Supersonic Trebuchet | Hacker News Reader

Supersonic Trebuchet

hackaday.com

650 pointswyrm236 comments

Yes he ends up with a cute toy that flings a marble but sounds like a rifle. But in the video he explains the engineering process to start with the desired goal and then optimized and derive the parameters he needed to achieve that result. It was a good example of engineering.

Interesting how different his device is, from historical trebuchets. In particular, they seem to maximize arm length, while he seemed to go for the shortest arm he could. I wonder if it is entirely the materials difference, or if his optimization was also much better than theirs. It would be cool to see his design built in period materials.

He built his arm first out of plywood and I’m pretty sure you could easily build the same design out of a high quality single piece of wood. It really comes down to an extremely well-optimized design.

Right, but I think its the high quality elastic bands that would be difficult to replace.

They could be replaced with either a weight, steel springs, or perhaps rope or sinew cords like the Romans used. Rubber is more effective and compact, but watch the video and you see there’s nothing particularly special about using rubber & he could’ve picked something else. The rubber puller itself does not exceed 10 m/s.

I agree he could have used some other spring, but I don't think he could have used a weight. The acceleration of springs/rubber is potentially much greater than a weight (at least one starting at rest).

You can use mechanical advantage to get the same result.

Which means a giant physical structure to support that giant lever arm. When you scale up to things that can knock down castle walls I suspect the difference is incredible.

I think there is something particularly special about rubber in the design. He stretches it extremely tight. Even with the leverage of a crank, he strains to compress it to the length required, and the rubber holds up. It might be possible to achieve the same result with older materials, but the design would likely be much less compact.

No, I don't think so. His design respects the inherent latency to accelerate a looooog set of elastics. Instead of stretching a set of elastics over a long distance, he uses a heavy bundle and stretches them over a short distance. It still takes a lot of effort to get that stretch, but all that energy is released extremely fast..as the elastics don't have to travel very far to do so.

Didn't that break?

It might have to do with the fact that he's using rubber bands to turn the arm, while historical trebuchets used weights. A weight is limited in how fast it can accelerate so you need to use a lever (the arm) to increase the speed.

Springs could be used instead of rubber bands. While "coil spring" is a relatively modern invention, other kinds of springs are known from the Bronze Age, accoding to https://www.designingbuildings.co.uk/wiki/History_of_the_Spr...

Springs are much older than that, the bow is perhaps the most famous example from antiquity.

Good point. I somehow was thinking about metal springs (Romans used springs from lead, for instance), but there's no reason to restrict it to metal.

Lead springs? They'd have to be absurdly oversized by modern standards to hold up to any kind of force without deforming, I'd imagine. I'd love to read more about them, if you happen to have a cite handy.

I don’t reckon.

Surely the above comment is confused, misremembering leaf spring as lead spring.

Leaf springs appeared as early as the Roman empire as a form of primitive suspension for chariots and other two-wheelers. Instead of steel, savvy Romans used flexible wooden sticks.

https://connect2local.com/l/112910/c/478618/examining-the-fa...

Ah, yeah, that makes a lot more sense.

Thank you. This is exactly the mistake I made.

It raised some flags in my head when I read it, but then again - Romans did so many things out of lead that I did not challenge it.

And they can be quite formidable, a full size wooden (yew, maple, ash) bow takes a lot of arm strength to properly pull. Just a few shots and you'll be wishing for stronger muscles.

You can double the muscle power available by using a foot-bow.

Torsion springs are another option.

https://en.wikipedia.org/wiki/Torsion_siege_engine

You could use a weight combined with pulleys or gears to get a faster acceleration.

Well, this might be supersonic but doubt it would throw heavy boulders at fortresses anytime soon to be useful as a siege weapon.

A scaled up version of this would be terrifying... I'm sure with the right materials (and budget) it could launch a cannonball at those speeds.

I was thinking it would be more fun with a rotten cow.

A rotten cow might disintegrate before it leaves the launch site. I wouldn’t classify that as fun.

That strongly depends on your vantage point. Definitely not from the point of view of the cow.

Rotten cows have limited perceptions.

Sourcing?

[citation required]

Rotten cows ... write no research papers.

Assume a spherical, rotten cow...

A sphere is a perfect form, and hence, immune from corruption.

"Fetchez la vache?"

Pink mist with subsonic bone shrapnel would still be a terrifying siege weapon.

There's some square-cube issues in scaling this up. Specifically, material strength is proportional to cross-sectional area, while the masses of the structures (and projectiles) involved is proportional to volume. The video hints at this with the mention of the difficulty being proportional to the length of the arm to the 1.6th power, slightly more than 3/2.

wouldn't need to be a boulder at those speeds. energy goes up with the square of velocity

50g @ 460m/s = 5290J

25kg @ 50m/s = 31250J

that is to say, the small but fast projectile is 1/6th the energy but only 1/500th the mass

edit: formatting

edit2: a 300g projectile at 460m/s would have approx the same energy as a the 25kg boulder at 50 m/s. at point blank range anyway.

At a certain point, wouldn't the tiny fast pebble just drill through the fort and leave a small, hot hole in its wake? Like a big neutrino?

A small projectile like a pebble would be vaporized pretty quickly by friction with the air at those speeds (and of course a bigger projectile would take more energy to launch).

You fire the weapon. The tiny pebble breaks up on impact, creating a little crater in the rock. To you, the medieval siege engineer, this is annoying. The frenchmen taunt you from their parapet, yelling maternal insults and farting in your direction. Your supersonic trebuchet is a failure, tactically speaking.

You can't take their mockery anymore, so you set about fixing your projectile.

The pebble lost much of its energy in flight - drag is a huge issue with any distance to target. You make a projectile in the shape of a rod. This has the effect of decreasing the cross-sectional area of the projectile relative to its mass, minimizing drag losses. There's a limit to the length:width ratio, however- go too skinny and it'll deform under firing forces.

The projectile's length also allows it to burrow deeper into the rock -- when material at the tip breaks off on impact, it's got the rest of the shaft behind it, just as mad.

