Prompted by Chris's efforts to produce cutters for clock wheels and pinions I thought I'd have a go myself.
I bought a Eureka relieving attachment via Homeworkshop.org some time ago and it's sat in a box ever since. It's much better finished than Chris's but still I had to make a few improvements. One problem was that the arbor ran out by 0.2mm. The guy who made it decided on 3/4" bore cutters rather than 1/2". I set up the arbor in the 4 jaw and got the bore running true and then skimmed it down to 16mm. (The clamping thread was 5/8" so didn't need recutting.) The arbor was bushed with, I think, brass and the bush was a very very loose fit so I remade it a press fit in the arbor and a nice running fit on the eccentric shaft.

Many years ago at work we used KE672 oil hardening low distortion tool steel and I managed to get an old colleague to find me 12" of 1.5" diameter. I sawed this into slices and faced them up in the lathe. So I could face the thin disks parallel I made a support which fitted in the bore of my 3 jaw and was just under 3mm shorter than the chuck jaws. I faced this holding it in place with a running centre with the chuck jaws removed. The slots are marked with the jaw numbers so it always goes back in the same orientation so should produce perfectly? parallel parts.

I made 20 or so blanks which were drilled and bored to 16mm. (No 16mm reamer!)
The button method of machining the profile on the cutter for involute gears isn't so straightforward for cycloidal cutters as the flanks of the cutter have to be radial with a radius at the root. I also chose to make the cutters with a radius at the tips so the finished clock teeth will have round bottoms. You can't achieve that with a button tool. (Antique clocks always had square bottoms but modern practice is to have round bottom teeth which are stronger.) As Chris found the button method would require buttons of less than 1mm diameter which wouldn't be stiff enough unless they were tapered. Having a CNC miller, I thought that it would be possible to machine a form tool with the tooth form to put the profile on the cutter. This form tool would also need a relief angle so that it didn't rub on the cutter. I chose 5 degrees as that's what seemed to be specified for the button method. To get the relief I angled the form tool blank at 5 degrees to horizontal. This would give a constant profile which would allow sharpening without losing the profile. I drew the form tool in Autocad and then generated the GCode in Cut2d. With a 0.8mm end mill at 10,000 rpm I cut the form into 6mm gauge plate which was then hardened (in old cooking oil) and tempered (in the deep fat fryer). The top surface of the tool looks dog rough but this was taken through a microscope and is the gauge plate finish before heat treatment and polishing. The radiuses on the tool are just over 0.4mm.


To remove the bulk of the material from the cutters I chamfered either side of the blank which would also help centre the form tool on the blank. The method of slotting the blank given in Ivan Law's book is to drill 12 holes and then saw into the holes. This may be OK for one or two cutters but I thought the CNC could do this job with ease.

The first attempts at forming the teeth on the blanks with the Eureka were very disappointing. I seemed to be getting nice little chips for a while but then the cutting seemed to slow up. I thought that perhaps the form tool was dulling too quickly but, on examination, it looked fine. After a few abortive attempts I decided that 5 degrees clearance on the form tool wasn't enough. The Eureka puts the relief on the cutter and I guess that this was also about 5 degrees. So, once the tooth started to form, it very soon started to rub on the cutter. I remade the form tool with 7 degree clearance and had another go. This worked OK for 8 and 12 tooth pinion cutters but when I tried to profile the wheel cutter (which has a much deeper form) I again had problems and had to make yet another form tool with 8 degree clearance.
Here are a few photos of the finished hardened cutters (just awaiting the deep fat fryer tempering and then sharpening).



I've marked the cutters with an Actograp engraver - not as neat as Rob's method of stamping but much quicker and I don't have any letter stamps. Perhaps I'll engrave them on the CNC if I make any more.
Phil.