In the realm of high-end manufacturing, the assembly process stands as a critical phase, often consuming a significant portion of the total production workload. Precision components, such as K-type precision lock nuts, are extensively employed in advanced machinery like five-axis linkage machine tools and CNC machining centers to ensure assembly accuracy and reliable positioning. These nuts are characterized by their robust locking capability, making them indispensable for secure fastening in precision mechanisms. However, their manual installation and removal present challenges, including low productivity, difficulty in confined spaces, and reliance on multiple specialized tools. To address these issues, the integration of industrial robots for automated handling emerges as a compelling solution, offering the potential to alleviate labor intensity, enhance precision, and improve overall efficiency. This article, from my perspective as a researcher in advanced manufacturing technology, delves into the design and comprehensive analysis of a dedicated end effector tailored for the automated assembly and disassembly of K-type precision lock nuts. The end effector is conceived to meet the specific工艺 requirements of these nuts, leveraging robotic capabilities for tasks such as recognition, positioning, and torque-controlled fastening.
The K-type precision lock nut, a standardized component detailed in specifications such as GJB859.37-1990, encompasses a range of models from KM18×1.5P to KM300×4.0P. For this study, the focus is on the series from KM27×1.5P to KM42×1.5P, which exemplify the structural nuances and loading demands typical of these fasteners. The nut comprises a front and rear section separated by a deep groove, and its installation involves two sequential steps: pre-tightening the nut body to a specified torque and then securing four internal hexagon socket head cap screws. Conversely, disassembly requires first loosening these screws before unscrewing the nut body. The structural parameters for this series are summarized in the table below, highlighting key dimensions and load capacities.
| Thread Designation | Outer Diameter D (mm) | Inner Diameter d (mm) | Pitch Circle Diameter c (mm) | Height h (mm) | Number of Positioning Holes | Positioning Hole Diameter (mm) | Internal Hex Screw Specification | Screw Tightening Torque (Nm) | Axial Load Capacity (kN) |
|---|---|---|---|---|---|---|---|---|---|
| KM27×1.5 | 46 | 43 | 37 | 20 | 4 | 5 | 4-M4×14 | 3.5 | 153 |
| KM28×1.5 | 46 | 43 | 37 | 20 | 4 | 5 | 4-M4×14 | 3.5 | 153 |
| KM30×1.5 | 48 | 45 | 39 | 20 | 4 | 5 | 4-M4×14 | 3.5 | 160 |
| KM32×1.5 | 50 | 47 | 41 | 22 | 4 | 5 | 4-M4×16 | 3.5 | 164 |
| KM35×1.5 | 53 | 50 | 44 | 22 | 4 | 5 | 4-M4×16 | 3.5 | 164 |
| KM38×1.5 | 56 | 53 | 47 | 22 | 4 | 5 | 4-M4×16 | 3.5 | 142 |
| KM40×1.5 | 58 | 55 | 49 | 22 | 4 | 5 | 4-M4×16 | 3.5 | 218 |
| KM42×1.5 | 60 | 55 | 51 | 22 | 4 | 5 | 4-M4×16 | 3.5 | 229 |
As observed, these nuts feature four positioning holes with a diameter of 5 mm, arranged symmetrically at 90-degree intervals on a pitch circle diameter (c) ranging from 37 mm to 51 mm. The axial load capacity varies from 153 kN to 229 kN, and the internal hex screws each require a modest tightening torque of 3.5 Nm. The presence of these positioning holes offers a strategic engagement point for an end effector, allowing force application in a manner that mitigates spatial constraints. Moreover, the variability in pitch circle diameter across nut sizes necessitates an adjustable feature in the end effector to accommodate different specifications without frequent tool changes, thereby boosting automation efficiency.
