Film at night, put LEDs in a ball of some sort so it glows or whatever. EZPZ

You get cricket balls in 113g, 136g and 156g like you had... They make lighter easier grip ones for junior cricket. Definitely recommend you try one of them and they'll be easy to get hold of...

The tennis ball wasn't dropping straight down at the end. Remember that your camera isn't directly overhead the ball for the hole trajectory, and that it's just at a point. Think of a triangle with the ball falling along one of the legs. Great video, I'm glad someone is out living life and having fun 😉

I think you should remake your trebuchet to be... 3 times bigger?

Being a maths nerd I love these videos! Just a quick point, just because a graph is curved, doesn't make it exponential. Keep up the great work!

1:12 What a catch

Magnus effect. Try a release mecanism that uses a pivot and a screw to adjust easely on field. Turning the screw modifies the angle., and the system is quite robust. You might want to try a ball with dimpled texture on it, like golf bals... This will help it to travrl further... And you might be able to modify travel direction if you dimple it with different paterns (just a dimple band and non dimpled sides) Good luck!

and thats how you do real science.. calculate some theory..trial and error.. gather the info.. tweak the components.. and try to figure it out mathematically later....well done

You can actually see the weighted tennis ball with backspin rise slightly from the backspin.

Trebuchet cricket. Want

Hey Tom, love these videos! I recently observed a full scale reconstruction of a medieval trebuchet in action, and the engineers operating it said that the ideal counterweight/projectile mass ratio is 100-1. So for a 5kg payload, a 500kg counterweight would be rigged. I'm not sure if this is optimal, merely anecdotal, or simply the manner in which the machines were configured historically, but I thought it would be worth mentioning. I believe your counterweight is 15kg? If so that would put your 156g Cricket ball slightly over the (alleged) optimal mass (which would be 150g) for your current counterweight. I assume though that the optimal mass ratio would vary based on the other parameters of the machine, though it seems not by a great deal, given the results you've shown here. Also of note: the trebuchet I observed was fixed (as I imagine most historical examples would have been). Perhaps as the mass of your projectile and counterweight approach the optimal ratio, the benefit of the wheels is reversed and becomes a detriment? IIRC the wheels altered the stall angle of the main arm, and you said in this video that raising the mass of the projectile changes the release angle as a result of a faster transfer of energy into the projectile. Maybe you could try the Cricket ball without the wheels, as the plateau you observed at ~130g could be a result of the wheels stealing away some of the energy from the projectile.

It makes sense that the Ke change wouldn't be exponential, I mean, the larger ball is being accelerated for about 10-20% longer by the same force. The speed won't be quite as high, but it won't be solely based on that kinetic energy equation, since that can't account for the change in inertia of the rotating body, which is now significantly greater. Also, I'd love to see this thing launch a fin-stabilized projectile. Maybe a go-pro embeded into a nerf football?

I think this is my favorite of all your projects yet. I will never build one but your process of investigation is great and your results have been really surprising, like I would never have thought wheels would increase efficiency! This is better than anything on Netflix. You're a legend!

I love that you have both the curiosity and competence to follow your exploration of this wonderful machine. It's a fascinating journey!

So impressed by your adherence to the scientific method. I have watched four of your videos and you sir, deserve the subscription.

... and he juggles!! This is now the best channel on youtube

nothing better then some classic catches with a trebuchet on a saturday

This is such a cool project. I love optimization

Welp, time to go to bed it's like 2am BAH GAWD, NOW HE'S OPTIMIZING THE PROJECTILE

In electrical engineering we understand maximum energy transfer with matched impedance (Load). Your experiment here parallels this axiom in mechanical Physics as well. Thank you and well done

Hi Tom, just a thought. You did one video explaining how efficient the Trebuchet is, and the result was lower than you expected. As a cyclist this reminded me of the marginal gains of cycling adding up to a huge efficiency difference between an old bike drivetrain and a new drivetrain with sealed bearings, often ceramic and in general a very watt efficient transfer of energy with the same input. What if you did a video replacing all the pivot points with sealed bearings, and similarly looking at other materials for the sling and other contact points. This could be very interesting to see how much different these reductions in friction can make and effect projectile velocity. Thanks For the great videos.

that drone shots around the 6 minute mark are awesome and I would love to see more. And for a projectile, just 3D print one and add some metal weights in the center. Maybe a "2 part shell screw together design" with even spots for steel marbles to customize weight?

3d print a projectile! Adjust the infill accordingly. I don't know if the density is high enough though

Enjoyed the juggling at the end way more than I should have done

The drone shots of launches are just so satisfying. Nice one!

At 10:30 that doesn't look like backspin. It looks like topspin!

I built a trebuchet as a high school project for my daughter and then couldn't leave it alone. I also found that tennis balls weren't ideal due to wind resistance and weight. Problem solved by cutting a small slit in the ball and inserting pea gravel. Weighing each ball on the scale for consistency. The gravel in the ball reduced bouncing and rolling on impact which helped recovery. Pink is easier to find in the grass than green. Changing the length of the sling also alters release time so heavier projectiles need to be tuned to the sling.

If you decide to use a baseball, keep in mind that the placement of the seams can make a big difference in where the ball will go. This is how pitchers can throw fastballs, curveballs, sliders, etc. Search for curveball and you will find a lot more information on this.

Tom, I look forward to seeing you run out of variables to test, and the only option is to double the size and weights so you can compare the two data sets! Thanks for making these videos.

I think it would be cool if you could try to find the optimal release angle (or better yet, initial projectile trajectory angle) between 30 and 45 degrees with backspin. Then once you know the best release angle hold that constant and optimize on the ball weight. RE: Dimples: The only point of dimples is to trip the boundary layer so it becomes turbulent and the flow can maintain attachment reducing pressure drag caused be the wake. The dimples increase viscous surface drag, but for the right Reynold's number the decrease in pressure drag is greater than the increase in viscous surface drag. If the larger tennis ball (relative to a golf ball) and fast speeds of the trebuchet can put a smooth ball into the regime (i.e. high enough Reynolds number) where the BL has transitioned to turbulence and remains attached (for longer) then the drag coefficient will be much lower. See the plot for C_d on the right hand side at en.wikipedia.org/wiki/Drag_coefficient. (Link to image: en.wikipedia.org/wiki/Drag_coefficient#/media/File:Drag_coefficient_on_a_sphere_vs._Reynolds_number_-_main_trends.svg)

1. E=mv^2 is a second degree polynomial in the velocity, also called a quadratic form. Functions that have the variable in the exponent are called exponentials. 2. At 9:19 and 8:07 it looks like that the arm is flexing quite much under the force, perhaps try for a more stiff arm? I would guess that the flex is messing up the projectile release. 3. At 8:07 you state that that the arm is moving far less with a heavier projectile, that should be due to the higher energy transfer efficiency of a heavier projectile. It is simply less remaining energy in the system. Edit: your trebuchet is also a double pendulum which is a chaotic system which means that very small deviations in initial values (starting position of the projectile, skewness or give in the machine construction e.t.c.) will affect the behaviour dramatically. Therefore the behaviour is very hard to predict.

I think this is maybe one of your best projects

Nice way to science things up!! Really nice job with the testing and consistency doing so.

I discovered the channel yesterday and ended up spending most of my Sunday watching your videos. Fantastic work on the airpowered engines btw... and today i ended up here. at the trebuchet videos. I'm still in the middle of watching them all, but a thought that i had was. Maybe you can get even further by making your own prejectile in the style of a golf.... and i reached the end of the video and you mention exactly that idea. And here i thought i waited long enough to share that idea :D

also, have you considered a steel ballbearing. heavy and small

A baseball's dry weight is supposed to be 5.0 ounces, or ~142 g. However, they absorb water over time, from humid air, or damp ground and grass. Worst case, a ball that has been used or left in the rain can be overweight by 0.5 oz/14 g or more after drying in normal household conditions. Of a sample of 5 used baseballs I happen to have handy that were dried in a dehumidifier, three weigh 147g, 1 weighs 146, and 1 was at its 142g initial weight. It also looks newest. So, if you're looking to really control the weight of your projectile for consistency, a baseball will get a bit heavier over time as it absorbs water into its leather covering.

That's awesome that you have that big field right there! Makes it more convenient for these kinds of experiments haha. Great video.

To make tuning the angle of release less tedious, you could erect a large net to catch the balls as they leave the sling. That way, you could reload and shoot within seconds, and still get some accurate slow-motion footage of the angle of release.

I've subscribed because of these trebuchet experiments. The cricket ball was the decisive factor! ;-D Thanks Tom.

Great channel you have👍 When we hit the golf balls on the driverange, we hit with topspin, to make the balls float on the air, when the balls are hit just right, there is a pretty decent gain. Good luck

Neat videos! One variable worth considering would be the launch arm length; if you increased the length you’d increase the rotational moment of inertia of the launch arm, which would delay the stall and compensate for the tennis balls’ low weight. Would make life easier not having to procure specialty projectiles.

Experiments like this are the best way to figure stuff out... or at least the most interesting. But I'm pretty sure research is goes just as far!

In junior school we used 133g cricket balls, that is right in he money I would say. I am sure they will be easy enough to find. I love these videos, keep up the good work 👌

You are a really lucky man with the British weather this year.

What you've discovered is what electrical engineers call impedance matching. The sling length, counterweight mass and lever geometry all affect the ideal mass of projectile. If they are matched, you get a newton's cradle kind of effect where all energy is transferred from one moving object to the other. Alternatively, have you seen a newton's cradle with unequal balls?

I’m very late to the party, but what you’re looking for (if you haven’t already found it) is a pitching machine practice ball. They’re dimples like golf balls and approximately the weight and size of a baseball. As a bonus, they’re also not quite as hard as a baseball. So if you DID happen to hit someone in the head, it likely wouldn’t kill them. Juggs (yes, that’s actually the name of the company) makes the most ubiquitous ball of this type. I believe they come in yellow and white. On amazon they’re called “dimpled pitching machine baseballs”

"The back spin didn't change the range much", but as Barnes Wallace found out (or rather already knew), inducing back spin on the bouncing bombs enabled them to be released in time for the Lancasters to get clear of the dams before they came to a stop and sank to the base of the dams before exploding.

I am glad to see that you're not done with the trebuchet as I'm really enjoying it. I would be interested to know what your efficiency came out to be with the last ball that you were using.

That is really awesome. This video should be on trending.

Rolling Whippers are the most complex Trebuchet as far as timing goes, frame stall, CW stall, arm stall, Gap, sling length, pin angle, mass ratio, beam ratio, cradle elevation all come into play.

great work with the data analysis and optimisation. now you ahve the weight pinned down, now its time to optimise for release angle

Awesome, you build a trebuchet that can shoot around corners!

I have built my own trebuchet copying your design and having fun. It started as a lockdown project during our New Zeland level 4 lockdown. I am trying a tapered arm made of radiata pine rather than an arm of constant rectangular cross section to reduce inertia moment. I am considering making the sling itself from thin cord as well as the using the same thin cord for the fixed and release strings. And then making the pin as light as possible-maybe from nylon. All this to reduce the inertia of the arm so that more energy goes into the projectile. Going further a tapered carbon fibre arm might be better for better strength to weight ratio.

If the tennis ball is spinning quite fast you have to consider that an amount-significant or not- of its kinetic energy is in the form of "rolling kinetic energy". Not considering this could mean that your calculated efficieny is actually somewhat lower than it really is. Not sure if is negligible or not, but it's something to consider. Keep up the good work!

Love the video! Great graphs and analysis, very interesting to watch. Always love me a good trebuchet video

keep the fuzz on the ball, believe it or not but it actually decreases drag, look at the workings of a golfball. to increase distance you can increase weigt of the counter balance, increase the arm of the trebuchet, change the release angle of the projectile and maybe give the projectile a spin to generate a bit of a magnus effect. maybe you can also make a stop for the counterweight so that it remains extended while upright.

Since you're probably not going to do much more with that cricket ball, why don't you drill a series of holes in it's surface to bring it down to your target weight. But spread them out so that it looks more like a golf ball. The dimples on that are to make it fly further and faster. Of course, the real reason I want you to try that is to see if the ball will whistle when you launch it. I think that would be really cool!

I like how the GoPro is spinning faster than a complete scan so you can see the sky twice in some of the frames.

Late to the party but I reckon that the bottom spin does not add to the range because, as you said the horizontal velocity of the ball diminishes quickly due to friction. Because of this the angle of the trajectory itself is less relevant. Great stuff!

Love your vidoes! That ending was great! Keep doing what you're doing! Your great at it!

You can either 3D print or CNC the right projectile. That way you can make any combination you want. Maybe even add something to it so it's better visible on the drone shot?

You probably know this, but the traditional trebuchet used a slide for the shot to track in before being lifted. They would move this slide to aim the shot. I made a trebuchet about 18 years ago. The frame was about 3 feet tall (1 meter) the first day of playing with it I smacked myself in the head with the arm and I then beat the shit out of it and threw the pieces into the trash. I haven’t thought of making another since.

thank you for correcting the previous KE calculation error, and for good science in this video. The tennis ball is quite furry which may have caused significantly more drag

In America, the youth baseball leagues use a pitching machine that works on a spinning wheel. Because of the spin the wheel imparts to the ball, they have to use a baseball sized polyurethane ball that is dimpled like a golf ball. That may be something you want to look into. And because they are made for firing out of a machine toward children, they are available in softer or lighter than a baseball versions.

part of this is they used to control the amount of weight by getting people to shovel rock and dirt in the weight to adjust. in this way they could affect the angle of release very precisely

You can use the knowledge to adjust your aim. Tilting the ball left or right, you can aim about 6 degrees left or right. For a huge war trebuchet that can,t be moved, that would be quite handy.

Your intelligence never ceases to amaze me

Pay very close attention to the stitching on the balls that have it. The stitching can interact with the spin on the ball and affect trajectory.

Hey Tom, I am loving your work. Keep it up!

This is a really cool series thanks for doing it!!

Some of the gravitational potential energy of the weights is converted into trebuchet motion: first backward motion, then forward, then backward again. The mass of the trebuchet is not trivial. Thus, I think you could maximize the distance of the projectile by locking the wheels of the trebuchet, or otherwise guaranteeing that the trebuchet is stationary during the launch of the projectile.

You should try making a projectile with holes in it so it whistles. You could even add several so that it whistles more.

What a wonderful juggling outro!

If you can juggle those dissimilar weights so well you should be easily capable of Mill's mess. Great video. I am currently working on a UK legal sea fishing bait launcher, and your videos are of real practical interest (mainly because they rule out trebuchets as too bulky given the range and weight of projectile). So thanks!

You are using 'exponentially' wrong. Just because it's a curve, it's not necessarily a exponential. x^2 is not exponential, 2^x is.

When optimizing something (such as projectile mass) you have to actually try values of range high enough so that there's an extremum point. For example in your projectile weight vs distance graph it just increases which means that you could have further increased projectile mass and got even longer shots before you actually reached maximum. So instead of monotonously increasing/decreasing curves, all of them actually have minimums and maximums if you plot them on wide enough range of input values.

I know of a paper that describes the optimal ratio of mass of a projectile to the mass of the counterweight as 1:133 respectively. Although the paper regarded a traditional style trebuchet, not a "whip", I would recommend testing.

Engineer AND a juggler. Well done!

I love this series! Great videos 👍🏻

Tom, did you know there is a trebuchet simulator website. You can enter in all of the variables, and it calculates everything for you. It even animates the model! Good luck, and keep up the good work!

What i am interested in now is how the shape/weight/movement of the frame impacts efficiency. My instinct says that allowing the pivot to move vertically would help, but i don't know how to do that. Installing wheels for horizontal movement is simple, but vertical is trickier, especially since you'd want it adjustable to allow for some experimentation. Maybe some sort of suspension could work. Fascinating project!

You could add this to your trebuchet playlist

If reducing the video data is time-consuming, just do a series of release pin adjustments and throws, then reduce all the data in a batch. This will let you find the best pin setting after the fact.

You can throw the heavier projectiles, but you have to advance the timing of the arm by opening the gap. Get the adjustable timing prop installed.

Hi Tom ! To understand the influence of spin on the trajectory, you could try the experiment with a ping pong ball shot in different styles with a bat. "flat" spins should give a parabola that's steeper in the end (because of drag) Back spins should give an almost linear trajectory in the beginning, and then vertical drop (like the one you experienced) Top spins make the ball want to go downwards with a trajectory that is more curved Once again it is very easy to try this yourself with a ping pong ball, and there are many simulators out there! Comparing the ranges of the three in your next trebuchet video would be awesome! Keep up the great content!!

Try experimenting with the sling length, If i'm not mistaken lengthening the sling should slow the energy transfer and allow you to use heavier projectiles

You would want a in swedish "bandy" ball. It's texture looks like an orange with acne. Also I do belive that in order to accelerate the hevier ball the arm and sling needs to be longer, not shorter, to maintain transfer efficiency.

Time to integrate the paraglider project into the trebuchet project - and use it to launch vertically ;)

A lacrosse ball is mostly smooth rubber, it is 145 - 147 grams or so (5 - 5 1/2 ounces). With the baseball you will get spinning because of the laces/stitching by design.

Time to try a golf ball. I don't know what they weigh but the dimples, weight, and smaller diameter should add a lot of range and reduce wind influence.

great new advance. love these vids i always learn something new. cheers Tom!

Man I love this project! You could try getting closer to a release angle of 45° to maximize the range.

If you want a more difficult yet fun project concerning catapults, I would recommend trying your skill at a Ballista (Grecko-Roman Catapult).

In archery, heavier arrows shot out of the same bow will always have more energy, because their slower speed allows more time for the bow to accelerate the arrow. I would assume the same is true for trebuchets.

You might want to start experimenting with projectile diameter. Drag is proportional to cross sectional area so a smaller projectile will go further.

One factor why the heavier projectiles go further (even though the launch angle is worse) is the reduced effect of air resistance. Namely the higher the average density of the projectile the smaller the effect of air resistance per gram, since the force of air resistance only depends on its shape. So a smaller projectile of the same mass should go further

I guess to find the optimal projectile you would need to balance a couple of things: air resistance, efficiency of energy transfer from trebuchet and any other physical limitations. Air resistance could be minimised by minimising the size of the projectile and may or may not include inducing turbulent flow (it's dependent on a variety of factors, in the case of golf balls, turbulent flow decreases air resistance but this isn't a universal benefit) and energy transfer is a function of mass as you calculated. You may find other physical factors effect the projectile, especially if the projectile gets too small and you have trouble releasing cleanly from the sling. It would be interesting to try a variety of different size, same mass projectiles such as a 16mm steel ball (~138g), 23mm Aluminium ball ( ~138mm), weighted golf ball (~41mm dia) and weighted tennis ball (67mm dia), in theory the small you go the further the range should be unless you experience any other issues.

You should 3D print a master for the projectile and make molds of it. You could experiment with different types of materials like epoxy, concrete, etc for the projectile and you could add different materials like steel ball bearings, sand, etc into them to get the weight right

I was going to suggest throwing something like a billiard ball. It would be interesting to see what would happen throwing something completely smooth!

10:30 If top part of the ball rotates forward, then Magnus force acts downwards, reducing its range. Rotation should make top part of the ball move slower compared to bottom part relative to airflow in order to increase range.