A few notes about the electronics for Peter's mill conversion.
There's a driver box and a manual control box plus a 24 volt power supply.
The driver box (not in Peter's photos) is basically a packaged up MD3AXSI8435N controller from
www.mdfly.com which is intended to connect three steppers up to a PC running CNC software. He sells a few variations on the theme, and not all of them all the time. Eg the board above isn't on his site at this moment, but will probably return at some point. There's several other people who sell boards like this, which really just wire up three Toshiba 8435 driver chips which do all the work, and usually add some optoisolators plus inputs for limit switches etc. There's a 25 way D connector and a cable to connect to a PC parallel port. Mods I did to this board include:
- fit a decent size heatsink and a fan on the outside of the box
- replace some of the ic's Mdfly decided not to fit, mostly optoisolators but also a voltage regulator
- change a diode in the power feed which wasn't properly rated
- fit some pullup resistors so the unit sits in a guaranteed safe state if you pull the D-type plug out/power it up without the PC connected
The manual control box replaces the PC and is homebrewed to Peter's requirements.
- a 100-step rotary wheel which can drive X or Y axes. With Peter's 2:1 belt drive, 2mm pitch lead screws on the mill and the right settings in the driver box (half-step I think) you end up with one rev of the wheel = 1mm table movement
- alternative to the rotary wheel, buttons for +/- X and Y movement (speed variable)
- buttons for Z movement
- a few switches to select the above, and some lights to show what's going on
- a big button for the spindle motor (see below)
Mostly on account of the rotary wheel, the manual control box contains a microprocessor (PICAXE 28X1 - see
www.picaxe.com) and about 15 ic's. The circuit is not optimum and there could have been less ics if I'd not used the Picaxe which contains a Basic interpreter and is thus slow enough that I had to do some functions in hardware that could have been done in software.
If you don't require a rotary wheel, it could be done with a few ics and I'm currently (err rather slowly) building one of these for myself. Incidentally this will run a different driver board to the Mdfly one as he stopped selling them for a bit. As all the stepper control work is done in the Toshiba 8435 chip that's not a problem.
I've reverse-engineered most of the X1 control box (nb for MY model, I'm sure there are several variations) so I have some inkling of how it works, including the zero-volt safety feature which is particularly devious.
If anyone is interested I'm happy to let them have the circuits and software for the Picaxe which I'd release under the Gnu Public licence ("once free always free"). I'd post it all them on a website but I don't have one (=too lazy to set one up and look after it).
I've got some photos of the innards somewhere which I'll post when I've found some which don't demonstrate my chassis bashing skills too vividly (Peter has been *most* polite about this

)
Chris