To pack the most punch into your projectile, you craft it from tungsten (then known as Wolfram). In order to keep it from yawing in flight, you add some rigid vanes at the back. It looks like a small, superheavy arrow.

Now you have the issue wherein it doesn't fit well into your cannon anymore. No mind -- you set it in a cup of sorts that seals the barrel. Wait, where did you get a cannon? This is the year 1217 AD...???

Congratulations: you have constructed an armor-piercing fin-stabilized discarding sabot munition (APFSDS). You load it into your M256A1 120mm smoothbore cannon. You peer through the sights.... now where were those sac à merde?

This would be a highly unexpected turn of events in 'The Holy Grail', at the same time there were already some temporal discontinuities so in a way it fits right in.

They are not knights of the round table, they're very naughty boys!

when SHTF I would like to be in your team

Well, the recruiter'll tell you you can be a tanker if you want, but as with any salesman you'd be a fool to believe him.

Great writing

Afaik that's exactly what happened with railguns: they were tested at something like mach 11, but it turned out that they just made nice holes in the targets, instead of delivering the energy to the targets themselves. So the speed had to be dialed back.

(Frankly I don't know if this makes much sense from the physics standpoint, but it's what I read in passing.)

The railgun test videos I've seen were a lot more than "just make nice holes."

At that energy level you have significant damage from spalling from anything the projectile passes through, the superheated metal the projectile actually passes through is launched everywhere, there's the shockwave of the projectile, etc.

That's easily enough to be anti-personnel, incendiary, and obliterate electronics and cables and pipes and hoses. And have lots and lots of penetration, which on a ballistic arc could mean exiting below the waterline, and I imagine something traveling at supersonic speeds hitting water probably does some pretty impressive things.

If you send one of those projectiles through a magazine, missile battery, or fuel bunker - a lot of people are going to have a very, very bad day.

This is something that's hard to appreciate with any energy-transfer mechanism: If you reach energies that exceed the capacity to transfer energy, you're just wasting effort.

For kinetic systems, the projectile-target interaction is critical. A flimsier target actually needs an impact to be spread over a larger area. Think of a pellet shot through tissue as opposed to, say, a blast of sand or salt.

There are some analogues with subatomic particles and fast/slow neutrons (slow neutrons are far more effective at sustaining nuclear reactions, as they effectively "stick around" where they're more likely to interact with other nuclei for a longer period of time, and the apparent cross-section of the target nuclei appear larger).

https://www.radioactivity.eu.com/site/pages/Slow_Fast_Neutro...

Weirdly, that's the idea behind naval guns from the age of sail.

Solid shot goes through the ship, perhaps injuring anyone in the way. Solid shot hitting the structure of the ship on its way through creates massive, dangerous splinters.

https://youtu.be/lIDu7NPLbwc

https://youtu.be/6jRhEiibhCc

https://youtu.be/dGd5HLl3GwE

https://youtu.be/EpNS0JpnUNY

https://youtu.be/9XMOrbOW9DM

https://kurage.files.wordpress.com/2018/03/hocker-isbsa-14-p...

This happens with armor piercing shells hitting tanks or other armored vehicles, too. And even if they don't penetrate, they can still generate spall from the inside of the armor that's still deadly. (This was at least true with ~WW2 tanks, modern armor is more advanced.)

I'm a little doubtful. I think your standard pebble would vaporize from the friction with the air. A pebble certainly wouldn't survive re-entry into the atmosphere, so there's an upper bound on the speed before it disintegrates. If you made a vacuum between you and the wall, it might work?

One reason to use larger projectiles is to deliver similar amounts of energy without having to fight things like that.

Smaller projectiles lose more energy to air drag.

You mean objects with lower surface area to mass ratio, AKA objects with lower density. If he wants some high-density objects to throw I think there are stores in the US that sell small, dense metal objects he could use.

Mass is not a factor in drag equations.

https://www.grc.nasa.gov/www/k-12/VirtualAero/BottleRocket/a...

Velocity-squared is the factor in drag.

But very much in inertia, which is acting against the drag, trying to sustain its initial velocity.

https://en.wikipedia.org/wiki/Ballistic_coefficient

Density is important when comparing two spheres of the same mass.

https://www.grc.nasa.gov/www/k-12/VirtualAero/BottleRocket/a...

Where frontal volume is determined by mass and density for a sphere.

So, two speheres with the same mass, but different densities will have different drag forces.

That's density of air, not of the object.

https://www.grc.nasa.gov/www/k-12/VirtualAero/BottleRocket/a...

I wonder if you could do something like this[0] on the projectile to reduce drag:

"It is reported that femtosecond-duration laser pulse can be deployed to reduce drag for blunt-body vehicle in high-Mach flow field of air by generating laser plasma and shockwave. The interaction of plasma shockwave induced by femtosecond-duration laser pulse and bow shock over the head of the blunt-body vehicle is investigated numerically in the flow field of 30 km apart from the surface of the earth at Mach number of 5 and the mechanism of deploying the femtosecond laser plasma to reduce the drag of the vehicle is analyzed. The Navier-Stokes equations are exploited to compute the drag reduction for different femtosecond laser energies. The present numerical experiment proves that the femtosecond laser pulse has a better drag reduction effect than the nanosecond laser pulse under the same condition. When the femtosecond laser energy is 0.06 mJ, the femtosecond laser plasma can reduce the drag by 98%. And the higher the energy of femtosecond laser pulse, the higher the drag reduction ratio and the longer the time of low drag. Deploying three femtosecond laser energy point to reduce the drag of hypersonic vehicle is much more obviously. This energy-deposition mode can improve the optimum drag reduction ratio and save the laser energy."

Probably out of the scope for a $200 build, but maybe in the scope of a $2000, $20000, or $200000 build?

I also wonder if iterating on the above, combined with a concentrated solar power plant of arbitrary sized solar field that could provide enough energy for a railgun to launch payload with femtosecond laser plasma capabilities out of earths gravity well?

