The Big Think

December 21, 2005

Project Trebuchet

Filed under: Mad Science — jasony @ 3:30 pm

(lots of big pics, please be patient)

On monday and tuesday this week my good (and too-infrequently-seen) friend Sean and I constructed a working model of a medieval trebuchet. I had mentioned the possibility of building one to Sean a couple of years ago and he got the wild-eyed look that told me he was a fellow mad scientist.

Sean brought two of his great kids with him (Liam and Emily) and we had a really fun two day slumber party. The first order of business was a trip to Home Depot for supplies. Instead of using carefully drawn out plans, we opted for the “print a picture off the internet and just wing it approach”. It worked great. We took the pic to the Depot and used it to gather 16 2×4’s, a couple of lengths of 1/2″ threaded rod, and some bolts and washers. The whole bill came to $58 but we ended up having ten of the 2×4’s left over (many of which I had already cut before we realized that we didn’t need the wood). Total price of materials used was about $35, however I did use some scrap that was sitting around the shop.

Lesson Learned: good tools make a huge difference. We built the entire trebuchet in about 6 hours (not including an hour off for lunch). This is far faster than any of the other builders online have mentioned. I attribute this to the fact that there were two of us working on the project, and that we had access to the right tools. While you could definitely build your own without a shop full of tools, I think it would significantly add to the time required. We used a table saw for ripping the 2×4’s, a compound miter saw for general cross cut duties, a drill press for the critical holes, a cordless drill for reinforcing screws, clamps, squares, and other miscellaneous things. But by far the most useful tool was my pneumatic nailer. This wonderful thing let us glue and nail boards together in seconds. We went through probably 100 nails on this project and it was very helpful to just pull the trigger and move on. I’d put this tool in the top 5 of the Pantheon Of Useful Tools. If we wanted the thing to stick around for years we’d go the bolt-and-nut route, but nails were plenty strong for what we had in mind.

Step 1 of the build was the assembly of the base. Since we didn’t make any measured drawings we decided that the overall size of the treb would be dictated by the length of the base, and the maximum base length was dictated by the longest 2×4’s in our pile. We found one that was 8′ long, ripped it on the tablesaw into a pair of 2×2’s, and had our base rails. Once that was done we determined a good width (18″… no real reason other than that it seemed like a decent compromise between strength and portability) and nailed the stretchers together. Here’s a pic of Sean on the compound miter saw (CMS):


Sean told me he didn’t have that much experience working with some of these tools, but like any male, he took to it fast and very naturally. He was using the CMS, nailers, etc with ease and speed. He’s a great shop buddy. Sidenote: it’s nice to have solitary shop time and get into that state of flow that makes time pass so fast, but sometimes it’s really, really good to work on building something with a partner.

Once the base and stretchers were done we did some figuring and started assembling the sides:


We came up with the height by basically pulling a dimension out of the air (or somewhere else maybe?), holding our hand at the approximate height, and then adjusting until it looked right. Then we measured from the base to our imaginary point in the air and cut the main vertical post.Before nailing it in place we had to glue and nail the vertical sidepieces to reinforcing plates made of scrap plywood. We then put the long side pieces (Sean said it looked like a picket plaque) on the drill press to drill the hole for the main pivot axel (a piece of 1/2″ threaded rod). Once the side pieces were nailed in place we measured for the diagonal beams, cut them, and nailed them in place as well.

Next, boxmaking! This was the main place that we could have improved the final range of the trebuchet. I made the box based on the clearance width between the sides of the trebuchet, but I let the height of the box be dictated by the scrap piece of plywood I used. If I had made the box deeper it could have held more weight and we’d have gotten a longer shot. Or if we could have found a teaspoon of neutronium we’d have had enough weight. Anyway, the box was easy to make: I basically cut four pieces out on the tablesaw and used the drill press to cut a couple of carefully aligned holes:


then I got a piece of scrap hardboard and nailed/glued it onto the bottom. If we were building for a more permanent structure I’d have wanted to insert the bottom in grooves or otherwise reinforce it, but for our purposes this was good enough (I did, however, use a lot of glue and nails).

It was starting to come together!

Our next step was making the main arm. We picked a good piece of ripped 2×2 and decided on where we wanted the hole for the pivot point to be using the ratio of 4 to 1… four units of arm in front of the pivot and one on the weight-box side. This seems to be the accepted ratio. I’m sure there’s a more precise point trigonometrically speaking, but this was close enough for us. We then cut two short lengths of 2×2 and used them to reinforce the arm on the weight box side to about a foot past the pivot. We glued the three pieces together. We could have used the jointer to cut flush joints but they were pretty close to begin with so we just went with it, reinforcing the arm with some long nails. Here’s Sean’s seven year old son Liam learning the rules of guydom:


Next we drilled the main pivot hole on the drill press and reinforced it with some metal plates that happened to have a 1/2″ hole. This was probably overkill, but we wanted to be safe:


Here’s a picture of us attaching the weight box. The arm has been bolted onto the main axle. Note that our holes weren’t perfectly perpendicular through both the main arm or the weight box so the alignments were slightly off, but not enough to interfere with the pendulum motion of the main arm. You can just tell that things are slightly misaligned


Here’s me testing the arm travel:


Closeup of the main pivot point:

Next step: the firing pin! I wrapped some string around the end of a long 44 cent nail and fed it through some eye bolts. We also screwed a metal plate onto the main arm to reduce friction. It made a cool “shhhhink!” sound when the nail was pulled. And that’s important.


While I was doing this, Sean cut up a pair of his old jeans and started sewing the pocket for the payload. His daughter Emily helped out by reinforcing the pocket with some really good stitches. Here’s Sean trying to sew. Remember Sean, you thread the NEEDLE, not the thread cutter thingie.


We’re almost done! In this shot you can see that we’ve added the side outriggers for lateral stability. These were detached for transport and screwed in at the firing range (the local park). We also built the trough where the sling line and payload pocket is laid out before firing. That way as the arm moves through its arc the payload has a smooth runway to get up to speed.

Before we took it up the street, we made a quick stop inside to finish preparing the pizza I had started earlier in the day. Homemade pizza. yum.


We christened the trebuchet Fafner, after the giant from Wagner’s Ring operas (see what you get when you have a couple of music geeks build something? With the addition of a Pirate Flag (“Fafner! Arrrr!”) we loaded it up into my truck and were off to the park!


Group shot with me, Sean, his kids Liam and Emily, and the Very Understanding Wife Erin:


Loading the projectile potato:


At this point most of the documentation was done with Sean’s video camera. He has edited the footage together and posted the file here. At some point he’ll also submit it to Google video for free hosting and I’ll change his link to go easy on his bandwidth.

Our first shot was a pretty big dud. The potato only went about fifteen feet, due to the fact that we only had 30lbs of scuba weight in the counterweight box. Plus, the string release on the sling was late. We added more weight and used Sean’s handy multi-tool to straighten the release pin (a spare hex wrench I’d bent up and screwed onto the end of the main arm). You fasten one end of the string sling to the arm with a screw, and slip the other end around the bent pin that sticks out of the arm (I had tied the string to a small carabiner so it would slide off smoothly). When the arm flings through the air the ‘biner slips off the pin, opening the sling, and the projectile sails through the air. It takes a bit of adjustment to get all the adjustments just right so you can achieve the optimal 45 degree release angle on the sling.

The second shot was over 70 feet! Hmm… still not far enough. We added still more weight by filling the box with all the scuba weights as well as all 25lbs of Erin’s yoga weights (what did I say about an understanding wife?). With some more tweaking of the release pin and some experimental shortening of the sling string (at our neighbor Jenny’s suggestion), our longest shot of the day was just shy of 110 feet. It started to get dark and we had to hustle back to the house for pizza and juice for the kids and pizza and well-deserved adult-beverages (Hefe Weizen!) for the big kids.

What did I learn from this whole project? Well, for starters, we were both amazed at how easy it really is to build a working trebuchet. It took us 7 hours of working time including going to the store for the supplies. A poorly equipped builder could easilty do it in a day or two, assuming a reasonable level of handiness. A trebuchet is also a much safer weapon since there isn’t anything under stress like in a catapult or ballista. As long as the treb is built strong enough to hold the counterweight, and as long as you carefully set it and keep out of the way of the swing, there’s little chance that it could hurt somebody. It was also surprising that it was so cheap. Less than $50.
Another lesson: build the box big enough! The biggest determiner of range is how fast the arm is moving when the sling releases, and this directly correlates to the amount of weight in the box. If I’d have made the box bigger we could have increased the speed of the arm, and the range would have been much farther. I’m considering making the modification just to see how far it’ll go.

The trebuchet is a wonderfully forgiving design as long as it’s built strong enough. As we proved, it’s possible to build a working model with nothing more than a picture from the internet and some modest mechanical ability. Considering the fact that, properly built, all of the forces are in a for-and-aft direction, it’s pretty easy to see where the main stresses are going to be, and build accordingly. It’s easy to build a really tiny trebuchet, or, with a lot of work, use the same principles to scale the design up to be really honkin’ big (I’ve seen this last model fling a grand piano a quarter mile on a NOVA special).

Finally, I relearned what I’ve always known. With a bunch of tools and a crazy friend, you can really tear down some walls.

Sean’s video is here (Great edit Sean!), and if you happen across this post sometime in the future and it helps you out, please drop me a line at my gmail address. I’m JasonY at gmail. Thanks also to Paul Carlisle for the pointers and email.

Whee!!! Now go make your own!

1 Comment

  1. Read this Aussie’s note about bucket size – it totally syncs up with what you said:

    I wanna see it after you build a bigger bucket!!! 😀

    Dateline Austin, TX and San Antonio, TX: Man throws potato from his backyard to the Alamo Dome…

    Comment by Patrick — December 22, 2005 @ 4:09 pm

RSS feed for comments on this post.

Sorry, the comment form is closed at this time.

Powered by WordPress