Work in Progress! I started this project to gain more in-depth knowledge of complex mechanical assemblies, material selection, and electronics. I recently received $300 from NYU's prototyping fund to implement and expand this project.
Amazon Result for "Small Desktop CNC machine"
CNC machines are expensive. This is the unfortunate reality I faced when I looked for a homemade CNC machine for project use. Entry-level desktop CNCs could reach costs of several hundred to thousands of dollars. This is not a price point that is reachable for a college student living on a tight budget. Affordable machines are available, but they often sacrifice rigidity, precision, or work area to hit a lower price point.
I want to build an open-source DIY CNC machine that closes this gap: a durable CNC machine built with 3D-printed components available in almost every MakerSpace and affordable hardware readily available online. The goal is to design a machine that can be assembled for a fraction of the cost of a commercial unit that can be purchased online. The goal isn't to match the level of tolerance precision, work volume, or material capabilities of a full-scale industrial CNC. It is, however, to hit the sweet spot that hobbyists and students actually need. It must have enough precision and rigidity to reliably cut through desired materials (wood, plastics, and soft metals), have a small enough footprint to fit on a desk or in a garage, and have a total cost low enough that it's genuinely accessible.
To better define hardware requirements and optimize for price, I first started with a list of project specifications and verification methods. The CNC that I'm hoping to design must be:
Cheap
Precise
Repeatable
Easily assembled
and easily manufacturable
I then looked for hardware requirements for the CNC, such as the linear guidance, drive mechanism, frame material, and bed material. To optimize for trade studies and formalize a way to determine the best option, I conducted a trade study. I weighed the functional requirements from most to least important and provided a 1-5 rating of each option in the categories, providing a weighted score at the end.
Based on design choices made with trade studies, I started to develop an initial CAD.
WIP CAD (Sep 2026)