Cornelius Drebbel's Perpetual Motions       
Cornelius Drebbel probably ranks as one of the most under appreciated figures of the Renaissance. Artist, engraver, alchemist, inventor, showman & manufacturer, his rude demeanor and odd manners seems to have kept him from true recognition as the brilliant mind he was. His genius was admired by the off-beat Emperor Rudolf II, but treated as a lesser amusement, despite signs of tremendous promise, by King James of England. At the funeral of James, Drebbel was there, but walked with the court jesters.

Before his foray into invention, alchemy and optics, he was a talented engraver for Golzius. He married Golzius' sister, and worked in his shop. During and just after his tenure there Drebbel released several fine engravings. I am lucky to have one of them, "The Judgment of Solomon". The detail of light and expression is extraordinary (and not represented by the below scan).



Drebbel invented the thermostat, the world's first feedback device. He discovered the first permanent, mordant, dye for fabrics. He isolated the properties of oxygen, a century before Priestly. For King James he built a 12 man submarine, which he navigated beneath the waters of the Thames river in 1620. He understood that wind was created by the heating of air, and could re-create it in his own chambers. Drebbel built camera obscura, using lenses he ground from glass he purified with his own processes, for clarity. To aid in his grinding of lenses, he constructed automatic lens grinding machines. He used these lenses to build a magic lantern, which he used to put on shows. An assistant would put glass slides of feathers or scales or fur in the device, while Cornelius acted out the animal, the texture projected on his white robe. The first compound microscopes introduced to Europe where models he built, between 1619 and 1622, which caused the art of microscope construction and the science of microscopy to blossom in Europe soon after. The legacy of fine 17th century Italian microscopes can be traced directly to Drebbel's refinements of the device. Huygens purchased both his first telescope and first camera obscura from Drebbel, in fact.

Drebbel is also well known, and probably most famous in his time, for his perpetual motion clock. This was an achievement which operated by taking advantage of changes of air pressure over time, turning them into a rotary motion. See below.

He is like the scarecrow of OZ, bits of him scattered through the history of Europe, known here for this, and there for that. Biographies often deal with one aspect of him, or several, but not all. The most comprehensive work on Drebbel was written by Dr. G. Tierie in 1932, and stands as the best I have been able to find. It is my personal theory that Drebbel may have been responsible for the enigmatic Voynich Manuscript, and I wrote an article outlining my reasons. This is The Voynich Manuscript: Drebbel's lost notebook?, which appears in issue #53 of Renaissance Magazine, March 2007. It has been lamented that Drebbel's notes have been lost to history, and I find the possibility that they might just exist, although hidden in code, very exciting.
Drebbel's perpetual motion machine was actually a device powered by changes in air pressure.  He would somehow use this motion to operate a clockwork, which would tell the time of the day, days of the week and months of the year. One model he made was reset to 12 noon by the sun striking a device (some expansion chamber arrangement, it seems). His patent says that it could run for decades, only stopping when the mechanical parts wore out.

I have long wanted to build such a device, and looked over the two illustrations and descriptions I could find. From the descriptions I have found, I've come to the conclusion there were two basic types of clock drives, both operating on the differential between the pressure in a sealed chamber, or chambers, and the ambient room pressure. But after that they differ in basic form. One type has a series of concentric tubes, connected in some manner, and partially filled with fluid. I'll call these the tube-type clocks. The other type had a spherical chamber. I'll call these the sphere-drive clocks.

Sphere-Drive Clocks

This is Antonini's drawing of Drebbel's "perpetuum mobile", when describing it to Galileo in a letter. Paraphrasing the description, we see a sealed sphere in the center, around which is a glass tube e-f (this corresponds to C-C in Tymme's illustration). The tube is open to the inside of the sphere using the tube D. The other side of the tube is open to the outside air, at c. The fluid in the tube, A-B, is therefore displaced towards B when the air in the sphere expands, and retreats toward A when it contracts. From what I gather... and this is an assumption on my part... this would cause the sphere and ring assembly to tip, due to the weight imbalance caused by the shifting water.

But whether or not this would set up a rocking motion, or simply find a new balance and stay there until the next change, I was not sure. So I wanted to build one, and test it out.

The sphere I used is a 10" diameter glass ball. The tube is sealed, via the white "cap", so that no air can go into or out of the sphere except through the tube. The axles, epoxied to the sphere, are small steel pins which afford limited friction. The liquid is dyed water.

