Page 1 of 5

Antikythera mechanism

PostPosted: Sun Jul 06, 2014 10:08 am
by minimoa69
I'm sure that most of you good folks here will be familiar with the Antikythera Mechanism - the 2100 year old gear driven orrery like machine found in a ship wreck off Greece.
For those that only have vague recollections of this incredible device, it was a machine the size of a shoe box that could accurately predict where in the heavens to look for all the major heavenly bodies known at the time. It included retrograde motion for all the planets (the machine represents what can be seen looking out from Earth, rather than being a birds eye view of the galaxy like a traditional orrery), and it even went as far as showing the apparent speeding up and slowing down of the Moon's orbit. It's predictions were depicted on both sides of the box by pointers for the five known planets of the time (and of course the Sun and moon), and with time being measured on the opposite side with a spiral scale that spanned 235 months, after which time the pointer on this scale would have to be reset to month one again. Other pointers predicted likely eclipses, moon phases and measures of time. All this pre-dated the first gear driven clock by over 1000 years. Put simply, it was an incredible feat of engineering for it's time.
So why am I rambling on about it you may ask? Well after building many orreries over the past few years - all with wooden gearing - I've finally got around to having a go at building one of these amazing machines myself, and again, it will be built with wooden gears. All the gearing has been worked out by minds far greater than mine, so with a little luck I shouldn't encounter too many problems.
I'm not sure if anybody here will be particularly interested - it's not exactly astronomy after all - but I thought I'd share my efforts anyhow, just in case. The machine has been built before, but not recently, and not including the latest thinking on driving the superior planets, so mine will be a first if I can pull it off.
I've attached a photo of the first few gears that will drive the pointers for Mercury, Venus and the Sun, all sat on their carrier gear.
I'll post the odd update as time goes on.
Dave.

Re: Antikythera mechanism

PostPosted: Wed Jul 09, 2014 11:04 pm
by minimoa69
Pivots test fitted and all runs as it should! Next is to fit the pin carrier discs to the three outboard gears that will drive the retrograde motion pointers for each. All will become clear...
:-)

Re: Antikythera mechanism

PostPosted: Sat Jul 12, 2014 10:17 am
by dave.b
I've seen a documentary about this device. I seen to recall that some of it is missing and the rest of what we know has been extracted from CT scans of the remains. This has led to a lot of speculation, but a very impressive working model was built to demonstrate what it was thought to look like and show how it is believed to operate.

It just goes to show how knowledge was gained and lost and gained again over the centuries by different cultures all because of poor communications.

How are you cutting your gears? What type of wood are you using?

Dave B.

Re: Antikythera mechanism

PostPosted: Sat Jul 12, 2014 7:30 pm
by minimoa69
Hi Dave,
You are absolutely right - there is a lot of speculation and guess work, especially when it comes to the drive for the superior planets. Most of the folk that have done serious research into the mechanism over the years have pretty much unanimously agreed that the original machine would have featured the movements of the superior planets, but unfortunately this is the portion of the machine that is missing.
The latest research has concentrated on tying in the gearing that they do have, and the techniques used by the machines creator and applying those techniques to the superior planets and at the same time only using the space available in the remainder of the box (they do know exactly how deep the box was as they still have the support posts that separated front from back).
What they came up with - and this was only in 2012 - was a beautifully elegant set of gear trains that fit with the theories displayed in the existing part of the machine, and the whole new assembly neatly fits in with some 'mysterious' lugs and brackets that had never been properly explained before.
We may never know it's true workings, but this version is good enough for me to have a go at, and will hopefully look pretty impressive when done.
I'm hand cutting all the gears from 6mm ply wood - it's stable, doesn't warp (at least not at this scale), and I have about eight years experience of cutting it to make gears for my wooden orreries. The gearing in this type of machine is all low speed, low stress, so wood works brilliantly.
:-)

Re: Antikythera mechanism

PostPosted: Wed Jul 16, 2014 10:19 pm
by dave.b
Looking forward to your updates!

Dave B.