Maybe, we could escape from the Goddard age eventually lol

[0] https://www.researchgate.net/publication/282709902_Exploring...

The laser in the study is moving with the projectile, so I guess even for $2000 you won’t be able to have a 30 km altitude, Mach 6 flying object with perfectly controlled laser launched from a trebuchet.

I'm talking about a projectile that has the laser capabilities in itself (i.e the electromagnetic slug is emitting the femtosecond laser pulses after it is launched), not launched via a laser (it could be launched however, but I was assuming later on the comment that both the rail gun and the slug itself will need to be charged up).

Yes, it's 'moving with the projectile'. How does that make it easier?

I never said how easy it would be only that there is a difference between a payload that is launched via a laser and a payload that has laser capabilities in it (yes, 'moving with the projectile', like i was talking about from the beginning where you assumed otherwise).

But for one, the energy source would have to be on board the payload to power the laser (localized around the payload), and for the other (that I was not talking about) the laser is not on the payload and as the payload travels further away from the launch site, you would use an non linear increasing amount of energy and follow the path of the payload (ionizing air not only around the payload continuously but along the path where the payload no longer is, which would require more energy than needed).

Try to read my comment, I never 'assumed otherwise', the difficulty of having the laser as payload is exactly my point.

> so I guess even for $2000 you won’t be able to have a 30 km altitude,

Does seem like it references some difficultly-ness

> flying object with perfectly controlled laser launched from a trebuchet

Does seem like it 'assumes otherwise' to (or its at least ambiguous to me and would be less so if there was a perhaps some separation between the laser and launched, but maybe that's a bit nit picky)

> The laser in the study is moving with the projectile,

And would be even more difficult than such (at least going by how much energy would be required).

But regardless, I'm not sure that 30 km altitude and first achieving mach 5 is a major constraint beyond that is probably not achievable at the $2000 price point (or its not clear to me that it is from the paper) and would be something that could be worked around with more iterations of what level of ionization of surrounding air around a projectile with a given density would give the desired amount of drag reduction (perhaps zeptosecond[0] pulses would work even better than femtosecond ones that were compared against nanosecond, or achieve similar results with lower mach number and higher density of air being ionized around the payload).

[0] https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.11...

Also try to read my other comment, the difficulty is precisely my point.

I don't see how 'laser in the study is moving with the projectile' is in any way, form, or shape ambiguous.

And the $2000 price point is not from the paper but from your own post.

> flying object with perfectly controlled laser launched from a trebuchet

Is what I considered to be ambiguous because it wasn't clear to me that you were assuming the system would be either 'laser launched' or 'launched by a from a trebuchet' with the laser part only applying to the object, regardless of your mention of:

> The laser in the study is moving with the projectile,

Your comments after that made it clear that we are both on the same page regarding the laser part is referring to the payload.

The difficulty being your point is not that interesting to me when i made the post, but the possibility of being able to send something into LEO while minimizing drag effects on the payload to something negligible along the flight path of the payload.

The $2000 came from the cost of the trebuchet system, I'm more interested in at what price point would a system be possible (Ideally below the cost of hypersonic missiles [0] now that I assume use something like this) with a rail gun kind of system.

[0] https://www.scmp.com/news/china/science/article/3142581/chin...

The frontal area of two spheres of the same mass will be determined by the sphere's density, and that drives the air resistance.

If frontal area didn't matter to drag, wouldn't all spheres would have the same drag coefficient?

From that page:

Cd = D/(rAV^2/2) where A is frontal area.

What am I missing?

> Cd = D/(rAV^2/2) where A is frontal area.

You're citing a page that uses the word "density" differently than you do.

"D" is density-of-the-air. The density you're talking about appears to be the density-of-the-sphere.

"Appears to be"?

They explicitly said so.

Twice.

It absolutely is, and will then drive differences in reference area, thus affecting the drag of the object.

You’re missing nothing, you’re correct.

Doubly so when you consider the supersonic barrier.

Air drag is proportional to the drag-coefficient * velocity^2. So doubling your speed quadruples the drag.

Except... the drag-coefficient itself is a complex curve over velocity. It barely changes from 0mph through 300mph, but once you reach transonic and supersonic speeds, the drag-coefficient skyrockets.

As a result: 250mph is more than 4x more efficient than 500mph from a drag perspective (4x predicted from Drag equation, but in practice might be 8x or more). This is mostly a problem for airplane efficiency and fuel consumption, but I'm sure it applies to war-machines and projectiles too.

--------

But modern airplanes fly at very high heights, so the air is thinner, reducing drag, but requiring a pressurized cabin lest everyone gets hypoxia.

> It barely changes from 0mph through 300mph

Tell that to cyclists :-) You can barely go above 25 km/h (~15 mph) without feeling like you are just pedaling to overcome the drag.

Reducing the Cd on a bike, ie going recumbent and putting a aerodynamic shell on it, ie switching to a velomobile allows you to go 40 km/h (~25 mph) with relative ease.

Just in case you misunderstood the parent post -- your Cd as a cyclist is pretty much constant around your normal cycling speeds. Your total air drag goes up with the square of velocity, that's for sure, but if you went transsonic on your bike, your Cd would ALSO increase dramatically on top of the already brutal velocity-squared multiplier.

It’s almost worse, because you’re working against this quadratic force, at a rate proportional to speed. So power output, the usual limiting factor for a cyclist, scales like 3rd power of speed.

Fair enough ... I really did not get that :-)

Only when they are flying to normal atmosphere..

One can use a laser + charge to pre-boil the flight-path into a plasma channel. Sort of like cavitation, but with gasses.

His trebuchet is quite high off the ground, so the arm and sling has room to accelerate. This wouldn't really be practical in a medieval trebuchet, especially if you wanted to move it to a siege.

Medieval Trebuchets didn't move. You hired engineers to cut down trees and build Trebuchets for a particular siege.

Honestly, "Castle Drops" in AOE2 are more akin to how Trebuchets were used. You build an army, the army gains control of an area, and then you use your engineers to build a long-range siege device at that location.