I first balanced the device so that it would just barely sit at rest in an upright position. I did this so that any shift in the water would cause a weight shift sufficient to rotate the sphere on it's pivot.

It works! Well so far, the sphere moves to new positions according to the relative pressure of the air in the sphere to the air in the room. I gently warmed the sphere with a hair dryer... at such a setting and position that the air flow from the dryer had no effect on the sphere. This is of course "cheating", and only to demonstrate the action in a more immediate way. As it warmed, however, the water moved (to the right in this picture), and the sphere rotated clockwise. When I took away the dryer, and the sphere was allowed to cool for a few minutes, the sphere rotated back to the position shown.

When not cheating with the hair dryer the machine of course moves very slowly.

I've begun to think of the slowness of the rotation in a different way, and not as a detriment anymore. Rather than think of how Drebbel might have made a faster engine, I started to think of how the slow action would be an asset to him. If the sphere were attached directly to an escapement mechanism of a clock, this slow, steady force would always be running the clock. This, because it would take time to find it's point of equilibrium between the inside and outside pressure. But that point is constantly changing... so the clock would always have a force, as the engine was always searching for that point. The only times were there would be a problem would be when it found the point, or was passing through it. But perhaps this was infrequent, and of a short enough duration, so as to not throw off the time that often. But a clue that accuracy was an issue is found in that Drebbel installed one clock so that when the sun reached twelve, it would reach a hole in a wall, and so actuate a "re-setter". The  clock's hand would reset to 12:00 noon.

But the first possibility remains, that there was some far more simple method to measuring the movements of this engine, and translating them into some readout over time... like the simple "nudged" pointers I suggested earlier.

4/6/07: A couple of weeks after building this version, I saw a couple of painted versions (thanks, Mike!). The fact that the water level was shown "uneven", as in the Antonini drawing, reinforced the nagging sense my model could be entirely wrong, and Drebbel never intended the sphere to tip in order to affect his driving force. Perhaps the moving water alone was enough to thrill spectators. But we are still faced with the description of gear works, and clocks, in more complicated versions. So if the sphere did not tip, how did he drive the clock? Antonini shows one end open to the atmosphere, so eliminating the possibility of tapping into any increased pressure on that side (by a piston or other). I am going to carefully study the painted versions, and see if I cannot answer some of the questions.
I made it down to Baltimore to get a shot of one. Here is a picture of the Drebbel spherical clock from the Walters Museum in Maryland. If I had had this image before the Antonini drawing, I would not have made my model rock, but rather, rotate.

From this picture we learn several things. First of all, we can see what an extraordinary glass worker Drebbel was. To make a tube like this, so perfectly round, would have been quite a feat at a time when most blown glassware was very uneven, and inconsistant.

There is a base on this unit. There is another version of the same scene in the Prado, which does not have the person to the left. In that shot, there is a device of some kind, linked to the base somehow. It appears to have metal or glass tubes, tilted to face the sun coming in the window. I believe something may be in that base, but cannot be sure.

Also, we have a clue as to the way the indicators worked. There appears to be separate brass disks, possibly representing the sun and planets, in the center of the sphere. I believe the clock drive may have been in a chamber in the sphere, but separated from the inner pressure chamber. There is not room here to go into all the reasoning behind this, but it will be my starting point.

I've begun to build a model of this unit. I will assume at first that there is a piston type drive unit in the scrollwork at the top of the tube. Drebbel was quite familiar with pumps, so I will proceed this way at first... using the changing pressure on one side of the tube to move a piston. This will rotate a ratchet mechanism, and wind a light spring. This will rotate the sphere. Did the sphere rotate, or just the hands move?


Tube-Type Perpetual Clock

This is an engraving of the clock in it's entirety, from Thomas Tymme's 1612 Dialogue Philosophicall.

This elaborate model was made for King James I. What are the odds that bits of it are crated away in a London basement? Tymme's description stated that "A", the globe, represented the Earth, "C-C" was the glass tube, with water in it, to represent the movement of the tides, and "B" was a rotating model of the phases of the moon. He also states there was a clockwork in "A", and that the movement was so slow so as to limit wear on the parts. From this and other descriptions, it seems to me... purposefully or not... that the daily variations of air pressure was being confused with the cycle of the tides. This because the fluid in C-C would only be affected by the pressure in A, and not by the moon's gravity (of course). Then Drebbel somehow used the level in C-C, or the changing pressure of the air above it, to rotate the moon phase indicator at B. So the question is: Did Drebbel make a mistake in associating daily fluctuations in barometric pressure with tides, or only sell his clocks as doing so, while knowing the truth of the matter?

