This study introduces a novel and simplified method for generating the cutter flank of an internal-meshing skiving cutter for machining male screw rotors. In this approach, the cutting edge is first positioned on the conjugate surface of the rotor. The cutter flank is then generated by extruding the cutting edge along an oblique direction, which simplifies the cutter geometry and enhances the feasibility of practical tool manufacturing. Subsequently, the complete cutter structure is constructed. A mathematical model employing the proposed skiving cutter for machining a male screw rotor on a dedicated CNC machine is then presented. Simulation results obtained using VERICUT software demonstrate that the proposed cutter can successfully machine a complete rotor with high surface accuracy. Furthermore, the distribution of the working clearance angle indicates that the clearance angle remains positive along the entire cutting edge during the skiving process, thereby preventing interference between the tooth flank and the finished surface.
This contribution presents an advanced unified robot description format and its usage in the Robot Operating System (ROS). The description uses a tree structure with additional constraints included by a new “constraint”-element. The constraints are defined using two cut frames, two constraint axes and a constraint type. This enables the description of closed-loop mechanisms in arbitrary configurations. Using a new element to describe the closed loop has the advantage that the existing packages do not break when using the proposed format. These packages just ignore the closed loops and use the underlying tree structure. However, the integration of parallel robots in the ROS Framework lacks the closure of the tree structure. A Joint State Transformer package has been developed to address this issue. This package solves the constraint equations using a given set of minimal coordinates to close the open tree structure. The solver uses the Newton-Raphson algorithm to solve the constraint equations. The proposal’s results are validated using the parallel kinematics kit PARA-ENGINEER.
Efficient cleaning methods for eyeglass lenses are required for longevity of lenses and for eye hygiene. Inconsistencies in manual cleaning of eyeglass lenses using microfiber cloth has led to the development of several commercial products, including those that use ultrasonic chambers. To mimic the cleaning action of a human user, a mechanized system has been proposed in this paper, which is driven by a pair of slider-crank mechanisms powered by a DC stepper motor. It produces controlled elliptical brushing motion patterns while reducing the chances of scratches. This design also incorporates a simpler technique for pre-cleaning and aims to be environmentally friendly by reducing the need for disposable wipes. Concepts from the principles of mechanism design and kinematic evaluation were used to finalize the architecture and to obtain the parameters required for several standard shapes of the eyeglass frames. After carrying out simulations using GeoGebra, one set of parameters was used to develop a physical system to test the mechanized system.
The manual production of Banh Gio in small- and medium-scale food processing facilities is time-consuming and highly dependent on workers’ skills, leading to low productivity and inconsistent product quality. To improve production efficiency and reduce reliance on manual labor, this study proposes the design, fabrication, and experimental evaluation of an automatic Banh Gio making machine. The system integrates feeding, wrapping, and tying mechanisms into a synchronized operating cycle. Performance evaluation experiments were conducted to compare the machine with the traditional manual method. Experimental results indicate that the proposed machine achieves 5.22 times higher productivity than manual production while maintaining operational stability and product shape uniformity. This study provides a practical mechanization solution suitable for small and medium scale traditional food-processing facilities.
Rotor-bearing-coupling systems frequently suffer from complex mechanical faults, notably shaft misalignment and mass imbalance, producing strongly nonlinear vibration signatures. This paper provides a systematic overview of dynamic modeling techniques, from classical analytical theories to advanced 3D multibody and Finite Element Methods (FEM), for multi-fault diagnosis. A key novelty of this study is highlighting the critical role of high-fidelity physics-based models as robust synthetic data generators. To demonstrate this, a generalized 12-DOF 3D multi-fault model is presented, successfully capturing complex nonlinear signatures. By serving as a high-fidelity synthetic data generator, the proposed model provides rich, multi-dimensional (12-DOF) labeled datasets for extreme multi-fault scenarios, addressing the critical industrial bottleneck of data scarcity. While the current study specifically bounds its scope to the mathematical formulation and synthetic data generation phase, it establishes the essential computational groundwork for future research to train emerging hybrid AI frameworks, such as Physics-Informed Neural Networks (PINNs) and FEM-assisted Transfer Learning.
