Quadcopters can steer without any linkages because they have four motors. Changing the relative speed of those rotors is all the controller needs to do. Take away three of them and you have a problem that helicopter designers have handled for decades with a swashplate, a precise assembly of bearings, links and servos that changes blade pitch as the rotor turns. Software engineer Joshua Bird took a different route. As Tom's Hardware reports, his small 3D-printed drone flies on a single motor. It gets pitch and roll control from a hinged propeller and firmware that responds in under 0.2 milliseconds, and it has no swashplate at all.
Bird documented the build in a YouTube video titled "I Built a Drone With Only One Motor." He has also published the design files and firmware forks in a public GitHub repository. That makes this more than a curiosity. Makers with a printer and some patience can open the geometry and see how the trick works.
The problem: one rotor, too few controls
The repository describes the project as an "Under-actuated drone with only one motor." Under-actuated is the important term. A free-flying body needs control over thrust, roll, pitch and yaw. A plain single-motor craft has one input, motor speed, which sets thrust and nothing more. Every other axis has to come from somewhere else.
Bird's answer, according to the repository, is "a drone that flies on a single propeller, mounted on top of a finned body." The propeller's hinges and motor timing handle roll and pitch, and the fins join in on yaw. Neither job uses a second motor or a servo-driven pitch mechanism.
How the hinged prop replaces the swashplate
In a helicopter, the swashplate applies cyclic pitch. Each blade's angle of attack rises and falls once per revolution, so the rotor makes more lift on one side of the disc than the other and the craft tilts. Doing that mechanically means tilting a plate under a spinning rotor and driving each blade through pitch links. That is a lot of small, precise parts for a palm-sized printed airframe.
Bird moves that work into the hub geometry and the motor controller. Tom's Hardware reports that each rotor blade sits on a 45-degree hinge that makes it pivot when the motor accelerates. As that happens, one blade pitches up to produce more lift while the other pitches down and produces less. By accelerating and decelerating the motor as needed, Bird can change the drone's roll and pitch, so a single spinning shaft gives directional control.
This only works if the control is extremely fast. The firmware has to command speed changes again and again while the rotor spins. Bird wrote custom firmware for both the flight controller and the motor controller, and Tom's Hardware reports that they let the motor react in under 0.2 ms. The repository links Bird's two forks: jyjblrd/ardupilot for flight control and jyjblrd/moteus for the motor controller. Bird changed existing open-source projects instead of writing a flight stack from scratch.
Yaw gets help from printed fins
One motor spinning one propeller also pushes the airframe the other way through reaction torque. Multirotors cancel that by running props in opposite directions, and helicopters use a tail rotor. Bird's drone uses neither. Tom's Hardware reports that yaw control comes from combining the main rotor torque, motor speed adjustments and 3D-printed stabilizing fins. That fits the repository's description of a "finned body."
Debugging with a home-made high-speed camera
The most useful lesson for printer owners may be the materials problem Bird hit. Tom's Hardware says most of the build video covers fixing issues. To see what the rotor was actually doing, Bird built his own high-speed camera from a Sony camera and a flash unit controlled by a microcontroller, which takes its signals from the prototype's motor controller and camera shutter. The rig showed that the drone's propellers were too soft. Bird replaced them with 3D-printed propellers in a sturdier filament.
The repository includes two propeller versions, v1 and v2, alongside the body and a prop plate. It also shows a rotor spinning on a test stand.
Where it flies, and where it doesn't yet
This is not a finished product. Tom's Hardware reports that Bird launched several test flights from his bed, many of which ended in a crashed drone and some wall damage as he tuned the programming. Flight is currently confined to his bedroom because the drone navigates with a fixed-base infrared beacon system. Bird says he eventually wants to add a linear camera so it could take off in other environments.
At the time we read it, the repository had six commits, 228 stars and 48 forks. That fork count suggests a fair number of people want their own copy of the files.
What It Means for Makers
For most readers, this is a design to study rather than a weekend kit, and the published files are the reason it is worth your time. The repository provides:
- Body as Fusion 360 archive (.f3z) and STEP
- Prop plate as Fusion 360 (.f3d) and STEP
- Propeller v1 and v2 as .f3d and STEP
- Test stand shown in a repository animation, a model of how to do bench work before free flight
Native Fusion files let you open the parametric models rather than only the final shapes. The STEP exports open in almost any parametric CAD package if you don't use Fusion. Comparing propeller v1 with v2 is a good exercise on its own, since a controller that depends on predictable blade behavior benefits from stiff blades, which is the problem Bird's high-speed camera exposed.
Three practical points stand out:
- Filament choice matters here. Bird's soft original props were a problem that only the high-speed camera made visible. If you print the propellers, treat material stiffness as a design parameter, not an afterthought.
- Start on a stand. The repository shows a rotor on a test stand. Getting rotor behavior right on a restrained rig is likely far cheaper than finding problems through crashes, and Bird's test flights did produce several.
- The firmware is half the design. Bird's approach relies on his modified ArduPilot and moteus forks and a motor that reacts in under 0.2 ms. Without that, a body and a hinged prop would likely have no way to steer.
Bottom Line
Bird's single-motor drone swaps mechanical complexity for timing precision. A 45-degree hinge and a fast motor controller do the work of a swashplate, and printed fins help replace a tail rotor. It flies only in one bedroom for now and has crashed along the way. Still, the core idea is demonstrated, and the CAD and firmware are public. For makers interested in control systems, rotor mechanics or simply what a desktop printer can produce, it is one of the more instructive open designs available.
