The paper presents an automated system for embedding brass threaded inserts into 3D-printed parts, reducing the variability of manual methods (misalignment, insufficient heating, excessive force) and the limitations of directly printed small threads (M4 and below). Target applications include rapid prototyping, low-volume production, and functional parts in aerospace and automotive sectors, where reliable joints enable repeated assembly. The device reads insert coordinates from Onshape (Parametric cloud CAD) via mate connectors (custom orientation systems in Onshape), ensuring faithful design-to-execution transfer. Its kinematics use a compact H-structure with a single-belt XY stage driven by stepper motors, plus a four-screw Z axis for uniform vertical motion and high stiffness. Inserts are automatically fed from a gravity tube through a servo-actuated lever and separator, then embedded with a heated tool head monitored by a thermocouple—tuned to 225 °C for PLA and 265 °C for ABS. A capacitive touch sensor provides real-time depth control (6 mm for M3, 8 mm for M4), while an inverted vise secures the workpiece above the tool head for optimal access. Integrating CAD data with Python motion-control scripts yields a low-operator workflow that improves repeatability and scalability. Tests on 3D printed parts of PLA and ABS with 3 to 6 perimeters show positional accuracy of ±0.05–0.07 mm, cycle times of 3.8–4.6 s per insert, and repeatability with 0.025–0.035 mm standard deviation—improving efficiency over manual methods by 73–75%. Thanks to modular design, compatibility with standard CAD platforms, and potential for adaptive tool heads, the system is well-suited for integration into automated additive-manufacturing post-processing lines and broader industrial adoption.
Cazacu et al. (Sun,) studied this question.