A few years back I was going though some of my junk, mainly odds and sods of electrical items and electronics and decided to put some of it to good use and I ended up with a fully functional CNC mill in my own workshop being a manual mill/drill press I decided to automate it a little.  This has been the ultimate project and has paid off in saving me time in making bits and pieces for my little engineering escapades or when feeling a little inventive. 

I have made automotive parts, motorcycle parts, computer parts mainly for liquid cooling and also replaced some OEM items with my own sometimes improved parts. and not only for me but family and friends and some paying customers which has been a help in further developing the machine. 

I use Mach 3 as the software to control the machine and Solidworks and MasterCam to design and code the products to be made, if you like you can think of the machine as a printer that makes actual usable parts. You draw it then if you like what you see you print it with some stock you specify be it, wood, plastic, metal or composites. 

I plan to make some CNC control boxes for other budding CNC’ers of the hobby market down the track once I further develop my machine, next on the agenda would be either Servo motors over the current steppers or a lead screw upgrade to a ball system just for added accuracy and repeatability. 

Mill during tuning and setup

Mill during tuning and setup

So there, shown above is my modified mill – now a fully operational 3 axis CNC machine. 

The Build 

This machine is based on the Hafco Metalmaster HM35 Manual Mill/Drill Press. 

I had chosen to go with Steppers with this machine because I knew at the planning stage I would opt for gearing down because of the nature of the compound table.  At the time speed was not much of a concern just needed a functional CNC mill. 

Being and Chinese copy the quality of the leadscrews weren’t  up to scratch there was heaps of backlash which I did manage to adjust most of  it out, only to discover the leadscrews are not uniform across their entire length which caused me a lot of grief with binding and sticking during rapid table moves.  So then I had to adjust the leadscrew nuts with a compromise between backlash and binding. 

With much use the gap of compromise has closed up as the leadscrew and nut bed in.  My next modification is Ballscrews and Ballnuts which cost a small fortune because of being  high precision parts, but improve friction and overall accuracy than the current trapezoidal jobbies in the machine presently.  Even though the machine did have a degree of backlash the improvement over milling by hand was tenfold in regards to speed and accuracy. 

My main reason to do this CNC mod was to basically take the mundane and human error factors out of making small parts and moulds by hand.  The benefit of being able to design your part on a PC and run a simulated toolpath to make sure all is good is a real bonus and also depending how you look at it, it can be quite fun. 

Assembly 

The machine when new or “out of the crate” so to speak is a Geared Head manual milling machine and drill press, there was a belt driven model but the risk of belt slipping during a cut and breaking tools was at a far more higher risk than a Geared model, the only real benefit with belt driven was that they run a little quieter.  

Now I need to get drive to the 3 axis X, Y, Z, the X and Y were my table controls and the Z being the “quill” or spindle.  I had to first determine the mechanics of driving each axis I had already decided to gear down the drives because I needed to increase the torque of the steppers I had already, 430oz/in NEMA24 Unipolar 1.8° Step Motors because of the gearing down ratio was 3.33:1 meant that 3.33 x 430 = 1431.9oz/in, yep that was plenty I thought. 

Now seeing that the manual wheels and graduated dials were now not needed as a human interface it was time to yank’em off and toss over the shoulder getting ready to measure up and start fabricating the bracketing to firmly hold the step motors in place while applying enough tension to the timing belts I planned to use.  At this stage I knew how I wanted the motors to mount to the machine but had no idea on the pulley sizes and belt lengths, all I knew is that this had to be well thought out so everything would fit together as I did not have the pulley or belts these were yet to be purchased once I knew the sizes and configuration that would work.

I took measurements off one end of the table where the leadscrew bearing housing mainly the hole centres of the bolts which pin the housing to the table, I preferred this idea as there was no need to drill and tap extra holes for mounting my custom bracket.  Once all the information was collected I turned to Solidworks (Parametric Modelling Software) which I’m fortunate enough to have access to and design a prototype of the idea I had of the assembly, from this and other data I could the determine the pulley sizes and belt sizes I would need to complete the model.  I followed the same principle for both table X and Y axis.  The Z was also quite similar except I took advantage of the broad space in the bracket I had to incorporate some control items such as the E-Stop (emergency stop) just in case your tie or scarf got caught in the spindle, your forehead as it was dragged into the machine would bump the E-Stop thus saving your neck.  You’d think it was strategically placed but as it turns out it was only a fluke that it ended up there.

Stay tuned…