TubbaBlubba
Knave of the Dudes
- Joined
- Jul 9, 2010
- Messages
- 12,942
The Antikythera device is an absolute marvel. I almost feel that it could stand as a symbol for the prowess of engineering of mankind.
The Antikythera Device
- Thread starter JoeTheJuggler
- Start date
shadron
Philosopher
- Joined
- Sep 2, 2005
- Messages
- 5,918
Invoking necromancy spell here.
I became interested in the Antikythera mechanism couple of months ago when I came across the NOVA/BBC program on it (cited above). I went and looked up the Wikipedia page on it, and the Antikythera Mechanism Research Project website.
In operation, the handle on the side of the box is turned through one revolution once per day (that is a speculative rate; the wheel and crank are, I believe, not a current part of the mechanism, so Prof Michael Wright, who built the magnificent model most often used to demonstrate what the mechanism once was, using 200 BCE metal working technology no less, decided to use that rate in his model). That in turn moved the internal gears, resulting in the movement of five rear dials in the calendars and the two (and perhaps as many as seven) planetary pointers and a moon phase indicator on the front. These indicated the placement of the day in the metonic and callipic lunar/solar calendars, the 4 year olympic cycle, the saros and exiligmos solar/lunar eclipse cycles and the location along the ecliptic of the sun, moon, and possibly the other five known planets, and the moon phase. The moon position is corrected for the moon's elliptical orbit (moves faster at some times of the month than at others), and for the precession of that elliptical orbit around Earth on a 9 year cycle. Simply breath taking - take it from me, from one who has programmed the celestial mechanics to equal the heavenly body motions against the stars.
I have one question I wonder if anyone can address, and one speculation that I've not seen addressed anywhere. The question is: how would one go about setting such a mechanism for the day on which it was completed? This doesn't consist only in setting the moveable dials, but also in such internal things as the phase of the moon's elliptic precession, and the epicyclic gears of the outer planets. It seems to me to be key to its use. Such a thing must have occasionally been required, after being dropped or cleaned, as well. I would suppose it would have to be returned to the craft shop who would have an astronomer on staff for just such tuning and adjustment.
The speculation: I saw in one x-ray of the real mechanism that the calendar hands on the back were in place inside the corrosion of the original mechanism. I wonder if anyone might have taken note of them and tried to see if it could produce an accurate date for its loss near Antikythera (estimated by found coinage at about 50 BCE). Presumably the item was in use, at least up to the point where the Roman treasure ship annexed it. Then again, it might have been just a knick-knack, dusty on a forgotten shelf of the J C Penny's Five and Dime on Rhodes, that caught the eye of a drunken middie, and had been unused for years before.
I thought I'd bring these things up, in hopes that this corpse might be brought back to life.
I became interested in the Antikythera mechanism couple of months ago when I came across the NOVA/BBC program on it (cited above). I went and looked up the Wikipedia page on it, and the Antikythera Mechanism Research Project website.
In operation, the handle on the side of the box is turned through one revolution once per day (that is a speculative rate; the wheel and crank are, I believe, not a current part of the mechanism, so Prof Michael Wright, who built the magnificent model most often used to demonstrate what the mechanism once was, using 200 BCE metal working technology no less, decided to use that rate in his model). That in turn moved the internal gears, resulting in the movement of five rear dials in the calendars and the two (and perhaps as many as seven) planetary pointers and a moon phase indicator on the front. These indicated the placement of the day in the metonic and callipic lunar/solar calendars, the 4 year olympic cycle, the saros and exiligmos solar/lunar eclipse cycles and the location along the ecliptic of the sun, moon, and possibly the other five known planets, and the moon phase. The moon position is corrected for the moon's elliptical orbit (moves faster at some times of the month than at others), and for the precession of that elliptical orbit around Earth on a 9 year cycle. Simply breath taking - take it from me, from one who has programmed the celestial mechanics to equal the heavenly body motions against the stars.
I have one question I wonder if anyone can address, and one speculation that I've not seen addressed anywhere. The question is: how would one go about setting such a mechanism for the day on which it was completed? This doesn't consist only in setting the moveable dials, but also in such internal things as the phase of the moon's elliptic precession, and the epicyclic gears of the outer planets. It seems to me to be key to its use. Such a thing must have occasionally been required, after being dropped or cleaned, as well. I would suppose it would have to be returned to the craft shop who would have an astronomer on staff for just such tuning and adjustment.
