1
$\begingroup$

I was flying quads today and it just suddenly popped up into my mind — do the propellers all have to be the same size in quadcopters? Mine are, but What if two propellers were larger and the other two smaller? The drone would still be symmetrical, but would it have any effect on the performance of the drone? Could you explain the physics behind your answer?

$\endgroup$
  • $\begingroup$ The control loop in the craft basically thinks "I'm changing attitude in such a way that I need to increase the thrust from motor #3" (for example), and it will keep increasing it until either the craft is under control, or the limit of the motor is reached (and will probably result in a crash). It may also decrease other motors to help compensate, but lift is limited. Note, however, if the control loop was tuned with one set of props, and you switched two to a different prop, the tuning will probably be off (it won't fly as well). $\endgroup$ – Steve Oct 22 '18 at 21:38
6
$\begingroup$

The main point of a quadcopter is that it can be controlled by engine power using much simpler fixed-pitch rotors¹. Assuming this construction, you need to be able to separately adjust power of

  • forward and rear set of propellers for pitch control,
  • left and right set of propellers for roll control and
  • clockwise and counterclockwise-rotating set of propeller for yaw control.

If you had two bigger and two smaller rotors, then one of the combinations would have the two bigger ones in one of the sets and the two smaller ones in the other. That would make it really, really hard to balance all the parameters to maintain appropriate controlability.

I think it might not be completely impossible, but it certainly makes no sense.


¹ Of course once you add cyclic and collective control, you can fly with one rotor only (well, two, or one with anti-torque rotor or jet) as most full-scale helicopters do, but then you've lost the simplicity of fixed propellers.

$\endgroup$
  • $\begingroup$ In addition to cyclic and collective control the full size helicopters need a means to cancel the rotation of the body. Most commonly a tail rotor. $\endgroup$ – Peter Green Oct 22 '18 at 15:16
  • $\begingroup$ @PeterGreen, single-rotor ones do. Twin-rotor ones don't, because the two rotors cancel each other—as long as they can have differential cyclic (for tandem rotors) or differential collective (for coaxial rotors) applied. $\endgroup$ – Jan Hudec Oct 22 '18 at 16:22
2
$\begingroup$

No, they don't have to be the same size. You could even fly with just two rotors - the Chinook does - the other pair would be needed for control only.

But having them all comparable makes it easier to design the quadcopter and keeps it generally more maneuverable.

$\endgroup$
  • 1
    $\begingroup$ Note however that Chinook has cyclic and collective controls on both rotors, which is something that quadcopters generally omit. $\endgroup$ – Jan Hudec Oct 22 '18 at 5:26
  • $\begingroup$ … and I think this is actually significant. You would have problems with yaw control. $\endgroup$ – Jan Hudec Oct 22 '18 at 5:32
  • $\begingroup$ It‘s also no longer really a quad copter with just rotors $\endgroup$ – JustSid Oct 22 '18 at 6:19
  • 1
    $\begingroup$ Yes. But the principle of a quadcopter is varying thrust only. While you need it to be symmetrical about the yaw axis for that, it doesn't have to have quadrilateral symmetry - it's just easier and a lot more practical that way. $\endgroup$ – Therac Oct 23 '18 at 6:04
0
$\begingroup$

No, you can fly a quadcopter using different rotor sizes. The smaller rotors would have to have a longer moment arm relative to the Centre of Gravity, and as @JanHudec points out the total rotational impulse in yaw direction must zero out as well.

So two pairs of counter-rotating rotors, correctly placed relative to CoG, would work.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.