Challenges: 

It needs to be lightweight.

I need to balance weight and power when choosing a battery. Ultimate Frisbee enthusiasts wouldn't use it if the flight was influenced.

It must be balanced.

Final design will be a circular PCB, with weight distributed evenly around the board. 

It needs to be robust.

Some people catch the frisbee by clapping the top and bottom. The PCB & peripherals will need to withstand a slap.
Picture
The back of Rev.1: Designed in EAGLE. You can see the IMU centered in the middle.
Picture
Soldering the front of Rev 1. Passives were all 0805. These are small enough to create a compact board but large enough to ensure I don't go blind.
Picture
After soldering the 9DoF IMU and supporting components: this will detect acceleration in x/y/z, angular rotation in x/y/z, and magnetic force in x/y/z.

2019-08-19: Improvements to mull over...
  1. I want to switch out the GPS module in this next version - maybe Adafruit's Ultimate GPS Module! It is much smaller and would not use SIM. This means I'd have to think of another way to store or communicate my data. Maybe SD card? I'm quite nervous about that though, since high impact from being thrown around may mean the SD card will fall out (apparently a common problem, as I've learned from rocket flight computers.) Bluetooth is another option, but the working range leaves me jaded.
  2. It's time to try out a circular form factor as well! Figuring out balance without compromising frisbee play was one of the biggest challenges and reasons I decided to do this project. Of course, a circle isn't going to solve components of different weights and a big rectangular LiPo. I also am not sure how I'll measure weight distribution and calculate the center of mass accurately. What even is the proper tolerance for how much extra mass a frisbee could hold? Many questions that need answers...

2019-05-14: How do I supply power to the board?

Now that the IMU and GSM modules both work, I need to find a way to supply power without a power supply!
My 3.7 V LiPo battery came in the mail today, but there's still a glaring issue.
The GPS may work with a battery anywhere between 3.4 - 4.4 V, which meant my 3.7 V battery was good to go. However, the microcontroller uses 5 V logic!

Would I need to hook up an additional battery greater than 5 V to accommodate the microcontroller?

I explored some possible solutions:
  1. 3.7 battery + boost converter (for microcontroller)
  2. 9 V battery + buck converter (for GPS)
  3. 3.7 battery + 9 V battery
A boost converter will step up the voltage from 3.7 to 5 so the microcontroller could run.
A buck converter will do the opposite, pushing the voltage down from 9 to 3.7 so the GPS could run.
​
Q: I was concerned about using a higher voltage battery and pushing it down with a buck converter. Why would that be dangerous?
A: The microcontroller circuit includes an FTDI module, which has its own built-in voltage regulator. This would draw too many amps, leading to high dropout.

Note: 9 V battery is interchangeable with anything above 5 V. However, if we connect an exact 5 V battery to the voltage regulator in the microcontroller circuit, the dropout voltage is too high and less than 5 V will end up being delivered to the microcontroller.