Desktop Mill Acorn Build (DIY)

[Jason3]

Hello all,

I thought I'd post this machine build here, for your interest and also to provide me with the motivation to get it done with due despatch!

I'm build a small, desktop milling machine. It's not my first one, but the aim is to improve on the previous ones where possible and make it more usable for most of the parts I like to make. Overall dimensions are roughly 800 x 700 x 700 mm, or 34" x 28" x 28". Travels will be approx. 440 x 240 x 240 mm, or 17" x 9.5" x 9.5". Weight will be around 140Kg, or 310lb. The structure is aluminium, for several reasons including weight and availability and the awful mess machining cast Iron makes...!

The control system will be Acorn based. I've used Mach3 in the past, and considered Mach4, USBCNC and most of the open source alternatives. I chose Centroid because I was happy with the support I received on this forum for my recent retrofit of a previous machine, the software is well considered and logical without bloat, the hardware was straightforward to wire and configure, and because I have a Cyclematic lathe running Centroid controls so it gives consistency across my favourite machines.

Axis motors will be Teknic's Clearpath. I originally intended to use Teknic's 'Hudson' servos, but inadvertently sold too many of the servo drives I had earmarked for the machine, and had since been having good success with Clearpath so I decided to switch. The new motors are the next frame length down from and therefore slightly less powerful than the Hudson versions, but space is tight and I'm moderately confident these will be ample. At this stage the power supply will be a single Teknic IPC5, but another might be needed if my using-a-servo-for-the-spindle-motor experiment seems promising.

Lead screws are NSK's Compact FA range; 20 x 10mm pitch X and Y, 20 x 5mm for the Z axis. Linear bearings are NSK's RA series, roller type. The X and Z axis units are huge, I cheated and went for one long type bearing on each rail over the conventional 2 bearings per rail configuration for this kind of machine because this allows me to keep the machine more compact, and was actually slightly lower overall cost.

Way covers for X and Y will be the same Gortite accordion style rubbery stuff I used for previous machines. I have my doubts about whether they do any keeping of crap out of the linear gear, but they make it look tidy and finished so that's something at least. The Z axis will get a sliding hard cover, probably 3d printed as by then I'll be wanting to just get it finished.

I'll detail my spindle plans in a future post, as I'm still undecided which way I'll go for that.

[Jason3]

Next I wanted to show my 4th axis unit that I'd like to use with this mill. I built this some time ago, but it's seen very little use due to studies, moving house, kids arriving and life in general causing an unexpected hiatus in my CNC hobby.

It's a direct-drive unit with a large frameless motor built in. The 'spindle' is 440c Stainless steel, running in a large cross-roller ring bearing. It's got a Renishaw LM13 327,680 cpr rotary ring encoder on the back. It's very fast at it's max. speed around 4,400º per second, but I'll need to do some experimenting to see what step rate feels best with Acorn - the drive can handle 4Mhz, but I don't think Acorn can approach that. And it doesn't need to anyway, I can't imagine a scenario where I'd need it that quick. It's not a torque monster, more appropriate for small parts < 75-100mm diameter, I'd say.

The through bore is 40mm with the 5C collet nose (fitted in the photos) removed. The servo drive is an older Granite Devices unit, with a 126Vdc power supply. I'm undecided whether I'll switch that out in the future, but for now I'm planning to use it as is.

My original intention was to anodise the housing black like the rest of the mill, but I made the mistake of trial assembling it first. When I went to break it down again later I couldn't get it apart without (a lot) of heat, more than I wanted to subject the bearing to. It might've been possible to press the motor stator out, but I was very reluctant to try because it was a bit of a unicorn - runout on the bore with a Mitutoyo 0.001mm indicator was zero. That hasn't happened before or since, and I didn't want to risk mucking it up. So it stayed shiny...

[martyscncgarage]

Keep us posted on your progress.
On your 4th axis, I suppose it will be ok until the motor can't hold up against the cutting forces on the part. Usually they are worm drives, have pneumatic locks or both.

[Jason3]

Keep us posted on your progress.
On your 4th axis, I suppose it will be ok until the motor can't hold up against the cutting forces on the part. Usually they are worm drives, have pneumatic locks or both.

Yes, very true! Max. Torque is about 30Nm. I used the Hardinge DD100 rotary as a sanity check when I designed it, and it’s worked well (admittedly with limited use) so far. As long as I’m sensible about how I use it I think it’ll do ok. I did have some doubts before I finished it, and even bought a bigger motor to build a stronger unit which I might do one day.