[Leo Todeschini]

Hi All,

Last year I was involved in running the build on a TV show where we built a 6 span catapulta in a serious hurry. This was my first exposure to these weapons and my understanding was helped by the input from Alan Wilkins and Tom Feely and I really got the bug.

The simplicity and purity of design just hooked me and so even though I make all day for a living, I decided to make one myself, just because, and so those odd moments of spare time have recently been dedicated to making a 3 span machine and I will document some of the work here.

Catapultas are the Roman arrow shooters and were developed from early Greek artillery and they went through continuous development from their origins through to the demise of the Empire. We have been incredibly fortunate in that two sets of documents have been assembled from parts to create a partially complete set of design sizes from Philon, a Greek scholar from the 5thC BC and Vitruvius a Roman architect from the first C BC. Between these two incomplete sets of documents we have been able to reconstruct nearly all the dimensions of these machines. There are some differences between the two sets of data, but most sizes are the same and so we can use these, with a little thought and guesswork, to make pretty accurate machines.

A great deal of the development of our understanding is thanks to EW Marsden and his fantastic treatises and then the continuing works of his friends Alan, Tom and Len Morgan.

The sophisticated simplicity of these machines is what hooked me. As we all know, engineering does not directly scale up or down and yet these machines seem to.

The motive power of these machines is generated by twisting two bundles of cord by drawing back the arms that pass through the centre of the bundles. By twisting the bundles you are in fact stretching them and so they want to pull back and to do this they must return the arms back to the rest position.

Start with the bolt size you want to shoot, divide this length by 9 and that gives you the 'hole' size. This is the size of the hole cut through the top and bottom plates of the head frame that the ropes pass through. Every dimension of the machine is a multiple of this hole size.

The Romans built machines from 1 through to 3 span, a span being 231mm and this refers to the length of the bolt, so a 3 span shoots a bolt 693mm long (28") and the therefore the hole size is 77mm.

What is lovely about this system is that an artillery officer need not even have experienced men working for him, he can just hand him a measuring stick 1 hole long, and ask for a timber 1 stick wide by 1 stick wide by 19 sticks long for example. (main stock, called the 'case' in Marsdens treatise). In fact Roman artillery officers were issued books with tables in them, so that when they arrive at a new site, a fort for example, they can pace a room and measure the ceiling height and after consulting their tables they can see what the maximum size of engine they can fit the space. Genius.

The Romans used sinew ropes for the torsion bundles and the real secret of these machines is that they were fitted to the machine and pre- tightened even before any twisting and were stretched so much, that the rope reduces to 66% of its starting diameter before the next pass of rope is fitted. By doing this, even before twisting the bundles, the arms come under enormous torque. We cannot use sinew rope and so I have used a modern synthetic that is not likely to be as efficient as sinew but still I hope for a good result.

The Romans listed the shooting distance of a 3 span machine as 440 yds, being the most efficient size. The one I made last year was a 6 span, a size the Romans never built and is likely to be pretty inefficient and it still got 319m (340yds).

So my plans are as follows.

The wooden parts are all green oak, except the arms, which are laminated ash and olive, which is required to get the curvature. The curve allowed for around 30% more power as it increases the power stroke and was a later Roman innovation. The head frame will be heavily reinforced with steel and again this was a later Roman innovation and allowed for a longer lived and safer machine. The washers (top and bottom metal components used to twist the bundles) have been cast in brass and need to be machined. I will also be fitting a brass front cover plate that protects the bundles from incoming missiles and also looks impressive. I will likely make the string in Kevlar as this is a component I do not want to risk breaking.

The earlier stages of the build mainly involved cutting wood and so is not so exciting, but now that things are starting to come together I will post up pictures as I go and try to be as coherent as I can through a very detailed subject.

I hope you enjoy.

Tod

[Isaac D Rainey]

Amazing work, I can't wait to see it completed!
What kind of projectiles are you going to use with this machine?

[Rune Vildhoj]

Great initiative! - those torsion powered shooters of the antiquity, be it any kind of ballistas really, are fascinating indeed. So a good deal of praise to you for taking the effort to actually documenting a build of this kind Tod.

I am in complete agreement with your statement about the beauty of design - really efficient despite, or probably in part just because of the deceptive simplicity. A build like this has appeal both as an attempt of recreating a functional historical design as well as it is an ingenious contraption in its own right - thus speaks to both the passion of history as well as the engineer in me.

Rest assured that this build-along will be followed with avid interest even if it ends up spanning a longer period of time as a result of the need to earn a living in between..

[Marc Ritz]

-You mentioned that sinew rope cannot be used by you? Why is that?
You also mention that your replacement is probably less effective. What string material do you use and why would it be less effective? One could probably quantify these differences.

-Would it be possible to get a video of you prestretching the ropes? I find it hard to visualize in my head.

-When you spoke of the advantage of having greater power-stroke due to the curved limbs, I believe you misused the term. Power-stroke is the difference between draw length and brace height AFAIK. It would not be affected by any curvature of the limbs.
An increase in stored energy results from (if it is in any way similar to bows) the force needed to stretch the string being over-proportionally greater during the early draw. To test this and the efficacy of the build, could you produce the f/d-curve?

