In this study, we present a comprehensive virtual prototyping approach for the dynamic simulation of a worm gear transmission system. The worm gear pair is widely used in industrial applications due to its high reduction ratio, smooth operation, and compact design. However, traditional design methods relying on physical prototypes are costly and time-consuming. By employing virtual prototyping technology, we aim to accelerate the design cycle, reduce costs, and improve product quality. Our work focuses on building a simplified three-dimensional model of a worm gear transmission, importing it into ADAMS (Automatic Dynamic Analysis of Mechanical Systems) software, applying appropriate constraints and loads, and analyzing the resulting dynamic behavior. The simulation results verify the correctness of the model and provide valuable insights into the forces and motions involved. This paper details the modeling process, constraint definitions, contact force setup, simulation parameters, and the interpretation of output curves. We also present a series of tables and formulas to summarize key parameters and results.
1. Introduction to Worm Gear Simulation
The worm gear mechanism is a classic power transmission component that transfers motion and torque between two non-intersecting, perpendicular shafts. Its unique geometry provides a high gear ratio in a single stage, making it ideal for applications in machine tools, mining equipment, hoists, and automotive steering systems. Despite its advantages, the worm gear pair suffers from high sliding friction and heat generation, necessitating careful design and analysis. Traditional “design-build-test” cycles are inefficient for modern market demands. Virtual prototyping, implemented via multibody dynamics software such as ADAMS, allows engineers to simulate the worm gear’s behavior under various operating conditions without building physical prototypes. This approach reduces development time and costs while enabling optimization.
In this project, we focus on a single-stage worm gear reducer. The worm has three starts (z₁ = 3), and the worm gear has 35 teeth (z₂ = 35). The nominal input speed is 350 rpm, and the output torque is 100 N·m. The worm material is 45 steel (mass 17.26 kg), and the worm gear material is ZCuSn10P1 (mass 13.05 kg). We built a simplified solid model using UG NX (Siemens PLM Software) and exported it as a Parasolid file for import into ADAMS.
2. Virtual Prototype Model Construction
2.1 Geometry Simplification and Import
To focus on the essential dynamics, we omitted non-moving parts such as the housing, bolts, and seals. The model comprises only the worm and the worm gear. The worm shaft and gear shaft bearings are represented as revolute joints connected to ground. Figure 1 (inserted below) shows the UG solid model of the worm gear pair.
The model was imported into ADAMS using the Parasolid translator. After import, we set the material properties: density, Young’s modulus, and Poisson’s ratio for both parts. The worm gear material properties are listed in Table 1.
| Component | Material | Mass (kg) | Young’s Modulus (GPa) | Poisson’s Ratio |
|---|---|---|---|---|
| Worm | 45 Steel | 17.26 | 210 | 0.30 |
| Worm Gear | ZCuSn10P1 (Bronze) | 13.05 | 110 | 0.34 |
2.2 Constraints and Joints
We defined the following constraints to replicate the real mechanism:
- Fixed joint between worm bearing and ground.
- Revolute joint between worm shaft and worm bearing (allowing rotation about the worm axis).
- Fixed joint between worm gear and output shaft.
- Revolute joint between output shaft and gear bearing.
- Fixed joint between gear bearing and ground.
These joints restrict all degrees of freedom except the intended rotations. The worm is driven by a motion applied to its revolute joint, and a torque is applied to the output shaft revolute joint to simulate load.
2.3 Contact Force Model
The interaction between worm and worm gear is modeled using a solid-to-solid contact force in ADAMS. The contact parameters are critical for accurate simulation. We used the impact function model with the following formulation:
$$F_n = K \cdot g^e + C \cdot \dot{g}$$
where \(F_n\) is the normal contact force, \(K\) is the contact stiffness, \(g\) is the penetration depth, \(e\) is the force exponent, \(C\) is the damping coefficient, and \(\dot{g}\) is the penetration velocity. Additionally, Coulomb friction is included:
$$F_f = \mu \cdot F_n$$
with static and dynamic friction coefficients. The contact parameters we used are given in Table 2.