Re: Antikythera mechanism

PostPosted: Wed Oct 01, 2014 9:18 pm
by minimoa69
Okay, so progress has been a little slow (due to work and summer holiday to the USA!), but I have started to make progress again, and it all works! Unfortunately time is never really on my side with this kind of thing so it'll still be about the odd late night hour here and there. So please bare with me, but progress should now be a little quicker....

So anyhow, the latest shots show the three Pin Followers for the Sun, Mercury, and Venus that give the machine the ability to accurately reflect their retrograde motions. The triangular frame provides a tube that is fixed to the rotation of the large main gear wheel that will provide a fixed date pointer that all the other pointers can be seen moving against.

So I now have a central core that will provide the moons movements (no gearing done for that yet), followed by the nested tubes to move the pointers fur the Sun, the fixed (mean) date, Mercury and Venus.

Next up is the slightly (lots!?!) more complicated assembly for the pointers for Mars, Jupiter and Saturn (also showing retrograde motions, but in a mechanically more complicated way... yay).

When done, this new assembly will be mounted above the bit that I've just done via spacer rods. This next bits complication comes from some very clever differential gearing that I may just understand once I've actually built it...

Watch this space.
:-)

Image

Image

Re: Antikythera mechanism

PostPosted: Sun Oct 12, 2014 3:04 pm
by minimoa69
Hello folks,
Here's another quick progress report for you on this daft project of mine.
All in all it's going really well - one minor hiccup when I realised that I'd got the build order wrong and had put the little triangular frame in at the wrong level, a trivial thing that would made the planet output pointers in partially the wrong order, easily fixed though.
Other than that, all is good. The photo shows the adjusted unit with rejigged assembly order, as well the newly cut gears for Mars, Jupiter and Saturn, they're just awaiting their spokes cut. These have a clever slot and pin drive system that provides a variable speed output from a constant speed input - that should then give us the apparent non uniform and sometimes retrograde movement of the planets as the months go by - I hope...
More soon.

Image

Re: Antikythera mechanism

PostPosted: Sun Oct 12, 2014 9:08 pm
by dave.b
Dave,
I'm staggered by how good this looks in your pictures... And you're only part way there!

Can you provide some details of the process it takes to make a component like a cog? It looks like you've glued a paper drawing of the cog to the wood in your pictures. Does it include all the teeth outlines or just their position indexes? Do you use a gear cutting jig and file or is this all freehand fret saw work?

Finally, I wish I had your patience!!

Regards
Dave B.

Re: Antikythera mechanism

PostPosted: Mon Oct 13, 2014 1:50 pm
by minimoa69
Hi Dave,
Thank you for your kind comments - I've got to say that I'm probably more excited about this project than anything I've built for several years (and I do build all sorts of things).
The gears are all hand cut on a fret saw, with no jigs or cutting aids other than the computer print of the entire gear itself, all teeth included, that is simply glued directly to the wood and then carefully cut around with the saw. I generally cut out the disc of the gear first, just skimming past the tip of each gear, I then go round and cut down the side of each tooth, and then around one more time to cut the other side of each tooth and then, in the same action, across the bottom to remove the waste between each tooth.
It's quicker than it sounds and probably within an accuracy of around 0.2mm. The tooth pitch on some these is only 2.4mm, so you can't drift too much!! However, to put this apparent big headedness of mine into some sort of perspective, my version of the machine is over one and a half times bigger than the original - and that was built with incredibly simple hand tools over 2000 years ago, so with modern tools I've got no excuse really!
As for patience, it's probably my only natural gift - all the rest is hard work, lots of homework and not giving up after the not infrequent mistakes. Yes, I'm stubborn too!

Re: Antikythera mechanism

PostPosted: Fri Oct 17, 2014 6:35 am
by minimoa69
Morning chaps,
Well that's all the gears cut for the back end of the machine - the bit that drives the planet pointers. Next job is to create some frames to house them in and see if they actually work!  
Fingers crossed....
Once this little lot are working properly it'll be time to start the front end of the machine which appears to have enough gearing in it to sink a small battleship - and I can't wait - should be fun to do.  This is also the part of the machine that is fronted by the double spiral display of 235 months. That's going to be a challenge to cut - and possibly not quite so much fun to do... However, that's weeks or even months away yet - one step at a time Dave... one step at a time....

Image