How hard can it be. I mean, building a dasktop CNC Mini Mill from scratch, without access to an existing mill or lathe. If nothing else I'll learn something along the way...
There are a selection of bits of kit that I've picked up over the last few years, but one that's been looking for a home is a motor out of the back a fried tumble-drier which was lying in the garage of my in-laws. At the time I didn't know whether the motor was at fault, or whether it was some other part of the drier, but on a little inspection there doesn't seem to be any major signs of damage.
It's an single phase AC induction motor, which means it has the big practical upshot of being effectively constant speed - and fairly powerful compared to the DC motors used in most small DIY projects. The question is what do do with it?
For lots of the projects I've been working on recently I've been using 3D Printing to prototype lots of the parts - but that means in the "final" versions of those projects the parts are still made of 3D printed plastic - which limits what's possible in terms of both surface finish and materials. It's also doesn't help with the electronic side of things where being able to quickly prototype PCBs would also be immensely valuable.
An aside on engraving PCBs with a 3D Printer.
I've attempted milling PCBs before. It didn't go well.
The printer design I have is based on the Hypercube design by Tech2C. There's even a hint in the name that it might be convertable into a CNC mill in the title:
HyperCube 3D Printer/CNC
So I designed a different carriage mount for a dremel tool, and given that without the print head, a 3D printer is just a three dimensional positioning system which accepts G code, it's very similar to a CNC Mill. You can find the
stl files for the mounting bracket and original OpenSCAD files on thingiverse if you want to remix it, although I wouldn't recommend it for actual use to mill parts.
The design of the hypercube is aimed at being really, really fast. It achieves this by keeping the weight of the X carriage and rail really low, so that the moving mass is small enough to allow fast movement, and fast acceleration with minimal force and inertia. As part of this tradeoff, the X rails are made of carbon fibre, and while stiff enough to move the relatively light print head around, they are nowhere near stiff enough to withstand the forces of a cutting head or the weight of an AC motor and cable hanging off the end. To make things worse, the cutting head is offset quite far forward of the rails themselves (70-80mm), which also produces quite a considerable bending moment too. All in all - not nearly stiff enough.
Tech2C has done a really really excellent job in his hypercube design. It prints beautifully. That is precisely because he's made the right tradeoffs for a 3D printer - and not for a CNC mill.
The same reasons that make the Hypercube a great printer essentially make it a poor design for a CNC mill.
Back to the drawing board
For all those reasons, there's been a gap in my collection of tools for a small tabletop CNC mill for a while, one that can:
- Mill woods and soft metals (aluminium, brass or similar)
- Act as a finishing step to 3D printed parts
- Engrave and drill circuit boards
To do that, I'm going to make some different trade-offs to the ones that Tech2C made in designing the Hypercube. If you're interested in learning more about the choices he did make, then I can highly recommend his YouTube channel where he documents the build. The plan is to optimise the build around:
- It's got to fit on a tabletop. I don't have a big shed to work in, so this mill has to fit on a shelf, which practically means no larger than a 500mm cube (roughly the same size as the hypercube). As part of this I'm going to aim for a working area of roughly 150mm x 150mm x 100mm.
- It's got to use the motor to hand. Otherwise that AC motor is just going to sit around on the shelf.
- It's got to be really, really stiff. I want to be able to use this thing to mill PCBs. This standard thickness of copper on a PCB is ~35µm which means we've got to have a positional accuracy in the Z axis of at least ~10µm - including under tool loading. Ideally similar levels of accuracy in the X & Y axes. Achieving this is going to mean trade-offs in speed, which given this is only even going to have low volume workloads is acceptable.
- It's got to use relatively off the shelf electronics and firmware (at least to start with). If I don't have a way of making PCBs without this, then it's going to be a bit chicken and egg if the design requires custom PCBs. Practically I'll aim to use grbl as the firmware as it's pretty standard - and some off the shelf electronics which work with that.
- It's got to use either off the shelf components or 3D printable components (for the same reasons as #4 above). We'll find that this is going to be one of the trickiest requirements to stick to, but with a little creativity I hope we won't have to break it. We can always come back once the mill is functional and retrofit some stronger milled parts.
- It can't break the bank - this is a hobby project after all. If we stick to rule #4 & rule #5 this should be pretty feasible. It means we'll be ordering most of the parts we need from AliExpress or similar.
And that's the plan. Let's see how we go!
Next time we'll cover some of the electronics, before we start diving into the mechanical parts of the mill...