« Previous PageNext Page »

I have been sitting far too long in my red chair, staring out across a vast expanse of green pool table, preening my laurels and just generally feeling very pleased with myself.  We are now up to 14 shots through a 3 inch circle at 55 yards.  There has only been one flyer in this group due to …. actually I don’t know….who cares?  The point is, all this self imposed regalia has turned me into even more of a douche than usual.   It really is time to get back to work.

Step 1,   …………..

Uhh…..Yawn!  Who knew that bliss could be so traumatic?  I can’t seem to stop visions of the Mk II from floating in.

……  No more!  No more! …… take it down Jimmy…..take it down.

I take the liberty of quoting some recent correspondence with Dr. Tracey Rihll, after I sent her some videos of Firefly’s latest exploits.  Tracey is the senior lecturer in the Department of Classics and Ancient History at the University Swansea and also author of that excellent book,  The Catapult.   Her extensive treatment of the subject is pretty much required reading for all catapult geeks.

—– Original Message —– From: Rihll T.E. To: Watts Unique Sent: October 19, 2011 12:58 AM Subject: RE: Firefly in action.

These are really useful Nick; thanks. I assume in the second one when you seem to touch a small device in front of the slider with a wooden pole, shortly before pulling the trigger, you are checking a velocity meter or something similar; please correct me if that’s not what’s going on.

Re: accuracy, the passage that springs to mind is Caesar, siege of Avaricum (Gallic War 7.25; see also 8.41, discussed in my book p. 97).

Have you tested range yet? It would be nice to know whether the figure of 400 yards/metres has any substance, and if so, what exactly!

All the best


Dear Tracey,

Yes, that procedure with the wooden pole is where I initiate the chronograph.  I have 30 seconds to make the shot before it stops transmitting the microwave radiation for the Doppler reading.

I have not tested for range yet.  That should be coming right up. However, my previous ballista  (the one I call the Gallwey and appears in the beginning of my blog),  had the same power of 1200 + foot pounds and 300 feet per second, and it would toss a 5, 000 grain bolt almost exactly 400 yards .  Lighter bolts made it out to 600 + yards. That old outswinger machine has springs more than twice the size of Firefly,  so it looks like the inswinger design really does favor high power in a compact package.

I realize now that utilizing the full 90 degree rotation of the arms and longer draw, are not necessarily advantageous in the inswinger design.  The power seems to come from inducing high levels of twist into the bundles when they are at rest, while, of course, keeping them as straight as possible.  Ideally the sweep of the limbs is short and violent, rather than long and sustained.  The draw lengths on Firefly have dropped to 33 1/2″  from the 50″ inches I had been using at the start.  This (and the fact they wouldn’t have needed space for the dynanometer) means that the Orsova engine could have been a good two feet shorter than Firefly, giving much more room to operate it in a cramped  guard tower.  It may also mean I need to make Firefly, Mk. II.

A quick word on the subjects of power and accuracy:  I feel that demonstrating how accurate these machines would have been is one area where experimental archaeology can make some valid contributions.  When we talk about “power” as it relates to these full scale reconstructions of torsion engines, we are invariably talking about the power derived from nylon springs instead of sinew springs.  Many assumptions about the equivalence of these two materials have to be made.  On the other hand,   the accuracy that the machine can produce is dependent more on quality construction, good mechanical design, and careful tuning, than it is anything else.  When it comes to the shooting accuracy of these machines, the modern experimenter can have a high degree of confidence that he is not straying beyond the bounds of what the Romans could have achieved.

Thank you for helping to keep me inspired these last few years as I worked on all this.  It made all the difference.  Nick.

What is that thing?

It’s a catapult. More specifically, it’s a viable reconstruction of an Ancient Roman torsion engine from the late fourth century.

Oh.   …… what’s a “viable reconstruction”?

That means it’s based on a very close copy I made of some stuff they* dug up.  It is also accurate enough to knock an ice cream cone out of your hand at fifty five paces, and powerful enough to staple you to, or through,  harder objects.

………..The Roman’s were mean.  They didn’t have ice cream did they?

Yes, the Romans were mean.  Probably no,  they didn’t have the ice cream.

Click for vid.   20111018115657(1) .    Range 55 yards, 286 fps, 1200 fpe.

