BSD Battleship Sean Mcevoy


  • For this project, my team was tasked to design, build and test a tennis ball air cannon to compete in a game of ‘BATTLESHIP’ on the football field. Our air cannon will have adjustable pressure, elevation, and z-axis rotation. Our air cannon will be mounted on a base of our own design that will allow for rapid testing and evaluation of accuracy and precision.
  • As an additional project, we were tasked to understand how Arduinos™ work and become proficient in circuitry design as well as the coding functions of the system.


We faced several challenges in the use of Arduinos. Personally, I had a hard time understanding the coding, at first, and just kept to the hardware side of the tool: building and customizing our circuits. I learned the function and how to properly use many common circuit components (resistors, relays, multi-color LEDs, etc). Probably the most interesting parts, and the ones where I liked to experiment with their code, were the multi-color LED (the ability to change/cycle through its colors by increasing or decreasing the input to the individual pins) and the buzzer (the ability to change the frequency of the sound produced based on strength of input).

Above is an example of one of the challenges we had to complete, a working relay switch. Visible is the circuit (with wires, relay, LED and resistor) and Arduino board. The relay acts as an electronically controlled digital switch.

This is the program we used to digitally design, assemble, and code our Arduino circuits before we did so in real life. This is Autodesk Circuits, which allowed us to assemble using identical parts as found in our Arduino kits as well as use identical codes.

Overall Cannon:

PVC Construction

For this segment of the project, creativity was banished and uniformity was in. Every group needed to fabricate the same PVC cannon (same materials and dimensions) so that there would be a fair competition in the final test stage. Fabrication of the parts was simple: cut a 10' length of 2" PVC into a 4' barrel and 3' pressurization chamber using a chop-saw. The two tubes were then fitted into a store bought connector piece which had a 2"-to-1" adapter with PVC primer and glue. Then, two 1" threaded adapters were attached to a sprinkler valve with Teflon tape to reinforce the water-tight seal (the sprinkler valve is triggered by an electrical current, which opens the pathway between its two sides, permitting an instantaneous trigger). The threaded adapters were primed and glued to elbow points, which were glued and attached to 1" PVC pieces that fit inside the 2"-to-1" adapters.

Barrel Support

At the start of this segment of the project, we came up with several designs for how we would both stabilize our cannon and, also, give it a vertical range of motion. Once we got approved to scrap an old bike for the project, we came up with the idea of using the front wheel, fork, and head tube to give us all our degrees of motion. For our actual cannon supports, we decided to create a collar-like design (a wooden frame with holes for the barrel and compression tube) that will slide down the cannon.

For the barrel supports, we created a simple design: a rectangular shape with two holes. Once designed in OnShape (a CAD program), I converted the files to PDF and used our laser etcher to cut the file out of thin laminate. I made a huge mistake, I placed what should have been the distance between the two holes of the support (the distance between the barrel and compression tube) as the total distance of the part. Need less to say, it was too short. But, the holes fit the 2" PVC tubes. Then we switched over to our CNC machine and used the edited file. While we got the distance correct, we did not correctly configure the offset in our cutting file (from V-Carve) to compensate for the tolerances in the machine: the holes were too small for the PVC. We quickly adjusted the file and re-ran it which was a success. To not just scrap the incorrect files, we sanded the holes until they fit unto the pipes.

Once we had the supports on the cannon, we needed to attach them to the bike wheel. To do so, we drilled two holes on each of two opposite sides to bolt the wooden supports to. This created a secure connection between the wood and metal, creating a secure mount for the cannon. Next, we had to create the base.

Base Construction

To make the base, we had several designs that used a wide wooden structure to hold the bike fork. However, our teacher gave us an opportunity we could not pass up. He said that he could get the school district's welder to come into our class and teach us how to weld. We drafted a design of a tripod that used the remaining bike parts. Before the welder came in, we cut the parts we needed sanded the paint off of the parts (head tube-main tube construct, body tube, and wheel supports). Once the welder came in, we practiced on other scraps before finally welding our base. To do this, we welded the body tube to the front of the head tube (look at photographs below) to create one large leg and then the wheel supports to the other end of the head tube-construct to create the other two legs for our tripod-like base. However, I made a huge mistake. When fastening the wheel supports on pre-weld, I did not make sure the would-be legs were aligned correctly with the first leg, creating a base that would never be leveled. But, thanks to some clever thinking, we came up with the solution of cutting the main tube in half (perpendicularly and width-wise) and then rotated the halves until the legs lined up how we needed them to. We then welded the two halves together.

Once the welding was complete, we decided to manufacture our horizontal lock. This we did not do by trial and error, we succeed on the first try. We just took a bolt from the shop that was of reasonable dimensions and used a tap set to tap a hole into the head tube. Now, with the assistance of a wrench, we could just use friction/pressure from the bolt to restrict horizontal motion.

Final Touches and Modifications

After the cannon was fully constructed, we had to make the cannon look attractive and bold. My teammates spray painted the cannon black and green, making sure the sprinkler valve was not touched. I decided to spray paint a spoke on the wheel bright orange so that we could use it as our angle indicator on our adjustable vertical protractor.

After a couple sessions of testing, I really disliked how cumbersome and awkward the battery was. I decided to make a mount for it by the back of the cannon, close to the sprinkler valve. I made this from a couple pieces of plywood and then decided to wire the battery and valve to a switch for ease of use.

Final Performance

To modify a quote by von Braun, the cannon performed perfectly except for landing on the wrong planet. There was never an issue with firing the cannon: the air tank filled without leaks, the wiring of the battery and sprinkler remained connected and was always triggered when the switch was hit, and the ball always released. We were even able to load a dixie cup with paintballs (my idea) and fire them precisely at a target. Precisely is the key word for the firing performance, though.

Aiming and accuracy were our shortcomings. Hitting a target seemed daunting to us, we often aimed too far right or too far left due to a slight offset to the right with the barrel in reference to the overall cannon (prevented accurate by-sight aiming). The locking mechanisms for the two directions of motion had varied performance. The vertical motion lock (basic bike brake that is locked into place with a velcro strap) never failed. The horizontal motion lock was not always secure because to fully tighten it, we needed a wrench. Also, because of the limited bike parts, our base was pretty small in width while our structure was tall and top-heavy, this resulted in a high tipping danger and impeded some of our aiming.


Here are two pics of the PVC assembly. The left shows the crucial sprinkler valve attached to two adapters. The valve allowed us to instantly open and leave open the pathway between the chambers for the most power. The right shows the full assembly, complete with excess PVC primer stains (purple). The end cap on the pressure chamber was not attached yet because the cross supports needed to be placed.
Here are the first two mistakes with our barrel supports: first the inner diameter was too little, the second was the length of the support. The issues weren't limited to the prototypes, when we cut out the supports on the CNC machine, the inner diameter was too small. So we adjusted the next few cuts, but we had to sand the interior of our already made pieces.
Here we were attaching the cannon supports to the wheel of the bike. This was done using two bolts passed through the wood and the aluminum wheel rim. After the supports were attached, we slid in the cannon until it was well balanced (using the axle as the center of gravity).

Here is a video of me welding one of the wheel supports onto the end of the main body tube to create a leg of the tripod design.

The left picture shows the bottom of the cannon after being welded and painted. The small wooden disk is a protractor we designed to measure our horizontal angle change. The right picture shows my personal creation, I disliked our old method of resting the battery in the nook of the elbow joint to I quickly designed and added on a battery holder from scrap wood and screws. To improve the firing mechanism, I added a switch and wired the battery and valve into a complete simple circuit.
On the left is a picture of myself on our second day of testing the cannon (separate from the class), I was pushing the ram rod in to ensure that our ammo (a miniature tennis ball) was properly placed in the back of the barrel. The right is a picture of me filling the pressurization chamber with air through the bike valve we epoxied on.
Here are 2-D PDF's of our CAD files for the project. We only needed files for the horizontal protractor (top left), vertical protractor (top right), and cannon cross-support (bottom). Both protractors were cut using a laser on thin wood and angle measurements were also etched. The cannon support was cut out using our CNC machine on 1/2" plywood.
Created By
Sean McEvoy

Made with Adobe Slate

Make your words and images move.

Get Slate

Report Abuse

If you feel that this video content violates the Adobe Terms of Use, you may report this content by filling out this quick form.

To report a Copyright Violation, please follow Section 17 in the Terms of Use.