Talk:South-pointing chariot

Comments

Wouldn't such a vehicle need a flat surface to ensure the mechanism always points in on direction? And I don't mean changing the vertical direction of pointing. —Preceding unsigned comment added by 89.161.27.195 (talk • contribs)

Yes, the South Pointing Chariot only works on flat surfaces. More specifically, it doesn't work on a sphere :-) Shinobu 21:48, 10 June 2006 (UTC)

Proof?

@Two scholars prove before the court that such a vehicle is impossible:

Is this about the curved-surface thing, or did they miss the point and submit a flat-surface-based proof-with-a-hole-in-it? Shinobu 00:04, 11 June 2006 (UTC)

Steering

Note: This type of device can be used to create a "skid-steer" type of vehicle being Totally mechanically driven.

Copy vio?

The table seems have been cribbed from [1]. --IanOsgood 02:01, 26 January 2007 (UTC)

From looking at the disclaimer here it's most likely the other way around. Garion96 (talk) 11:24, 3 March 2007 (UTC)

I just updated the thread with information about Ma Jun from Joseph Needham's Science and Civilization in China.

--PericlesofAthens 05:48, 21 March 2007 (UTC)

dubious nautical use claim

well, this device just won't work on ships. if somebody historically significant really claimed it is actually used in that capacity, well we should state he did claim use, cite&verify and state it is actually impossible.--Calm 12:26, 2 May 2007 (UTC)

Dk just took all the historical information from Ma's article and placed it in the South Pointing Chariot article. The passage at the tail end of the Song Shu explains this somewhat, and I will refer to Needham again to get a clearer picture of this.--PericlesofAthens 12:49, 18 May 2007 (UTC)

dubious accuracy

Because of the difficulty of making the two wheels of very precisely the same circumference, and perhaps more because of the random bumps on even very flat plains, it seems that this vehicle couldn't be at all reliable even at a distance of the order of 10 km. - Tuomas —Preceding unsigned comment added by 89.59.12.109 (talk) 21:43, 11 September 2007 (UTC)

Crafting the wheels with the precise circumference was not a problem; I fail to see how bumps in the road affect the device. What does 10 km have to do with anything? I don't think you understand the pointing chariot very well and what it is used for. Furthermore, are you gaining this idea from a scholar? If not, Wikipedia does not endorse original research.--Pericles of Athens^Talk 03:38, 3 March 2008 (UTC)

Without wanting to get bogged down in a lengthy argument I just want to say that Tuomas' observation is absolutely correct. If for example the circumference of the wheels is about 1m, and the difference of circumference is about 1mm, then the SPC will point in the complete opposite direction if moved over completely flat ground for 1km (if a 2:1 differential is used). Also, if the ground is not flat then that introduces further errors. If the unevenness of the ground is assumed to be fairly random, than the pointer of the SPC will follow Brownian motion, with standard deviation dependent on the relative unevenness. If you need a reference on how the SPC behaves if the ground is not flat just read the referenced article by Santander. --ShanRen (talk) 08:38, 17 June 2008 (UTC)

Wouldn't the drift due to differing wheel size be constant, and as such possible to adjust for? 85.166.66.182 (talk) 20:52, 10 August 2008 (UTC)

Yes, it is in principle possible to adjust for this. (But what is the inaccuracy in your adjustment mechanism?) But even if you manage to perfectly adjust for this, and you can assure that your wheels are rolling without slipping, then this will only work on perfectly flat ground. If e.g. the unevenness on the ground is such that when moving along a straight line on average one wheel will follow a path that is 3mm longer per revolution (1m), you get a random error of roughly 1/600 (of a full turn) in the direction of the SPC per meter travelled. Then the expected distance after which the SPC will point in the opposite direction is about 10km. --ShanRen (talk) 21:27, 10 August 2008 (UTC)

Image

This article would be vastly improved if it had a photo of one, and/or a cool diagram like the ones at Differential (mechanical_device) with colored gears. Huw Powell (talk) 21:30, 21 June 2008 (UTC)

