Desktop CNC – Work In Progress

To make my projects smaller and more durable, it would be better to make custom PCB’s rather than jamming a wad of wires into already small cases.

To make this possible, I’ve made a desktop CNC. To keep the costs down, it will be made using mostly surplus parts. To cut back on costs further the sides and 3D printed parts where redesigned to fit 8mm smooth rods, rather than 12mm, which are considerably more expensive and harder to find.

There are many revisions of the CNC design I chose and I had to piece together parts from 2 or more of them, notably the X-carriage – most of them where designed as a single piece, which would need supports.

I increased thickness of most of the CNC parts to 9mm (from 6mm). During the stepper motor testing, I noticed that I should have done the same for the front and back, possibly eliminating the need to use aluminium rails at the top and bottom of them.

Unlike the controller used in the instructible, I’m using a Ramps 1.4 board to control the steppers, mainly due to familiarity and flexibility. At this stage I’m not sure how the gerber files, and how well they will be converted to g-code.

Initial tests with the motor running, chuck and engraver bit installed where slightly unexpected. because the bit is ground in half, it’s off center so the tip spins in a circle rather than on it’s tip. Making thin precise lines impossible. I tried another balanced bit and it was far better, but not perfect.

UPDATE: 15-8-2019

After buying another motor, 3 different chucks, and a few different router bits, I’ve replaced the spindle with an old Ozito Projex with an an extension ‘hose’ to lower the weight and increase the precision of the spinning router bits. With all 3 chucks and both motors, the router bit tip would spin in a small circle and get larger with higher speeds. This resulted in large grooves where it should be engraving a fine line.

New tool mount for the Ozito Projex hand piece.

The first tests looked good and the ‘Z’ end stop – a couple of alligator clips completing a circuit between the router bit and copper plate, worked great.

Next is to test the auto leveling and fix the slack in the threaded rod couplers – all 3 axis’s have a 1-2mm play on them. The Z axis is quite bad.

This is a test design, the red areas are machine or milling faults, blue is design flaws.

Leave a Reply

Your email address will not be published. Required fields are marked *