Hulot, père "L'Art du tourneur mécanicien" 1775
Prior to the Industrial Revolution, the two most common styles of lathes were pole lathes and great wheel powered lathes. Pole lathes featured a string looped around work held between centers. One end of the string was fastened to a springy wood pole and the other was attached to a pedal. The turner would press down on the pedal and the work would rotate in one direction. The work would be cut and then the turner would let up on the pedal so that the pedal would be pulled up, rotating the work in the wrong direction, and concurrently resetting the string for the next cut. This method worked well -- and a pole lathe was very easy for a capital-deprived turner to afford. The image above features a pole lathe in action at a French shop for wood and ivory turning. Interestingly, while the person in the back is sawing out stock, two people, one looking pretty young, are roughing out stock before it gets turned. Pole lathes work with an interrupted motion, and rounding the stock saves oodles of time. The lathe itself, with a springy pole attached to the ceiling, was pretty standard. This image from 1775 isn't the earliest image of turning, but it is representative (and also makes for a great opening picture for this blog post).
Joseph Moxon's "Mechanick Exercises" 1680
The second way of powering a lathe was using a great wheel (as in the picture above). A large wheel - as large as the turner had space for - was mounted in line with the lathe and connected with a string belt. A second person turned the wheel, powering the lathe while the turner cut wood. This model worked very well, giving the turner continuous motion, but had the downside of requiring a second person's involvement.
At the time of the picture (1680) water and water mills could have been used to gear up a wheel to power a lathe, as it was used for all sorts of larger machinery. I do not know know of a specific case where this occurred, and I am not sure if the mechanics at the time were precise enough to drive a line-shaft fast enough to be practical for a lathe.
Many permutations of the pole and great wheel lathes existed. The spring of the pole could be replaced with a bow, and all of the early books on lathes show pole type lathes with small crankshafts on the ceiling which could be pumped by a second person with the cord attached to the crank.
The main problem with a pole lathe was the interrupted rotation. The main problem with a great wheel lathe (and its derivatives) was the need for a second person. You could, I suppose, also use a treadmill apparatus for an animal, but I haven't found any examples of lathes being powered in such a way.
There exists a drawing from 1480 by Leonardo da Vinci that shows a treadle lathe. Was it something he saw, dreamed up but didn't build, or actually built what he designed? There is no proof one way or another for any of these options. The earliest image I have of a treadle lathe dates from 1680 and appears in Joseph Moxon's "Mechanick Exercises." That might be the first published image of a treadle lathe in literature, but I'm not sure. Considering how many pages are devoted to pole lathes and great wheel lathes in the few lathe books that exist from the 17th and 18th century, you have to make an assumption that treadle lathes were not that common. Moxon shows the treadle (below) on a special lathe for oval turning. It's doing fairly light work, but Moxon says that the treadle is useful for turning oval work, as is a great wheel lathe, because you needed continuous motion. Joseph Moxon's "Mechanick Exercises" 1680 Charles Plumier’s "L’Art de Tourner" 1701
Plumier, writing in France in 1701, has drawing upon drawing of pole lathes, great wheel lathes, overhead drive mechanisms. The first illustration in his book is of a large lathe which has an overhead hand cranked wheel and a pole option. But a few pages later, a large illustration (above) of another main lathe shows the same hand-pumped flywheel mounted on the ceiling as before - but no spring pole option. Later in the volume, he detailed the parts of this lathe, and the detailed drawings shows something interesting.
Charles Plumier’s "L’Art de tourner" 1701
Whereas in the main drawing the pull cord for the flywheel is clearly meant for hand action, the drawing above shows how to connect the flywheel on the ceiling to a floor pedal. As near as I can see from the drawings, the flywheels are just heavy wheels of solid wood and would work but are pretty light for a general purpose lathe. This is a big deal. Part of the advantage of a flywheel is that the flywheel stores energy and evens out the footwork so the lathe is powered all the time, not just on the downstroke of the pedal. The pedaler might be coasting on the way up, but the lathe is still drawing power throughout the entire stroke.
Later in the book, Plumier shows a light duty treadle lathe. This makes sense: a lot of the fancy work Plumier described required continuous motion and many of his readers were hobbyists who would want to work alone.
But these circumstances changed. By 1800 (and certainly by 1840), treadle lathes became all the rage - from the high end of Holtzapffel, to random low-end lathes for homeowners and hobbyists. I found many advertisements for these lathes. If you were a small shop or a hobbyist, the treadle lathe was the way to go. Great wheel lathes disappeared and factory line shafts appeared in mills. Pole lathes remained in use at least until the middle of the 20th century by bodgers working in the forest, where it was easier to turn chair parts in the forest then haul out raw wood.
What changed?
What happened?
It's easy to point to the nature of invention, to say that things have to be invented piece by piece by piece by piece. And the treadle is a natural evolution for small lathes. But I don't think that's the real case. If you're in a production environment, a great wheel lathe allows you to have continuous turning - as long as you have someone to do the powering. With the overhead drive motion seen in Plumier and Moxon and even earlier in Felibien (below), a turner could be very productive cranking the overhead crank if you had somebody else helping. In a production environment, a great wheel made a lot of sense. Pole lathes made a lot of sense as well because they entailed very little capital to build one, and they worked well for a one-man shop. Moxon's treadle lathe is a dainty little thing, with Plumier is not far behind. Plumier's overhead flywheel powered by a treadle could be operated by one person. The wooden flywheel could be big and store more momentum, but it would far from ideal. Plumier also illustrates a fairly lightweight treadle lathe, but it had a wooden flywheel.
