hey again all! I thought it would be cool to have a dedicated thread for guys who build bigger hex and octocopters with apm where we can share tips and advice regarding tech specs, builds, code, firmware and whatever else comes up.
Firstly, how many of you are building bigger AP rigs with APM? We've been building all our commercial SteadiDrone RTF kits with APM with great results.
so, who else is out there, lets get chatting.
Yeah, I always flew APC on my planes, never a problem. But the APC props I got with my 3DRQ don't seem to be the same. They are not stiff at all.
So exactly which APC do you use?
And what about Graupner E props? They certainly look the business.
we use the APC SF 12 x 3.8 and 14 x 4.5 SF
APC EP's are great on planes (after balancing), but in my opinion SF's (Slow Fly) are to flexible for a heavy lift type of copter. After all they are meant to be used on light weight foam planes. I would think you get a lot of blade warping (brrr, brr sounds) when you punch the throttle or lift heavy payloads?
I agree. When I first got them, I thought wow, these are flexible!
What we need is a SF profile, but with stiffer blades. As Brad has said many times, propeller development has seriously lagged in this industry.
Well, I am very happy with these ones.
Here is a test flight I did yesterday in a strong 10m/s (22mph) wind with 14m/s (31mph) gusts.
That looks really good. The only problem with those is, 15" is a bit bigger than most want. 12" would be nice.
Interestingly, I just watched that video about National Geographic's camera systems they use. They have an Octo they've spent a lot of time and money on. But, they showed a video clip where they lost a motor and... it crashed. I wonder what their system specs are? Which FC? Etc.
To me, if your Octo is going to crash anyway, you might as well just build a quad with BIG motors, so you have half as many failure points.
More thrust units gives you lighter disk loading (higher aerodynamic efficiency) and less rotational inertia to fight for control response. Since the inertia of a rotating disk is equal to the square of the radius, the only way to keep the disk loading down and maintain a suitably brisk control is to add more thrusters.
This is mitigated somewhat by the inertial airframe moments added by the motors out on the periphery of the disk area, but there is a net gain.
That, gentlemen, is the great challenge facing EMC designers - the tradeoff(s) between flight endurance, control response, and added complexity (or is it redundancy?).
Brad, you seem to be well versed in flight physics. I have a question for you. If you look at a multicopter, you obviously have center of gravity (CoG). But when the copter is flying and/or hovering you also have a center of rotation (CoR).
My question is: Are CoG and CoR the same, or does the multiple trust vectors (propellers) make them differ?
CoG and CoR are not the same; in fact, they're not even related. The center of gravity is the point on the copter in which gravitational force is in equilibrium. In other words, from that point in all directions, the pull of gravity is the same.
The center of rotation has everything to do with applied torque forces, and in the case of the electric multi-copter, refers to the sum of all forces which cause the copter to change its orientation along the pitch and roll axis. If, for example, the left prop produces more thrust in response to a stick movement to pitch right, and all other thrusts remain the same, then the CoR will be at the right-most thrust vector. Here's a good explanation of the basics with pictures:
Somehow, though, I get the feeling that you asked the wrong question. EMC designers prefer to concentrate the mass near the center of the ship because they expect most of the control CoRs to be near there. However, if you look at the clips of most model EMCs flying, rarely do they swivel about on the pitch and roll axes as a fixed wing airplane might. That's because of the thrust vectoring involved; as the higher and lower props are not entirely used to create lift any longer, the speed of all of the props has to go up a bit, including the lower (right) one, in order for the vertical vector (lift) to remain constant. This means that as a practical matter, the CoR is always quite far-flung from the CoG.
Why do you ask?