The speculation: I saw in one x-ray of the real mechanism that the calendar hands on the back were in place inside the corrosion of the original mechanism. I wonder if anyone might have taken note of them and tried to see if it could produce an accurate date for its loss near Antikythera (estimated by found coinage at about 50 BCE). Presumably the item was in use, at least up to the point where the Roman treasure ship annexed it. Then again, it might have been just a knick-knack, dusty on a forgotten shelf of the J C Penny's Five and Dime on Rhodes, that caught the eye of a drunken middie, and had been unused for years before.
I thought I'd bring these things up, in hopes that this corpse might be brought back to life.
Last edited:
Imagine how far the ancients would have progressed if they'd have had Lego:
http://acarol.woz.org/antikythera_mechanism.html
Seriously though, the BBC documentary was fascinating. I particullarly liked:
* The bit where they demonstrate how easy it would have been to make a gear, and how the clever bit was actually the maths behind it, rathr than the device itself.
* The idea that the technology got passed from the Greeks to the Islamic cultures nearby, where they developed their own maths, and this idea of using gears got passed to the Europeans, who eventually used it to develop their own technology, and ultimately the Industrial Revolution. Or something like that.
http://acarol.woz.org/antikythera_mechanism.html
Seriously though, the BBC documentary was fascinating. I particullarly liked:
* The bit where they demonstrate how easy it would have been to make a gear, and how the clever bit was actually the maths behind it, rathr than the device itself.
* The idea that the technology got passed from the Greeks to the Islamic cultures nearby, where they developed their own maths, and this idea of using gears got passed to the Europeans, who eventually used it to develop their own technology, and ultimately the Industrial Revolution. Or something like that.
I read a book which had that contention as its central plank, it was pretty convincingly argued:
I can remember the argument that
The Romans did have water mills, but they weren't as economic as slavery. The Early Middle ages involved a lot of innovation. (with examples)
I can't remember the book because I got it out of the library sometime in the early 1990s...
HansMustermann
Penultimate Amazing
- Joined
- Mar 2, 2009
- Messages
- 23,741
Well, one thing you have to remember is that most machines were expensive, even if you knew how to make one. Metal was extremely expensive in the ancient world for example, and anything that involved precision fitting (e.g., if you wanted to make a steam engine) with manual work would be even more expensive.
That said, the Romans did use machines wherever it was economical to do so. E.g., water power was extensively used for such things as grinding grain (see, Pliny The Elder), but also for such things as the Hierapolis sawmill that used water power to saw stone blocks (also earliest known instance of a crank and rod mechanism to transform rotation into linear movement), or kneading machines (cf Vitruvius), stamp mills for crushing ore, and even a possible water-powered furnace (as in, the bellows part of it.) Most of that stuff is even older than the Romans, e.g., using water power to pump/lift water appears in Egypt long before the Romans ever heard of it.
So, yes, they used machines, where it was economical to make one.
That said, the Romans did use machines wherever it was economical to do so. E.g., water power was extensively used for such things as grinding grain (see, Pliny The Elder), but also for such things as the Hierapolis sawmill that used water power to saw stone blocks (also earliest known instance of a crank and rod mechanism to transform rotation into linear movement), or kneading machines (cf Vitruvius), stamp mills for crushing ore, and even a possible water-powered furnace (as in, the bellows part of it.) Most of that stuff is even older than the Romans, e.g., using water power to pump/lift water appears in Egypt long before the Romans ever heard of it.
So, yes, they used machines, where it was economical to make one.
shadron
Philosopher
- Joined
- Sep 2, 2005
- Messages
- 5,918
The original mechanism has gearing to drive a sun/moon ecliptic display (in terms of the Egyptian (sothic) year, which the Greeks used, which could be adjusted against the zodiac display to adjust for leap days) including moon phases and adjustment for lunar ellipticity (moves faster half the month than the other half) and a 9 year cycle of elliptic precession). It ran a Metonic clock (translated the lunar calendar to solar, essentially) and a solar/lunar eclipse clock. Finally it has a small pointer indicating the year in the 4 year Olympic cycle.