* Messrs.   Gudea & Baatz,  Orsova dig, 1968-69.

Yes indeed.  Lipstick on a coconut.  Click for vid. 20111015144301(2)

Range from the coconut’s nose to Firefly’s nose is now at 55 yards.  This shot used a  6,651 grain bolt with a velocity of 285 fps.  Energy  was  1,199 foot pounds, about the same power as a hot .44 magnum load.   It was number nine in a consecutive series of shots that have struck inside a three inch circle at fifty five yards.     Note: other than the five shot group, and a  shot through the bulls-eye to confirm the readjustment of the sights, there are no other holes in the cardboard that would indicate multiple attempts  to make this shot on the coconut.  In other words, there was a high state of confidence that this would be a one shot affair.  Here is how the target looked directly before  finally cracking this nut.

Keeping in mind that each shot has to be individually aimed with aperture sights (a 65″ sight radius makes precision pretty easy),   and then triggered without any bobbles on my end, this is the kind of performance I’ve been dreaming about these last three years.  Here’s a vid from some of yesterday’s shooting.   20111014115447(1).    I have 30 seconds to make the shot from the moment  the chronograph is initiated with the broom handle.  Sure glad I haven’t flubbed the release yet.

Nine shots made one after the other into what is essentially a single ragged hole at fifty five yards.  (Not shown are two other shots on another piece of cardboard falling into this same pattern.   A sixth shot that was not marked also went through the center of the five that are indicated. )  Our continuing investigations here will probably not end this debate about whether or not inswingers existed.   They should, however, indicate that the inswinger design is a perfectly viable interpretation of the Orsova artifacts, and that it could clearly have been made  into a powerful and accurate weapon with a uniquely flat trajectory for the ancient battlefield.

I think I’ll take the rest of day the off.

P.S.  Draw weight 3400 lbs.  Draw length 33 1/2″.    More detailed records of shots 18 through 26  will appear in the posting titled, “Mk. VIII,  Trials & Tabulations”.  These lower draw weights and draw lengths are certainly making Firefly more of a pleasure to shoot than the over-cranked extremes we have toyed with in the past.

I’m too exhausted to warble on very long about this right now.  Suffice it to say that today was brilliant.

Four shots in something the size of my fist, and a fifth shot (#2) just touching that group.  Range is fifty five yards.  Bolt weight is 6,651 grains.  Velocity was within 10 fps of  200 feet per second for all five shots.  Draw length is 33 1/2 “.  Draw weight 3400 lbs.  My respliced string has slipped a bit.  Flexing the bowstring by hand reveals a definite decrease of tautness.  This, and the lower draw weight,  are probably the reasons for the lower velocity.     I suspect that the slippage in the loop splices has been arrested by the limbs finding their stopping point more by the stanchions than by the string.  Cross stitching the loop splices needs to happen.

The five shot group in this photo was preceded by two shots with velocities of 260 fps and 258 fps.  Both of them also went through the same hole, although on a  different piece of cardboard.  Then the string slipped and the velocity dropped, but not apparently the accuracy.

You can imagine my emotions at this moment.  Blissed out.

Sporty, Hey?

He’s not thinking what I’m thinking he‘s thinking, is he?

I’m rather afraid he is.

Does the Rebecca know about this?

It was her idea.

This is bad.  Very bad. You’re kind of trapped aren’t you?

Only if the short draw lengths pan out.

…………  Bad all around then?

Yup.  A sequel.  Cursed by the Catapult Gods and held in thrall for eternity.

Well, boo-hoo Cry me a river caveman.

I believe that Firefly’s springs are a bit too squat to utilize her maximum length of draw.  That is too say,  the springs would need to be longer to make use of her full 60″ draw capacity.  Her current springs would be pushing draw weights near 7,000 lbs if I were to take the draw to the full sixty inches. These short springs seem ideally suited to shorter draw lengths.

Shorter draw lengths have many possible advantages.  It may be possible to reduce the overall length of the machine by a couple of feet.  The original Orsova machine was unearthed in the remains of a tower.   If standing duty in some kind of cramped guard tower was a major function in their original conception,  these late model, iron framed ballistas may have been quite compact little units.  ……… No worries on Firefly’s long legged stature, though.  She was built for research.  Overcapacity is a plus, at the moment.

