VG Hart paper - Hatra superiority..

[Warhammer1]

I have not been able to bring up any threads discussing this paper. ( hatra ballista - a secret weapon of the past?) The short version or summary is that the Hatra (inswinging) is 48% more effecient than the Greek palintone outwsinger models. While this news is nothing new to inswinger fans, one would think the implications would go a lot further than that regarding the current debate on inswinger or outswinger design for both Hatra and Orsova machine.

I'd like to add that the paper has a co-author and contributer - an MJT Lewis of the UK. Mr. Lewis build two identicle machines both of the same power and limb length, one an outie and one an innie. He performed the ranging experiments and others, while Mr. Hart did the mathematics with both coming to the same conclusions. In other words the math supported the testing, or the testing supported the math. You figure it out.

The bottom line here folks is that while both the Hatra and Orsova COULD have been constructed as an outswinger, its performance in comparison to using it as an in-swinger makes the argument kinda mute?

For those who might disagree with Mr. Harts math, I too at currently in disagreement with him on a proposed Orsova design I am touting as superior to the Hatra, and that this new design is the most likely configuration of the Orsova frame providing a significant increase in obtainable velocities. While their is no academic evidence to support this claim, it is made clearly on the superior performance of it. Secondary and an independent agreement has been obtained from Nick and his work with his Orsova build and a number of tests involving the chronograph.

Its hard to argue with a math whiz with multiple degrees but then again I always like a good challenge. Especially when Im right LOL. Mr. Hart gracefully has accepted the challenge, confident in his mathematical computations, while I stand with reproducable upon demand evidence.
There can be only one person right, and with future testing and reporting of stats to come, we will soon see.

So take heart outswinger proponents, perhaps sometimes math loses out? Is Mr. Hart and Lewis wrong in their evaluations?

Lets have some lively discussion!

[Warhammer1]

Hey Duncan,

Heres a little evidence to the possible existance of hand held ballistas that could be used by soldiers, charioteers, and those who just want to raise a little mayhem LOL. http://www.youtube.com/watch?v=63on5rLS ... =1&index=9

My inswinger ballista bow is half that size and works just fine! Dont listen to the naysayers haha.

[D B Campbell]

I have not been able to bring up any threads discussing this paper.

Journal of Engineering Mathematics! On RAT! You're hopeful. hock:

Heres a little evidence to the possible existance of hand held ballistas that could be used by soldiers, charioteers, and those who just want to raise a little mayhem LOL. http://www.youtube.com/watch?v=63on5rLS ... =1&index=9

Mmmmm ... Ristorante pizza.

[Warhammer1]

Actually I believe the article is or was reproduced in full by Lewis SOMEwhere on the internet, and is written in German. I came across it sometime back, remembering the stats and the pictures. At the time the stats did not mean much because I dont speak or read in German.

After Mr. Hart sent me an off-print, I recognised and remembered the two ballista's. Identical except for the arms. Pity I did not bookmark the website. Vincent seems to be quite an agreeable fellow and did not mind me challenging some of his work. He may not agree with me but didnt write me off.

I seem to take great delight in challenging accepted works and work of others, but only after a lot of work and research. However, pure science does not seem to count for much here on RAT. Nick is about the only one who seems to respect my thinking, and is great to bounce ideas off and discuss things, trying some of my designs and theories. I will tell you though - he is no pushover. Every little change is hard won and he is an extremely clever fellow with great methodology and focus. He causes me to re-think things many times, and his videos allow me to observe rope bundle technology and how it relates to the design (as my springs are steel).

My steel ballistas allow me to change things very rapidly and inexpensively. Thus it is I believe I can successfully argue against the need for further stanchions in at least the Orsova which I believe contradicts Aitors work which I have read and studied as much as I can. When I say further stanchions It means stops. Aitors work involves having the heel of a limb come into contact with a (secondary/) stanchion to arrest movement and prevent string breakage.

Another misconception about optimal ballista performance is that the string must be under tension. In one of NIcks videos I noticed the string after a shot was slack (stretched out really) and slowly contracted over three or four seconds until it appeared to have some tension once again.