What it really needs is an animated diagram. I just added the "reqdiagram" tag accordingly...Tempshill (talk) 21:58, 14 May 2009 (UTC)

Timeline

Antikythera mechanism is not directly related to the chariot. It could provide a reference point, however it belongs to a completely different culture. Do we need it here at all? Cema (talk) 16:20, 9 January 2010 (UTC)

You're right, it is not needed here. I don't feel strongly about it either way; if you want it deleted, go for it. — Sebastian 07:45, 6 April 2010 (UTC)

Did the deletion. DOwenWilliams (talk) 23:19, 6 January 2011 (UTC) David Williams

Lack of precision, and implications

I've added a couple of paragraphs to the article entitled "Lack of precision, and implications". I said that simple geometry shows that if the track-width of the chariot is 3 metres and the wheels differ in diameter (or circumference) by one part in a thousand, then during a journey of one kilometre, the "south pointing" figure will rotate nearly 20 degrees. Here's a summary of the geometrical calculation.

Suppose, to start with, that the wheels are exactly equal in size. If the chariot goes around a circle, the outer wheel will travel a distance of 6pi metres further than the inner one. In order to keep pointing south, the figure (the doll on top) must rotate 360 degrees relative to the frame of the chariot during this circular trip. So the gearing must be such that the figure turns 360 degrees if one wheel rolls 6pi metres further than the other.

Now suppose that the wheels differ in size by one part in a thousand, implying that the diameter of the larger wheel is 1.001 times the diameter of the smaller wheel. If the chariot travels one kilometre in a straight line, then the number of rotations of the smaller wheel will be greater than the number of rotations of the larger wheel by an amount that equals the number of rotations that the larger wheel would make if it rolled just one metre. (Note that the actual diameters of the wheels don't come into the calculation. Only the ratio between them is important.) So the effect on the gearing of moving one kilometre in a straight line with these unequal wheels is the same as the effect of going 1/(6pi) of the distance around a circular path with identical wheels. The doll will therefore turn 360/(6pi), or 60/pi degrees, which comes to 19.1 degrees. I approximated this to "nearly twenty degrees".

My stepson could do this calculation in Grade 10. It's simple high-school math. Of course, nobody is going to write quotable learned papers about it.

Personally, I think it's extremely implausible that a chariot worked by a mechanical differential or anything similar was ever used for navigation over significant distances. The errors would have been far too great. Coupling the south pointing doll to the differential only when the chariot was going around turns would have helped, but leaves the problem of how the chariot was kept on a straight course without using the doll. When driving across Antarctica (where a magnetic compass is unreliable) in 1958, Sir Vivian Fuchs left a trail of flags, and kept his course straight by looking backwards and keeping the flags lined up. Possibly, the Chinese did something similar, but there is no record of it. Maybe the chariot existed only as a toy, or as a device intended to impress foreigners with the wonders of Chinese technology without ever being used in practice, except for public demonstrations - something like nuclear weapons today. The fact that there was no diffusion of these chariots outside China, except to Japan, suggests that most foreigners who discovered their limitations decided they were not worth copying.

DOwenWilliams (talk) 16:30, 6 January 2011 (UTC) David Williams

Mathematical Approximations

From the main page:

---

After this initial description of Yan Su's device, the text continues to describe the work of Wu Deren, who crafted a wheeled device that would combine the odometer and South Pointing Chariot:

....................

“ Left and right, too, were double gear-wheels (lit. tier-wheels), a pair on either side. Each of the lower component gears was 2.1 ft. in diameter and 6.3 ft. in circumference, with 32 teeth, at intervals of 2.1 inches apart. Each of the upper component gears was 1.2 ft. in diameter and 3.6 ft. in circumference, with 32 teeth, at intervals of 1.1 inches apart. On each of the road-wheels of the carriage, left and right, was a vertical wheel 2.2 ft. in diameter, 6.6 ft. in circumference, with 32 teeth at intervals of 2.25 inches apart.

---

Apparently, in ancient China, the value of pi was 3. Dividing the circumferences of the wheels by their diameters gives this value. But, for example, 32 teeth 2.25 inches apart would fit around a wheel with a circumference of exactly 6 feet, not 6.6 ft as stated.