To design an effective end effector, a thorough understanding of the torque requirements for both tightening and loosening the nut body is essential. The analysis employs the “torque method,” considering the frictional moments in the thread pair and at the bearing surface. For the largest nut in the series, KM42×1.5P, which represents the worst-case loading scenario, the tightening torque T is the sum of the thread friction torque T1 and the bearing surface friction torque T2. The thread friction torque is derived from the mechanical model of a wedge slider on an inclined plane, representing the nut moving relative to the screw thread. For a non-rectangular thread with a thread angle α = 60°, the equivalent friction coefficient f_v and equivalent friction angle ρ_v are introduced:
$$ f_v = \frac{f}{\sin \delta} = \frac{f}{\cos(\alpha/2)} $$
where f is the coefficient of friction (approximately 0.04), and δ = 90° – α/2. Thus,
$$ ρ_v = \arctan(f_v) $$
The thread lead angle λ is calculated from the pitch p and the pitch diameter d2:
$$ λ = \arctan\left(\frac{S}{\pi d_2}\right) = \arctan\left(\frac{n p}{\pi d_2}\right) $$
where S is the lead, n is the number of thread starts (typically 1 for these nuts), and p = 1.5 mm. For KM42×1.5P, d2 = 41.026 mm, yielding λ ≈ 0.687° and ρ_v ≈ 2.634°. The thread friction torque during tightening is:
$$ T_1 = Q \frac{d_2}{2} \tan(λ + ρ_v) $$
where Q is the axial load (229 kN). Substituting values gives T1 ≈ 272.5 Nm. The bearing surface friction torque T2 is given by:
$$ T_2 = \frac{1}{3} μ_W Q \frac{D_0^3 – d_0^3}{D_0^2 – d_0^2} $$
where D0 = 55 mm (outer contact diameter), d0 = 40.376 mm (inner contact diameter), and μ_W = 0.04 (bearing friction coefficient). This yields T2 ≈ 220.21 Nm. Thus, the total tightening torque is:
$$ T = T_1 + T_2 ≈ 492.71 \, \text{Nm} $$
For loosening, the thread friction torque reverses direction, and the loosening torque T’ is:
$$ T’ = Q \frac{d_2}{2} \tan(λ – ρ_v) + T_2 $$
which computes to T’ ≈ 64.37 Nm. These torque values, particularly the substantial tightening torque, underscore the need for a robust end effector capable of transmitting high rotational forces while maintaining structural integrity. The significant difference between tightening and loosening torques also informs the control strategy for the robotic system.
Guided by these technical insights, I developed a comprehensive design for the end effector. The primary objectives were to ensure access to confined installation spaces, provide adaptability to multiple nut sizes, minimize weight and dimensions to reduce wrist load on the industrial robot, and integrate seamlessly with robotic operations via a flange connection. The end effector is a modular assembly consisting of a base body, actuation ends, torque sensors, and a transmission mechanism, all engineered for precision and reliability.
The base body features a cavity structure to reduce mass, with guide slots for precise linear movement of the actuation ends. The transmission system employs a symmetric design utilizing bevel gears and lead screws to enable adjustable spacing between the actuation ends. Specifically, a main motor drives a right-handed bevel gear that meshes with two left-handed bevel gears, converting rotational motion into linear displacement via lead screws and nuts. This allows the distance between the two actuation ends to be adjusted according to the pitch circle diameter of the target nut, ensuring alignment with the positioning holes. Each actuation end is equipped with a torque sensor and is driven by an auxiliary motor for tightening or loosening the internal hex screws. The entire end effector mounts to the robot’s wrist flange, facilitating easy integration and control through robotic programming.
In operation, the robot uses 3D vision to identify the nut’s position and orientation. The end effector first adjusts its actuation ends to engage the positioning holes. The robot’s sixth axis then rotates to pre-tighten the nut body to the required torque, monitored by the torque sensors to prevent over-tightening. Subsequently, the actuation ends are repositioned to align with the internal hex screw holes, and the auxiliary motors drive them to tighten each screw sequentially to the specified torque of 3.5 Nm. The disassembly sequence reverses these steps. This design not only streamlines the process but also enhances repeatability and reduces human error, making it a viable solution for automated production lines.
To validate the structural adequacy of the end effector, I conducted a static structural analysis using ANSYS Workbench. The model was simplified by removing non-essential features like small fillets and threaded holes to streamline meshing and computation. Materials were assigned as follows: the base body from aluminum alloy 6061 (yield strength ≥ 110 MPa, density 2.75 g/cm³, Poisson’s ratio 0.33), the actuation ends from 0Cr18Ni9 stainless steel (yield strength ≥ 110 MPa, tensile strength ≥ 520 MPa, density 7.93 g/cm³, Poisson’s ratio 0.3), and other components from structural steel. A fixed support constraint was applied to the mounting surface, and a moment load of 492.71 Nm, corresponding to the maximum tightening torque, was imposed on the actuation ends to simulate the worst-case scenario.