Should one use Philons wedges to tighten to springs instead of twisting them, arresting the movement of the arms is not nec. and not rec. for maximum performance. The wedge technique should always produce a superior velocity if done correctly. As the arms are not under any torque force, it always one to maximize effeciency by using a greater percentage of the stored energy. The more one twists the bundle, the less energy as a percentage can be extracted from it as energy is locked up. It also increases internal friction in proportion.

My little manuballistas do not emply any stanchions whatsover, and torquing the axles in two used to be a problem. But by allowing the spring to expend all of its rotational energy (slightly slack string) except inertia. My stage two inswinger (compounding) 18 inches wide allowing 180 degrees of arm rotation and 24 inches of draw I can dry fire repeatedly without damage if I allow the spring to expend all its energy as it has a natural and stationary rest point. No stops at all just a ever so slightly slack string.

When Nick fires up his machine again, I believe you will see this in action unless he tightens the springs by twisting them. Even so, no heel stops or needed or required - part of the original design. While NIcks machine does employ a stanchion, I believe it functions more for structural integrity and extreme compressional forces. The curved part of the stachion is there to allow extra rotation, and is why the tangs on the kam are offset. It can only be assembled one correct way, presumably the correct way is the way that produces superiour performance and ease of construction and operation. (What happens with no stanchion to arrest heel motion and the string is slightly slack? Simple. The curved stanchion becomes merely a fulcrum and the bundle and string bring the arm to a stop. The stanchion fails to arrest movement in other words and aside from acting as a fulcrum, merely serves as structural support for the crushing pressure exerted. It does however allow the arm to return to a consistant placement for accuracy.)
A greater energy potential means more velocity potential.

I agree that while our work done either together or separately may not be historically correct, it will help to understand the physics behind the machine and designs, so that in the end, both inwinger and outswinger ballista builder will ultimately benefit. In the future, some of NIcks work might involve improving bundle construction so that when twisted (tightening) the internal friction is lessoned considerably.

There is also work to be done on the proper size rope bundle crossbar, thickness and shape (concave or convex).

For some reason I get the feeling that perhaps this new build for the Orsova may be the last (for a good while anyway). Perhaps it is in his writing, or the fact that this ballista-ing and engineering will never end. I am for sure quite confident now that Firefly will indeed sail right past that elusive 400+fps ballista he has set his sights on.

For me, NIck taught me new respect for the simple rope bundle spring catapult, that I previously had simply dismissed as being inferior or flawed tech. and or engineering and that better and faster machines could be made with steel springs. Now that I have worked out where and why some of the flaws (room for improvement) exist, it is a simply matter of correcting them.

W.

[Warhammer1]

Anyone else have access to Hart and Lewis paper?

In studying both the machines and methodology, I do believe the results may be somewhat skewed although I still find favor with the inswinging design.

Both machines at maximum extension/rotation (full draw) were adjusted to be identicle at 10 kgs. All things being equal this MUST mean any pretensioning at rest would have skewed the results. We know for a fact that with the basic inswinging design of the Hatra, that during the first 33 - 35% of rotation, approx. 50% of the machines draw is acquired. Because the Hatra rotates exactly twice that of the experiments palintone, which maxes out at 52 degrees.

It is my argument to see if the inswinging arrangement TRULY is superiour to that of the palintone, both machines must be pretensioned exactly the same at rest. In this way at 52 degrees of rotation, both machines should have identicle loads (and draw) IF the extra rotation of the Hatra is the sole influence and not design dynamics.

I've yet to find someone to agree with me, but to my way of thinking, the experiment does not take into account the difference in the rotation dynamics, or the differences in lever movement to that of the string (amount of drawlength per degree of arm rotation).

Are the dynamics of the inswinger truly different? I think so. I wonder if both machines were pretensioned identically while arms are at rest, that at 52 degrees of rotation, would the draw tension remain identicle for the entire movement duration. I would have to say no. It is interesting to note that while the palintone had only 52 degrees of rotation and the Hatra 104, the total distance the missile was pulled back (total draw) on slider the Greek machine at 140.5mm and the Hatra 216mm.


