Having been one of many who had the idea that bigger was more stable and prop. lift efficiency was the key factor, The penny finally dropped and I realised that the FC computer determines the level of stability and all of the other factors just cloud the issue by being less than ideal. I made my reluctant u turn and never looked back. I came across this video which pretty much confirms my most recent thoughts. What you don't want in a successful design is Heavy motors with heavy props. long arms, and a lot of global inertia. What you do want is fast response......https://www.youtube.com/watch?v=R3NLWxJ1N5U
An interesting combination with amazing results.
Replies
Some valid points there but at the end of the day. You cant compare the bandwidth response from a variable pitch constant speed system with a variable speed motor driving a fixed pitch propellor. I do think, however that the norm with multirotor will eventually become a variable pitch control function. better response and autorotation being the key issues.
As things are at the present, the heavier the motor and the prop. the more latency in the response due to the inertia of starting and STOPPING the corrective motion.
> Larger heavier motors that turn large props have more delay in resolving any given output from the controller.
Large, heavier motors develop more power, so are driven by a combination of higher voltages/currents. Why would they be slower to respond ? Not saying you are wrong, but I am not finding any theoretical back-up for this, (which does not mean there isn't any, of course) given the relationships between mechanical torque, prop speed, voltage and current, and both mechanical and electrical typical efficiencies.
Just guessing, but I bet that pilot could fly a much larger quad just the same, as long as said larger quad had the right combination of props and motors, with comparable electrical power efficiency. It just would cost him more ...
Speaking of larger props, and although it's not exactly comparable, how is that for response? https://www.youtube.com/watch?v=p8t41avFuCc. 23" blades!
Also, the larger the copter, the smaller a fixed given payload (camera) weight is with respect to the overall weight. This would seem to actually favor a "bigger is better" view for AP.
If what you want is super acrobatic RC performance, just strap a three-axis gyro to a quad and you'll get this. It's a wonderful thing, and is pretty much what a KK board and MultiWii is. Great fun. The code is crazy simple (it's literally a few hundred lines), but you'll have to fly it yourself the whole time.
But if you want autonomy (a multicopter that will fly by itself for a whole mission), you'll have to move to a whole different level of control laws and math. It really is a trade-off: super-fast manual control or highly processed sensor fusion for autonomous control. There are probably some ways to get both, but in general systems are tuned for one or the other.
Well, the goal of Arducopter is not to be able to duplicate this. As Jan said this copter was tuned for acrobatics. It's basically like balancing on a marble.
You're right though, the FC has a lot to do with response. On my new minicopter, I actually dialed down all the P stability values to get less response, since I was flying indoors.
Amazing. Arducopter developers have their work cut out for them.