This paper presents a comprehensive investigation into a novel vibration damping structure for straight spur gear transmissions, which incorporates slot-holes machined in the tooth region. The motivation stems from the persistent challenge of reducing vibration and noise in high-speed, precision gear drives. Traditional approaches such as tooth profile modification have limitations in manufacturing complexity and cost. Our work focuses on an alternative structural design that introduces controlled flexibility in the tooth body through a combination of longitudinal slots and circular holes, aiming to mitigate meshing impacts and reduce vibration. We systematically study the influence of slot width, backlash, and hole diameter on the dynamic response of straight spur gear pairs using both numerical simulation and experimental validation. The findings reveal that the proposed slot-hole structure can reduce vibration acceleration by 20%–33%, with backlash being the most influential parameter. A three-factor three-level orthogonal experimental design is employed to identify the optimal parameter combination, and the results are verified through vibration testing on a dedicated test rig. The study provides practical guidance for designing low-vibration straight spur gears in applications requiring high stability and silent operation.
The vibration reduction mechanism of the slot-hole structure in straight spur gear is multifold. First, the longitudinal slot cut along the tooth tip introduces a local reduction in tooth stiffness near the contact area, which softens the meshing impact when a tooth pair engages. This is particularly beneficial for straight spur gears due to their typically low contact ratio, leading to abrupt load transfer. Second, the slot decouples the deformation of adjacent teeth, reducing the mutual interference caused by tooth bending under load. Third, the circular hole located at the root of the slot relaxes stress concentration and further alters the local stiffness distribution. When combined with a negative backlash (interference fit), the slot-hole structure provides a compliance that accommodates the geometric mismatch, effectively eliminating backlash-induced impacts. In the following sections, we detail the structural design, the dynamic simulation methodology, the orthogonal optimization, and the experimental confirmation.
Structural Design of Slot-Hole Straight Spur Gear
The proposed slot-hole structure for straight spur gear is illustrated conceptually. A longitudinal groove is carved from the tooth tip downward, connected to a cylindrical through-hole drilled in the tooth body. The groove width \(w\) is varied from 0.2 mm to 0.4 mm, the hole diameter \(d\) from 4 mm to 8 mm, and the gear backlash \(b\) from −0.03 mm to 0.2 mm. The basic geometric parameters of the straight spur gear pair used in this study are listed in Table 1.
| Parameter | Pinion (z1) | Gear (z2) |
|---|---|---|
| Number of teeth | 34 | 34 |
| Module (mm) | 6 | 6 |
| Pressure angle (°) | 20 | 20 |
| Addendum coefficient | 1 | 1 |
| Clearance coefficient | 0.25 | 0.25 |
| Face width (mm) | 60 | 60 |
| Profile shift coefficient | 0.3945 | 0.3945 |
| Center distance (mm) | 208.4 | |
All straight spur gear specimens are made of 20CrMnTi steel, case-hardened and ground. The reference backlash for a standard straight spur gear is 0.2 mm; we deliberately test three levels: 0.2 mm (normal), 0.07 mm (reduced), and −0.03 mm (negative, i.e., interference). The negative backlash is achieved by increasing the tooth thickness beyond the standard, which is possible only when the slot-hole provides enough compliance to avoid jamming.
Dynamic Simulation Methodology
The straight spur gear pairs are modeled in SolidWorks and imported into Adams via the X_T format. The simulation setup includes:
- Units: MMKS (mm, kg, s).
- Material properties: density 7.85e-6 kg/mm³, Young’s modulus 2.1e5 MPa, Poisson’s ratio 0.3.
- Revolute joints at gear centers relative to ground.
- Impact contact parameters: stiffness K = 1.1e6 N/mm^(3/2), damping C = 50 N·s/mm, force exponent = 1.5, penetration depth = 0.1 mm.
- Input speed: step function from 0 to 154 rad/s (1470 rpm) in 0.1 s.