While some may care to disagree on my theory on the Orsovas correct frame design interpretation (physical evidence) where on the original Orsova find, the longer tangs on the kam were to face forward, thus creating an offset creating some extra 15 to 20 degrees of rotation. Such an arrangement would have been perfectly in line with the earlier tech. advancement of assigning the earlier palintone and extra degree of arm rotation by either offsetting the intial half spring angles, or, introducing outwardly curved arms.

It only serves then to reason, that if the Hatra was first generation with 105 degrees of rotation, and given that movement past 90 takes up string slack and reintroduces it into the release, advancing the entire 105 degrees of rotation another 20 degrees to the outside would totally eliminate this perceived ineffeciency, as well as changing the entire range of arm rotation to amount of string draw. Given that any bundle material however wound has a terminal velocity assigned to it, peformance gains must lie in the design and engineering.

I will study the paper more, but without someone to argue/discuss/dispute this paper (and my arguments), Im afraid my work and writing are fairly pointless. If Samuli is here, please join in with your thoughts. It is clear to me that the formulas for the outswinger MUST vary to that of the inswinger. The inswinger MUST have a longer spring to take full advantage of the extra rotation available.

Thus, constructing identical machines with the exception of the direction of pull the arms take, for this test seems flawed somehow, and does not properly address the true advantages of the inswinger dynamics superiority. Inswingers, because of the extreme amount of rotation possible should have a much longer spring. It need not be larger in diameter, but certainly it should be taller, while the arms should decrease in length.

Without taller springs if both machines were equally pretentioned at rest, the Hatra at 105 degrees would generate a final draw weight expressed as a multiple to that of the Palintone. I doubt if the little machine could withstand the structural load, and would almost certainly result in the "evil huddling" (not to mention stuctural failure) Philon spoke of as the rope gets the snot stretched outta them, and knocks the axis off center.

The unusually tall spring bundles depicted on Trajans column are not needed for outswingers with their relatively small rotation, and the extra height would detract from performance as more rotation would be required to generate and adequate twisting action. Im also pretty sure some lacked sliders, but instead had a groove for the sliding and winched trigger block. This is a first draft, will edit later if anyone cares as Im sure I made a few mistakes to correct on a later edit.

I hope that there are a few here who will challenge the content of this post, and help identify where I have wandered off the path with incorrect assumptions or generated flawed reasoning within this posting. I do not expect to be right all the time Otherwise I wont be able to learn much more as a result of proceding with flawed data or reasoning.

[D B Campbell]

In my mind the conclusion that the Hatra is more effecient than the Palintone by 48% seems best explained by the fact that the Hatras draw length at 104 degrees of rotation is exactly twice that of the palintone (double potential energy factor). I cannot conclude that it is any more effecient simply because rotation is twice that of the palintone.

Hart's claim to greater efficiency seems to be based on something else, but his mathematics have defeated me.

I have always assumed that the inswinging design's greater power is simply owing to the greater angle through which the arms recoil. I think that is what you are suggesting, too. But Hart seems to imply something more. I'm just not sure what it is.

[Warhammer1]

More than anything else i think the paper goes to show that in the case of two machines having equal power upon release but varying drawlengths, the one with the longer draw or acceleration pathway, will always have the greater velocity unless the basic laws of physics have changed. This can be proved with two latex rubber bands of varying thickness who acceleration pathways are indentical.

In this instance, all the movement in lineor and parralel. There is no horizontal or secondary direction to provide acceleration. Thus, no deviance. The final outcome would be the same. A Thinner latex tube (slingshot) with a final pull of say 20lbs will have to have a longer stretch or elongation to reach that draw weight than a thicker heavier matertial. The thicker one will have less draw length, than the thinner one at the same draw weight.

Accelerating the same mass with the smaller tubing over a greater distance would have a greater velocity than the short thicker one with less of a draw. On youtube, there is a slingshot builder JoergS who has a slingshot channel. A video of his "hi-tech" slingshot with pulleys goes to show this quite well I think. He uses the pulleys to change the distance of the pull, but overall draw weight remains the same. The result is an approximate 20% increase in velocity. I will see if I can pull it up and post the link. Hang on...

