- Joined
- Jul 26, 2006
- Messages
- 658
- Reaction score
- 49
Following an earlier topic http://www.muzzleloadingforum.com/fusionbb/showtopic.php?tid/212463 I found the article mentioned by Dave Person yesterday and it is story to warm the cockles of our hearts. Well done Foley et al.!
Leonardo, the Wheel Lock, and the Milling Process
Author(s): Vernard Foley, Steven Rowley, David F. Cassidy, F. Charles Logan
Source: Technology and Culture, Vol. 24, No. 3 (Jul., 1983), pp. 399-427
Stable URL: http://www.jstor.org/stable/3104759
I am sure there are a couple of people who already have this, indeed I wouldn't be surprised if one of you knows (or is one of) the authors. The paper can be had by going to a university or other library with access to JSTOR and downloading it like I did. If you can't do this, an email to thinkshard at yahoo.com with your email address may be helpful.
Leonardo, the Wheel Lock, and the Milling Process
Author(s): Vernard Foley, Steven Rowley, David F. Cassidy, F. Charles Logan
Source: Technology and Culture, Vol. 24, No. 3 (Jul., 1983), pp. 399-427
Stable URL: http://www.jstor.org/stable/3104759
The article has photos, etc, but having to click the copyright agreement means I won't post the whole thing....the scholarly firearms literature, most notably by Claude Blair, Marco Morin, and Arne Hoff. It centers on a firearms ignition system called the wheel lock, and at present the debate is polarized between two camps. One inclines to the view that Leonardo invented the lock, the other that it was developed by some anonymous German gunsmith.4 In attempting to strengthen the former view, we will argue that this device can shed important new light on Western machine-tool history. The full argument is too complex for this paper, but its leading fea- tures can be summarized.5 Like other common gunlocks of the preindustrial era, the wheel lock is an assembly mounted on the gunstock at the rear of the barrel. On its chief piece, a metal plate, is mounted a powerful mainspring connected at its free end by a short flat-link chain to a shaft carrying the wheel that gives the lock its name (fig. 1). A wrench or spanner can be applied to the squared end of the wheel shaft to twist it through nearly a full turn, winding up the chain and compressing the spring. The wheel can then be held in this position by a subassembly of catches, the sear system, which can be released by the trigger when desired.
Also attached to the lock plate is a spring-loaded swinging arm carrying at its end a miniature vise gripping a piece of very hard stone, iron pyrites (FeS2) being preferred. When this is fully lowered, the stone bears against the rim of the wheel. Surrounding a portion of the wheel rim is a metal block hollowed on both top and bottom, the powder pan (fig. 2). Fine-grained priming powder is put into the top cavity and secured there by a tight-fitting lid which is opened, in most cases automatically, by an arm-cam system driven by the wheel shaft, just before ignition occurs. In the lower part of the pan is the slot in which the wheel turns. The two cavities intersect, so that the wheel rim peeps up through into the powder cavity, where it can be touched by the stone.
When the trigger is pulled, the wheel begins to spin, and sparks are struck by the stone from the wheel rim. The action of the lock thus superficially resembles that of a modern cigarette lighter, but there is an important difference.
In a lighter, or in the similar welder's sparker, a serrated, hardened steel surface, toothed rather like a file, is rubbed against a material that is called a flint but which is in reality a very soft alloy of rare earth metals, Mischmetall, composed chiefly of cerium.6 This element oxidizes so rapidly that when it is machined at high speed, the chips make sparks. Hence, in the cigarette lighter, it is the wheel which machines away the "stone."
In the wheel lock, it is the reverse. Pyrites have a Mohs hardness as high as 6.5, or higher than many carbon steels.7 When the gunlock wheel spins against the stone, it is bits of steel that are scraped away, so fast that the heat generated by the friction causes them to incandesce and so ignite the powder. Afterward one can find microscopic spheres of fused steel in the pan.8
Hence, if we are looking for historic highlights in machine-tool history, it is already interesting to notice that the wheel lock in its firing mode acts as a miniature, all-metal, spring-powered, high-speed, metal turning lathe. Its cutting tool harks back to the Stone Age, but the remainder of its design and operation anticipates the era of Maudslay, Brunel, Bodmer, and Hall.
In order for the gun to function, the priming powder in its pan must not be dribbled away, as by the bumping gallop of a warhorse. Since priming powder was specially fine grained, with pieces about 0.1 millimeter in diameter, or about like coarse flour,10 the fit between wheel and pan had to be a close one.
That Renaissance gunsmiths succeeded in this task is confirmed by measurements taken by us on more than forty locks in several museums and private collections."1 On the best-made gunlocks, the clearance between the wheel and the sidewalls of its slot can run as low as 0.07-0.08 millimeters, or close enough to confine the powder. Between the wheel rim and the slot bottom, even tighter fits are found-0.04-0.05 millimeters is not un- common. These approach modern industrial standards and raise the question of how the slot was made, for the wheel can run as thin as 1.8 millimeters,'2 and the pan may embrace up to a third of the circum- ference of a wheel whose diameter may range from 25 to 40 millime- ters.
The slot can thus be 5 or 6 millimeters deep and is quite cramped for any kind of hand filing or chiseling. The profile of the wheel rim further complicates hand fitting, for nearly all the wheels we have seen, read of, or heard about show a rim which at first glance appears to have threads cut on it.13 Closer in- spection will show, however, that instead of one continuous groove cut helically around the disk, there are typically two (sometimes more) which are completely independent of one another because they all lie in planes parallel to those of the wheel itself. In other words, the helix angle of these vee grooves is zero. On most wheels there will also be small cross grooves cut more or less at right angles to the threads and down as far as their roots. These may be much narrower than the latter. One can see immediately that for a close fit between pan and wheel, each side of each thread must be independently fitted, so that the piercing where the wheel slot meets the pan will have sawtooth ends.