Clearly, these descriptions should not be taken literally.

DOwenWilliams (talk) 22:33, 14 January 2011 (UTC) David Williams

Using automotive differentials

Differential gears are used in the drive-train of almost every road vehicle. It is quite easy to use a scrap automobile differential as the basis for a functioning South Pointing Chariot.

For example, take the rear axle from a rear-wheel-drive car. It will have a differential at or near its middle. Mount it on the chariot frame with the coupling that was attached to the drive shaft coming from the car's engine facing upward. Mount a road wheel on one end of the axle. On the other end, mount a 1:1 gear to reverse the direction of rotation, and mount another road wheel, identical to the first, on the gear. If you now roll the wheels along a road in a straight line, the coupling facing upward from the differential will not rotate. If you roll the assembly in a curved path, the coupling will rotate, and the angle through which it rotates will be proportional to the angle through which the chariot has turned. If you know the tooth-counts of the gears inside the differential, you can work out the proportionality constant. If not, you can determine it by experiment. Then you can make a gear assembly onto which you can mount the south pointing doll to complete the chariot.

DOwenWilliams (talk) 23:32, 6 January 2011 (UTC) David Williams

Inconsistent pointing

In the section "Geometrical properties", on the main page, I said that if two chariots leave from the same starting point with their pointers aimed in the same direction and then travel *by different routes* to the same finishing point (so both chariots start and finish at the same places), usually their pointers will *not* point in the same direction at the finish. This result, which many people find surprising, is easily demonstrated in an extreme case.

Suppose the starting and finishing points are both on the equator, and are 180 degrees apart in longitude, so they are on opposite sides of the earth. One chariot, starting with its pointer aiming south, travels eastward along the equator, so the pointer is at 90 degrees to the direction of motion. The pointer aims south the whole way, so it still points to the south when the chariot arrives at the finishing point.

The other chariot, with its pointer initially aiming south, goes northward from the starting point, with the pointer aiming backward. When the chariot gets to the north pole, it carries straight on, still with the pointer aiming backward, but as it now goes southward from the pole to the finishing point, the pointer aims to the *north*. It still aims to the north when the chariot reaches the finishing point.

Thus, at the finish, one chariot's pointer aims south, and the other north!

This extreme case is exceptionally simple. It is more difficult to visualize chariots moving between arbitarily located points on the earth. However, it is generally true that (with rare exceptions in which quantities cancel out) the pointers of two chariots that travel by different routes between the same two points will not aim in the same direction at the finish even if they did aim in the same direction at the start. If the starting and finishing points are close together, the discrepancy will be small, but it will not be zero.

Thus this type of chariot does not function as a perfect compass.

DOwenWilliams (talk) 21:00, 16 February 2011 (UTC) David Williams

Dimensions and gear ratios

A Han Dynasty stone rubbing of a horse-drawn odometer cart, showing the upper part of the chariot overhanging the wheels.

In the main article, I said that Yan Su's chariot would have worked as a compass if it was always turned so that one wheel was stationary. This, I said, would have been true if the pointing doll was attached directly to the large horizontal gear wheel, and if the track-width of the chariot (the space between its wheels) was the same as the diameter of the road wheels.

The calculation that shows this is simple. The large horizontal gear wheel had 48 teeth, according to the Song Shi description. The vertical gear wheels attached to the road wheels had 24 teeth. Thus the gear ratio was 2:1. (The fact that small gear wheels were used to connect the main gears is irrelevant.) Thus, if the pointing doll was directly attached to the large horizontal gear, it would have rotated, relative to the body of the chariot, half as fast as the rotating road wheel turned. This implies that the radius of the curve around which the wheel rolled, which equalled the track-width of the chariot, was twice as large as the radius of the road wheel. So the track-width equalled the diameter of the wheel. This equality of the track-width and the wheel diameter was common in Chinese chariots, which increases my confidence that Yan Su's chariot was like this.