[Owen Bush]

I am exited that I am going to see this today.
Having worked on the 6 span machine last year . I am most interested in what this will do and am fully expecting it to be quite an awesome machine.
The machine last year was a most impressive beast , despite being rushed for a fast build television show.
I am expecting good things from this build (having had a couple of sneak previews of whats coming up) and think you are going to be in for a treat.

[Leo Todeschini]

Marc Ritz wrote

[Leo Todeschini]

I have started to plate the frame now and the central stanchion has been plated and through riveted and the two side stanchions have also been plated and riveted.

This system is shown very clearly on the Xanten machine and will make for a frame that is tough as hell and as I want to load the bundles on this machine until it screams, this is a good thing.

I have also plated and riveted the windlass hubs. The texts do not mention the Romans doing this, but again as I want to load the windlass levers heavily it just felt suitable to do this to ensure longevity.

The next step is to fit the front anti-missile plate and the top and bottom caps and I have something special planned for the anti-missile plate...........

Tod

[Mark Griffin]

Hi Tod,

a while back I had made a lot of horsehair rope for a project but it wasn't used so i have a good many meters in stock. If you have a project where you need it and are having trouble getting it (I had to buy horse hair in china, have it sorted elsewhere I can't recall and then woven up in Denmark) then drop me a line....

Griff

[Leo Todeschini]

Thanks Mark.

I have made some fairly slow progress since the last post as I have had a lot to get on with but the top and bottom trays have been cut and welded up and the washers machined.

The trays look modern, but looking at the Xanten machine they were pretty much like this.

The front armour panel is steel covered in brass repousse. I went this route as many of the fronts were decorated and the screaming face of Jupiter seemed appropriate. The face was taken from a fountain and I added the thunderbolts. Much as I would like to take the credit for the repousse, I did not make this part.

The winch is underway and should be finished in the next week or so and then onto the trigger.......

Tod

[Matthew Bunker]

Veeeery cool.
Baggsy on the crew if we can take it to Maiden Castle to skewer some Britons?

[Leo Todeschini]

Matthew, I am sure you will get to play one day, but thanks or the thoughts.

I have just finished the trigger system and made a start on the winch system - it is getting close now.

As soon as the winch is finished then I can find the balance point and get the whole lot mounted up.

And after that it is time to rope it all up!

Tod

[Leo Todeschini]

I have finally roped the machine up and have some more tweeking to do but so far so good.

I will post up some better pictures next week and do some testing.

Tod

[Tim Lison]

That is outstanding Tod! Please post some Youtube video if you can when you fire it! And, please blast some pumpkins or watermelons or a big bag of flour! Something that wil make an impressive spectacle!!!!

[Leo Todeschini]

Here are some better pictures of it.

I have been ironing out a few teething problems, adjusting string length and so on and am now happy with how it all feels. Until I draw it back fully I won't know how hard it will be, so I might have to up or down the pretension of the ropes, but at present all feels good.

I have not measured it accurately yet, but at around 12" of draw it is pulling around 250-300kgs.

Weather permitting I will have a proper shot on Thursday when I have a helper in who can help me spot as I have to shoot across a neighbouring field.

I will unfortunately not have time to mess about with it properly and do chrono tests etc on Thursday, but hopefully will get back to before too long - either way I will keep you posted.

Tod

[Martin Francis]

Very, very pretty Tod..... Purposeful and aesthetically pleasing. There is something in your soul that renders you incapable of producing anything other than elegant perfection I am pleased to say.

Perhaps it needs a home somewhere up on the Roman Wall..... Coincidentally, round about where Flora and I live..... I mean it would be an interesting counterpoint to the latchet bow; the balestrino; the windlass bow

Martin

[Ralph Grinly]

I've always been fascinated by these machines. So very good to actually be able to follow the construction details. Have you had a chance yet to actually run some live fire tests ? If so, there's a couple of things I've been curious about. At the moment of firing, I'd imagine there would be a certain amount of ?forward bounce? in the arms and bowstring before it returns to it's 'rest' position. When this occurs..is there any danger of the arms actually contacting the semi-circular notches in the frame ? If this happened, I'd imagine, given all the stresses involved, there could be a danger that the sudden shock could snap the arms of the bow ?
Also the ?slider" that the arrow/bolt rests on/in as the machine is cocked seems to project a long way out beyond the front of the machine when it's fired. I'd imagine that this would be rather a vulnerable point and prone to accidental damage or breakage when the machine is not in use . Is there any practical reason for it being this long ? Does it aid in the machines accuracy ?

[T. Kew]

[Leo Todeschini]

Ralph Grinley wrote

[T. Kew]

Hm, nice to know it's possible. Not that there's really any possible application I can think of for wanting less power. Maybe if you were wanting to arc shots through a window at a specific point on the ground below, but tha's incredibly niche.

Will you have this with you at TORM? It would make a rather lovely centerpiece for your stall (and I'd like to see it in person)

[Gregg Sobocinski]

That is a beautiful and powerful looking piece of work. Thanks for sharing the steps of this adventure!

(Yes, I also want to know when you expect to shoot it.)