The Shay, Part 4: The Crankshaft, Take 4 I honestly didn't anticipate having to make a fourth crankshaft, but since the third failed, I'd no choice. So, I elected to make the best of the misfortune and try some different fabrication techniques. Two expand my options, I placed an order for the complete range of nickel-silver sheet thicknesses from Thunderbird Supply Company. For the amount of metal I'd need, it was quite economical, especially since they sell by weight and will cut as little as one inch of stock. This time I decided to try making the webs from .012-inch thick nickel-silver sheet instead of .015-inch. It was closer to scale—the actual webs are 2.5 inches thick, or .011 inches at 1:220. I also decided to try drilling without a jig, using just scored marks as a guide, as I had a feeling the jig was creating more problems than it was solving. I was tempted to try dropping the wire size from .020 to .015 as well, but this would require drilling #78 holes, and I didn't care for the idea of drilling so many holes that size. Plus I feared it might make the crankshaft too delicate. Another change I decided to take in my assembly approach was to invert the way I was making the cranks: instead of fabricating each one start to finish, and then attaching them to the main crankshaft, I decided to "flip" them such that the crankpin was soldered on the long end of the webs, which would allow me to leave the webs uncut until after the cranks were soldered to the crankshaft; this would afford much more control over the assembly process.
After a couple of months away from modeling, it was slow going at first; just cutting a work piece seemed like a chore. But I soon got back into the swing of things. After scribing some guide lines (above left) and loading up the pin vice with a #76 bit, I quickly learned that my decision to eschew a drilling jig was a good one; the thinner metal was easier to drill, so I just kept drilling holes until I got six pairs that had consistent spacing (above right).
Leaving a generous amount of metal around the holes, I cut out the rough web parts with a jeweler's saw (above left), bent the tabs away from the work piece, and clipped them off with flush cutters (above right). Next, I loaded a diamond cutoff disc in the Dremel and fired it up. Once again working in silhouette—foreground light causes distracting reflections that lead to judgment errors—I slowly ground the parts down to their final width (below left). This alone occupied about an hour of a muggy, gloomily-overcast Saturday afternoon (padded out with frequent coughing fits).
As hoped, the nearly-finished webs (above right) were finer then their predecessors. Pressing onward, I re-drilled the holes with a #75 bit in preparation for soldering, then inserted two scraps of wire through the holes of a pair of web parts. I taped the web parts to either side of a scrap of .015-inch thick sheet metal, and clamped everything in a vice (below left). After removing the wire from the outer holes, I soldered the wire to the inner holes of the webs with my spanky-new needle-point soldering iron tip (the finest tip I could find on the World Wide Web).
Aside from a hole filled with solder—easily remedied with a quick drilling—the first crank went together without a hitch (above right). The problem I have, however, is that whenever things go well, I get excited and start to rush, which leads to mistakes; indeed, I ruined half of the web parts and one finished crank assembly. So, I worked on this web page for a while in between each of the three crank assembly procedures, which stretched the project out through the rest of the afternoon. At the end of the day I had three crank assemblies that still had nice long "handles" to simplify the tricky process of soldering them to the crankshaft—particularly in getting them accurately oriented 120 degrees apart. The long handles not only made them easy to clamp in the vice (below left), but acted as heat sinks to keep them intact for the final critical soldering. After grinding off the excess crankpin material from each crank, I began the crankshaft assembly process.
All of the steps leading up to this point paid off, yielding perfect solder joints, with fine fillets around the outer edges of the joints (above right)—a sure sign of good solid bonds. Which quickly brought me to the final nail-biting step: cutting out the excess crankshaft between the webs. Previously I'd done this holding the crankshaft in one hand and the Dremel in the other; given my increasingly unsteady hands, I decided to secure the Dremel in my vice so that at least one of the two objects wasn't going anywhere.
Alas, all of this good fortune once again led right up to failure at the very end. As I was clipping off the handles and filing the web parts to shape, my finger caught on the end of the crankshaft and ripped the end crank assembly apart (above right). And once again, given the way it's assembled, there's no repairing it; I must start over from scratch for the fifth time. Contrary to what one may think, it's only slightly discouraging; I've learned so much up to this point that the iterative process of "try, try again" can only improve the outcome. In particular, the lesson learned here is that it may not have been such a good idea using .012-inch thick nickel-silver for the web parts; they seemed to be too delicate. I may need to go back to .015-inch stock for the strength, and live with a slightly fat crankshaft. Three steps forward, two-and-three-quarters steps back. Such is the way of the hobby when you're pushing the envelope. (Casualty report: one #76 bit and one jeweler's saw blade forfeited their existence for this attempt. Amazingly, however, I'm still on my first .009-inch thick cutoff disc. I'm not sure how that happened!)
|
Continue
to Part 5Copyright © 2010 by David K. Smith. All Rights Reserved.