The analysis yielded stress, strain, and deformation contours. The equivalent (von-Mises) stress distribution indicated that the maximum stress of 61.86 MPa occurs at the connection region of the actuation ends where rotational force is applied. This value is well below the yield strength of the materials, confirming that the end effector possesses sufficient strength to handle the operational loads without plastic deformation. The equivalent elastic strain was minimal, with a peak value of 0.0003 mm, signifying negligible deformation under load. Total deformation and directional deformations along the X, Y, and Z axes were also examined. The maximum total deformation was 0.0015 mm, located at the interface between the actuation ends and the base body, while directional deformations were as follows: X-axis: 0.000478 mm, Y-axis: 0.00148 mm, Z-axis: 0.00034 mm. All deformation values are extremely small (less than 0.15% of critical dimensions), ensuring that the end effector maintains dimensional stability and accuracy during nut fastening operations. These results affirm that the mechanical design meets rigidity and strength requirements, though attention to material selection and surface treatment at high-stress zones is recommended for long-term fatigue resistance.
Beyond static performance, dynamic behavior is crucial to avoid resonant vibrations that could compromise precision or cause premature failure. Therefore, I performed a modal analysis to determine the natural frequencies and mode shapes of the end effector. The governing equation for undamped free vibration is:
$$ M\ddot{y} + Ky = 0 $$
where M is the mass matrix, K is the stiffness matrix, y is the displacement vector, and \ddot{y} is the acceleration vector. Assuming harmonic motion y = y sin(ωt), the eigenvalue problem becomes:
$$ (K – ω^2 M)y = 0 $$
The eigenvalues ω_i^2 yield the natural frequencies ω_i. Solving this via finite element analysis in ANSYS provided the first six modes, summarized below.
| Mode Order | Natural Frequency (Hz) | Maximum Relative Displacement (mm) | Mode Shape Description |
|---|---|---|---|
| 1 | 41.061 | 1.306 | Swinging deformation along the X-axis (up-down) |
| 2 | 45.695 | 1.215 | Swinging deformation along the X-axis (left-right) |
| 3 | 53.425 | 3.416 | Torsional deformation around the X-axis |
| 4 | 68.679 | 3.445 | Tensile deformation along the Y-axis |
| 5 | 74.015 | 1.127 | Torsional deformation around the X-axis |
| 6 | 89.940 | 1.762 | Vertical vibration along the X-axis |
The results show that natural frequencies increase with mode order, ranging from 41.061 Hz to 89.94 Hz. The largest displacements occur in the third and fourth modes, with values of 3.416 mm and 3.445 mm, respectively, associated with torsional and tensile deformations. To assess resonance risk, the excitation frequency from the robot’s drive system must be considered. The end effector is rotated by the robot’s sixth axis, typically powered by an AC servo motor. For instance, a 5 kW motor like the TS0500C33, capable of delivering the required 492 Nm torque, operates at a speed n derived from:
$$ T = 9559 \frac{P}{n} $$
where P = 5 kW. Solving for n gives approximately 97.05 rpm. The corresponding excitation frequency f is:
$$ f = \frac{n}{60} = \frac{97.05}{60} ≈ 1.6175 \, \text{Hz} $$
However, this is the rotational frequency; for vibration analysis, harmonic frequencies related to motor pulsations or control signals might be higher. In practice, the dominant excitation frequencies during robotic motion are often below 10 Hz. Comparing this with the lowest natural frequency of 41.061 Hz reveals a substantial margin, indicating that the end effector is unlikely to experience resonance during normal operation. This separation ensures stable performance and validates the dynamic design of the end effector structure. Nonetheless, in applications where higher-frequency vibrations are present, further tuning or damping might be considered.
The design and analysis of this end effector demonstrate a holistic approach to automating the assembly of K-type precision lock nuts. By leveraging adjustable actuation ends, the end effector accommodates a range of nut sizes, reducing tooling changes and enhancing flexibility on the production floor. The integration of torque sensors allows for precise control over fastening operations, critical for maintaining joint integrity and preventing damage. From a structural standpoint, the static analysis confirms that the end effector can withstand the high torsional loads involved in nut tightening, with stress levels safely within material limits and deformations being negligible. Dynamically, the modal analysis reveals natural frequencies sufficiently distant from typical excitation sources, mitigating resonance risks and ensuring operational reliability.
In conclusion, the development of this specialized end effector addresses key challenges in automating the handling of K-type precision lock nuts. The design prioritizes adaptability, strength, and precision, aligning with the demands of modern manufacturing environments where robotics play an increasingly pivotal role. Future work could explore optimization of the end effector’s weight through topology refinement, incorporation of advanced materials like carbon composites, or enhanced control algorithms for adaptive torque management. Additionally, experimental validation through prototyping and testing would provide further insights into real-world performance and durability. As automation continues to evolve, innovative end effector solutions like this will be instrumental in boosting productivity, ensuring quality, and advancing the frontiers of intelligent manufacturing systems.