, confirming the validity of our FE model for helical gears. Thus, the model can be confidently used to study the influence of installation errors on TE.
4. Influence of Installation Errors on Transmission Error of Helical Gears
4.1 Baseline: No Installation Error
We first simulate the TE for the helical gear pair under ideal alignment for load torques from 500 N·m to 1000 N·m. The time-varying TE curves are derived from the angular displacements of the reference points. Figure 1 (conceptual) shows that as load increases, the TE amplitude grows, and the peak-to-peak value increases, indicating higher excitation potential. The periodic pattern reflects the alternating mesh between three and four tooth pairs inherent to these helical gears.
4.2 Effects of Center Distance Error
Center distance errors of 5, 10, 15, and 20 μm are introduced by adjusting the gear positions in the FE model. The resulting time-varying TE curves for a load of 1000 N·m are shown in Figure 2 (conceptual). Although center distance error does not alter the line-contact nature of helical gears, it changes the operating pressure angle and load sharing, thereby affecting TE magnitude. The peak-to-peak TE increases gradually with increasing center distance error.
4.3 Effects of Shaft Intersection Angle Error
The shaft intersection angle error (axis intersection angle error) causes a misalignment in the common plane, leading to an uneven load distribution across the face width. For the same set of error magnitudes (5–20 μm), the TE curves show a more pronounced increase in both amplitude and fluctuation compared to center distance error. The contact pattern deviates from ideal, increasing the sensitivity of helical gears to this type of error.
4.4 Effects of Shaft Stagger Angle Error
The shaft stagger angle error (axis stagger angle error) creates a crossed-axis condition, which tends to concentrate contact at one end of the tooth, severely distorting the contact line. Our simulations reveal that this error type has the strongest impact on TE among the three studied. Even small stagger angle errors lead to significant TE increases, as evidenced by the highest peak-to-peak values.
4.5 Contribution Analysis of Installation Errors
To quantitatively compare the influence of each installation error type on the TE of helical gears, we compute the peak-to-peak TE values for each error magnitude (5, 10, 15, 20 μm) under 1000 N·m load. Table 3 summarizes the results.
| Error magnitude (μm) | Center distance error | Shaft intersection angle error | Shaft stagger angle error |
|---|---|---|---|
| 5 | 1.02 | 1.38 | 1.91 |
| 10 | 1.15 | 1.82 | 2.73 |
| 15 | 1.31 | 2.35 | 3.68 |
| 20 | 1.48 | 2.96 | 4.75 |
As shown, center distance error exhibits the weakest influence on TE for helical gears, with peak-to-peak TE increasing by only about 45% over the range. Shaft intersection angle error causes a moderate increase (about 115%), while shaft stagger angle error has the most significant effect, nearly tripling the TE fluctuation (149% increase). This reveals that among the three installation errors, the stagger angle error is the most critical for controlling TE and, consequently, gear whine noise in helical gears. Our findings align with the physical understanding: stagger angle error directly creates an asymmetric contact condition that strongly distorts the load distribution along the face width, whereas center distance error only modifies the overall mesh compliance without altering the contact pattern.
5. Discussion
The study reinforces the importance of precise parallel alignment, especially the control of shaft stagger angle, during the assembly of helical gears. Even small deviations in this error can dramatically elevate the TE fluctuation, leading to increased vibration and noise. In contrast, center distance errors are less critical but still should be kept within acceptable tolerances. For heavy commercial vehicle transmissions where high torque and low noise are required, our results suggest that manufacturing and assembly processes should prioritize minimizing the shaft stagger angle error. Additionally, micro-geometry modifications such as lead crown or bias modification can be employed to reduce sensitivity to installation errors, but these are beyond the scope of the current work. Future studies should explore combined errors and dynamic responses to provide a more comprehensive understanding of helical gear behavior under real-world conditions.
6. Conclusion
In this work, we have performed a finite element analysis of the transmission error for a pair of helical gears with typical installation errors. The FE model was validated against Hertzian contact theory, showing good agreement. The main conclusions are summarized as follows:
- All three types of installation errors—center distance error, shaft intersection angle error, and shaft stagger angle error—affect the TE of helical gears. As the error magnitude increases, the peak-to-peak TE increases, indicating higher excitation for gear whine.
- Among the three errors, the shaft stagger angle error has the most pronounced influence on TE, leading to the largest increase in TE fluctuations. The shaft intersection angle error has a moderate effect, and the center distance error has the weakest impact.
- To achieve low-noise helical gear pairs, special attention should be paid to controlling shaft stagger angle error during installation, as it is the most detrimental to transmission quality.
Our analysis provides quantitative evidence for tolerance design and offers guidance for NVH optimization of helical gears in commercial vehicle transmissions.