------

https://en.wikipedia.org/wiki/Trebuchet#/media/File:Ms.Thott...

See? No wheels. Those things couldn't move at all.

My impression was that they were assembled on site, but they were pre-made and could be dissembled. From this thread, it looks like both methods could have been used:

https://www.reddit.com/r/AskHistorians/comments/3z597l/in_me...

> Interesting how different his device is, from historical trebuchets.

It's more of an onager really: it's a torsion engine, whereas the trebuchet you'd think of is a gravity (counterweight) engine.

> It would be cool to see his design built in period materials.

"Period material" would be limited to sinew and / or hair, wood, and relatively small pieces of metal (mostly fasteners).

Onagers do not have slings.

You are wrong.

> It is commonly depicted as a catapult with a bowl, bucket, or sling at the end of its throwing arm.

https://en.wikipedia.org/wiki/Onager_(weapon)

I stand corrected!

He actually explained this in the video.

You missed the best part.

The book Structures describes how trebuchets are quite inefficient compared to the Roman Ballista, which the author alleges the medieval folks forget how to make. Much of the energy is wasted in the counterweight swinging. By contrast, he alleges bows, including hand-drawn, crossbows, and ballistas (or scorpions, or springalds for you AoE fans)are close to 100% efficient.

This is relevant, since as the original com mentor pointed out what the video-maker built is arguably closer to one of those torsion weapons than a counterweight trebuchet.

Seems like efficiency is relatively less important (peasant labor is essentially free) and weight of fire more relevant. Trebuchets were usually siege weapons.

That cute toy would probably shoot the ball straight through you at appreciable distances.

It's about the same muzzle energy as a .38 special - definitely not a toy!

So with that in mind, his standing behind the loaded device, was akin to him muzzling himself.

I would think the four rules of firearm safety would apply to supersonic trebuchets.

I would think a prototype supersonic trebuchet would come with a lot more rules than a standard firearm.

They would be much closer to the safety precautions for a prototype firearm, which includes things like being behind a blast shield because there is no safe zone if it fails catastrophically.

I don't even trust that the sides of that thing are safe from shrapnel or the rubber bands whipping parts around. If one side of the machine gave way for some reason, it could absolutely swing sideways. That tiny string at the end is probably going fast enough to slice through skin and veins, and it's concerningly close to neck height.

He should have parked his car off to the side and used the engine block for cover. The paneling of the car will stop minor wood shrapnel, and the engine block should be able to stop any metal pieces that come off. Ballistic barriers would be better, but at least you're not standing there tempting fate with your squishy and easily separable limbs.

I found his lack of safety-glasses and ballistic shields disturbing, especially given that one of the rotors had already disintegrated.

A piece of 3/4" plywood to stand behind would slow down a lot of lightweight shrapnel.

Alas, one of the few useful things that comes with age and experience is an awareness of consequence...

My biggest worry would be the joint at the bottom (the four bolts visible in the video) giving way because the tension is across the top of the wooden beams. That won't end well because suddenly the winch becomes the projectile.

I would definitely not watch Colin Furze

I can't stand that guy due to the way it is all presented, it's like watching 'Blippy for adults'.

You obviously haven’t seen the Blippy Harlem shake meme.

Also, there's not a huge angle from "where the ball is supposed to go when it fires" to "where he's crouching adding energy to the system".

Yes, that seemed a disaster ready to happen, especially as it was already loaded. He must be very confident of his release system.

It already surpasses the muzzle velocity of many handguns with more flexibility in terms of projectile. I would not want to be on the receiving end of one of these.

I’d like to suggest linking directly to the video (https://www.youtube.com/watch?v=gdXOS-B0Bus) instead of the hackaday page, which doesn’t even pretend to add anything and is largely incoherent besides.

Hackaday is just blogspam. They happily rip off your work.

On the other hand, I would never have heard of any of the cool projects they feature if not for Hackaday. Do you know of better aggregators for such content?

You're not accounting for all the high quality content they produce. Like Creating a PCB in Everything, or Logic Noise.

https://hackaday.com/tag/logic-noise/

https://hackaday.com/tag/creating-a-pcb-in-everything/

they also have a kind of community website full of maniacs (in the good hacker sense)

people making discrete custom CPUs for instance

I admit I don't like hackaday a lot but they have some value

They've been bought out some years ago, before that it was mostly just interesting articles at a much lower rate.

A sad but regular occurence of business mindset.. quantity .. quality and the usual.

I agree this is a bad textual article, but am I one of the few people who prefer text and images over video? I literally never watch a video link here but I'll read almost any textual content. Not judging others, but I'm sad that technical videos are so prevalent and most don't have any textual counterpart.

I absolutely prefer text to video for tutorials or documentation. I don‘t think this preference is unusual. But the linked article is not a text version of the video. It has, in fact, no content. Everything is in the video. And the final bit, where you hear the snap of the sonic boom, is worth watching the whole thing.

Agreed, thanks for responding. I guess it was more bitching than I should have done and off topic here; I agree with your assessment of this "article".

The video is very good, actually, for once.

I hope the paper that was explained in the video is available published somewhere.

The text was sufficient to prompt me to watch the video, which I wouldn't otherwise have done.

Direct link to the timestamp of first full power shot: https://youtu.be/gdXOS-B0Bus?t=806

Second shot: https://youtu.be/gdXOS-B0Bus?t=871

The most striking aspect is that he was explicitly optimizing for project difficulty. I smiled when he said, in effect, 'let's mathematically calculate How Hard Can it Be'. Sure kid...

But that's what he did. Fantastic job determining the criticial criteria and engineering to it.

3 weeks total project time.

Kid's got a future...

Fun fact, you can accelerate an almost-everyday object to supersonic speeds: whip tip [1,2].

1. https://en.wikipedia.org/wiki/Whipcracking

2. https://www.youtube.com/watch?v=AnaASTBn_K4 (very much recommended)

Anyone interested, amazon sells Bullwhips for $23. Almost got one.

