Mini Analog Quadcopter

Who needs bits and bytes to fly a quadcopter.

Similar projects worth following
Had a bunch of broken mini helicopter and decided to make a quadcopter out of it. And why not make it powered by Op-amps (I had tubes of tiny TLC2254 lying around). So the analog quadcopter was born.

The basic idea is to perform a PID (or maybe just a PI or just an I) control loop in opamps. Then use the analog values to control a PWM generator to drive the motors. The final form is aimed to be 50x50mm or less. Powered by the original mini Li-Po pack from the helicopters, it should be entertaining for a few mins. 

As this is purely an exercise in analogue electronics, it makes sense to focus on the control loop. After much thought, it seems easiest to make independent drive modules that will maintain a certain parameter such as altitude or velocity. I hope to add a purely analogue FM controller if the project persists.

  • 2 × Mini helicopters There must be a few victims of careless flying burried in some cupboard or draw.
  • 20 × Op-Amps Quad packages are great for this.
  • 4 × power FETs anything that will sink more than 1A should do.
  • 1 × foam structure anything that will hold the motors.

  • Height Sensor Prototyped

    JLAM12/16/2014 at 22:36 0 comments

    Just a quick update, The sensors are prototyped but due to a few noise coupling and filter tuning issues, I don't have any conclusive results. The primary source of EMI was in the ribbon cable carrying the LED drive current which is parallel to the photodiode wires. As the current in the LED wires are orders of magnitudes greater than the photodiode wires, there is a lot of noise in the same frequency as the signal coupled in. Will look into shielding the wires.

  • Design update

    JLAM08/18/2014 at 14:03 0 comments

    The pendulum design was getting quite troublesome with need to balance the multiple motors, so I've decided to try creating individual modules that can maintain a constant height. Then hopefully by joining multiple modules I should have a flying object. 

    The module will use light reflectance to determine height as it is the smallest method I can think of. It will use an arbitrary frequency to drive an LED which is then sensed by a photo-transistor and passed through a bandpass filter to help reduce the effect of ambient light. This should provided the feedback signal to control the motor driver. 

  • Inital test of pendulum

    JLAM06/04/2014 at 21:46 0 comments

    Made a two propeller version to test the pendulum design in one axis. Found that the dampening was not enough and caused the device to oscillate. The next step is to use a spring instead of a free hanging pendulum.

  • Pendulums and hall sensor tested

    JLAM05/16/2014 at 13:25 0 comments

    Made a pendulum with a magnet attached to the bottom and tested the hall effect sensors. One thing to note is that by moving the magnet in the right direction I was able to get an increase and decrease of the motor speed with varying angles. Another thing I tried was dampening with a sheet of Al plating. This worked well even with a hole cut in for the hall sensor.

  • PWM stage tested.

    JLAM04/29/2014 at 11:25 0 comments

    So the idea behind the motor drive stage is a triangle wave generator with a op-amp acting as a comparator. The voltage reference signal will determine the PWM duty cycle. All this fits onto a quad opamp package. And to save weight I just soldered it dead bug style with SMDs. I also had dual N&P channel FET packages and had to drive one motor with a N and other with a P channel. It all worked out.

View all 5 project logs

Enjoy this project?



Shasdo wrote 09/13/2019 at 13:36 point

Don't quit unsterstand what you are trying to achieve with the "pendulum" but just in case, I'll leave this here:

  Are you sure? yes | no

Adam Fabio wrote 07/03/2014 at 05:25 point
Love this analog quadcopter, JAM! If you run into trouble with your spring material, give sheet mylar a try - It saved my bacon in a similar project years ago.
You're doing great so far - but keep the schematics, pictures, and amazing micro mechanical work coming!

  Are you sure? yes | no

Adam Fabio wrote 07/03/2014 at 05:26 point
And thanks for entering The Hackaday Prize!

  Are you sure? yes | no

JLAM wrote 07/03/2014 at 05:35 point
Thanks for the tip. I'll give it a go.
More updates will be coming soon.

  Are you sure? yes | no

frankstripod wrote 05/01/2014 at 01:25 point
I just posted my quadcopter project from my "wish list", then stumbled on this. Maybe my dreamcopter is not that far off. Thank you for the inspiration. Looking forward to updates!

  Are you sure? yes | no

dave.m.mcdonough wrote 04/30/2014 at 14:26 point
I like the pendulum idea.. at first i was thinking accelerometer on a chip, those have all 3 axis. But I don't know of any that wouldn't leave you with a load of digital to analog conversions to do.

  Are you sure? yes | no

JLAM wrote 05/02/2014 at 14:21 point
I happen to have a few old sparkfun sensors with analog output. That was the initial idea, but the challenge of going more old-school was too appealing.

  Are you sure? yes | no

Mike Szczys wrote 04/29/2014 at 15:18 point
I think you need an IMU in there to keep the thing from going nuts. Now if you want to go hard-core analog, try do this with mercury switches rather than solid-state gyro/accelerometers.

If you don't add an IMU, please still make a video as I'd like to see the behavior. Nice hack!

  Are you sure? yes | no

JLAM wrote 04/29/2014 at 22:20 point
Not sure about the mercury switch. They have too much inertia and too slow. But I was thinking about using 4 proximity sensor around the motors to maintain stable flight.

  Are you sure? yes | no

JLAM wrote 04/30/2014 at 08:42 point
After a bit of thought about not using a solid-state IMU, I have come up with a magnetic pendulum and hall sensor approach. I'll put up log about it when I build one. Thanks for suggesting a non-solid-state approach.

  Are you sure? yes | no

Mike Szczys wrote 04/30/2014 at 14:07 point
That pendulum idea is brilliant! If it works, you're a genius. If not, you need to try implementing the concept with a 2 wheel self balancing robot. This is going to be good!

  Are you sure? yes | no

Similar Projects

Does this project spark your interest?

Become a member to follow this project and never miss any updates