Personal Project CFD 312 km/h

300+ km/h Quadcopter Speed Drone

A 3D-printed quadcopter designed for top speed and easy production from stock 5" parts. I took it from CAD and airflow simulation to a hand-built aircraft that flew 312 km/h on its third flight.

Hero video or photo of the drone in flight.

312 km/hGPS top speed, third flight

880 gAll-up weight, printed in PLA

9.1Peak lift-to-drag ratio, at 5°

3.4 kWInstalled power, of which 1.0 kW holds 300 km/h

Skills

SolidWorks
OpenFOAM (CFD)
ParaView
Aerodynamic analysis
3D printing (PLA)
Electronics integration
Soldering
ArduPilot
PID tuning
Flight testing

The goal

I wanted to see how fast a quadcopter could go if every part of it was designed around speed, while still being simple to build: a printed airframe, off-the-shelf 5" components and nothing exotic. I also built in GPS so the same airframe can fly autonomously later.

Designing the airframe

I built the airframe in SolidWorks around the electronics and the camera stack, so the body is only as big as what it has to carry. The print includes a mount for the GPS, internal supports, scoops that feed cooling air to the electronics, a camera cutout, motor end caps, tail fins and an internal structure for the flight-controller stack.

Labelled CAD render of the airframe.

Simulating the airflow

I solved the external flow in OpenFOAM across angles of attack from −4° to 13° and speeds from 10 to 100 m/s, then inspected pressure and flow separation in ParaView. I reduced the results to a drag polar, a pitch-stability slope and an efficiency curve, compared four fuselage shapes, and checked the power each one needed against what the motors could deliver.

Drag polar with parabolic fit.

Lift-to-drag against angle of attack, peaking at 9.1 at 5°.

What the simulations showed:

Drag polar: C_D0 = 0.029, k = 0.122.
The drone is passively pitch-stable, with a moment slope of −0.30.
It departs controlled flight at 12° (C_L,max = 0.865), which sets the limits for the flight controller.
The camera cutout alone raises the drag coefficient from 0.109 to 0.172.
A longer body separates earlier at the tail junction.
Holding 300 km/h takes about 1.0 kW against 3.4 kW installed, so the top speed is limited by propeller pitch, not drag.

Surface pressure contours.

Streamlines at 0°: the longer body separates earlier at the tail.

Building and flying it

I assembled the drone with M3 screws and nuts, did all of the soldering myself and tuned the PID loops in ArduPilot. On its maiden flight it reached a GPS-measured 172 km/h at only 48% throttle. By the third flight it hit 312 km/h.

Maiden flight: 172 km/h at 48% throttle.

Electronics integrated into the final print.

Every previous iteration of the airframe.

What I learned

Write a few sentences here in your own words: what surprised you, what you would design differently, and what the project taught you about turning simulation results into real-world performance.

What's next

A maximum-thrust test once the new propellers arrive.
Autonomous flight using the onboard GPS.