If I apply torsion all around by another 7 1/2 degrees, and be careful to adjust the draw length to keep the draw weight around 3700 lbs, this will insure the machine is not overstressed.  No harm, no foul.  At the moment some experiments along this line seem like a no brainer.  The benefits might be exponential.

Modelmaking is revealing when not limited by too much theory.  Observation precedes creation.

In this photo taken looking down the muzzle,  the starboard limb  (the one on the left) shows itself to be tracking a tad above the plane of the bolt groove.

Corrections can be made by inducing more twist into the top of the bundle which tends to drive the limb tip downwards.   I will tighten the top of the bundle by 7 1/2 degrees and loosen the bottom by 7  1/2 degrees.  This should keep the overall balance between the two bundles the same.  More tomorrow…..

……It’s now “tomorrow” and the adjustments mentioned above have been performed.

It’s kind of hard to see the difference between these two photos, I know.  However, the two shots I tried this morning appear to be flying true and with minimal wobble in the tail.  The velocity on “Little Walloper” climbed significantly after performing this limb adjustment.  No surprise there.  Leaving the  string should never be unduly traumatic.  Bolts are such sensitive little creatures.

Shot #16 was with a new bolt that weighed in at 6,651 grains.  It had a velocity of 287 feet per second and was packing  1216 foot pounds of energy.  Flight of bolt 95% of perfect.

Shot #17 was done with “little Walloper” at 8966 grains.  It had a velocity of 245 fps and generated 1,184 fpe.  Bolt flight 85% of perfect.  So much better.

Both of these shots were done with a mere 36″ of draw length and 3700 lbs of draw weight.  I intend to rack up a good deal more shots at this particular power setting as it does not appear unduly stressful on the either the machine or myself.  More later.

After resplicing a loop on my single remaining bowstring,  I managed to shorten the string by 1 1/2″.  Shot #14  showed a velocity of 285 feet per second with a 6,867 grain bolt.  Draw weight was 4100 lbs.  With a few more adjustments I’m confident that we can turn in  300 fps pretty consistently.

I suspect that it will not be in the cards to get much more velocity than that out of these 10 1/2 lb limbs.  The way forward will be to lighten the Mk VIII limbs  by a couple of pounds or so.  …………. Or make the Mk. IX’s.

Ain’t ballistaing fun!   At this rate I could stretch these mods into a job for life.  It don’t get no funner than that.

Little Walloper (that heavy 9,000 grain bolt made yesterday),  gets kicked out of Firefly at a fairly sedate 200 feet per second.  Click for video:     20111010120415(2)

Something must have changed between shot # 12 and shot #13.  Overall projectile energy is down to 796 foot pounds from 1300 odd foot pounds when we were shooting that 7,000 grain bolt.  I suspect the bowstring has loosened due to some slippage in the eye splices.   They really need to be cross stitched.  None of this can progress until I break down and order some more double braid for bowstrings.  Mr. Pat B has offered these insights into how imperative it is to have a tight string.

“If you remember the figures I  posted back when I first found your blog, they show that the velocity
ratio of missile-to-arm-tip doesn’t even hit double figures until the
last couple of degrees of movement. Then it shoots up exponentially.
The difference between inswingers and outswingers is that tiny
time-slice. The outswinger wins the contest for 99% of the draw length.
The inswinger relies on its performance over the remaining 1% to turn the
advantage the other way. But that dramatic turnaround depends on a
bowstring capable of tautening to the max. At the moment its slackness
is losing you a lot of velocity, and all readings would be thrown out
of kilter.”

I couldn’t agree more with his analysis.  You can still sense from the sound of her firing that Firefly is not actually breaking out in song yet.   Strumming the string yields no tone.

On another note:    I need to remount the electric winch.  The hand winch would scale much better if there was a selection of stout legionaries for the task ,  but as can be imagined,  these heavy draw weights get a bit much for his Lordship at times.  Click for vid.   20111010120415(3) .      And this is just part of what it takes to get it up to 3400 lbs.   This particular shot ended up at 4200 lbs of draw weight.  Talk about Pilgrims Progress.

« Previous PageNext Page »