My interest in drone design began with a desire to create custom components for a racing drone, starting with the front nose crush zone used to mount and protect the onboard camera. After building my first 3D printer, I wanted to move beyond simple prints and begin producing lightweight, structurally reliable parts for high-speed flight. However, I quickly discovered that the entry-level printer I was using lacked the precision, rigidity, and material capability required for performance-critical components.
Instead of working around these limitations, I applied what I had learned to design and build a more advanced 3D printer from the ground up. I developed a CoreXY system with linear guide rails to improve stiffness and accuracy, integrated 32-bit motor drivers and higher-quality stepper motors, and optimized the machine for faster, more precise motion control. With this improved setup, I was able to experiment with different propeller designs, materials, and structural thicknesses, carefully testing strength and weight trade-offs to shave off as much mass as possible without compromising performance. Through iterative testing and refinement, the final drone design reached a top speed of approximately 90 mph. Although the drone was ultimately destroyed during a high-speed landing, the project taught me invaluable lessons about real-world engineering trade-offs, failure modes, and the importance of testing at the limits of performance.
As the project progressed, I began grounding my design decisions with simple calculations to better understand performance limits. The final drone used a 15 cm (0.15 m) propeller, a 4S battery (~14.8 V nominal), and had an all-up weight of 283 g, or about 2.78 N of weight. Reaching a top speed of roughly 90 mph (≈40.2 m/s) meant aerodynamic effects dominated the design. Using standard sea-level air density (ρ ≈ 1.225 kg/m³), the dynamic pressure at top speed is
This helped me understand why reducing frontal area and smoothing airflow around the camera mount and frame had such a noticeable effect on stability and efficiency.
I also used propeller and motor estimates to sanity-check structural and electrical limits. For a 15 cm prop (radius 0.075 m) spinning near 30,000 RPM, the blade tip speed is approximately
which is well into the regime where material strength, print quality, and balance become critical. This motivated repeated testing of blade thickness, infill patterns, and materials to avoid failure at high RPM. With a 4S battery, a power draw on the order of 500–700 W at full throttle corresponds to currents of roughly 35–45 A, reinforcing the need to carefully manage efficiency and weight. Despite the drone weighing only 283 g, it experienced forces many times its own weight during acceleration and maneuvering.
Through this project, I gained practical experience with aerodynamics, CAD modeling, basic simulation, and data-driven iteration. Although the drone ultimately reached its performance limit and was destroyed during a high-speed landing, the process taught me how calculations, simulation, and real-world testing come together in engineering design—especially when operating near the edge of what a system can handle.