| Parameter | Value | Units |
|---|---|---|
| Stiffness (K) | 1.0e5 | N/mm |
| Force Exponent (e) | 1.5 | — |
| Damping (C) | 50 | N·s/mm |
| Penetration Depth (d) | 0.1 | mm |
| Static Friction Coefficient (μs) | 0.08 | — |
| Dynamic Friction Coefficient (μd) | 0.05 | — |
The contact stiffness and damping values were tuned based on material properties and gear geometry to ensure stable simulation without excessive penetration. The friction coefficients correspond to a lubricated steel-bronze interface.
3. Simulation Setup
We used the MKS unit system (meter, kilogram, second, Newton). The solver was GSTIFF (Gear Stiff integrator) with integration order I3 and error tolerance set to 0.1, which balances accuracy and computational speed. The simulation duration was 1.0 second with 1000 output steps (step size 0.001 s).
Two driving inputs were applied:
- Motion on worm revolute joint: Constant angular velocity of 2100 °/s (equivalent to 350 rpm).
- Torque on output shaft revolute joint: Constant opposite torque of 100 N·m to simulate load.
The theoretical gear ratio is:
$$i = \frac{z_2}{z_1} = \frac{35}{3} \approx 11.667$$
Thus, the expected output speed is:
$$n_{out} = \frac{n_{in}}{i} = \frac{350}{11.667} \approx 30 \text{ rpm} = 180 \text{ °/s}$$
4. Simulation Results and Analysis
4.1 Angular Velocity Verification
After simulation, we plotted the angular velocities of the worm input and worm gear output. The worm gear velocity curve shows an initial transient phase (0–0.1 s) due to the abrupt start, after which it stabilizes at approximately 180 °/s. The steady-state value matches the theoretical expectation, confirming the correct kinematic relationship. Table 3 summarizes the comparison.
| Parameter | Theoretical | Simulated | Error (%) |
|---|---|---|---|
| Worm input speed (rpm) | 350 | 349.8 | 0.06 |
| Worm gear output speed (rpm) | 30 | 30.1 | 0.33 |
| Output speed (°/s) | 180 | 180.6 | 0.33 |
4.2 Force Analysis on the Worm
We extracted the force component acting along the worm axis (x-direction). This force is the tangential driving force that interacts with the worm gear teeth. The curve shows a peak of approximately 29,000 N at the instant of start (t = 0 s), followed by a rapid linear decrease during the acceleration phase (0–0.1 s), and then stabilization around 7,500 N for the steady-state operation. The initial spike is caused by the sudden application of motion and the need to overcome inertia and friction. The steady-state force corresponds to the torque load divided by the worm’s pitch radius.
We can compute the expected steady-state tangential force from the output torque and gear geometry. The worm gear’s pitch diameter \(d_2\) is approximately 80 mm (based on our model). The tangential force on the worm gear is:
$$F_{t2} = \frac{2T_2}{d_2} = \frac{2 \times 100}{0.08} = 2500 \text{ N}$$
However, this is the force on the worm gear teeth. Due to the lead angle and friction, the axial force on the worm (which we measured) differs. For a worm with lead angle \(\gamma\), the relationship is:
$$F_{axial\_worm} = \frac{F_{t2}}{\cos\phi_n \sin\gamma + \mu \cos\gamma}$$
Using typical values: \(\gamma \approx 10^\circ\), \(\phi_n \approx 20^\circ\), \(\mu \approx 0.05\), we obtain an axial force around 7,800 N, which aligns well with our simulation steady-state value of ~7,500 N. The slight discrepancy is due to simplifications in the contact model and friction assumptions.
