Iíve decided to build a small hobby type milling machine for circuit board apps. Iím in the process of gathering materials for the machine and learning Eagle circuit board software program.
I hoping someone might advise me briefly as to what I need to learn to achieve this.
Building the machine, stepper motors and drivers should not be a problem, but Iím a little fuzzy about computer generated control of the motors. Iím assuming there are software programs for this.
If someone would explain the steps between circuit board design and computer output to the drivers, would be a great help.
Any response to this regard would be greatly appreciated.
You'll want to use Mach 3 software for your CNC control. There's a group on CNCZone for this and another good one on Yahoo. Start reading.
Circuit board software outputs files as gerber design for the traces. I found a freeware program on the web to convert this to a .dxf file. Then used a CAM (Mastercam in my case) program to write the G code program for the CNC control.
The drill holes are output as excellon type files. These can be quickly manually edited into a CNC Gcode program.
I did this once for experience, then found out you can just send your design out and have four copies of your board made for $50.
I am using TurboCNC running on a old laptop under DOS and it works ok for me.
For Eagle there is a wonderfull ULP from John Johnson called PCB-GCODE. After designing your PCB in Eagle just run this ULP and you get good G-code for milling and driling your PCB. Look at http://group.yahoo.com/group/pcb-gcode/ for this ULP. There, after you register, you can find pictures of the machine I use and the PCB's I made with it under "Albert's Pictures".
Thanks Karl, AbSat, for the reponses. Good information. I followed up the links provided which led to other links and came up with a wealth of information. Looking at the Mach 3 and turbocnc software, there is several versions Mach 3. For x,y,z milling machine which ver will work best for me? The turbocnc is free for $60 ??
Bill, the mach3 mill program will work. If you want to minimize the work you need to do, you can buy a ready made cross-slide table for less than $100 from Enco. Then all you need is the stepper drives and vertical axis. PCB's don't tend to lie flat, so a floating vertical axis would be ideal.
Absat, here's a very simple implementation I'm playing with. The z axis nut will be spring loaded. The short screw on the left of the cutter is adjusted so that the bit goes the proper depth into the pcb. The stepper motor "tries" to lowers the bit about .05" below the pcb, but the depth stop screw will stop the motion. The spring loaded nut allows the head to float up and down.
This mechanism is far from ideal, but it's good for testing.
Thanks for the answer H500. I was playing with that idea too but I could not find a good solution for the tip of the screw yet. There must be some kind of roller ball at that tip, maybe more than one because, while working with 90 degrees milling tips, 0.5 mill difference gives to much difference in the milled isolation path already. This "sensor" has to sense on milled area so there will be difference in height. I was thinking of a circle of roller ball points around the milling tip but this will give problems with removing the dust I think so it must have a vacuum cleaner integrated.
Till now I mount the PCB at a underlay with double sided tape and adjust that as flat as possible on the milling table. When necessary I adjust the Z axis position in the G code for specific areaís by hand. I was thinking of measuring the surface with a sensor before milling and automate that Z axis correction, but if I ever find the time for that, Who knows ?
A project I tried to get going a few years ago needed some way to have a tool follow an uneven surface without actually touching it. I never built it but I came up with the idea of what I called an air shoe; I uploaded a picture trying to show the idea. The blue part is the shoe which has an annular groove supplied with compressed air. Above this is a housing in pink connected to a suction line to take away dust with everything connected by a bracket to the toolhead. The theory is that the air supply lifts the shoe up from the surface while the suction pulls it down with the gap between the shoe and surface being determined by the pressure/suction difference. The air passing from the shoe in to the suction housing sweeps dust and chips inwards and keeps them away from getting caught between the shoe and surface.