Here it is. http://www.youtube.com/watch?v=DO_Meq0Hiac - from the comment section regarding pull weight, draw length and velcity as measured by his chrony: "The Chrony shows a speed increase by about 15-20% at the same draw? weight. I can get about 90 m/s max speed, with smaller bullets. It is still too cold for record shots, under 10 degrees centigrade. Rubber needs warm temperatures for best performance."

So you see, in my mind the experiments are somewhat inconclusive.

With the sliingshot there is only one directional acceration force acting, which I call a vertical or paralell movement. Ballistas, and most bows have two components to the acceleration factor. The is the forward movement of the limbs (vertical), and a sideways movement (horizontal).
Each of these movements contribute to the end velocity.

A perfectly balanced ballasta system will have a true 50/50 horizontal vertical make up. Increase one component over the other and the net velcity changes. More horizontal movement will result in faster velocities. More vertical than horizontal will produce less than optimal results.

Paralell limb tech has made its way into sports and hunting equipment over the past few years, resulting in an ever increasing average velocity as the tech. catches up. It really matters very little whether the limbs actually face forward or backward, as long as it is paralel limb tech, as there is a limit to much draw one can actually use with the current accepted designs.

JoergS and I got into a pretty debate on a forum, where I had claimed higher velocities than what he could achieve with a regular slingshot design in my compounding slingshots thread on slingshot forum. So Joerg set out to make his own compounding slingshot forays. You guys are not the only ones I constantly challenge. Im not prejudised. I manage to piss off or insult everyone LOL.

Of course, you know my torsion assisted compounding slingshot (inswinger of course) would be much faster than JoergS machine on the video.
While Kooi and Marsden might argued an elasticated string would only detract from attainable velocities, I would challenge that as well. You must remember how I got started with inswingers. I took a simple slingshot as a starting point, analylized where energy was wasted and creating the most effecient design in utilizing the available energy was the singular goal.

I ended up with a slingshot with the same inswinging limb arrangement as the Hatra, a stage one and most basic inswinger design.

Over the next 10-12 years it began to morph into something else on its own.Simply by addressing ineffeciencies of the current design, and continually improving.

The entire rotation cycle of the inswinger can be likened to a car or truck transmission with the rotation being degrees of rotation (engine rpm), but instead of the gears changing throughout the acceleration process, the amount (ratio) of string draw measured in inches or centimetres per degree of arm rotation constantly changes. Actually with the Hatra design, it is the number of degrees the arm moves to produce an inch of draw. It constantly changes through out the draw, and upon release. In the rest postion at 12 oclock, there is virtually no vertical component left. (Only horizontal. Any further movement past that 12 oclock position is actually a negative vertical movment,meaning the arms have now reversed their direction having passed the apogee. This may in fact add another acceleration factor (parametric resonance?) into the equation, but I am not sure. At worst, it cancels out some of the effect of the additional horizontal movement by an equal percentage of the negative vertical movement in relation to the continuing horizontal movement.)

(In any case the lever changes direction which simple science tells us causes an object to undergo further acceleration. You experience with your bicycle every day. This is just another juicy tidbit adding another small percentage of velocity gain, and right at the end of the stroke as well helping to create a flatter accel. curve. The extra 15 degrees is important. 20-25 works well, on up to about 35 max. From there on out too much neg vertical movement to offset horizontal gains. Huge untapped potential there, but I do not expect nor want Nick to completely rebuild everthing! He just finished the new Kam LOL Were he to add another 30 degrees instead of just 15, you would be tapping that chrony yet once again.)

It is the same with an outswinger, the amount of degrees to produce one inch of draw is radically different. So which is the most effecient?

In the basic Hatra config. the horizontal component is exactly twice that of the vertical. It is definitely not so with the outswinger. When you add pulleys and increase the amount of draw length in the Hatra, the effect is doubled according to the mechanical advantage you assign to it.