- Load torque: step function from 0 to −150 N·m (output gear) in 0.1 s.
- Simulation time: 0.5 s, steps: 4500, integrator: GSTIFF & SI2.
Figure 1 (conceptual) shows typical angular acceleration response of the output gear for the baseline (no slot-hole, backlash 0.2 mm). The steady-state vibration is extracted after 0.15 s. The time-domain signal is transformed to the frequency domain using FFT. The meshing frequency for 1470 rpm is \(f_m = \frac{1470}{60} \times 34 = 833\) Hz. The amplitude at this frequency is used as the primary vibration indicator.
We also compute the power spectral density (PSD) using the correlogram method. The autocorrelation function \(\hat{r}(m)\) for a discrete signal \(u_N(n)\) of length \(N\) is:
$$
\hat{r}(m) = \frac{1}{N} \sum_{n=0}^{N-1} u_N(n) u_N^*(n-m), \quad |m| \le N-1
$$
The power spectrum estimate \(\hat{S}(\omega)\) is:
$$
\hat{S}(\omega) = \sum_{m=-M}^{M} \hat{r}(m) e^{-j\omega m}
$$
With sampling frequency 5500 Hz and 1024 points, the total power (integral of PSD) is used as a secondary metric. For the baseline straight spur gear, the PSD total power is \(2.515 \times 10^8\) rad²/s⁴.
Orthogonal Experiment Design and Results
Three factors are considered: slot width \(A\) (mm), backlash \(B\) (mm), and hole diameter \(C\) (mm). Each factor has three levels as shown in Table 2.
| Factor | Level 1 | Level 2 | Level 3 |
|---|---|---|---|
| A (slot width, mm) | 0.2 | 0.3 | 0.4 |
| B (backlash, mm) | 0.2 | 0.07 | −0.03 |
| C (hole diameter, mm) | 4 | 6 | 8 |
Using an L9 orthogonal array, we simulate nine combinations (Table 3).
| Run | A (mm) | B (mm) | C (mm) | Angular acceleration (rad/s²) | Total power (×10⁸ rad²/s⁴) |
|---|---|---|---|---|---|
| 1 | 0.2 | 0.2 | 4 | 201 | 2.683 |
| 2 | 0.2 | 0.07 | 8 | 195 | 2.590 |
| 3 | 0.2 | −0.03 | 6 | 162 | 1.984 |
| 4 | 0.3 | 0.2 | 8 | 224 | 3.247 |
| 5 | 0.3 | 0.07 | 6 | 184 | 2.338 |
| 6 | 0.3 | −0.03 | 4 | 156 | 1.859 |
| 7 | 0.4 | 0.2 | 6 | 211 | 2.921 |
| 8 | 0.4 | 0.07 | 4 | 177 | 2.244 |
| 9 | 0.4 | −0.03 | 8 | 173 | 2.185 |
| Baseline | — | 0.2 | — | 194 | 2.515 |
Runs 3, 5, 6, 8, and 9 show lower angular acceleration than the baseline, with run 6 being the best (156 rad/s², 22.7% reduction).
Range analysis is performed on the angular acceleration at meshing frequency. The average values for each factor level are given in Table 4.
| Factor | T1 | T2 | T3 | Range S | Order of influence |
|---|---|---|---|---|---|
| A | 186 | 188 | 187 | 2 | B > C > A |
| B | 212 | 185 | 164 | 48 | |
| C | 178 | 186 | 197 | 19 |
Backlash (B) has the largest range (48 rad/s²), followed by hole diameter (C, 19 rad/s²), while slot width (A) has negligible effect (2 rad/s²). The signal-to-noise ratio (S/N) for smaller-the-better is computed as:
$$
\eta = -10 \log\left( \frac{1}{n} \sum_{i=1}^n y_i^2 \right)
$$
For each run, \(n=1\). The S/N values confirm the same order: B > C > A. The optimal combination from S/N analysis is A=0.2 mm, B=−0.03 mm, C=4 mm.
