A Hinged Propeller Lets One Motor Fly and Steer a Drone

Oct 19, 2026 Last reply: 1 hour ago 3 Replies
Jump to replies (3)

A drone that lifts off, hovers and steers using a single motor sounds impossible, and that is exactly what makes this the most surprising build I have come across in a long while. The trick lives in the hub: the propeller hub is hinged and 3D printed from PPA-CF, and when the motor accelerates the blades lag behind, so one pitches up while the other pitches down. Pitch and roll come from the same motor that makes the thrust. The rotor turns at almost 20,000 RPM, and torque is pulsed 300 times per second, timed to the rotor angle. At that speed a 0.2 ms control loop already trails the rotation by 20 degrees, so custom firmware computes the torque pulses directly on the motor controller.



As hardware it is a tidy package. The cylindrical red body is printed in flexible TPU and carries anti-torque fins that keep the body's counter-spin in check. Thrust comes from a brushless RS2306 motor, driven by a moteus r4.11 board with an integrated magnetic rotary encoder, alongside a flight controller. Credit to Joshua Bird for making the whole arrangement look almost simple.



The road there reads like proper engineering. A test stand of aluminum extrusion, weighted down by a food pouch, hosted the early bench runs, followed by crashes into walls and furniture before stable hovering and steering under radio transmitter control. A digital camera with an external flash, triggered by microcontroller pulses, captured strobe images of the propeller at precise rotor angles. Heading comes from an external infrared lighthouse beacon that the onboard receiver picks up once per revolution, because the body slowly spins.



Two things remain open. Vibrations saturated the accelerometer near 160 Hz, and the blade flutter behind that was cleared only by fitting stiffer propellers. The external beacon also reads as a stopgap, since the plan is an onboard linear camera sensor for future yaw estimation. Code and CAD files are released open source.



Project by Joshua Bird. Watch it on their YouTube channel:

formatting link


Moving the pulse calculation onto the motor controller looks like the key call to me. With a rotor near 20,000 RPM, a 0.2 ms loop already trails the blade by 20 degrees, so pushing that timing off a standard loop is the only way 300 pulses per second land where they should. It trades simplicity for a much harder firmware job, but the stable hover under radio control suggests it paid off.

I am curious about the hinged hub itself. Was the hinge printed in place as one piece with the PPA-CF parts, or assembled from separate printed pieces? With the blades lagging and leading 300 times a second at those speeds, the fit of that joint seems like it would make or break the whole design.

Since the aluminum extrusion test stand was weighted down by a food pouch, I would swap that for bolted ballast before pushing the rotor harder. A pouch can shift or rupture if a blade lets go, and given the crashes into walls and furniture during testing, anything that keeps the stand planted seems worth doing.

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required