Make It Fly

Just sent my oscilloscope PCB for manufacturing, and it will take about a month, so I’m thinking about my next project now :D.

I’ve always wanted to build something that flies. That’s my childhood dream.

I think it’s about time.

Did quite a bit of reading over the weekend, end up deciding on a quadrocopter.

There are basically 3 types of flying machines –
1) Lighter-than-air (LTA) vehicles – things that use hot air or helium (or *gasp* explosive hydrogen) to float in air, and motors just for maneuvering. Like an airship.
2) Fixed wing aircrafts – airplanes. Aerodynamically stable (if engines fail, you can still glide it back home).
3) Rotary wing aircrafts – helicopters, tri-copters, quadrocopters. Aerodynamically unstable, and requires precise electronic control to maintain stability (if engines fail, it will enter free fall immediately… bloody!).

Airplanes would be too hard on the mechanical side, since I know nothing about aerodynamics.

I was debating between airship and rotary wings… airships are just not cool enough =P and they are heavily influenced by wind, which may suck.

And quadrocopter is apparently the easiest to build/control among all rotary wing crafts, so I’ll start with that.

In my dream, it has a range of about 1km, sends a live video feed, and allow remote control, as well as GPS based auto-pilot (auto return home and waypoints).

Major challenges –


For the thing to be stable, the rotors have to make sure there is no tilt on all 3 axises of rotation (roll, pitch, yaw). And to do that, it must know its own orientation, which is a lot more difficult than it sounds, requiring an inertial measurement unit (IMU), usually consisting of a 3 axis gyroscope and a 3 axis accelerometer, along with a few other things (GPS, magnetometer) to correct long term drift. This alone will cost >$100. Also very complicated algorithms and crazy math.


High frequency = high bandwidth and low range. Vice versa.
To stream video, I’ll need at least a few hundred kbit/s, which means it will probably have to be 2.4GHz.
That will limit my range to about 1km with line of sight, but I’m hoping that won’t be a problem…
Lower frequency radio can do a few kilometers, but at tiny fraction of the bandwidth.


I looked at some motors used by other people. They draw about 10A (yes! 10 AMPS!) each, for a total of ~40A.
That means I will probably have to ditch my favourite NiMH, and go lithium polymer. Both for the energy:weight ratio and output current.


They (the government) don’t want whatever you build to crash into a 747. Which sounds reasonable.

Transport Canada doesn’t seem to have any regulations for “model aircrafts” except it must be <35kg. I have already emailed them for clarification. I want this thing to be classified as a "model aircraft" because it's the only class that doesn't require any license/certification to fly. In the US, they also have a restriction of a maximum altitude of 100m. That would be fine with me, too. I'm not going to launch my chopper into low earth orbit or something =P. For comparison, commercial airliners fly at about 10km - 100 times higher. They also have a line of sight requirement, meaning you must be able to see the aircraft with naked eye at all times. And if it's autonomous, you must be able to do manual override to get it out of trouble any time. I'll get to work =D.

3 thoughts on “Make It Fly”

  1. Quadrocopters are easily the most stable form of lightweight helicopters.
    My dad’s friend owns a hobby shop, feel free to ask for quotes on any parts u need!

  2. Haha thanks! I’ve already ordered my stuff online.

    They certainly look easier than regular heli. No varying blade pitch to worry about.

  3. hmm, interesting. A friend of mine and I were planning on building something similar. He already has an IMU (minus gps) built for an arduino. (we are getting a pcb made with everything and gps).
    We also have most of the math figured out for the IMU. I haven’t even begun on the crazy PID control algorithm required to make it be stable though.

    radio communication: 1km while maintaining bandwidth is (imho) extremely optimistic. We might try 3G since the bandwidth is certainly good enough and range is not an issue.

    But good luck, how much are you spending on this thing?

Comments are closed.