When you increase the amount of horizontal movement by adding an additional 15-20 degrees of movement to the Orsova design, you end up with a discrepency in performance between the two. Clearly the Hatra with only 105 total degrees of movement is at a disadvantage in range and velocity not only as the Orsova now has a greater usable rotational range. Using only 105 degrees of rotation in the new Orsova limb position (design) does not take the limbs past the point where they begin to move away from each other, as would the Hatra at 105. Because of this, the horizontal and vertical (dynamics) have also changed - the string has more movement per degree of rotation of the arms. It has been assigned more velocity potential.

From here, it is merely a matter of using the correct mass as a projectile to extract max. performance from the design. So in the end, using the older design palitone with less rotation than the offset springs offering up to 65 degrees, doesnt really matter in my way of thinking. The testing itself is flawed. The calculations may be correctly done, but it still does not conclusively prove the Hatra's superiourity as a design improvment.

I may be wrong because the math is all Greek to me (pun intended LOL), it may have calculations dealing with the differences and rate of change throughout the entire loading and release cycle - I just dont know. Only the methodology seemed to be not right, and the change in dynamics may not have been taken into account at all because of the complexity they would add to the calculations.

yikes, I think I even confuse myself with all that gobblety gook. So despite whatever our good college and friend in arms Mr. Hart says, here is what Nicks Firefly has to say about the entire affair:

The final sixth shot of the day, seen on the far right, is noticeably more deformed than the first five. This is because its velocity was boosted up to 270 fps by another major change that was introduced into the limb set up. I removed the bronze hardstops from the curved stanchion, and that allowed the limbs to rotate an extra 15 degrees past the usual 90 degree position they have always had relative to the string. This allowed for a longer string and a longer draw before a 4800lb draw weight was achieved. That represents a 50 fps increase basically for free. One may wonder why this 270 fps is remarkable, given that 285 fps was achieved the other day with the limbs in their normal 90 degree at rest position. Probably the pretensioning in the bundles was much higher the other day and that pretty well overwhelms any efficiencies that might be apparent from other factors. (Stupid me didn’t record the draw weight from then because the dynanometer was not zeroed properly , so now there’s no way to know for sure) Today’s testing was done with a deliberate attempt to keep the pretensioning of the bundles consistent throughout the experiment. Not maxed out, just consistent. That being the case, it really does seem that having the limbs rotate past 90 degrees boosts the velocity quite a bit."



Anyways, if you think JoergS slingshot rocks, add some inswinging levers to it properly proportioned. The same pull that loads the latex powerband on the Sprave slingshot, also loads an additional amount of energy which is then stored in the spring, effectively doubling it upon release. When you add in the horizontal movement (additional accel. component) AND the amount of extra stored energy in the spring to the total stored in the latex powerband, no way can JoergS design begin to come close. The laws of physics dont change on my say so.

Thats my story and Im stickin to it. At least for now LOL.

[D B Campbell]

When you add pulleys and increase the amount of draw length in the Hatra, the effect is doubled according to the mechanical advantage you assign to it.

The draw-length of the Hatra design can clearly be lengthened -- Mike Lewis has restricted it (to 103/105 degrees) by the positioning of the rear stanchions, but these can be moved. (I have not measured the theoretical maximum.)

When you increase the amount of horizontal movement by adding an additional 15-20 degrees of movement to the Orsova design, you end up with a discrepency in performance between the two. Clearly the Hatra with only 105 total degrees of movement is at a disadvantage in range and velocity not only as the Orsova now has a greater usable rotational range.

This is more tricky, as the Orsova design has a built-in restriction -- the bow-arms can only move within the range dictated by the vertical iron struts. (I have not measured this, but hopefully Nick based his kambestria -- the vertical spring-frames -- on the actual Orsova example.)

[Warhammer1]

While it is possible to get more rotation by moving the stanchions, it corrupts the original design of that from the Hatra dig or orig. iron frame.
Altering the machine to extract more rotation from backside would not have the same effect, if the rest position of the arms remained the same.

As you rotate the arm past 90 degrees from a 12 oclock rest position, you enter into what I call for lack of proper term negative draw. The string is at its very slackest when the arms at 90 are closest together. Going past that only serves to tack up the slack in the string, so that upon release all that slack it has taken up is released BACK into the movement so that once again, the string is completely slack as it again passes 90 degrees.

