Chapter 4: The Barrel & Arbor

November 29, 2004

I finally got to begin work on the mechanics! In this chapter, you build the barrel which the weight line wraps around, and the arbor which holds the barrel and great wheel. Machining the arbor was interesting, because it's the first serious turning I've done with steel. Turning the steel was a pain in the ass. A lot of material needed to be removed, and I had a very tough time keeping the tool from chattering the whole way. I only managed to prevent it for short stretches, or at very high speeds with miniscule cuts (e.g., .001"). I eventually switched to a carbide cutting tool, but it only helped slightly since it's a really cheap tool. I'm now wondering if a significant part of my problem was that I was machining it from cold rolled steel. I picked that originally because it's harder than hot rolled steel, and I thought that would be a positive quality for an arbor... but it also means that there are various internal stresses that make it less dimensionally stable, and might lead to difficulty when machining. Next time I'll try hot rolled steel, I guess.

Here's a shot of the finished arbor (although I still need to trim it to leave the proper distance between plates, and burnish the pivots). Using the knurling tool on the middle section was a bit of an adventure. My knurling tool was built such that I couldn't bring the knurl all the way up to the flange. I actually cut a part of the tool away to get a bit closer, and then trimmed the un-knurled portion of the arbor slightly in an effort to prevent any problems. If you click the thumbnail to see the closeup picture, you'll see that the knurl actually looks a little messy. I'm not sure why that is, but I hope it doesn't cause problems. I'm a little worried, because that part of the arbor is slightly undersize (by a few thousandths), and then the knurl didn't enlarge it quite as much as I expected (it increased the arbor diameter by about 7 thousandths). The book didn't say how large a knurl to use, so I just picked something I thought seemed reasonable. I hope it holds.

[BarrelArbor]

[TurningEndcap]

I tried and failed to use a circle cutter to cut the end cap blank. I finally bought a new blade for the bandsaw (I never knew a blade could be that dull!) and used it and my disc sander to rough out the end cap blank from 1/4" brass plate. I drilled the center hole undersize, as the book suggested, but I didn't have a 3/8" reamer, so I mounted it in a 4-jaw chuck, and bored the hole to size. Regular twist drills tend to drill oversize, and this allowed me to actually leave it a little small to account for the undersize diameter on the arbor. Between the end cap blank, and boring the inside of the barrel, I'm quickly coming to appreciate my 4-jaw chuck. It takes a little while longer to set up, but it lets you center things accurately enough that you don't have to worry much about re-chucking them off center like with a "self centering" chuck.

I pressed the blank onto the arbor, and machined a shoulder for a slip fit with the barrel. The end cap blank initially seemed very sturdy, but after the machining operations it now wobbles slightly. It won't turn around the arbor at all, but you can wiggle it back and forth disturbingly easily. I'm worried that I'll end up needing to machine a new end cap (or arbor!). My worst fear is that this will work just fine throughout construction of the clock, and then some day it'll give way, the weights will drop suddenly, and the clock will be damaged. But I went around the inside of the wheel with a punch to help tighten it a bit, and go will ahead with it as-is. Katherine tells me I worry too much, so maybe this'll show her! Besides, there will still be the other 1/4" thick end cap to help hold the barrel in place. I got a nice finish on the end cap - you can see a reflection of the knurls!


Machining the second barrel end cap went without a hitch. I made a superglue arbor as the manual suggested, and it worked like a charm. I realized that I couldn't easily follow the book directions for cutting the ratchet wheel teeth since I don't have a milling spindle that mounts on my lathe. I decided to do things out of order and machine the ratchet wheel teeth on my milling machine while the end cap was still mounted on the superglue arbor.

I made a fly cutter as described in the book. It was actually quite easy, and my first attempt is pictured here. Unfortunately, though, I made it backwards. Because of the direction my mill cuts, and because I wanted to cut the teeth towards the superglue arbor to avoid pulling the blank off, this cutter would only cut teeth pointing the wrong direction! Rather than reverse the entire winding mechanism, I made a replacement cutter. If you can believe it, I actually forgot to make sure that I didn't make the same mistake again, but I got lucky and the second cutter was a mirror image of the first. Making a holder for the cutter was actually quite easy, since I luckily had a "#1 Morse Taper Blank" I'd ordered from Sherline when I started this project. All I had to do was drill a big hole for the cutter, and a small hole for the set screws. Here's pictures of the cutter & holder.

[FlyCutter] [CutterHolder]


[RatchetSetup]

I think I must have spent a good part of Saturday messing around with the milling setup for machining the ratchet wheel teeth. Fitting everything onto such a small machine turned out to be a bit of a challenge, but this is what I finally decided on (along with making the new cutter, of course). The picture here shows two teeth cut - my very first horological wheel teeth! It's only a ratchet wheel, not a gear, so precision isn't very important, but it was a great exercise to cut my teeth on, so to speak. I suppose it may become mundane after a while, but for some reason wheel cutting is one of the parts of this project I've been most excited about.

Cutting the rest of the ratchet teeth went perfectly. The cutter worked beautifully, and the time spent on the setup was a good investment. Introductory machining texts frequently seem to say "Expect to spend more time on the setup than doing the actual machining," but they often neglect to mention: "And by the way, it's worth it to spend all that time, you'll be glad you did!" My ratchet teeth have chatter marks, but then so do the ones in the Hermle movement I'm currently cleaning for Katherine. I don't really know if it's reasonable to expect to machine them without chatter marks, especially when cutting the teeth a full depth on a small machine, but I'm pretty satisfied with the way it turned out.


[BarrelThreading]

Next up was threading the barrel. I bought an expensive thread cutting attachment mostly for threading this barrel, but as it turns out, the Sherline lathe can't be set up to cut 12 left hand threads per inch on something as large a diameter as the barrel! You need to use the headstock spacer so the barrel will fit over the cross slide, and then the 100 tooth spindle gear is too far away to mesh with the 50 tooth gear on the thread cutting attachment. You can replace the 50 and 20 tooth gears with a 100 tooth and a 40 tooth... but then the gears won't mesh with the 24 tooth gear on the lead screw. I finally settled on using 100 and 34 tooth gears in place of the 50 and 20 tooth ones, and a 20 tooth gear in place of the 24 tooth. This yields a main gear ratio of 34:20, yielding 11.76 threads per inch... close enough, right? Even after carefully picking the intermediate gears, they almost didn't mesh, and I was unable to use the hand crank with the collet draw bar. I ended up spending about 5 hours turning the spindle back and forth by hand, but the threads actually turned out quite well despite the problems. The round cutting tool I ground worked like a charm, and my thumbs were only sore for a few days. The threads are a few thousandths shallower than what the book suggests, but they are plenty deep enough to make sure the weight line falls into place properly.


Project Index | Clock Project