I think the key is that there's a technical problem with all treadle lathes before 1780. The flywheels that you see in Moxon and Plumier are of wood - and they are simply not heavy enough. Holtzapffel tells us that the flywheel should weigh anywhere between 30lb and 100lb (for a big metal turning lathe). The flywheel also has to be machined and balanced, otherwise the entire lathe would shake and wobble. The cast iron flywheel on our treadle lathe weighs about 30lbs. But the operative word is cast iron. Cheap cast iron is an 18th century invention, and the ability to machine the grooves easily is a very late 18th century invention. While a great wheel lathe can be huge and made of wood, it gets its momentum from its diameter, not its weight. A treadle flywheel that has to live under a lathe bed and has a maximum diameter. It gets its momentum from mass. While it was certainly possible, even in the 16th century, to make a heavy bronze flywheel, accurately machining and boring it would be very difficult before 1800. And if the flywheel isn't machined, it's useless.
Up until the late 18th century it was possible, but very difficult, to can make a balanced cast flywheel except on a onesie-twosie basis, so it wasn't until the 19th century that machining a cast iron flywheel was cost-effective.
We actually tested a wood flywheel in an early prototype of the lathe. It worked, but not as well as the same same flywheel did when we bolted some metal plates to it. There are of course tradeoffs in weight, and some basic physics involved. If you have a light flywheel, the lathe is easier to start and spin, but it doesn't hold a lot of momentum, so heavy cuts slow it down a lot. If you have a heavy flywheel, the lathe is hard to start and you have to pedal for a few seconds to get it up to speed, but then it can power through work and its overall performance is better. In other considerations, a heavier flywheel is harder to move around, and it and costs more. Flywheel design is a compromise between usability and cost.We thought the learning curve would be easier with a cast iron flywheel, ideally in the sweet spot of a good bit heavier than a wood one, but on the lighter side of the weight range.
Incidentally, there's almost no literature on the design of treadle lathes. Part of the interest fueling this project for the past 3 years has been understanding the engineering why something was done and then actually testing it out - i.e., reinventing the wheel.
The drawing of a treadle-powered grinder (below) from Hulot is interesting. Chronologically, Hulot is later than the other references in this blog. The book published in 1775, with parts written very significantly earlier than that. But the grinder, unlike the traditional pole turning lathe at the start of this blog, is high tech. I can't tell if the flywheel is metal or not, but it looks metal to me, and in 1775 or earlier making it would have been quite a trial.
Lastly, Félibien's book is slightly earlier than Moxon's, but the lathe is nearly identical to the other lathes shown here. And we have the combo power option of pole or wheel. I originally thought that this was a drawing convention with a weird perspective, but I wonder if having a pole made it possible for one person to do pretty heavy turning, but having a wheel option allowed for faster, lighter work, if a second person were available to pump. (Lighter work because it's a wooden flywheel)
Hulot, père "L'Art du tourneur mécanicien" 1775 André Félibien "Principes de l'architecture, de la sculpture, de la peinture, &c." 1676
Join the conversation
02/25/2026 Ian
Thanks! Great article.
02/25/2026 Wink
Thanks, Joel, as ever… I have a really rudimentary question! Is the drive-string crossed over on its way from the great wheel to the lathe so that it will stay seated more reliably in the grooves on the wheels?
All the snow you’ve been getting must be a trial round the shop…
It’s been raining here in CA, but we welcome it. Less fire danger, and no need to shovel!
Wink,
I don't know - interesting observation. but not all the wheels have the crossed belts - see Félibien. You might be right as the crossed belts would rub each other but also contact the wheels over more circumference.
03/02/2026 Terence Mc
Crossing the belt reverses the direction of rotation. If the drive wheel is turned clockwise the lathe will spin anticlockwise. Looking at the plates it appears that the belts are crossed when both the person on the wheel and the turner are on the same side of lathe. When they are on opposite sides the belt does not cross. Intuitively it seems natural for a right handed person to tern a cranked wheel clockwise.
This level of detail appeals to me. The illustrator understands the process.
Terence,
On the treadle lathe Moxon has the belts crossed, but there is no reason to - a treadle can pull in either direction.
03/02/2026 Terence Mc
In the case of the Moxon treadle lathe it could well be that the two pulleys are so close together, that the drive belt was crossed to provide sufficient contact, especially on the smaller one.. I should have said that reversing the direction of rotation could be another. Yet another reason could be just to make the belt longer and easier to fit (there is very little stretch in a natural fiber rope compared to rubber or even leather drive bands).
Joel's Blog
Built-It Blog
Video Roundup
Classes & Events
Work Magazine
All the snow you’ve been getting must be a trial round the shop…
It’s been raining here in CA, but we welcome it. Less fire danger, and no need to shovel!
All the best!
Wink
I don't know - interesting observation. but not all the wheels have the crossed belts - see Félibien. You might be right as the crossed belts would rub each other but also contact the wheels over more circumference.
This level of detail appeals to me. The illustrator understands the process.
On the treadle lathe Moxon has the belts crossed, but there is no reason to - a treadle can pull in either direction.