Wu Deren's chariot was a bit different. The main horizontal gear had 100 teeth, and the vertical gears attached to the road wheels had 32 teeth. The diameter of the road wheels was 5.7 feet. Thus if the doll was directly attached to the horizontal gear, it would have turned at the correct speed when the chariot was turned with one wheel stationary if the track-width was 5.7x100/(32x2) feet, which comes to 8.9 feet. The description says that the width of the chariot was 9.5 feet, so the track-width would have been a bit narrower than the chariot. The chariot was about twice as high as the diameter of the wheels. It is likely that the upper part of the chariot overhung the wheels by a small distance, so it is very plausible that this chariot, too, was designed to be turned with one wheel stationary.

DOwenWilliams (talk) 15:31, 24 March 2011 (UTC) David Williams

Uncertainties, and questionable motives

I've done quite a lot of edits in the past few months, January to March, 2011. One of my purposes has been to give the article a more balanced viewpoint concerning the use of differential gears in South Pointing Chariots. Before my edits, the article implied that all chariots used differentials, and that no other workable mechanism was possible. Now, it says that some chariots may have used differentials, but others did not. Some examples of other mechanisms are described. In fact, it is uncertain that any chariots with differentials were built prior to the 20th Century, when this mechanism was proposed, and the chariot reinvented, by people who were familiar with uses of differentials in automobiles. Some of the ancient descriptions are of mechanisms which definitely did not involve differentials. By acknowledging these alternatives, the article is now much more balanced than it was a few months ago.

Another purpose has been to raise the idea that the South Pointing Chariot may sometimes have served another purpose besides guiding people across the steppes of central Asia, or wherever. The mechanical designs which have been proposed for it, including the use of differential gears, are inherently inaccurate if used over long distances. There is no way to make a mechanical chariot (except one with a gyro compass) which could be used for long-distance navigation. The Chinese must have been aware of this limitation, so why did they persist in building these chariots? I have suggested two possibilities: that chariots, probably toys, were used for amusement, and that they were used fraudulently to impress spectators. Possibly, the people who built the chariots defrauded their own employers with them, which could have earned them fame and fortune provided nobody tried using the chariots for real navigation. Maybe it's not surprising that the chariots were "lost" whenever conflicts flared up! Apparently, misleading people with fake technology was done then as it is now.

Plus ça change, plus c'est la même chose.

DOwenWilliams (talk) 21:27, 28 February 2011 (UTC) David Williams

Degrees of scepticism

Some people are highly credulous concerning accounts of the South Pointing Chariot. They accept that it was invented and used a very long time ago, maybe by the Yellow Emperor, that it worked very accurately, and that it used gearing technology - differentials - that was unknown in the West until the 18th Century, and not widely used until the 20th. At the other extreme, there are people who seriously doubt the existence of South Pointing Chariots in ancient China, and who think that the whole story was a myth until 20th Century engineers built the first working chariots, and thereby made the myth come true.

My own opinion is somewhere in the middle, between these two extremes. There seems to be enough historical evidence to show that chariots did exist. On the other hand, there is strong evidence, much of it rooted in hard scientific fact, that shows that the chariots can not have been as effective as navigational devices as ancient accounts describe. The accounts may not have been outright lies, but they must have been greatly exaggerated.

To give a balanced viewpoint, I have tried to make the article show both points of view, and to allow the reader to form his own opinion. I have added material about differentials, but I have also added material that casts doubt on their use in chariots.

Nobody will be entirely happy with this compromise, of course. Such is the nature of compromises.

DOwenWilliams (talk) 22:45, 14 May 2011 (UTC) David Williams

Why south?

Nowadays the arbitrary convention is that a compass points north. Is there any reason why (or confirmation that) all of these ancient chariots were focussed on south? (Or for that matter, does anyone know why the convention has reversed since?) Cesiumfrog (talk) 04:35, 16 August 2011 (UTC)

I think you're misunderstanding how the thing works. There's nothing inherent in these chariots that makes them "focussed on south". It just (ideally) keeps on pointing in the same direction it started with, that's all. You can see that with the following thought experiment: Pick it up, rotate it 90° (without allowing the wheels to move) and set it back down. Voilà - you got a "west pointing chariot"!

Or maybe you meant "why is the thing called "south-pointing chariot", and not "constantly pointing chariot" (常指車 or some such)? My unscholarly reply to that would be that it probably was named in analogy to the compass, which in turn can only be seen as pointing north or south. The choice which end of the compass needle you regard as its front is basically just a toss of a coin. (It's not completely random; I guess the west favored north because that's where the north star is, which was used for navigation, but there are good reasons to see the south as the main direction, too.) — Sebastian 05:39, 19 August 2011 (UTC)

Timeline section

The whole section is highly questionable. No individual item is sourced and many are unclear and/or worded in a joking manner. Until each claim is verified/sourced and its relevance properly explained, I think it is best to exclude it from the article.

In fact much of the article (especially in the second half) feels like original research and speculations.

-su88 (talk) 00:14, 9 July 2012 (UTC)

Let's face it, this entire topic is nothing but speculation. No south-pointing chariots exist, except ones made in the last century or so to demonstrate postulated mechanisms. Ancient accounts of chariots are generally contradictory and unreliable. Some "experts" genuinely believe that the south-pointing chariot never existed, except in legend. They may well be correct. DOwenWilliams (talk) 00:25, 9 July 2012 (UTC)

sheer speculation:

perhaps the mechanical differential combined with a magnetic compass.. the differential adding mechanical amplification to move the statue to follow the compass pointer.

 I ask this: Why point it south when it could merely be pointed at the destination instead? Wouldnt that be easier for the people who could see the navigation pointer to interpret?  Think of it's purpose.. to be seen by many, from a distance.  

(I know, bigger wheels also improves accuracy). accuracy would be important in the mechanical geared section, since a magnetic compass would supply little corrective force. But perhaps it would keep up with 'slower' angular changes. What kind of 'rod' does the 'immortal' carry? immortal=permanent, as in a permanent magnet? (ignorant speculation)

For clarification what I mean is that if I built one I would try putting the statue on bearings, and have a strong permanent magnet in it so that it would act like a giant compass. when the wagon turns, both the differential AND the magnetic forces act on the statue. The magnetic force alone likely would not be enough at faster turn rates due to the mass of the statue. Errors in the mechanical part would be corrected for up to a certain degree(rate), as long as the magnetic force is enough to overcome friction in the bearings and move the mass of the statue. if the mechanical part is accurate enough, 20 degrees a kilometer doesnt seem like it would take a very powerful magnet, like the rare earth magnets we have now. perhaps a meteoric lodestone? like I said, sheer speculation.

Alternatively, a smaller compass mechanism could drive another gear chain, as the magnet would supply a small amount of torque when angular error from south starts to accumulate, that would drive a second differential. For a simple experiment, take two small powerful magnets and clamp them on the end of a string and dangle them down so the north and south poles hang in the horizontal plane.. the magnets will torque the string as they swing around to face north-south... that torque could be geared down to turn a larger object slowly, or drive one side of a differential mechanism only when there was an error.. that would correct for any error from the wheel mechanism as the driving torque would stop when there was no error. Fencelizard (talk) 20:21, 7 January 2016 (UTC)

References

External links modified

Hello fellow Wikipedians,

I have just modified one external link on South-pointing chariot. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:

• Added archive https://web.archive.org/web/20110707140731/http://www.ancientengineering.com/14201 to http://www.ancientengineering.com/14201

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

This message was posted before February 2018. After February 2018, "External links modified" talk page sections are no longer generated or monitored by InternetArchiveBot. No special action is required regarding these talk page notices, other than regular verification using the archive tool instructions below. Editors have permission to delete these "External links modified" talk page sections if they want to de-clutter talk pages, but see the RfC before doing mass systematic removals. This message is updated dynamically through the template {{source check}} (last update: 5 June 2024).

• If you have discovered URLs which were erroneously considered dead by the bot, you can report them with this tool.
• If you found an error with any archives or the URLs themselves, you can fix them with this tool.

Cheers.—InternetArchiveBot (Report bug) 22:36, 26 December 2017 (UTC)