My Site, My Stuff

4th Axis For The CNC Mill

Shortly after converting my Bridgeport mill to CNC, I realised how useful CNC was and wanted a small CNC lathe for short batches of a few turned parts I used for various products.

With nowhere to put a CNC lathe, the only option I had was to make a small 4th axis for the mill and use it as a lathe instead of the more conventional use for a 4th axis. I had a spare channel on the CNC controller so it should be pretty easy to add-on.

I decided to use 20mm 6061-T4 tooling plate as it is nicely finished on the faces and relatively cheap. A design was drawn up based around a 5C collet spindle – this is a fairly well known use for one of these. 5C collets can go up to around 30mm diameter which was plenty for my needs.

The Build

The frame was easy to machine on the Bridgeport, CNC is a wonderful aid once you get to know it, even in a minor capacity it can really help make complex parts easy. Above is the bare chassis for my build, registers are milled in the base for accurate alignment and its secured with 4 M8 socket head screws in each plate.

The 5C collet spindle which came from a Chinese “Spindexer” – a rotary indexing block, the tube just happens to be perfectly sized for some off-the-shelf bearings. The nose bearing butts against the spindle flange and the rear bearing floats.

The spindle is held in the chassis with a thrust plate, a clamp-ring and adjusted to give very low slop in the bearings, the 20mm wide toothed drive belt is then screwed on to the clamp ring.

The motor is a 750W AC Servo motor, I used these on the X, Y & Z axis of the mill conversion, they are extremely powerful motors and with it geared at 1:1 meant I had a 3000rpm drive capable of indexing and thread cutting as well.

Above is the finished build, I made a steel cover, powder-coated it and secured it with some M4 screws. The toe-clamps were later replaced with milled slots and the axis was bolted directly to the table.

The Tool Holder

Shown above is my 1st tool holder – it’s a fixture I made previously that held a 2kW 24,000rpm spindle on the mill – this was not used anymore so I pressed it into service to hold the tool post off of my manual lathe.

Wiring Issues

In order to do thread cutting I needed to feed the motor encoder signals back into another port on the CNC controller, this port was already used by the main spindle motor drive so I needed to build a small digital switch to swap the encoder signals when the 4th axis was used. It used a couple of 74LS257 switches to swap all 8 encoder lines from two encoders into the one port on the controller. The input was opto-isolated for safety.

I had some noise on the drive lines as shown in the left picture above, the right shows what they should look like. This problem was fixed by changing some of the cables around in the cabinet and it worked much more reliably afterwards.

Testing It Out

The first cut, shown above, was a nerve-racking moment – changing from a basic CNC mill to a CNC lathe with very odd axis designations was not easy to grasp.

Above is a test piece I made to check diameters, after getting the tool offsets set correctly it was extremely accurate and repeatable.

Some blocks I made in batches for a test-gauge tool I sold for the printing industry, very pleased with the wonderful surface finish – these were straight off the machine, no finishing at all.

A little brass thumbscrew for one of my products – the thread is machine-cut M4 and the head is only 12mm diameter.

Some Videos

Running the test part as shown above, worked very well indeed.

Making the stepped test part.

Making some bevelled aluminium washers for one of my products.


The project was a complete success, it made many small parts as needed. I had the controller programmed for manual tool changes and a small rack of tooling which enabled me to make all I needed in brass or aluminium, it was not built for steel and the tool mounted on the Bridgeport mill spindle would not be sturdy enough for the forces involved.

The downsides are are similar to many jack-of-all-trades machine adaptations – each time you remove the axis and refit it, you have to run tests to realign the axis to the mill table, you also have to re-run tool compensation for offsets or the parts will come out the wrong size etc. In the end a dedicated lathe would have been far easier to use but this served its purpose perfectly.