At this point however, one cannot ignore the limbs which have been accelerating under 5000lbs of pull for the last 15 degrees at 105 total rotation. At 120 degrees of rotation the problem is magnified. You have to keep in mind that most of the acceleration in an outswinger occurs very early on in the stroke, as it has the most application of available force before it starts dropping off.

Why going past 105 degrees of rotation on the Hatra is because of the "stacking" effect. The potential small gain is not worth loading the machine that high. Without altering the height of the springs, you also stand a very good chance of permanent damage to springs due to over stretching. You might get one or two shots (if no structural failure occurs), but then you would most likely have to change out ropes, or let them relax over a number of hours. Heres a link that shows what I mean about stacking. Look at the 36 inch spacing as it best illustrates what I mean. Going from 120 degrees of rotation to 135 serves very little purpose as we delve even deeper into the neg, it only gets worse and worse.

http://warhammer1.wordpress.com/in-swin ... s-ongoing/

Were one to alter the positioning so that it would allow extra rotation on the front end, you could still attain 120 degrees of rotation without ever entering that negative string feed zone. Modding the rear stanchion should also require further modding to let the arm rotate past the 12 oclock position. It then becomes something else and no longer the original design giving free artistic licence?.

Well, we know he is gonna make some new only slightly shorter limbs (increase draw weight for heavier projectiles) but uncertain on his design as he has entered stealth mode and not telling. It could be hes happy with just a 22% velocity increase. After all, he's had it up to 395 fps already.
A 22% percent increase in that is a lot more substantial than 425 fps. Nick has never stated he has ambition to go much higher than 400, as it was his initial goal.

If he feels he needs a little more, he can just builds it into the limbs by adding a slight curvature. The math looks good for 500 fps for a slightly lighter projectile than the monsters we are using. Also please remember we must have a built in safety margin, and the arrows bolts and spears he will be able to chuck outta thing have to be able to withstand some awesome acceleration. Any heavy steel points will only serve to add more compression upon release. RAther him be safe.

[Sean Manning]

Actually I believe the article is or was reproduced in full by Lewis SOMEwhere on the internet, and is written in German. I came across it sometime back, remembering the stats and the pictures. At the time the stats did not mean much because I dont speak or read in German.

After Mr. Hart sent me an off-print, I recognised and remembered the two ballista's. Identical except for the arms. Pity I did not bookmark the website. Vincent seems to be quite an agreeable fellow and did not mind me challenging some of his work. He may not agree with me but didnt write me off.

I seem to take great delight in challenging accepted works and work of others, but only after a lot of work and research. However, pure science does not seem to count for much here on RAT.

I think its more a lack of specialized skills! The same reason you don't see much textual criticism, or numismatics, or discussion of works not available in English. Only a few of us (definitely not including me!) are experts on ancient artillery with a strong grasp of mechanics and (I'm guessing) calculus. I have the math but not the physics and know little about artillery.

[Warhammer1]

I dont know if I can remember how to do even simple fractions and never did get the hang of algebra in school. It sure does make things harder I'll agree. Thanx for the comment.

W.

[D B Campbell]

While it is possible to get more rotation by moving the stanchions, it corrupts the original design of that from the Hatra dig ...

No, not at all. For the Hatra frame, all we have are the outside corners. And the washers, of course, marking the position of the springs. But Mike Lewis added another "back-stop" on each side (I simply use this term to give you a mental picture of what this stanchion is doing). This effectively stops the limbs as they are winched back. Without the back-stop, they'd swivel back to 6 o'clock, if you wanted them to!

As you rotate the arm past 90 degrees from a 12 oclock rest position, you enter into what I call for lack of proper term negative draw. The string is at its very slackest when the arms at 90 are closest together. Going past that only serves to tack up the slack in the string, so that upon release all that slack it has taken up is released BACK into the movement so that once again, the string is completely slack as it again passes 90 degrees.

From what you're saying, though, it sounds as if we don't want this extra rotation. Is that correct?

Maybe that's why Mike added the back-stops. Or maybe to try and replicate the iron frames (Orsova, Gornea, Lyon, etc.), which definitely do limit the total range of rotation.