I wonder at the clockworks, too, after reading the description. I have to wonder if there was some very simple method in this, of marking the rotational movement of the globe, without a complicated works. Like the pointers simply being nudged around the ring somehow. Until I find more descriptions, and play with my version a bit more, I can't know for sure.

Here is a blow up of the clock from the Tymme engraving. Although Tymme stated that C-C is the outer "crystal ring", I suspect that he is actually referring to the ring just inside, as it appears to be a glass tube, and connected to the cross-piece from C to C (c to shining c?). The outer ring appears decorated, and solid, in this drawing. A is described as the globe, which is marked in twenty four hour increments. Both Tymme and Drebbel state that this rotates, and Tymme states it had a clockwork built in. For these reasons, and it's small size, I suspect this is not the sealed air chamber as in the version in the Antonini sketch. I am tending toward the belief that the air chamber is actually in the base of this clock, which is unusually large in relation to the works.

Looking closely at the engraving, other important details come to light. For one thing, in the outside glass tube, just below the place the cross tube joins it on the left, you will see what appears to be a water level. The small tube above the inner ring, and attached to it, has two marks on it. Just above the lower mark there appears, faintly, a ball. My guess is this is the filler neck to replenish the water, and the ball and lines are a filler gauge. Periesc asked a friend, whom he insisted should see this very clock, to ask the attendants how often the water needed replenishing.

Another feature is the center man. He appears to be a decorative support, perhaps for the globe's axle, and maybe the cross tube also. It seems the seraphim were to support the outer (metal?) ring, and the man the axle and glass. Perhaps, if the air chamber was in the base, a tube also ran up the man to the mechanism.

I wanted to study the existing sketch and descriptions, and make a CAD model of the clock, in order to better understand how it might have worked. To start, though: as I said on the Drebbel page, this clock, or at least it's motor, was basically a mechanical barometer. As such, it would not have been able to sense or indicate the tides. So if this was a claim of Drebbel's, he was either being dishonest, misunderstood, or in error. I tend not to believe the latter, as the man did certainly understand that the force acting on his circular column of water was the difference in relative pressure between his sealed chamber, and the outside air.



While making the CAD model, I felt accuracy was very important. Although Tymme's written description is agonizingly incomplete, the artist who engraved the clock seemed to take pains to include much technical detail. I felt they likewise must have cared about proportion, and proportion is the make or break of this technology... the relative size of chambers and tubes, the angle of moments which would affect the effort exerted on the axle, etc.

To keep it accurate to the drawing, I textured a plane with the actual engraving, and built the clock on that.
This is where I am at so far. Basically, I have form but no function. Like a Pinto with the connecting rods missing. But a drawing like this, combined with the descriptions, the working single tube clock engine, the other clock technology of Drebbel's era, I may be able to fill in the missing pieces. At this point I cannot be sure if the outer ring and crosspiece rocked or rotated, or if the fluid in it alone moved. And I suspect the other tubes had fluid, too. Different rates and timing? Not sure yet... maybe never.

One possibility occurred to me as I was researching this clock. I came across Kircher's Clock, which also used a smaller globe to tell the time. Kircher's globe was also marked in 24 hour increments, but it floated or hung in a globe of water, with a fish model as an indicator. The claimed operation was that the globe was remaining still, while the Earth and the clock rotated under it. The problem? It was a trick! Kircher had a water clock hidden in the base, which turned the globe via magnetic traction. Not to impugn Drebbel, but I wonder if his machine was part reality, in the barometric action part, but a parlor trick in the rotating globe part. This would be an especially strong possibility if the glass tubes did not move in this clock. Kircher's clock relied on an attendant secretly replenishing the water of the water clock in the base when no one was looking... Drebbel's clock, if powered the same way, would have had the advantage of allowing a fill in front of the awed onlookers! A small amount could have run down a tube, into the base, and run a water clock. This could have driven the globe easily via a small shaft through the body of the man sculpture.

Again, I don't want to second guess Drebbel on this, but only want to point out what is a practical, technical and historic possibility... as one option.
  



Visitor Counter by DigitsWebcounter by Digits!