Examining the inside of the pan's wheel slot will show that it has a curved and vee-grooved bottom, which is the mirror image of the wheel rim. Fitting such a slot by hand thus becomes a challenge. The sidewall planes must be cut quite true, and the curved, grooved bottom must be radiused to perfection. Because the crucial surfaces are concave, there is very little room for chisel or file to work, and their strokes could be only a few millimeters long at most. Fortunately, examination of the wheel-slot surfaces under magnification shows that hand fitting was avoided. The job was mechanized, with the wheel rim utilized as a cutter to create its own cavity. The wheel-lock wheel is thus a rotary metal cutter, the most crucial component of the milling machine. As will be seen, its invention can be located half a century earlier than the oldest previously known example and linked closely with thought experiments under- taken by Leonardo. The cross grooves, on those wheels that have them, create triangular cutting faces which scrape away at the bottom of the pan when it is pressed against the wheel rim and the wheel is turned. All the locks we have seen have the pan fitted to the lock plate in such a way that it can be fed against the wheel rim, either by being slid down into a slot in the top edge of the pan or by pivoting around a conveniently located screw.
Locks exist which lack these features, but in them the pan is now brazed into configurations which could not have characterized the parts while they were being manufactured.14 In other words, the wheel lock has not only a cutter but also provision for feeding the workpiece, the pan, toward it. Robert Woodbury has defined a milling machine as a tool wherein the workpiece is secured to a table which is then moved past the cutter.15
Wheel locks lack the table, but they are obviously more akin to millers than to gear-cutting machines, the other obvious choice. Hence our title. Further evidence to support the use of the wheel as cutter in the fabrication of the lock comes from the geometry of the tool marks found on the wheel rim and on the surface of the pan slot. These are best seen under 5-15 x magnification. On the slot sidewalls, fine semicircular striations can be seen which are almost concentric (figs. 3, 4). They are not fully so because the pan advances toward the wheel as the cutting proceeds. On the bottom of the wheel slot, on each flank of each thread form, will be seen continuous and parallel scratches. It is the surface uniformity, continuity, and geometrical regularity of these marks which demonstrate that they were made by the continuous turning and cutting of the wheel. Hand-tool striations would be dis- continuous and nonparallel. (Let anyone who doubts this try to make them otherwise.)
Characteristic curled exit burrs will sometimes be found along the outer sawtooth edges of the wheel slot, where the In order to examine such marks for oneself, locks in very good condition must be sought out. The ignition of gunpowder creates compounds which are highly corrosive, and pan and wheel may no longer have their original surfaces. Wear to the parts caused by firing can also obliterate the striations, although usually not evenly (in our experience) over the whole surfaces of slot and rim. One can usually find a revealing spot if rust has not taken the original surface com- pletely away. After some experience has been acquired, it is not hard to discriminate between firing wear, which produces a buffed or frosted appearance, and the original tooling evidences.
Gouges pro- duced by foreign objects, such as fragments of the pyrites falling between wheel and pan, are also easy to detect. Their depth, direc- tion, and jaggedness usually contrast sharply with the crisp and reg- ular forms of the milling marks.
Armed with this evidence, we sought out the opinion of mechanical engineers about whether the geometry of the wheel was suitable for cutting. Several expressed doubts, citing such technical handicaps as zero back relief and suboptimal cutting angles.17 Nothing would do but to try it.
Accordingly, I constructed a lock using largely tradi- tional hand tools and methods, according to plans published by Georg Lauber,18 the only ones which could be found at the time. Lauber's text reassuringly stated that the wheel could be used to groove the pan, but since he suggested brass rather than the more traditional iron for the pan and turned the wheel with an electric drill rather than by a hand crank, it seemed prudent to try to check his work. Unfortunately, several of his critical dimensions were found to be wrong, and it appears doubtful that a functional lock could be con- structed from his plans as published.
Also, electric drill speeds are much too high for optimal cutting. In order to give the wheel rim as much benefit of the doubt as possible, its spindle was originally turned out on a modern lathe to insure full concentricity.
Unfortunately it warped in heat treating. To overcome this, the spindle was installed in the lock, and a wheel blank with slightly oversized rim was mounted on it and turned down to size and trueness by cranking the wheel by hand while paring away at its rim with a jeweler's graver supported against back thrust by a collar bearing against a convenient adjacent screw. This took less than ten minutes and gave a wheel concentric to less than 0.05 millimeters.
It was then realized that this must have been the process used during the Renaissance, when lathes for trueing the wheel spindle were perhaps not yet commonly available. Subsequent examination of wheel rims for tool marks confirmed this, for the flanks of their threads showed the same parallel striations as the wheel slots, with one slight difference. A slight waffling of the surface, familiar to machinists as "chatter marks," and sometimes forming ripples run- ning approximately normal to the line of the striations, could often be seen. These are caused by tool vibrations, caused in turn by the tool being hand held. It is not unusual to find wheels whose rim eccentric- ity is far less than the eccentricity of their spindles, confirming that they were turned after being mounted in place.
I am sure there are a couple of people who already have this, indeed I wouldn't be surprised if one of you knows (or is one of) the authors. The paper can be had by going to a university or other library with access to JSTOR and downloading it like I did. If you can't do this, an email to thinkshard at yahoo.com with your email address may be helpful.
Last edited by a moderator:
