How were lead screws improved over Maudsley’s originals? Or does the accuracy of all screws trace back to those original screws? I understand how you can originate a flat surface or a right angle, but I can’t figure out how you would originate a lead screw.
Thanks for your help,
Rob
I used "Maudsley’s Lead Screw to google to get a .pdf document. “020000001.pdf” Titled “Filing and Fitting to Flexible Manufaacturing: A Study in the Evolution of Process Control.”
Consists of 120 pages.
IIRC the first hand made screw is used with a long nut to generate a new screw with less irregularities that the first, then you use the 2nd to generate a 3rd etc.
I’ve no answer for you However after finding Wikipedia’s bio of Henry Maudsley unsatisfactory (it’s the wrong man), I did manage to dig up a piece on the ‘originator of modern machine tools’. Now I at least know what you’re talking about.
[minor hijack] I once asked a similar question on rec.crafts.metalworking (usenet) “How was the leadscrew for the first engine lathe made?” I recieved (in Albuquerque, NM, USA) a reply from a man in Syndey, Australia, with an extensive answer to the question. He had an unusual yet familiar last name, and a first name that paired nicely with the name of a fellow I knew from time I had spent in Vienna, Austria. Turns out they are brothers. My best ever “small world” story. [/minor hijack]
Unfortunatly, I have long since lost that reply. But I do recall a bit of it.
You can make a crude lead screw by wrapping a wire or tape around a shaft and soldering, brazing etc. By first wrapping two “threads” in paralell, then unwinding one, you can get fairly close to an even pitch.
If you then cast a long lead or brass nut around that thread, you can apply abrasives, and work the nut up and down the thread. This will lap the thread, averaging out irregularities in the pitch. If the nut is split, you can take up the wear, and keep lapping. Eventually you will obtain a fairly uniform pitch, but have probably distorted the threadform to something non-optimal.
While such a lead screw may not be durable, it can last long enough to be used to cut a replacement with, e.g. an acme thread. By using overly long (by modern standards) half nuts on the carriage, you can average the pitch of the lead screw in use, making the new screw better yet. If you iterate this a couple of times, you can end up with a very accurate lead screw.
Nothing to add at all, really, but I just wanted to say, Kevbo, that your description of the process was too cool for words. It stirred up the distant memories of my machinist past (before I turned computer geek), and I wish I could go out and make a lead screw by hand, painstakingly using each generation to improve on its ancestor until…
I suppose that would be a pretty fancy home machinist’s project: a precision metal-cutting lathe made absolutely from scratch, using nothing more than simple hand tools. It would sure beat all of those fancy steam engines and the like that folks build in their shops.
I am not sure if this thread is referring to some specific process relating to lead screws and generations thereof, but is the implication that you cannot make a good screw without a screw to start with? Can a screw not be made by rotating a piece of metal with a cutting bit horizontally moving at a fixed rate? Such a contraption could be made with or without screws and any imprecision that can be detected or measured can be compensated for resulting in a screw as perfect as you’re willing to make it.
Of course if you have a screw based machine of such a design that a better screw in the machine would produce an even better screw you can use successive generations for refinement, but that’s true of any tool that can be used to improve its own components - what’s unique about screws?
I’d like to lead the quest for the improvement of screwing…
I was using a hammer drill one day when someone (a coworker) walked by and told me I was screwing to loudly.
There are certainly many ways to get it done today, but the fact is that it was all done with only Victorian era technology.
I can certainly create a CAD drawing of such a machine. MAKING it without using any machine or measurement instrument which itself relies on a lead screw for accuracy is another matter.
Measuring instruments:
The dependence of a micrometer on an accurate screw is obvious. A ruler, or vernier caliper has no screws you say, yet the markings on that caliper were doubtless engraved on a ruling engine controlled by what? Yup, our friend the lead screw. A dial caliper depends on it’s rack for accuracy. How do you cut an accurate rack?
The fact that you can bootstrap a process starting from what amounts to blacksmithing technology, and ending up with precision measured in .001”, or better, units seems almost magical to at least this geeky gear head.
You have also trivialized something else in passing:
The fact that machines even exist that can create (at least parts of) themselves is quite interesting to many. The first time I used my lathe to turn a repair part for that same machine* I was grinning for a week over the accomplishment.
I certainly get that many won’t find this stuff even interesting, much less remarkable. Fortunately, however, I’m not alone, and the history and some of the artifacts are being preserved for those of us that care.
*a shaft used in the quick-change gearbox of a 10”Logan.
Rowland made some improvements to the process while working out a method to produce diffraction gratings:
No, that part I understand and I understand the fascination with machines that create parts of themselves in general (although I always thought of them as fairly ubiquitous, why there’s one walking by here right now carrying another one! 
). I was just wondering if there is somethign special about screws. There is a lot of Victorian era technology (real or hypothetical using available tools) that allows one, at least in theory, to make screws without screws to begin with. Using screws as a starting point of screw refinement is not necessarily the easiest way either (although it might be, I’ll need a cite )
It just seems that, using only very primitive technology, you can use screws to make a lot of things that have higher relative precision than the screw you started with. A lot of those things can be used to make more precise screws. It doesn’t matter what you start with, although some things are a lot more convinient than others, but you have to start with what you have and then walk the path of least resistance. So what’s so special about screws? How are levels, cylinders, flat surfaces, angles, bearings, gears, chain links or U-joints all that different in that regard? As far as I know all of those are Victorian or pre-Victorian…
Count me among those geeks that feel that way.
I can understand being fascinated by the how, but why does “the fact that you can” fascinate you at all? I mean, if there was some sort of a fundamental limit on how precise you can get starting with primitive tools, that’s about as precise as we’d be able to get. Anything that’s already been done can definitely be bootstrapped from nothing - we have proof of that all around us.
For all those who are interested in ‘bootstrappy’ technology and don’t already know about it, I recommend the Lindsay Publications website. The Gingery series on making your own metal shop starts with building a metal-melting foundry to make castings, which you then use to build a lathe, with which you build gears and pulleys, with which you build a milling machine, with which you build a drill press, etc. (although Gingery does let you cheat by using a threaded rod as a lead screw for the lathe). The Lindsay catalog itself is a great read, and has lots of books dealing with ‘industrial revolution’ level technology.
In the large view, yes technology as a whole builds, and has ever thus, upon older technology. This is typically a long term process, spread over generations and many often obscure fields cross contribuiting.
Getting back to your previous question (paraphrasing) “why are lead screws special?” I think they stand out because they are a case where not just a single technology, but a specific part is used to create an improved version of itself. This was done, not by the whole of humanity, over generations, but by a single artisan in the course of perhaps a few weeks.
A screw is a slight modification of one of the basic machines. (the ramp) How the fact that something so simple can be used to evolve itself to near perfection in only a few iterations fails to fascinate should be your question, not the converse.
My shop instructor once had to cut an oddball thread in some aluminum tubing and none of the lathes we had could do it, so he did it freehand!
Oh, and I second brossa’s mention of the Lindsay catalog. (He’d make a great Doper, BTW.) Great books and great service!
Optics are often shaped from scratch and made arbitrarily accurate, like generating spherical surfaces by rubbing two surfaces together in various patterns. Optical flats (which have very flat surfaces) are made by rubbing three surfaces together in a rotating series of pairs.
The gears that point the Hale Telescope on Palomar Mountain were cast and machined accurately, then split and shifted by a partial rotation and lapped, then reshifted and lapped, and so forth. This is a pleasant convergence of gear technology and optical methods, I think.