I have one at home. They are awesome fun. But you only get a few cracks off before people start poking their heads out of windows. They are easily as loud as a small firearm.

Funny thing: a friend of mine lives in rural Bavaria, with a very old, conservative land-lady. The kind that will call you once you exceed your lawful one hour of instrument practice time per day.

Still, she allowed rehearsals in the backyard from what can only be called a music formation comprising accordion ... and bull-whips.

Like literally whipping to the beat of the music, hours of staccato whip cracking. I honestly had no idea that that existed. Example: https://www.youtube.com/watch?v=QquXdrDcX7I

Wow. What.

As an Australian living very close to Bavaria I can tell you that ain't Bayern! Oida! ;) https://www.youtube.com/watch?v=yY6nminL9wY is what you are looking for.

At 13:12 in that video notice the tension in the frame and how the main spars are getting out of alignment with the part the winch is mounted on. I wonder how much spare strength there still is but it can't be much. Pretty scary machine!

No kidding; when he was cranking the final few inches of tension, I wonder if he realized what would happen if anything snapped: instant death was definitely one of the potential outcomes.

Exactly. It would be interesting to hook the whole thing up to an electric motor and to winch it (from a safe distance) to the point where it breaks just to determine how much margin there was. Probably not a whole lot past the point that he hand cranked it.

That would be extremely interesting. Would be great to put a dummy where he was squatting in the crank position while you did it too.

I think it makes sense for the project to be powered by human energy.

I guess you could power the motor by an alternator+crank :)

Well at that point he did say "One of the things I love about working with wood is all the menacing creaking noises it makes when you load it. Really makes you appreciate how precious your fleeting life is" so I think it's safe to say he realizes this.

The proper response to that realization is to unwind the tension, not to keep pushing. Really, he just got lucky, but that's fine, it's an interesting video. I take risks myself when building stuff but I would not trust my life to something that isn't engineered for strength and in spite of all of the work that went into the computations of how to get the projectile up to speed as well as the strength of the arm there is already some proof that he misjudged how strong wood is for this application (one of the arms desintegrated).

Wood is super strong, if used properly, this looks like a setup similar to the neck of a base guitar but without the engineering of the joint to go with it, and a much nastier angle between the direction of pull and the joint.

He spent all that time engineering the speed, he forgot to engineer the base.

Safety glassed would have been a good idea, any rubber can suddenly fail. But he is absolutely right that "creaking noises" are common in wood far far far below their breaking point. I suspect he calculated the strength of all these things if he went to that much trouble for the sling. Also the markings on the bolt heads showed they were grade 8 (those 6 radial lines) which are near the best common off-the-shelf bolt strengths you can get. The OG wooden sling breaking was likely a nuanced issue stemming from it's rotational speed, so not a risk during the cranking stage. The sling wasn't even under load at that point. High-speed rotational effects are many orders of magnitude harder to predict than the needed strength of a base structure. But yes, safety glasses would have been wiser.

I don't think it's the bolts shearing we should be concerned about. Rather, the wooden beams might be split in the middle by the torque on the four bolts.

This guy seems like he's probably a very qualified engineer, yes he should be wearing some safety equipment but he also seems like he knows how close he is to the limits of the device far far more accurately than the shade tree maker.

He may very well have designed in a 50% or more safety margin on the frame.

Reminded me of one of Ben Krasnow's videos where he ran a hydraulic cylinder out of spec and caused it to bulge noticeably. Um, dude, there's probably a few sticks of dynamite worth of energy in that thing, you know...

Without knowing the specifics of the incident:

There is actually very little stored energy in a hydraulic system, due to the incompressibility of the fluid. Hydraulic injection injuries[0] are nasty, but catastrophic failure isn't particularly dangerous. Pneumatic systems on the other hand can reach dynamite energy levels and near-enough dynamite energy discharge rates.

[0]A not-for-the-faint-of-heart google.

Hmm, that sounds like a good (if unintuitive) point. I withdraw the criticism. :)

Yep, that was pretty terrifying to watch.

Indeed, I wasn't comfortable watching it until I realized that he was the one that posted the video. But instant desintegration was definitely on the menu there and once stuff comes apart and starts flying it's impossible to predict trajectories. That's a very large amount of energy on what looks to be two run-of-the-mill boards, they're fortunately under compression but the joint at the bottom looks (far) less than ideal for this, especially with the holes drilled in line with each other (that's a definite 'no' if you want to avoid splitting the beams).

A++ for the theory, but the implementation is not something I'd want to be near when it is in operation, it is just too unsafe judging by my somewhat experienced DIY eye.

I'm sure the guy is aware of how much stress he's putting the wood under, and about how much it can withstand.

when I was a kid, my dad made an engine hoist frame out of wood. that thing creaked out some scary sounds when he lifted whole diesel trucks with it when load testing, and it didn't even faze the hoist frame at all.

it sounded like it was going to explode into a billion supersonic toothpicks and didn't even visibly move. we painted red lines across joints to see stress movement more easily and nothing ever moved.

his material of choice? 4x4 construction lumber.

it is amazing what wood can withstand when you understand what you're asking it to do, and design so that material properties are fully exploited.

my point: loud creaking is not necessarily indicative of impending failure.

Not necessarily, but it is the only warning you get.

That's true, but you also can't know every possible point of failure of an apparatus under load just from design principles. That's why they flex the wings of new airplane designs until they break, for instance. So I expect testing under extreme tension while squatting right beside the thing without so much as a pair of safety goggles is taking unnecessary risk. I'd go for a heavy coat and helmet with face shield at least, if I needed to be that close. (But I'd rather find a way not to be.)

I'm quite aware of what you can and can not do with wood, unless he's tested one to destruction the degree of variability alone makes this a particularly dangerous stunt.

It all looks fine right until it doesn't.

> I'm quite aware of what you can and can not do with wood

yes, I'm sure you are. you always seem to know everything about everything when it comes time to argue.