Analysis of variance (ANOVA) is summarized in Table 5.
| Source | DF | SS | MS | F | Contribution (%) |
|---|---|---|---|---|---|
| A | 2 | 1.71e4 | 8.55e3 | 1.77 | 0.13 |
| B | 2 | 1.15e7 | 5.75e6 | 1189** | 85.84 |
| C | 2 | 1.87e6 | 9.35e5 | 193.4** | 13.96 |
| Error | 2 | 9.67e3 | 4.84e3 | — | 0.07 |
| Total | 8 | 1.34e7 | — | — | 100 |
** indicates significance at α=0.05 (F0.05(2,2)=19.0). Backlash contributes 85.84% to the total variation, while hole diameter contributes 13.96%. Slot width is insignificant.
Optimal Parameter Validation by Simulation
The optimal combination (A=0.2 mm, B=−0.03 mm, C=4 mm) corresponds to run 6, which already shows the best performance. We further confirm by running a dedicated simulation with that exact geometry. The steady-state angular acceleration time history and spectrum are obtained. The amplitude at 833 Hz is 153 rad/s², which is 21.1% lower than the baseline (194 rad/s²). The total power from the PSD is \(2.0 \times 10^8\) rad²/s⁴, a reduction of 20.4% compared to the baseline.
Experimental Verification
To validate the simulation, we manufactured four straight spur gears with different base tangent lengths to achieve three backlash levels (Table 6). Gear C has the slot-hole structure with A=0.2 mm, C=4 mm (optimal). The gear materials and heat treatment are identical to the simulation.
| Gear ID | Base tangent length (mm) | Slot-hole | Role |
|---|---|---|---|
| A | 84.075 | No | Input |
| B | 84.101 | No | Output (pair with A: backlash 0.2 mm) |
| C | 84.192 | Yes (0.2 mm slot, 4 mm hole) | Input (pair with B: backlash 0.07 mm) |
| D | 84.199 | No | Output (pair with C: backlash −0.03 mm) |
The test rig consists of a drive motor, gearbox, torque-speed transducer, and magnetic powder brake. Four PCB accelerometers are mounted on the bearing housings (channels: 1 input radial, 2 input axial, 3 output radial, 4 output axial). Data acquisition is performed using an M+P system. Tests are run at two speeds (780 rpm and 1470 rpm) and three loads (20, 100, 150 N·m).
Contact pattern inspection is performed before measurement to ensure proper tooth contact. All pairs show acceptable patterns without edge contact.
Tables 7 and 8 show the vibration acceleration amplitudes at the fundamental mesh frequency and its first harmonic for the three gear pairs under different conditions. Here we present representative data for channel 2 (input axial) and channel 4 (output axial) at the two speeds.
| Pair | Load (N·m) | Ch2 442Hz | Ch2 884Hz | Ch4 442Hz | Ch4 884Hz |
|---|---|---|---|---|---|
| A-B (baseline backlash 0.2 mm) | 20 | 1.751 | 6.146 | 1.008 | 1.745 |
| 100 | 2.305 | 6.327 | 1.847 | 4.154 | |
| 150 | 3.277 | 4.776 | 0.483 | 4.352 | |
| B-C (backlash 0.07 mm, gear C slotted) | 20 | 1.208 | 4.601 | 7.721 | 1.804 |
| 100 | 1.877 | 6.970 | 8.861 | 6.859 | |
| 150 | 3.792 | 2.625 | 3.591 | 5.197 | |
| C-D (backlash −0.03 mm, gear C slotted) | 20 | 1.159 | 1.272 | 0.816 | 0.165 |
| 100 | 1.821 | 0.358 | 1.316 | 0.202 | |
| 150 | 2.491 | 0.777 | 0.387 | 0.575 |
| Pair | Load (N·m) | Ch2 833Hz | Ch2 1666Hz | Ch4 833Hz | Ch4 1666Hz |
|---|---|---|---|---|---|
| A-B | 20 | 1.686 | 2.462 | 2.969 | 2.873 |
| 100 | 0.194 | 1.007 | 0.161 | 4.502 | |
| 150 | 0.254 | 0.874 | 0.256 | 2.442 | |
| B-C | 20 | 3.029 | 1.925 | 2.030 | 6.138 |
| 100 | 0.286 | 0.747 | 0.225 | 3.131 | |
| 150 | 0.231 | 0.850 | 0.228 | 5.419 | |
| C-D | 20 | 1.349 | 1.625 | 2.079 | 0.642 |
| 100 | 0.145 | 0.388 | 0.117 | 0.492 | |
| 150 | 0.196 | 0.470 | 0.180 | 0.469 |
Comparing pair C-D (optimal slot-hole with negative backlash) against pair A-B (baseline), we observe consistent reductions. At 780 rpm and 150 N·m, the fundamental frequency amplitude at channel 2 decreases from 3.277 m/s² to 2.491 m/s², a reduction of 24.0%. At channel 4, the reduction is from 0.483 m/s² to 0.387 m/s² (19.9%). At 1470 rpm and 150 N·m, channel 2 drops from 0.254 m/s² to 0.196 m/s² (22.8%), and channel 4 from 0.256 m/s² to 0.180 m/s² (29.7%). The overall reduction range across all test conditions is 19.2%–33.8%, in good agreement with the simulation prediction of 21.1%.
Furthermore, pair C-D consistently outperforms pair B-C, confirming that negative backlash combined with the slot-hole yields the best vibration suppression. The slot-hole structure in gear C enables the negative backlash to be used without causing tooth jamming, thanks to the increased compliance.
The experimental results also show that the first harmonic (2nd order) is significantly reduced for the optimal design, indicating better meshing smoothness.
Discussion
The dominant role of backlash in straight spur gear vibration is evident from both simulation and experiment. Traditional wisdom suggests that increasing backlash reduces the risk of tooth interference but may increase impact vibration due to larger clearance. Here we find that reducing backlash toward negative values, when combined with a flexible tooth structure, actually suppresses vibration. The slot-hole provides the necessary local deformation to absorb the geometric interference, effectively acting as a built-in compliance element. The hole diameter also plays a secondary but significant role, likely because it alters the stress distribution and local stiffness gradient. The slot width, within the tested range, has virtually no effect; this suggests that any narrow slot that is deep enough to decouple adjacent teeth is sufficient.
The practical implication is that designers of straight spur gears for high-speed, low-noise applications can specify a small negative backlash (interference fit) together with slot-holes to achieve both zero clearance and reduced vibration. This approach avoids the complexity of active backlash control or expensive profile modifications.
Conclusions
In this study, we have systematically investigated a slot-hole vibration damping structure for straight spur gear through simulation and experiment. The main findings are:
- The proposed slot-hole structure reduces the vibration acceleration of straight spur gear by 20%–33% under typical operating conditions. The reduction is robust across speeds and loads.
- Backlash is the most influential parameter, contributing 85.84% to the total vibration variation. Negative backlash (−0.03 mm) combined with the slot-hole yields the best performance.
- Hole diameter contributes 13.96%, while slot width (0.2–0.4 mm) has negligible influence. The optimal parameter combination is slot width 0.2 mm, backlash −0.03 mm, and hole diameter 4 mm.
- The optimal design reduces the meshing frequency amplitude by 21.1% in simulation and 19.2%–33.8% in experiment, confirming the validity of the approach.
- The slot-hole structure enables the use of negative backlash without compromising gear operation, providing a practical solution for precision, low-vibration straight spur gear applications.
Future work will explore the influence of slot depth, tooth width variations, and the extension of this concept to helical and bevel gears.