If he feels he needs a little more, he can just builds it into the limbs by adding a slight curvature.

Good idea. But the description of the Cheiroballistra -- on which much of this research rests -- is pretty clear that the limbs are straight.

[Warhammer1]

Without the back-stop, they'd swivel back to 6 o'clock, if you wanted them to!

Of course they would if you provided enough force, but given the final string config is a triangle, the six oclock position would be kinda tough, but I get the point. The thing is, everyone pretensions the spring bundle as much as possible, which usually includes some twisting of the crossbar and rope. This would detract from the total possible rotation. This gets into another theoretical area for me. The real question is with the current formula for construction, how much rotation before structural failure of either rope, arm,or frame occur. I dont hear of anybody cranking up to the current allowable limits. Bad things happen even at lower power and rotation levels already.

Anyhow, removing the back stops is the wrong way, it should be the front ones, is more my point. The rotation should be added on the front end instead of the back. The only time I would add onto the back end is if I wanted to toss unusually heavy projectiles. I am a velocity addict and work on eliminating ineffeciencies. Since most of an objects acceleration occurs very early on in the powerstroke, why waste it in the backend of things where the limbs accelerate more than the string. The amount of addtional rotation needed to gain one inch of draw escalates wildly once going past the first 90 degrees.

It is far more beneficial to apply that initial energy where more speed is assigned to the string than the limbs. At extreme rotation the dynamics have changed so that its like a trucks granny gear. You can pull a lot of weight, but it just wont be very fast until you start shifting gears. As the arm accelerates and more rotation occurs, the amount of arm rotation (in degrees) to move the string one inch in draw changes. It decreases. MOre string travel per degree is assigned.

Thus it is you would engineer and operate the same machine differently for different purposes. Each would have different performance characteristics, according to how it is tuned. The Hatra with its arms at a 12 oclock rest position could/would be engineered differently altogether.

From what you're saying, though, it sounds as if we don't want this extra rotation. Is that correct?

Not for light bolt shooters where you are after velocity. I would refer you back to NIcks early experiments, extreme rotation, and an arrow that flew ahead of the string for about half the draw. Watch the videos. The arrows out accelerate even the arms as it gets the snot jerked out of it.
This you can clearly see on the slow motion vids. Where you do want extra rotation is on the front end.

Remember that as the arms pass the Hatras normal 12 oclock rest position, they reverse direction and begin to rotate back toward the shooter, even though the motion is curviliner. Even if the arm was not under power, it would still accelerate for another 15-20 degrees. In this case the heavier the arm in relation to the projectile, the greater amount of velocity added as the energy is soaked up and converted into acceleration.

Maybe that's why Mike added the back-stops. Or maybe to try and replicate the iron frames (Orsova, Gornea, Lyon, etc.), which definitely do limit the total range of rotation.

No. The string length (should) already limits that and an unnec. duplication. I will look up those other Gornea Lyon ballistas, as I dont recall them in my cyberspace wanderings. For me there are only two camps - inswingers and outswingers. The rest is academic. It is possible to load the arrow off centre of the string, or it is more than probable that the springs when tightened are not alike and a disparancy exists. The arm backstops would instantly indicate this. When perfectly set up, both arms touch the backstop evenly, similtaneously indicating everything is in balance and alignment and should provide for an accurate shot. I do not know how them Romans targeted stuff a quarter mile off, but the torsion inswinger is built for range and accuracy. It is also only half the width of the same power outswinger, but longer because of the draw.

If he feels he needs a little more, he can just builds it into the limbs by adding a slight curvature.

Good idea. But the description of the Cheiroballistra -- on which much of this research rests -- is pretty clear that the limbs are straight.[/quote]

Well...you got me there. The curved arms possibility stems right from Aitor site with the little manuballista cheiroballista or whatever the correct term happens to be. We seem to be getting back to academics instead of the physics part of, in which I have little to no knowledge. Maybe thats my fault for bringing up the hatra in that context. Maybe Aitor was talking about the outswinger when he suggested more rotation by curved arms, but I'd have to go back and look. Also I suggested it only as a possibility in the case Nick felt he needed more extra rotation than the original Orsova had built into it.

