A Software-Only Post-Processor for Indexed Rotary Machining on GRBL-Based CNCs

Affordable desktop CNC routers are common in education, prototyping, and makerspaces, but most lack a rotary axis, limiting fabrication of rotationally symmetric or multi-sided parts. Existing solutions often require hardware retrofits, alternative controllers, or commercial CAM software, raising cost and complexity. This work presents a software-only framework for indexed rotary machining on GRBL-based CNCs. A custom post-processor converts planar toolpaths into discrete rotary steps, executed through a browser-based interface. While not equivalent to continuous 4-axis machining, the method enables practical rotary-axis fabrication using only standard, off-the-shelf mechanics, without firmware modification. By reducing technical and financial barriers, the framework expands access to multi-axis machining in classrooms, makerspaces, and small workshops, supporting hands-on learning and rapid prototyping.

💡 Research Summary

The paper presents a software‑only solution that enables indexed rotary machining on low‑cost GRBL‑based 3‑axis CNC routers without any hardware modifications. Recognizing that affordable desktop CNC machines are widely used in education, prototyping, and makerspaces yet typically lack a rotary axis, the authors develop a custom post‑processor that transforms conventional planar X‑Z toolpaths into a sequence of discrete rotary steps interleaved with the original cutting moves. The core of the system is a Python script that parses standard GRBL‑style G‑code, extracts X and Z coordinates, removes any existing Y commands, and inserts a rapid Y‑axis move (G0 Yθ) before each pass, where θ is the angular increment calculated from the user‑specified number of indexing steps (N). This approach repurposes the unused Y‑axis of a typical GRBL controller as a virtual rotary axis, effectively emulating a fourth axis through stepwise rotation.

To make the workflow accessible, the authors provide both a desktop graphical user interface (GUI) and a browser‑based web interface. Users upload a planar G‑code file, specify parameters such as stock radius, desired angular resolution, feed‑rate overrides, and the system automatically extracts tool diameter, feed rate, and spindle speed from G‑code comments. Real‑time validation checks numeric ranges, and a 3‑D preview of the revolved geometry is generated alongside the original 2‑D toolpath, allowing users to verify the conversion before machining. The web service processes files in isolated containers, discards uploads after conversion, and imposes a modest file‑size limit for security.

Hardware implementation is deliberately minimal: a 2‑phase NEMA 23 stepper motor drives a timing belt and pulley attached to a 63 mm three‑jaw chuck mounted on a rigid aluminum fixture. A 1:1 pulley ratio ensures that each motor step corresponds directly to the angular increment computed by the software; alternative gear ratios can be accommodated by adjusting a scaling factor in the post‑processor. Safety routines automatically insert spindle stop and Z‑axis retraction commands during each indexing move to prevent collisions.

The methodology was validated by machining solid of revolution test parts from hardwood and copper. Both parts used the same Fusion 360‑generated X‑Z toolpath, which was processed with N = 80 (θ ≈ 4.5°). Measured dimensions showed a deviation of –0.20 mm for wood and +0.25 mm for copper, well within the ±0.25 mm tolerance claimed. A comparative experiment with a coarse indexing setting (N = 22, θ ≈ 16.4°) demonstrated visible faceting and scalloping, confirming the expected trade‑off between surface finish, machining time, and program size. The authors report that the 3‑D preview feature significantly reduced setup errors and improved user confidence.

In the discussion, the authors argue that while the approach does not replace true continuous rotary axes—lacking torque continuity, dynamic compensation, and the ability to produce high‑precision surfaces—it offers a cost‑effective alternative for educational and low‑volume prototyping contexts. By leveraging existing 3‑axis hardware and open‑source software, the barrier to multi‑axis machining is lowered, aligning with the broader trend toward democratized digital fabrication. Limitations include surface artifacts inherent to discrete indexing, dependence on accurate machine calibration, and the absence of real‑time feedback for adaptive feed control.

Future work is outlined to address these limitations: exploring semi‑continuous rotation schemes, integrating sensor‑based feedback for torque and vibration monitoring, extending compatibility to a wider range of stepper drivers and controller boards, and employing AI techniques to automatically select optimal indexing resolution based on part geometry and material properties.

The conclusion reiterates that a simple Python post‑processor, combined with a modest mechanical rotary attachment, can extend GRBL routers to perform practical rotary machining with dimensional accuracy suitable for most prototyping tasks. The contribution lies in providing an open, low‑cost, software‑driven pathway to multi‑axis capability, thereby expanding the educational and decentralized manufacturing potential of affordable CNC platforms.