I don't understand why he isn't wearing eye protection during the whole experiment. Made me incredibly nervous. Obviously won't protect you from a direct hit of the main projectile, but there are so many things that can go wrong unpredictably, your eyes are the squishiest part, and protection is easily available and not too annoying.

I really hope he gets to partner up with Smarter Every Day or with The Slo Mo guys, for high-speed recording. I'd love to see how the trebuchet parts bend/vibrate during a prep/launch.

Reminds me of the unlikely idea of using a whip crack to send something into orbit[0] maybe SpinLaunch needs to consider adding another joint to their arm. I'd love to see a massive trebuchet launch something into orbit.

[0] https://futurism.com/the-byte/amazon-patented-system-whips-s...

The idea of using whip cracks for acceleration features fairly prominently in Neal Stephenson's Seveneves. None of the instances in the books are sized for orbital launch but that was the first time I encountered the idea.

In general, it was fascinating to see how much of Seveneves was devoted to acceleration and orbital mechanics.

Not really a surprise, though, after Anathem.

Funny, earlier today I was thinking about how Anathem was fascinating but the least approachable book I've ever read

I'll admit I was a bit confused after the first section in which they interview Flec. I read that section again, and then everything made sense. It may have helped that I have read (numerous times) Wolfe's The Shadow of the Torturer, of which the initial chapters of "mathic" life in Anathem are a sort of pastiche.

Stephenson gets plenty less approachable. I have tried multiple times to read at least two of the three books in The Baroque Cycle, and every time I have failed to progress.

All they ways they used the energy from deorbiting mass was just so fascinating. Almost makes you wish the moon could shatter in real life.

It can't with that attitude.

Here is the local newspaper's coverage of the supersonic trebuchet (Manitoulin Island, northern Lake Huron in Canada):

https://www.manitoulin.com/island-engineer-designs-supersoni...

Amazing project! Although, I would probably err on the side of caution and at least wear some safety glasses when hand-loading a supersonic trebuchet.

I had the same thought! Earplugs for the sound might not be a bad idea either.

I think if anything in that machine snaps it would rip right through his body.

If you build one of these then practice better range safety than this guy. He apparently has the trebuchet loaded while tensioning the rubber bands downrange of the release point. Excellent engineering work though.

It is interesting to me how many engineers hop from English ("3 inches to go on the rubber band") to the metric system ("400 meters per second"). Andy Weir touches on this in his book "Hail Mary".

Except this person isn't American, as he lives on Manitoulin Island, which is in Canada.

Still missing the "football fields" unit of measurement.

I am from NZ and find myself doing this. The difference is whether something is estimated or measured. If it’s a rough estimate, and the value doesn’t matter at all, I use inches. So “six inches” is more of a way of describing “about this much”. It’s deliberately imprecise.

This post prompted me to google "last time trebuchet used in combat".

The answer appears to be 2013/14 by Syrian Rebels fighting against the Syrian Army.

https://www.quora.com/Warfare-When-was-the-last-time-a-catap...

There was also a trebuchet appearance during Ukranian riots in 2014: https://www.youtube.com/watch?v=IEY4WG7P0NQ

This is a perfect example of engineering. He says it at the end - its not something that could ever be arrived at by accident. (And I suspect its not feasible to launch sizeable weights (bowling balls) at near mach 1. The Ancient Romans would have been disappointed :-)

Hmm. I'm not so sure about the bowling ball. It seems almost straightforward to scale this up as far as you want. You wouldn't be able to use wood, and the hard part would be to find a winch that won't snap, but it seems like one must exist.

The limiting factor seems to be the arm, which would probably become impossible even with modern materials pretty soon as you increase weight.

The SpinLaunch[1] folks are betting on being able to do it.

[1]: https://www.spinlaunch.com/

Wait what. Seriously no they cannot mean it. Surely that's just Jules Verne's big gun in Florida?

Is this for real?

Having to stand downrange of it to windlass it is intimidating.

Real Men (tm) stand in the plane of rotation.

That's pretty awesome. May have to give a whirl. Thanks!

I remember years ago seeing some show about weapons seized in jail, and one of them was a crossbow made out of underwear that could shoot a plastic fork through a steel filing cabinet.

Next stop lunar surface! Something like this may be what you need to export lunar materials from the surface, say to the LaGrange point.

I know a large group of weirdos that are probably flipping their shit with this.

Not only the superior siege engine but the supersonic one too.

I have two words for this dude, and they are safety glasses.

I'm not sure that there exist 'safety' glasses that would actually be safer in this situation

I'd be less concerned about him shooting his eye out than I would be about something unexpectedly breaking and sending debris flying.

I know not with what weapons World War III will be fought, but World War IV will be fought with the supersonic trebuchet.

Excellent engineering video and it hope it inspires more young engineers than it lands in the ER. Stay safe kids!

Missed the opportunity to use the Trebuchet MS font: https://en.m.wikipedia.org/wiki/Trebuchet_MS

I've always thought the trebuchet cord should be wrapped around a wheel. Then the force would always be rotational - no component of fource would pull against the hub and be wasted.

The arm is just a large wheel and a small wheel on the same axle, with the unnecessary parts cut away. Round (not-cut-away) wheels will capture more of the energy as rotational inertia, leaving less for the projectile.

Sort of. A light-weight wheel compared to the load etc is negligible. And an arm is definitely not a wheel - the rope doesn't always pull completely tangential to the rotation, not at all. Nowhere near. That's the point.

You can use that to your advantage building flywheel trebuchet: https://www.youtube.com/watch?v=RVT5i4nhIGs

Instead of torsion-applied force for his videoed trebuchet, he could have gone with a set of spiral gears to obtain the progressively increasing speed using just the original gravity-based counterweight method.

Something like toroidal continuously variable transmission (CVT) in today’s automobile.

https://en.m.wikipedia.org/wiki/Continuously_variable_transm...

Built at trebuchet for physics high school class eons ago. This is that same project after completing an engineering degree and dosing it with steroids. Well done.

Kind of wish I had gone back to my trebuchet post engineering program.

