Gearing up!

John has decided that Chalupy acres has a tutelary god. That's the good news. The bad news is that his name is Sisyphus. He was the one that kept rolling the stone up the hill, only to have it roll over him before he got to the top. In this matter of gear-cutting old Si has been right up to scratch. OK, so far we have made a dividing plate. We even have a blank gear at hand. The big moment has arrived. We must make a cutter to the correct profile.

Now the subject of gear profiles is somewhat controversial and quite mathematical. But I propose to avoid this as much as possible, although the practical details are really not that complicated. I am trying to cut clock gears. These are "cycloidal" profile gears but you can ignore that. Thing is you have to do is shape a cutter to the correct profile. You can work out that profile. The tip of the tooth is curved. And so I went through a bunch of calculations and came up with a radius of curvature for which I happened to have a diamond tool. And behold, Sisyphus came in for a visit. I proposed to cut the cuttter with the same setup used to cut the gear;  i.e my DSO (Dremel-Shaped Object) in the vertical slide on the lathe. My carefully chosen cutter would'nt stay put in the DSO! The shank was defective. The shanks are 6.3 mm which I suppose is some fraction of an inch in RGU, but the shank was 5.95 mm and the DSO collet wouldn't grab. Curses!

So what I did was to take a radius much too large for the occasion. It was all I had; it was either that or go online and order some diamond tools, which I will eventually have to do. But not right now, please. What I wanted to do is convince myself I could cut 30 teeth in a 30 mm or so blank. So I ground up a cutter, in the lathe itself. Its radius is twice what it should be. But it will do for now. All I'm after is feasiblity. Above is the cutter as ground. It is being held in an arbor I had to make, held by a setscrew. The arbor is a piece of hardware store rod. The cutter itself is a piece of Dremel tool, broken or worn down, from my scraps. Dremel tools are very nice and hard. Just the thing for a cutter.

Now, Houston, we have a cutter. Next  step is to set up the divider. We must make sure the teeth of the gear are uniformly spaced. So the divider plate comes into action.

The divider plate is held steady by a springy piece (old strapping tape) with a pin affixed. We index to the #1 hole. Since I am cutting thirty teeth, and the divider has 60 holes, I have to skip one hole for every tooth in the gear. This requires some concentration. Note I have removed the drive belt from the lathe. This is a safeguard against accidentaly turning on the lathe! (I did that too. Disaster. Go away, Sisyphus!)

Next we use our previously made arbor (read "axle") and bolt our gear blank into it.

With the vertical slide micrometer feed we painstakingly adjust the cutter height to the center line of the lathe. The proto-gear is being held in the 3-jaw chuck. It should be held in a collet. Don't have a big enough collet. Too bad. Now we turn on, not the lathe, but the Dremel. With the cross-slide, advance the tool until it cuts. It cuts amazingly well. Since the cross-slide is calibrated you must note how far you advanced (actually retracted) it. This gives you depth of cut. I was none too careful about this. In about two seconds the tooth is cut.

Now it is routine. "Index" over. Skip one hole on the dividing plate, remember? Turn on the DSO. Do not turn on the lathe. That's why I took off the belt. Crank cross-slide in. Remember to what number you must crank it in. Oh, you forgot? Gear won't work. But this does not matter for now. We are, literally, cutting our teeth. When I had done all the 30 holes I had an object that looked remarkably like a gear.
 

I am very pleased. The whole lashup -- the DSO held to the vertical slide, the homemade divider plate, the primitive index pin -- it all worked. The cutter was all wrong for reasons given. But all the things I was worried about -- wobble, for instance -- did not materialize. Maybe brass would have failed, but since I am planning on plastic gears, this will not matter.