A pair of physicists at Universidad Nacional de Mar del Plata, in Argentina, have created a computer simulation of the famed Antikythera Mechanism and in so doing have found that manufacturing inaccuracies may have caused the device to jam so often it would have been very nearly unusable—if it was in the condition it is now. Esteban Szigety and Gustavo Arenas have posted a paper on the arXiv preprint server describing the factors that went into their simulation and what it showed.
In 1901, divers looking for sponges off the coast of the Greek island, Antikythera, discovered a mechanical device among the ruins of a sunken ship. The mysterious device was dated to the late second or early first century BCE, and from that time on there has been much debate in the scientific community regarding its purpose.
Some markings on the device suggest it could be used to track time and astronomical events and even predict some others, such as the arrival of a comet, courtesy of its intricate gears and pointing indicators, by turning its hand crank. Since only one of the devices has ever been found, some have suggested it had an otherworldly origin.
Sitting at the bottom of the sea has led to a lot of corrosion and some parts of the device have clearly been lost. Both problems make it very difficult to recreate the original device to verify its operational use. In this new study, the research pair sought to learn more about the possible purpose of the device by recreating it using a computer simulation.
To do so, they used data from work done recently by Alan Thorndike, who developed a means for taking into consideration the non-uniform way the teeth would move, and work by Mike Edmunds that helped to account for imprecise manufacturing of the distance between the teeth.
In running the simulation, the researchers noted that the odd triangular-shaped gear teeth did not seem to have a negative impact on how well the device would have run. But inexact spacing between the gears, they noted, would soon lead to jamming if the crank was turned. Such jamming, they note, would have made the device impractical for scientific use. But they also note that the irregular spacing between the gears might not have been there before the corrosion occurred.
The research pair conclude by suggesting that if the device jammed all the time, it might not have been anything more than a clever toy, made for a child, although they believe that scenario is unlikely. They point out how much time and effort must have gone into making the device, and the craftsmanship that was involved. That suggests the device must have been more reliably made than their simulation shows.