Off topic:

What’s up with the ”[Name]” quoting style they’re using? I’ve only ever seen it when changing names/words inside quotes[1], but here it is being used outside of quotes.

[1] Which I don’t mind when used within reason.

It's just a long time habit of Hackaday. They've been doing it since I started reading in like 2008. Just be glad they decided to move beyond black and white pictures!

I'm curious, could this be a realistic and cost-effective weapon for, for example, militias in third-world countries?

I can't imagine so. Even if it was somehow accurate, it wouldn't be very mobile or quick..

I'd imagine one person with a pistol could take out a good sized group of people shooting marble trebuchets

That heavily depends on range, pistols are not known for accuracy at range.

Still pistols have some barrel so they should be vastly more accurate at range. I forget the ranges they are good for, but I think within 100 meters you can still hit your target most of the time.

Though if this is a real war I'll take a real rifle. The ar15 (m16) and ak47 are the staples of modern armies for good reason, and they have much better sniper rifles that in the right hands are amazing.

> Still pistols have some barrel so they should be vastly more accurate at range.

Way misguided here.

Pistol accuracy has almost nothing to do with the barrel. Not in the sense that you think.

Accuracy is repeatability. That means tolerances. The bore, rifling, the chamber dimension, the jump distance from the bullet to the rifling in the chamber, the natural resonances… but then just or more importantly is your ammo.

You have you ogive (“oh-jive”) of the bullet, the sectional density, the tail shape, it’s ballistic coefficient (actual, not G1 or G7 artillery guesses), your powder burn rate, consistency of the powder, amount obviously, the case neck tension and length, primer hole and primer consistency…

Then you get into dwell time of the action and all the user variables you can’t account for without bolting the gun to a frame.

… so as to pistol accuracy, you can have the world’s most accurate pistol and barrel ever made, and still have shit long distance accuracy. A barrel isn’t going to help you.

Pistol bullets are bricks compared to rifle bullets. A ballistic coefficient of .1 is just never going to be very consistent because at distance it’s dumped so much of it’s speed via air resistance that it’s supersonic time is limited and will enter transonic and subsonic quickly. This isn’t what pistol rounds are designed for. You want bricks with high sectional density for shooting things up close.

Somewhat accurate. A 9mm Luger fired out of a 16 inch barrel will have better terminal ballistics then when fired out of a 4 inch barrel. Yes, actual bullet design matters a lot too, but rifles are much more ergonomic for distance shooting, and I'd argue that makes a lot more difference.

Also depends on your definition of "at range". 300m? 1000m?

I have 15 years of reloading competition rifle ammo. But, ok, your opinion that longer barrels are better and rifles are more ergonomic is something I’ll really have to consider.

I’m glad we can agree that the actual differences between rifle and pistol ammunition matter “too”.

EDIT: Also your example is… not good. Almost all 9x19mm handgun ammo is going to have max potential in a 8-12” barrel. A 16” is doing next to nothing. All a longer barrel is doing at that point is introducing worse harmonics for a negligible difference in muzzle velocity, probably not even noticeable outside of the standard deviation. It was a good example for giving me insight to your background on this topic.

Gunpowder has largely superseded these kinds of weapons.

How far does the projectile go?

I'd be very impressed if they were able to find any of the projectiles after firing

Wait for snow.

Smear bearings with orange grease.

Look for the orange smear. (A drone would help here.)

A friend's father used to build and test rockets using this principle, though he'd smear the outside of the rocket casing. During the Cold War. Up north.

That earned the family a visit from an air force colonel one day, apparently after pilots had put together the sighting of high contrails, orange snow smears, and the lone figure seen trudging toward the smear.

Ah, that's a good trick. I bet if he greased a dozen ballbearings in snow, could probably find at least a few.

I'd imagine these projectiles are going thousands of feet, would still be tough to find them all.

Depends very much on the launch angle, anywhere from a few hundred meters to well over a kilometer.

You can play around with this:

https://www.omnicalculator.com/physics/range-projectile-moti...

Note that it does not take projectile shape (and so drag) into account and that's the largest factor.

At 5:30 in the video he claims that the problem is infinite-dimensional. How so?

Solutions to differential equations belong to infinite dimensional function spaces unless discretized.

US will impose a ban on rubber exports for Iran and North Korea. :D

Cool, hadn't heard of this guy before. Judging from the headline, I would have assumed it was Tom Stanton, or even more probably, Jörg Sprave.

(Or, dammit, what's the name of that English guy who builds and sells medieval weapons?)

This is a perfect illustration of why I chuckle when people claim to be able to "ban guns".

He is projecting a 3/8" (.375 caliber) round steel projectile at ~3x the maximum velocity limit for an unrestricted air rifle in Canada.

Using a crank and some hefty rubber bands.

Sure, but this project isn’t really hand held and mobile and reloadable in a small amount of time. It took some smart enough person to operate as well. Some nuance and differentiation still appropriate; this is as close to an air rifle as vacuum tubes are to SoC.

"Sure, Mr. Wright, but the craft isn't reliable and easy to fly..."

I'm not sure what point you're trying to make... that machine is not particularly mobile. It takes him 5 minutes to load a projectile.

The UK has banned guns. People there generally don't build trebuchets to compensate.

Murder rate in the US is 18x higher than in the UK, and 23x higher than in Canada.

Edit: there are obviously other ways to murder someone, knives being the most common in the UK. If you're a good aim, guns are a lot less messy, and I think that makes a big difference to would-be murderers.

Your statistics are wildly off: UK Homicide rate 2018: 1.20/100k US Homicide rate 2018: 4.96/100k

And making something illegal is not the key factor in comparing countries: UK Drug Overdose Deaths 2020: 7.6/100k. US Drug Overdose Deaths 2020: 28.1/100k.

You're right. There is a difference between murder and intentional homicide, but the factor of 18 appears to be using the raw numbers of murders rather than murders per person.

The point is -- if a newbie engineer can create this in a few afternoons out of spare parts, any black-market weapon maker can contract some engineers and with a few CNC machines and some billet aluminum and steel, can start cranking out AR-15s.

