Building a Rocket Drone

For the last four years I’ve been building and flying rocket-propelled planes on the Project Air YouTube channel, but I wanted to try and build a rocket-powered drone... What could go wrong?

The first step was to build the drone that would become the test bed for this project.

For our first test, we needed to see how fast this drone would fly on electric power alone, so I fitted a small GPS device to the drone to gather data from the flight. With the stock drone, we reached 98kmh/60mph.

To test the same rocket engines we’d be using on the drone, we decided to try one out on a standard model rocket, to see how high it would go.

We needed to do some low-power experiments with the drone to see if the onboard flight computer would keep the aircraft stable as the rocket fired. I 3d printed a range of motor mounts for a range of different rocket motors, starting with some very small low thrust motors.

Instantly it was obvious the rocket engine firing straight through the centre of gravity was having little to no adverse effects, and the flight controller was easily able to keep the aircraft level.

Repeating the tests with some more aggressive flying showed the electric motors were easily able to compensate for any instabilities caused by the rocket engine.

We tried some slightly larger motors which had pulsating, irregular burns, but these were also kept stable by the flight controller.

The previous rocket engines had produced just a few hundred grams of thrust, but now we were going to see what would happen when we fitted a rocket engine with 12KG of thrust to the bottom of the drone, giving it a power to weight ratio of more than 12-1. The test was successful, the flight controller handled the influx of power very well and the drone was landed safely, ready for our planned improvements.

I decided the simplest idea would be to build an aerodynamic shell around the frame using model rocket parts which would cover the cameras, batteries, electronics, and of course the rocket motor, while some fins would help with stability at higher speeds.

First, I completely stripped the drone to it’s bare bones, removing the electronics and disassembling the frame to see what could be used for the new aircraft.

Now I could now set to work designing some new circular frame plates that would fit snugly inside the model rocket tube. Then, I reassembled the drone frame with the new 3D printed plates before reinstalling all of the electronics.

So, the next step was to modify the model rocket components, removing the base of the plastic nose cone and cutting the fuselage tube in half so it could slot either side of the drone core. I replaced the stock fins with upgraded ones I made from balsa wood.

I’d found with my previous project that gold was a good colour for these high speed vehicles as it stood out in the sky and also against the grassy fields of the test site, meaning it was easy to find in the case of a crash landing away from our base.

It was now time to add a battery mount and screw everything together to prepare for the first test hover with the aero shell.

Back in the workshop, I now needed to add the FPV camera and, importantly, remote rocket engine ignition system. I found the simplest solution was to reuse the system from the SR-71 Blackbird rocket plane, featured in a previous Project Air video, so I scavenged the parts I needed from that aircraft and then soldered them together with the drone electronics.

Finally, there remained the problem of where to stick the camera! - which was easier said than done.

My idea was to cut a hole in the nose cone that could be fitted with a transparent vacuum formed canopy from a model airplane kit. This would provide a highly streamlined bubble to fit the FPV camera in, facing at such an angle that I could fly the aircraft at an extreme angle of attack while seeing the ground and the horizon.

Now to the test field. After a successful launch, I could pull my goggles on and fly from the drone’s first’ person perspective - but immediately, I found something not right at all.

As I pushed the controls to go forwards, the drone went backwards.

All I could do now was try to make a crash landing in a nearby field.

Disaster.

Attempt 2

Now with the camera on the front of the drone rather than the rear, I was confident the reversed controls issue was solved. The aircraft was flying perfectly, and felt super stable.

As I’d anticipated, the large fins made it harder to change direction at high-speeds, but they were helping to keep the drone rock solid in a straight line, and I was relieved to find the powerful electric motors were able to rotate the drone normally as soon as the aircraft deaccelerated.

The improved aerodynamics allowed it to fly faster than it’s previous best of 60mph. The rocket drone reached a speed of 121kmh/75mph.

Test Flight 2

Now it was time for the second test flight with the goal to fire a rocket engine mid-air for the first time to go even faster than before. Just as on the previous tests, I took off line-of-sight and pulled my goggles down to fly in FPV mode, while Emma spotted for me from the ground.

The test flight was a success, achieving a speed of 190kmh/118mph!

So, we’d proved that a drone, flying at full throttle, could successfully be boosted to a higher top speed using a rocket engine.

If you enjoyed this article, check out our others on the website.