We know surprisingly little about Thomas Savery — the inventor of the first widely-used steam engine. Unfortunately, his achievements were almost immediately overshadowed by the engine of Thomas Newcomen, and so he’s often only mentioned as a sort of afterthought — a loose and rather odd-seeming pebble before the firmer stepping stones of Newcomen, Watt, Trevithick, and other steam engine pioneers in the standard and simple narratives of technological progress that people like to tell. I’ve often seen Savery’s name omitted entirely.

He is even neglected by the experts. The Newcomen Society, of which I am a new member, has an excellent journal — it is the best place to find scholarly detail about all steam pioneers, and about the history of engineering in general. Yet even it only mentions Savery in passing. In over a century, Savery has been named in the titles of just three of its articles, and the last was published in 1986.

I think Savery is extremely under-rated, and deserves to be studied more. Fortunately, we do know quite a bit about the engine he designed, which he intended primarily for raising the water out of mines. Steam was admitted into a chamber, and then sprayed with cold water to condense it. This caused mine water to be sucked up a pipe beneath it, by creating a partial vacuum within the chamber and thus exploiting the relative pressure of the atmosphere on the surface of the mine water. Then, hot steam was readmitted to the chamber, this time pushing the raised water further up through another pipe above. Two chambers, the one sucking while the other pushed, created a continuous flow. (Video here.) We have dozens of images and technical descriptions of Savery’s engine, as it continued to be used for decades, especially in continental Europe.

But we know essentially nothing about Savery’s background, training, inspiration, or profession, and most of the supposedly known biographical details we have for him are simply wrong. There is no evidence that he was a “military engineer” or a “trenchmaster”, for example — mere speculation that has been repeated so often as to take on the appearance of fact.^[1] Savery simply appears out of nowhere in the 1690s, with his inventions almost fully formed.

Yet in Savery’s own writings we can see a few tantalising hints of how he thought, including some flashes of brilliance. I kept coming across them by chance, when investigating various energy-related themes with Carbon Upcycling. Take the following aside, from Savery’s 1702 prospectus for his steam engine: “I have only this to urge, that water in its fall from any determinate height, has simply a force answerable and equal to the force that raises it”. Savery here seems to be hinting at some idea of the conservation of energy, and perhaps of a theoretical maximum efficiency — in a phrase that is remarkably similar to that used by the pioneers of water wheel theory half a century later, and to Sadi Carnot when he applied the same ideas to early thermodynamics.

Savery, frustratingly, doesn’t expand any further on the point, except to then casually mention the notions of both mechanical work and horsepower — over eighty years before James Watt. Savery noted that when an engine will raise as much water as two horses can in the same time, “then I say, such an engine will do the work or labour of ten or twelve horses” — ten horsepower, rather than two, because you’d need a much larger team of horses from which to rotate fresh ones while the others rested, to keep raising water as continuously as the force of a stream could turn a water-wheel, or his steam engine could pump. (Incidentally, Watt wasn’t even the second person to use horsepower — the same concept was also mentioned by John Theophilus Desaguliers in the 1720s and John Smeaton in the 1770s. Watt just managed to get all the credit later on. Perhaps watts should really be called saveries, even if he was less precise.)

Savery also alluded to further insight when comparing his engine to the various schemes of charlatans. Unlike most mechanical contrivances powered by muscle, wind, or water, he had managed to almost entirely do away with friction — “except a little inconsiderable friction of the water against the side of the pipe”. In fact, he specifically mentioned that this was a big problem in pistons, which perhaps explains why he avoided using a piston to exploit atmospheric pressure like Newcomen did.

Savery’s first public invention was actually to replace ship oars with a capstan-driven paddle wheel. It sounds like just another madcap seventeenth-century scheme. But his reasoning is solid, perhaps even ahead of its time. Savery noted how force was lost from having to do a return stroke, when no useful work was being done, as well as from the impact of the oars on water. As he put it, “the harder it is struck, the less the thing that strikes it penetrates; which is plainly seen by a cannon ball, which being easily let fall into the water, sinks; but being shot into it with great violence, rebounds as from a mountain of brass”. Thus, he said, “a solid, steady motion” was best. This may not sound like much, but had he applied this reasoning to water-wheels, he would have been at least fifty years ahead of his time. It was only in the mid-eighteenth century that Smeaton and others showed that water-wheels were far more efficient when driven with minimal impact.

Savery certainly seems to have intuited this, and I think he may even have thought about it for water-wheels too. But unfortunately all we have is the following tantalising hint, of more that he could have written: “what may yet be brought to work by a steady stream, and the rotation or circular motion of a water-wheel, it would swell these papers to a much larger volume than was at first designed, and frustrated my intended brevity.” Perhaps, had he not been so concise, Savery might have earned himself a place in the history of water-wheels too.

Oh well. If there’s a lesson there, it’s to write more — at least somewhere — for posterity to find.

1. ^{^}

   Rhys Jenkins, ‘Savery, Newcomen and the Early History of The Steam Engine’, Transactions of the Newcomen Society 3, no. 1 (1 January 1922): 96–118.