Jari has been busy planning and building a new spindle for our CNC mill. The project is now about half-way so I thought I'd take a few pictures and write some text.

So far we've been using the stock spindle on the Opti BF20 which has a Morse Taper #2. The original motor burned out at some point so we had to replace it. But as our machine has run for probably hundreds of hours already the bearings are showing some serious wear. We measured a runout of more than 0.05 mm...

On the left the new spindle, Morse Taper #3, from littlemachineshop, a bargain at $34.95 !. It's sold as a lathe spindle, but we thought we'd get the longest spindle they have to place the bearings as far apart as possible for rigidity.

This is part of a larger upgrade plan for the mill which will include replacing the dovetail ways of the stock mill with linear bearings, replacing the standard acme screws with ballscews, and upgrading the control electronics to use AC servos.

The linear bearings are 15mm wide and manufactured by IKO. We got the ballscews from metallstore (I think they are made by HiWin), and the Sanyo AC servos are from last years surpluscenter sellout of Sanyo AC servos. Three 400 W servos will direct-drive the ballscrews (I hope the torque will be more than enough, otherwise we will have to resort to gearing as is often done with servos). A 1 kW servo will be used as the spindle motor. I'm hoping that using a servo as the spindle motor will allow rigid-tapping eventually. Finding affordable servodrives for the AC motors (sinusoidal 3-phase commutation based on hall sensors and/or 2000 ppr encoder) is not easy, so I've actually had a go at building my own drive (also check out Tero Kontkanens effort).

Everything will be controlled by EMC, with a Mesa Electronics M5I20 PCI-card for I/O.

Some pictures of the spindle assembly and toolholders below...

This is the spindle box itself. Circular recesses for the bearigns were first milled and then bored for a tight fit. Runout measured from the inside of the morse taper is now about 0.01 mm or less. Two steel plates will be added to the sides, and another steel plate to the back where the spindle attaches to the Z-axis carriage.

The top and bottom of the spindle assembly. The top end will receive a toothed-gear and a tightening bolt. We are hoping to have two gear ratios to switch between, 1:1 for max 3000rpm, and 1:2 for max 6000 rpm. The higher rpm range will be useful when running finish passes with small mills, something typically done on the IOM fin/rudder moulds we intend to make. We also need to make an M12 draw-bar. The bottom end will require a tight seal to be added which will keep chips and coolant and all the other nasty things that fly around out of the bearing and the spindle box. The bottom end of the spindle has holes in funny locations for some reason...

Some toolholders we got recently from Franz Singer on ebay (you would not believe the kind markup domestic distributors have on these things !). To the left a larger ER32 collet chuck which will be fitted with a 20mm collet and hold an edge-finder, used for locating the stock material when setting up a job. In the row to the top right and standing with a 6 mm collet fitted with an endmill are ER25 collet chucks which will be the main type of toolholders used (although kestools sells weldon holders reasonably cheaply). A selection of collets in the bag at the back. To the front right a holder for a face-mill, something that we have often missed with the current setup - hopefully this one will allow us to quickly get stock material into the correct dimensions before beginning a job. Not shown in the picture is a toolholder for our boring-head, and we also need to get a B16 standard holder for our drill-chuck.