Introduction

Although I had no real requirement for CNC it was a subject that started to fascinate me having seen what others had achieved by either building from scratch or modifying an existing machine. I therefore decided to make this a project to have a go at for no other reason than the challenge of entering a new engineering field and of course having a machine that would potentially be capable of more accurate work …… or so I originally thought.

My existing mill is an A1S from Warco and my early foray into the subject of CNC was to enquire if this mill could be converted. As it turned out at the time I rang Warco to speak to them about this possibility they had just finalised a deal with SimplyCNC to modify one of their other mill products that they sell. Speaking to SimplyCNC threw some light on their modifications but they were unlikely to tackle the A1S for some while, if at all.

One of the other things I was contemplating at the time was buying a small precision drill, again one of the various types sold by the usual model engineering suppliers for around £150 or so but I then spotted the Super X1 mill with an extended table available from Arc Euro Trade and the light bulb lit up as I realised I could have my small precision drill and a CNC mill project all in one using this machine. And so the decision was made and the project commenced.

Now I am starting from scratch, having to learn about machine modifications to make it capable of being driven by CNC software, what electronics/electrics are needed, what software drives the machine, what software generates the code and indeed what CAD/CAM software to use? As far as CAD is concerned my experience is using Turbocad from IMSI but this is a low end cost package but quite adequate for my modeling activities and it can produce DXF files capable of being imported to a CAM system. However I thought a package especially designed for CAD/CAM would be the better option to keep me on the straight and narrow path.

I decided to start, having chosen the basic machine, with the CAM software and software/hardware needed to take the output from the computer and drive the motors on the axis of the machine. A search on the Internet produced a number of alternatives but in talking to those who have experience in the matter I was guided towards the software called Mach3 from Artsoft in the USA which is the bit of software that actually takes the G code and drives stepper motors in the right way, and a CAD/CAM package from Dolphin in Scotland that is a CAD/CAM package that will transfer its CAD drawing (or imported DXF files) into a Milling (amongst others) machining CAM package and generate the G code for exporting to Mach 3.

I had no experience with any of this software but fortunately both suppliers have free evaluation software available on their sites which can be used to at least get a very good feel of what is involved in using the software but cannot be used for cutting metal as the G code is restricted to a few lines only. Another plus was that both offer free video tutorials on their web sites too. More on this software later.

The next problem was how does the computer and software communicate with the electrics that drive the mill motors. This is achieved by the use of what is called a "breakout board". This is a small printed circuit board that connects via cable to the parallel port of the computer and takes the Mach 3 output and converts its signals into the appropriate signals to drive the stepper motor electronics. It provides an important optical isolation between the computer and electrics to protect the computer from nasty voltages and currents that might otherwise cause damage, and also ensures that the right level of signal is processed from and into the computer. The "into" the computer signals are those, at the simplest level, generated from the table limit switches and Emergency Stop button. There are several types of these printed circuit boards available from a search on the Internet but again using guidance from others I settled on one supplied from CNC4PC in the USA. Again there are choices to be made as these boards can handle a variety of signals involving many output and feedback signals from machine to computer and vice versa. I chose the simplest board that would handle the limit switch feedback from the machine and allow me to drive the four axis I planned to use.

So what is involved in the electronics/electrics to drive the mill axis? Well obviously there is the motor that drives an axis and it seems that stepper motors are favourite for this although servomotors are also used. These stepper motors commonly work "open loop" that is to say there is no positional feedback to verify the position actually achieved. Alternatively "closed loop" operation is possible, but for most this seems an unneccessary complication. A stepper motor requires an electronic driver that powers the motor with the voltage/current and the signal to make it turn in the right direction and at the number of pulses to make it run at the right speed etc. These drivers require a DC power supply that is quite hefty in terms of its output. These drivers also provide important current limitation to the motor as the motors are usually driven at higher voltages than rated to achieve torque and speed.

Also a small DC power supply is required to power the "breakout" printed circuit board at 5V, although it is possible to get this supply from the computer I understand.

That is a brief overview on the software/electronics side but what of the mill itself? Well here I enter unknown territory as well, as there are feed screws and backlash to consider, how to mount motors and drive the feed screws, whether to keep the manual handles or not, what to do about the Z axis, drive the quill or drive the vertical head slide? And last but not least does the mill require any attention to improve its accuracy on slide ways and minimise vibration?

All the articles I read seemed to talk about replacing feed screws with ball screws to obtain better accuracy through lack of backlash and better longitudinal pitch accuracy as well as more efficient power conversion from rotational torque to linear movement. However the ball screws come in various sorts, all of which seem to cost an arm and a leg and some apparently not necessarily offering better performance than a good ACME threaded lead screw. I will discuss this subject further.

The choice of driving quill or driving the head is also not straightforward. Driving the quill would only give 30 mm of movement and make tool changes a possible problem and the mechanics for adding the drive motor are more complicated than adding the motor to drive the vertical headstock on its slide. But the headstock is much heavier and not counterbalanced on its slideway so getting free movement may become a problem. Also more vibration can be expected to be present from an unclamped headstock. Again a subject to delve into more deeply. But at last we have the bones of the project. We know, in the main, what is required to be looked at and done in principle and decisions can be made on some things to move it along.