The velocity of projectiles from NIcks machine can go much much higher. However, designing an arrow with the proper characteristics to withstand sudden acceleration to 450 500 fps is another thing for roman days. This is why even today with modern aluminum arrows and carbon arrows, 400 fps is at the limit of arrow tech, perhaps with the exception of PSE TAC 15 arrows. The arrows must fly true and be accurate. Right now that is the limiting factor to speed crossbows. That and finding a suitable backstop to bring the arrow to a halt.

There is also a limit to the absolute or terminal speed of rotation for the rope bundle. HOwever consider 5000lbs of pull and a one lbs. arrow. Thats a hell of a power to weight ratio. Are we sure all of that power is being effeciently used? Are we exacting as much speed (flatter trajectory) as we possibly can?

We know that half the draw is achieved in the first 33 degrees or so, so the rest of the stroke is not as effecient in generating speed. During those last 33 degrees of rotation, the string must double its speed to catch up. It is this kind of potential velocity generating capability we are after, and where it makes sense to apply more power or torque during this last phase. By extending the rotation in this "speed zone" combined with a wiser use of available torque, is not a great problem to significantly up the end velocities. I may not have the articulation and proper language to describe it properly, but again I offer Nicks testimony with his Orsova 5000lbs machine as proof and vindications of my experimental ballista work.

Its one thing to play with my little metal rod machines, so its great that Nick helps prove them out on a proper ballista and provide confirmation of my initial findings. Sometimes as with the experiment with the shorter arms, results are not so clear at first and so more testing needs to be done. I have no control over that LOL. Most times gains are not so significant as a 50 fps per second increase in velocity, but when you make enough little improvements, the results add up. With Nick its simple. If it works and gives better performance without significant deviation, use it.

One other thing I will share with you for thought, concerns the fact that the extra 15 degrees of rotation where it changes direction and naturally accelerates. If it need not be under power to accelerate, how nec. is it for the bundles to be tight and pretensioned?

Could that cause any thing else to change? ONe could get 15 degrees of rotation before the ropes tightened up to where they normally might be IF the arms rested at a normal 12 oclock position. In turn that would add another 15 degrees to the maximum previous rotation capabilities. Or, with just a slight amount of pretention, he could still get 105 degrees of rotation without ever going past the normal 90 degree position and encounter the perceived problems with introducing string slack into the acceleration part of it upon release. These are questions and experiments for Nick and the rest if you care to play with them at all. It certainly does raise a question in my mind about pretension in rope bundles.

Last edit: Wow, everytime I see Aitors machine its like seeing an old friend. I have admired it many times over the year visiting Aitors site and is a good post to tide this thread over for a while.

Heres the latest update from Nicks site and progress on his new build, to be completed with stand and handwinch:

"As Firefly moves closer to completion it seems appropriate to examine her intellectual underpinnings. (Sounds serious, I’d better put the kettle on.) The great conceit of this project has always been that if we somehow accurately reproduce the original artifacts of the Orsova ballista and then design for them all of the ancillary components needed for a finished machine, and if we make and test the whole apparatus to encourage maximum performance; then that result would be as near perfect an analog for the original machine as this researcher can muster. (phew! burnt my lip on that one.)

This approach respects the original Roman artifacts in two ways. First, our close adherence to the dimensions of the parts dug out of the ground means that certain constraint dimensions will closely limit the size of the springs that will be allowable. This will insure that the scale of our machine is in keeping with the original. Second, by taking a ferocious interest in maximizing all issues related to performance, we can be confident our intent is exactly the same as the Roman engineers. It’s a fair bet to suggest that there is probably no level of honorable* innovation that I can bring to this project that would not have been mirrored and bested by the Ancients.

Blend all that verbiage together and it follows that the most humble set of assumptions about how our finished machine should look and perform is to suggest an apparatus that is as mind bendingly kick-ass as possible. We proceed accordingly."


Thanx for the great comments and questions. - W.

[Iagoba]

I cannot add more than a recent photo of Aitor's machine... I don't know if I have a photo of it loaded, if I found one I'll post it here too :wink: