Pressurized Air-Powered Fibreglass Toy Submarine

The main functions required for a successful toy were determined to be neutral buoyancy, hull, customizability, and propulsion systems. From the main functions, the design was broken into three distinct systems that each satisfy a main function.

The first prototype determined that mounting the syringes on the inside of the hull (a plastic bottle in this case) and using slots with handles to actuate them was not feasible. The core issue was that the arm would allow a large amount of water into the inner hull-which would be a clear failure.

The second iteration consists of 2 syringes attached to the inside of the hull and connects to the outside via small rubber tubes. To actuate it, the user pulls on the nosecone (the bottom end of a 2-litre bottle in this case) of the submarine and the syringes draw in water. However, a consultation with the ME 100 teaching team revealed a massive design oversight: while extending the nose might draw water into the syringes, the weight gained from that water would be negligible compared to the massive increase in internal volume. The second iteration thus demonstrated the impracticality of using a “piston-type” buoyancy system.

Following the two previous failures, the third iteration demonstrated that a simpler solution consisting of external tanks was the best path forward. The solution involves integrating the syringes into the structure of the hull, rather than trying to fit them inside. This approach not only eliminates the challenges related to increasing weight while simultaneously increasing volume, but also completely frees up the inside for the propulsion system.

We decided to utilize an Air Diaphragm Engine that was created in whole by Tom Stanton-a British aerospace engineer. We obtained the files and instructions from his YouTube channel and Thingiverse page. This provided enough mechanical complexity to our project without adding an unmanageable amount of difficulty.

The initial 3D print resulted in the first prototype. When the parts were dry-fitted to ensure compatibility with each other, several issues were found. One of the setbacks was that the pumped-in air was leaking out of several engine ports. This severely affected engine performance-so much so that the engine did not complete a full cycle. The first prototype was therefore disassembled, modified, and reassembled into a second iteration.

Using a pre-existing design presents unique challenges, as the hardware is difficult to obtain in a timely fashion. Many off-the-shelf parts are uncommon not only due to their small size but also due to the unavailability of metric sizes. Obtaining all required components presents a monumental logistical challenge.

Under normal use, Tom Stanton’s PLA connecting rod was too weak, and redesign was deemed necessary. To fix the connecting rod, a replacement was created using SOLIDWORKS and machined out of stock aluminum.

We spent many hours troubleshooting, testing, and modifying the engine as initially we could not get the system to turn over. Increasing both the size and weight of the propeller were two main fixes we attempted.

Fibreglass composite was chosen as the hull material as it fits the strength requirements and could be shaped around a mould of the required size. In addition, fibreglass was found to be a cheaper method of construction compared to 3D printing using ABS.

1 roll of 40”x 2 yd fibreglass was sourced from an online retailer, whilst epoxy resin and hard pink insulation foam were sourced from a local brick-and-mortar store. The foam boards were stacked and glued together using epoxy then trimmed according to prior measurements using an industrial filing tool.

This form served as a blank mould to form the fibreglass around. Using special epoxy, sheets of fibreglass were laid onto the foam core in a climate-controlled room. Special care was taken to coat the entire sheet with the epoxy and to reduce wastage.

The fibreglass hull was then left to cure overnight. We decided to not use acetone in the removal of the foam due to time limitations, so instead the decision was made to implement hand tools in cutting the foam out. After the foam had been removed, special fins were created by layering popsicle sticks and rubber bike tires together-multiple of these were made and hot glued onto the body.