Literally the only people affected by "gun bans" are the law-abiding. Any criminal gets armed with cheap black-market weapons. All the political elites and aparatchiks have armed guards. The rest of us dumb slobs get to be mowed down in the cross-fire between roving gangs, or defencelessly abused and robbed in a home invasion.

Or clubbed as we protest in the streets, by our ever-caring politician's brown-shirts.

Remember: every smokeless powder propellant powered firearm is basically an "air gun". The rest is ... refinement.

Most mass shootings in the US seem to be perpetrated by the "law abiding" gun owners. Just having less guns in circulation would go a very long way to reduce gun related homicides.

Having less firearms is not the solution.

2/3rds of all firearm deaths in the US are from suicide. The majority of the rest are from drug/gang violence or domestic disputes.

Having free healthcare in the US (esp. mental healthcare) would drastically reduce the amount of firearm deaths via suicide.

Couple that with free after school jobs and skills training for those most at risk inner city youth who are the most likely to join a gang or sell drugs, would also have a large impact at reducing drug/gang firearm deaths.

Those two things alone would make the largest impact. Not reducing the amount of firearms.

Heck even raising the min wage to $15 would help so those kids parents dont have to work multiple jobs and can spend more time with and raising their children.

because "mass shooting" is an artificial category that excludes most shootings with multiple victims (gang/drug violence).

Amen.

It's so annoying how people don't realize how manipulated the word "mass shooter" is.

"Mass" (i.e., 3 or more) shootings are <1% of shooting murders in the U.S. It's a silly thing to pay attention to, and I wish people would stop.

Most mass shootings in the US are gang related using illegally-owned pistols, IIRC. At least using the definition that gets you big numbers of mass shootings.

> Most mass shootings in the US are gang related using illegally-owned pistols

I used to think this, but then I did some research, and it does not seem to be true:

https://www.gvpedia.org/gun-myths/gangs/

In Canada, the number is around 20%, per:

https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=351000...

Neither most mass shootings, nor most gun murders, are committed by gang members. Although it is a non-trivial fraction.

The article you linked says that it's a myth that most gun violence is committed by drug gangs. OP didn't argue that, he argued that most mass shootings are gang activity related. At no point does it mention mass shootings.

You're correct. It looks like indiscriminate mass shootings are a relative rarity among mass shootings. Thanks for pointing that out. I think for those who would ban guns, they're probably more concerned with the overall rate of gun homicide, as opposed to specifically about mass shootings. But it's a valid distinction for some purpose I'm sure.

> Literally the only people affected by "gun bans" are the law-abiding. Any criminal gets armed with cheap black-market weapons.

Norway has pretty strict gun laws, requiring active shooting club membership and documented activity for being allowed to own non-hunting weapons and strict limits on number of firearms based on competition activity level.

Yet I read recently one can get a decent black market 9mm pistol for half what the cheaper models go for in the gun stores...

There's also been a record number of shootings here in the capital this year. Hope it's just a coincidence...

Once you have high strength steel tubes freely available in a HD store, the rest is pretty easy. Also, I'm amused by the lack of concern over crossbows and archery in general by busybodies.

Just point the busybodies at the slingshot channel[1] and they will ban them about 30 seconds after they let Herr Sprave "show them its features"

1: https://www.youtube.com/watch?v=BF_OuEba3a4

yeah instead of using a firearm in the UK, you'll get stabbed.

The US also has a way larger population.

When you take away suicides (which make up 2/3rds of firearm deaths), we really dont have a lot of firearm deaths.

Per [0], US firearm deaths (per 100K people per year) minus suicides would put them between Nicaragua and Paraguay, and about double the next country in the developed world. That's still a lot of firearm deaths.

[0] https://en.wikipedia.org/wiki/List_of_countries_by_firearm-r...

Knives are better for murder, but guns are better for self defense.

You can make an ar-15 receiver (legally the gun) out of plate steel from Home Depot and some spacers. The build is from the glory early years of the internet (90s?), so instructions are kinda hard to find. Why bother? CAD drawings for a $800 mill machine are available everywhere. Really, if you've ever held a receiver you'd realize how... primitive it is.

Europeans don't DIY guns because you're next to enough failed states to easily get fully automatic rifles. Why trouble your self with a building a semi-automatic when you can buy the full auto version? Except for Sweden and the UK. They prefer hand grenades and knives respectively.

> Europeans don't DIY guns because you're next to enough failed states to easily get fully automatic rifles.

This is not quite accurate. The [FGC-9](https://en.wikipedia.org/wiki/FGC-9) was made in Europe and has popped up in Myanmar. As I recall from the Popular Front documentary on JStark, the inventor of this weapon, other people were known by him to be making them in Western Europe.

I was being facetious to remind Europeans that their criminals have guns, better guns than criminals in the US do. The DIY scene in Europe is quite healthy, so of course someone will make a DIY gun.

Cute. It is around 8 feet long and per the comments, it throws tennis balls. Throwing a tennis ball at supersonic speed is nothing to sneeze at, but the really impressive thing about historical and modern trebuchets is how big some of them got. They could throw things like pianos, as in a famous Northern Exposure episode. Doing that supersonically would have really been something ;).

No, it throws ball bearings.

> They could throw things like pianos

One example: Trebuchet throws a piano that has been set on fire at Burning Man: https://www.youtube.com/watch?v=SgboQNcRN5w

Such a dense crowd doesn't seem to be the best environment to throw a burning piano with a trebuchet, but I guess I don't get the point of the burning man.

Literally just watch the video.

> per the comments

Why are you believing comments over what the content of a video/article is? And you're wrong.

The article didn't say either way. The comments gave info that I didn't particularly have reason to disbelieve. I didn't have any interest in watching the video once I found the info I wanted. The article could have just said "it throws ball bearings" but it didn't, making it useless.

The comment you're referring to is about the trebuchet that the commenter built (not the subject of the article).

The article itself is blogspam that adds effectively zero content to the video other than to capture some advertising traffic.

Reply on news.ycombinator.com