Neyland
Peterson

Me and my team designed a rocket intended to pass Mach 1. I was responsible for the avionics, programming the accelerometer and altimeter to compare real flight data against simulation, and ensuring reliable recovery. I also 3D printed the nose cone and the motor chassis. OpenRocket put the G80 configuration at 453 m/s with an apogee near 984 m. We chose not to launch because the high-thrust motor paired with an airframe that was not strong enough made flight unsafe, which looking back was definitely the correct decision considering we were using three G-size motors on plastic bottles. I hope to launch something more regulated in the future.

453 m/s simulated (G80T-10)  ·  ~984 m simulated apogee

This was a guided project, and the data was given to me rather than something I collected. I worked through a Python notebook that analyzed accelerometer data from four static rocket-engine tests, three 3D-printed thrust structures and one held together with zip ties. My understanding of the deeper math is still at a surface level, but two ideas stuck with me the most.

The first is power spectral density, or PSD. Instead of looking at the raw shaking over time, a PSD breaks the signal into the range of frequencies it contains and shows how much energy sits at each one. Plotting all four tests together made it easy to compare how violent each one was: the gRMS value next to each test is a single number for its overall vibration, and Test C came out the highest at 1.61 g.

The second is the spectrogram, which puts time back in. It shows frequency on one axis and time on the other, with color for power, so you can actually watch events happen, like the engine igniting and the parachute charge firing, instead of averaging them away.

I also learned the basics of filtering. Before any of that analysis, the raw signal was run through a Butterworth high-pass filter to strip out low-frequency drift and electrical noise, which is what makes the real vibration readable in the first place.

This was my first independent project. I modulated a laser with an audio signal, aimed it at a photodiode, and fed the recovered signal into an amplifier and speaker to play sound over light.