The latest paper I've been able to find is from 2012 (http://dlib.nyu.edu/awdl/isaw/isaw-papers/4/) by Freeth and Jones of the Antikythera Mechanism Research Project. It has been suspected that the Sun/moon display on the front had additional pointers for all the visible planets, which were run by gearing dissolved away while under the sea. Their motion would involve, of course, retrograde movements, so the gearing is not trivial ratios. This is the third proposal I know of to work out what that planetary gearing must have been; Michael Wright's physical model shown in the NOVA program was the first, and another was proposed by another team that required separate zodiac dials per planet. This third one also handles sun elliptical anomalies, since if they did it for the moon, the sun should have also been done. The picture of the front plate is striking: http://dlib.nyu.edu/awdl/isaw/isaw-papers/4/images/figure04.jpg
Freeth points out that the accuracy presumed by inclusion of corrections for lunar and solar ellipticity would have been largely nullified out by the roughness of hand-filed gearing and the use of triangular teeth. Presumably these corrections were added because they were known and could be modeled, the physical reality be damned. Freeth also points out that during Mars retrograde motion the pointer could be as much as 38 degrees off the correct setting, but that was not because of the mechanics but because they had no good theory of the ellipticity of Mars orbit, and Ptolomy's Almagest was still 300 years in the future. He notes that the use of circular gears for modelling required a good theory of epicycles, because that is precisely the model that has to be mechanically reproduced. Elliptical gears were just not in the cards
)).
I'm still wondering how they set the damned thing for, say, the day on which the manufacturing was complete. Setting the moon's current elliptical angle and angular velocity must have been a bear, let alone synchronizing all the retrograde movements to proper phases; then there's the five calendar pointers on the back. The guild of astronomers/mechanical engineers/craftsmen must have been a real wonk hangout.
It has been pointed out that Aristarchos had already theorized a heliocentric theory, and it is likely the people associated with the AM knew about it, if not believed in it. However, in 250 BCE, even then, Greek and Roman religious rigidity kept it a minority view and slowly squeezed it out of society. The AM (100 BCE) was likely the last gasp, stolen by the Romans and headed for a triumphal parade thrown by Julius Caesar when it rebelled and sank itself into the sea.
Sounds like a good opera scenario,anyway.
The latest paper I've been able to find is from 2012 (http://dlib.nyu.edu/awdl/isaw/isaw-papers/4/) by Freeth and Jones of the Antikythera Mechanism Research Project. It has been suspected that the Sun/moon display on the front had additional pointers for all the visible planets, which were run by gearing dissolved away while under the sea. Their motion would involve, of course, retrograde movements, so the gearing is not trivial ratios. This is the third proposal I know of to work out what that planetary gearing must have been; Michael Wright's physical model shown in the NOVA program was the first, and another was proposed by another team that required separate zodiac dials per planet. This third one also handles sun elliptical anomalies, since if they did it for the moon, the sun should have also been done. The picture of the front plate is striking: http://dlib.nyu.edu/awdl/isaw/isaw-papers/4/images/figure04.jpg
Freeth points out that the accuracy presumed by inclusion of corrections for lunar and solar ellipticity would have been largely nullified out by the roughness of hand-filed gearing and the use of triangular teeth. Presumably these corrections were added because they were known and could be modeled, the physical reality be damned. Freeth also points out that during Mars retrograde motion the pointer could be as much as 38 degrees off the correct setting, but that was not because of the mechanics but because they had no good theory of the ellipticity of Mars orbit, and Ptolomy's Almagest was still 300 years in the future. He notes that the use of circular gears for modelling required a good theory of epicycles, because that is precisely the model that has to be mechanically reproduced. Elliptical gears were just not in the cards
)).I'm still wondering how they set the damned thing for, say, the day on which the manufacturing was complete. Setting the moon's current elliptical angle and angular velocity must have been a bear, let alone synchronizing all the retrograde movements to proper phases; then there's the five calendar pointers on the back. The guild of astronomers/mechanical engineers/craftsmen must have been a real wonk hangout.
It has been pointed out that Aristarchos had already theorized a heliocentric theory, and it is likely the people associated with the AM knew about it, if not believed in it. However, in 250 BCE, even then, Greek and Roman religious rigidity kept it a minority view and slowly squeezed it out of society. The AM (100 BCE) was likely the last gasp, stolen by the Romans and headed for a triumphal parade thrown by Julius Caesar when it rebelled and sank itself into the sea.
Sounds like a good opera scenario,anyway.
Last edited:
shadron
Philosopher
- Joined
- Sep 2, 2005
- Messages
- 5,918
Actually, there are a series of hand-held machines from the Byzantine and early Arabic/Persian (pre-Al Ghazali, to be sure; 400 to 1000 CE, I think) cultures which displayed the day of the month and the lunar phase, involving 8 gears. Whether the Europeans got it from Persia, or by some direct link, or invented it largely themselves is not really known.