As promised, more progress has been made on the thrusters, camera, and electronics. Now that the design is becoming more complex, I’ll go into more detail on my process and mistakes. 

Firstly, I worked on the design of the two back thrusters, which I envisioned as two cylindrical units that would house the motor and propeller that I would be able to print separately and attach in a flush and dynamic way to the rest of the drone housing. If you remember from my last update, the only progress I had made on this was cutting out the area for the thrusters to be placed. However, this ended up being a bit of a mistake, as I realized that in order for the thruster to maintain some of the curved features that would transition smoothly into the drone, I would need to somehow keep some kind of reference to the original curvature of that part of the drone. I ended up solving this problem by instead designing the two cylindrical structures first, then using the boolean feature to intersect and separate the two chunks rather than deleting them, allowing me to add their geometry to the design of the thrusters. 

I then added all the necessary details to the thrusters, including the motor mounts, screw holes for mounting to the drone, and various fillets and other small features to make it all look nice.

I then decided to work on the camera module, which would consist of a small camera housed in a separate, smaller acrylic tube. The key features I wanted to include were as follows: 

• A structure to keep the camera in place inside the tube to prevent any unwanted movement

• Strong, small end caps

• A way to mount the camera module to the drone

First, I looked online for camera options, and came across the OV9782 camera, which would provide the high speed, color imaging and small size I wanted. I then found a small 50mm OD 46mm ID acrylic tube that I would be able to cut down to size that would fit the camera. Finally, I designed the camera mount, which would screw in around the camera and fit snugly into the tube, as well as the end caps, which also doubled as mounting structures.

Finally, I corrected some mistakes I made regarding the battery. Currently, the only thing powering the drone would be a single 6-cell battery pack, with each cell providing around 3.6V and a maximum amp draw of 30A. This means that a battery pack with 6 of these cells would have the adequate voltage for 6S motors, but would remain capped at a 30A draw, which is a problem considering each of my motors would be rated for a draw of 40A. Furthermore, the Raspberry Pi and Pixhawk would also need to receive constant power, taking away even more from my already limited maximum amp draw. So in order to bring up this number, I would need to connect another 6S battery in parallel, maintaining the same voltage, while doubling the max amp draw to around 60A. While still not much, based on a couple of assumptions, it should be adequate for this project. Firstly, only two of the five motors equipped on the drone will actually need to be spinning consistently, as the rest are for minor up and down adjustments. By making the drone neutrally buoyant, I can minimize their usage. This brings the total amp draw to a maximum of about 80A for the motors + 5A set aside for the electronics. After that, I should just be able to limit the speed of the thrust motors so that the maximum draw falls somewhere under 20A-25A, so that the average amp draw falls somewhere between 45A to 55A, under the 60A limit. Limiting the speed likely would have been a necessity anyway, as underwater, pushing water too fast leads to cavitation, meaning slower, higher torque propellers are often more desired. 

So I designed a new holder for a larger battery pack with 12 cells rather than 6.