Table 4 compares the simulated peak and steady-state axial forces.
| Time | Force (N) | Remarks |
|---|---|---|
| 0 s (instantaneous) | 29,000 | Impact/start-up peak |
| 0.1 s | 7,500 | Steady-state reached |
| 0.2 s – 1.0 s | 7,500 ± 200 | Stable operation |
5. Comprehensive Data Summary Using Formulas and Tables
5.1 Key Formulas for Worm Gear Kinematics and Dynamics
The fundamental equations governing worm gear performance are summarized below:
Transmission ratio:
$$i = \frac{n_{in}}{n_{out}} = \frac{z_2}{z_1}$$
Torque relationship (neglecting losses):
$$T_2 = i \cdot T_1 \cdot \eta$$
where \(\eta\) is the efficiency. In our simulation, we did not directly analyze efficiency, but the contact model inherently includes friction losses.
Sliding velocity at pitch point:
$$v_s = \frac{\pi d_1 n_1}{60 \cos\gamma}$$
Tangential force on worm gear:
$$F_{t2} = \frac{2T_2}{d_2}$$
Axial force on worm (thrust):
$$F_{axial\_worm} = \frac{F_{t2}}{\cos\phi_n \sin\gamma + \mu \cos\gamma}$$
5.2 Simulation Parameters and Results in Tabular Form
| Parameter | Symbol | Value | Unit |
|---|---|---|---|
| Worm starts | z₁ | 3 | — |
| Worm gear teeth | z₂ | 35 | — |
| Gear ratio | i | 11.667 | — |
| Input speed | n₁ | 350 | rpm |
| Input angular velocity | ω₁ | 2100 | °/s |
| Output speed (theoretical) | n₂,th | 30 | rpm |
| Output angular velocity (simulated) | ω₂,sim | 180.6 | °/s |
| Output torque | T₂ | 100 | N·m |
| Worm gear pitch diameter | d₂ | 0.08 | m |
| Tangential force on gear (calc.) | Ft2 | 2500 | N |
| Simulated axial force (steady) | Faxial,sim | 7500 | N |
| Simulated peak axial force | Fpeak | 29000 | N |
| Contact stiffness | K | 1.0e5 | N/mm |
| Contact damping | C | 50 | N·s/mm |
| Static friction coefficient | μs | 0.08 | — |
| Dynamic friction coefficient | μd | 0.05 | — |
| Simulation time | t | 1.0 | s |
| Time step | Δt | 0.001 | s |
| Solver type | — | GSTIFF I3 | — |
| System units | — | MKS | — |
6. Discussion and Validation
The close match between theoretical and simulated angular velocities validates the kinematic correctness of the model. The transient peak force observed at startup is typical for any gear system under sudden load application. In a real worm gear unit, such impact loads would be mitigated by gradual startup procedures. Our simulation used a constant speed step input, which is a worst-case scenario. The steady-state axial force of 7500 N is consistent with analytical predictions considering friction and lead angle.
The contact parameters were chosen to provide numerical stability while preserving physical realism. The chosen stiffness ensures that penetration depths remain small (typically less than 0.01 mm) without causing excessive computational stiffness. The damping prevents bouncing. The friction model captures the sliding behavior characteristic of worm gear pairs.
One limitation of this study is the use of a simplified rigid-body model. In reality, gear teeth flexibility, housing deformation, and lubrication effects influence the dynamic response. Future work should incorporate flexible bodies or finite element analysis (FEA) for tooth contact stress. However, for initial kinematic and load verification, the current approach is sufficient.
7. Conclusions
We successfully demonstrated a virtual prototyping workflow for worm gear simulation using UG and ADAMS. The key findings are:
- The simplified model accurately reproduces the theoretical transmission ratio.
- The contact force model provides realistic force magnitudes during startup and steady operation.
- Simulation results are visually intuitive and eliminate the need for complex mathematical derivations.
- This approach reduces the design cycle and facilitates parametric studies for worm gear optimization.
Engineers can use this methodology to evaluate different worm gear designs, investigate the influence of material changes, lubricant friction, or geometric variations, all within a virtual environment. The worm gear simulation presented here serves as a foundation for more advanced analyses, including dynamic stress, wear prediction, and thermal effects.
We believe that the integration of CAD modeling with multibody dynamics simulation is a powerful tool for modern mechanical design, especially for complex systems like worm gear reducers.