h a l f b a k e r yWarm and Fussy
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
So, the single bladed propeller is ~ 15% more efficient than the two bladed propeller. It swings 'round without the bother of an additional blade creating messy vorticies ahead of it. Fitting single bladed props to a quadcopter drone has been done, a simple google will demonstrate this. Single bladed
props deployed on a multicopter drone already have advantages. They will pack away nicely, by folding the only blade inward and there's half the chance of damage.
Now, what if we push it further? Moving the motors inward. The whole machine becomes much more compact, but the propellers are now within range of interference with one another, the rotors COULD be separated vertically, but that's boring, and they're operating in downwash. The precise control afforded by brushless motors means that with adequate computational power (which is light, cheap and small nowadays) even though control inputs demand different speeds from each rotor, the whole set of four (6-8) can be throttled independently, while subtle adjustments ensure they never run into one another. Now, the whole thing can be smaller. Combine the single bladed propeller with the bs0 intrinsically unbalanced brushless motor <link> and you save weight.
Intrinsically Unbalanced Brushless Motor
http://www.halfbakery.com/user/bs0u0155
[bs0u0155, Jan 05 2016]
assumptions busted
https://www.youtube...watch?v=CmhwROMBK8E
[pashute, Jan 07 2016]
Underactuated swashplateless cyclic pitch helicopter rotor
https://www.youtube...watch?v=aEPf0QHVuMM
As mentioned in my anno. [notexactly, Jan 11 2016]
[link]
|
| |
By moving them in, you lose the control leverage and you also lose the lack of turbulent interference. |
|
| |
How are they throttled independently yet manage to not
contact eachother if operating in radial interference? |
|
| |
//How are they throttled independently yet
manage to not contact eachother if operating in radial
interference?// |
|
| |
Well, brushless motors, or stepper motors as they
used to be known, can relay exactly where they are in
their arc. The motor controller can also command
exactly the degree and speed of rotation (given
enough perimeter magnets). So, say rotor 1 needs
1000 rpm and rotor 2 needs 1200 rpm, the motor
controller just needs to organize rotor 2 so that it is in
front of rotor 1 at the beginning of the interference
zone. All you need to do is speed up one rotor and
slow the other for PART of its arc and you can get the
order right going into the interference zone. Then the
motor controller has the other, say 300 degrees, to
modify the rotor speed so that the AVERAGE rpm
matches the commands for each rotor. |
|
| |
If one rotor needs to average 1200 rpm, then it can
pulse to 1300 for 5 degrees or so to ensure clearence,
then slow down to 1195 rpm for the rest. With 4 rotors
all sharing the same space, it's a MUCH harder
problem. But it's just an on-line computational
problem that's right up a microprocessor's alley. |
|
| |
//By moving them in, you lose the control
leverage and you also lose the lack of turbulent
interference.// |
|
| |
An empty Boeing has way more control leverage than,
say Concorde of the same weight, but keeping the
mass centralized gives better roll rates, for example.
Anyhow, why don't we make the booms telescopic,
there, now it can be a regular single bladed prop
drone, then tuck in for that small window. |
|
| |
//single bladed...more efficient |
|
| |
So, remove the single blade and get maximum
efficiency. |
|
| |
//All you need to do is speed up one rotor and slow
the other for PART of its arc// |
|
| |
But you're doing that several hundred times per
second (or at least several tens of times per second).
That's going to put a much greater load on the
motors, and therefore increase power consumption,
no? Even with energy recovery, I think you'd still
have problems. |
|
| |
Ok, so it looks like RPM is in the 10-50,000 RPM range. |
|
| |
So, 50k is ~ 600/second. Make it easy, lets have 3 pairs of
magnets for 6 steps. That's 3600 switches per second.
This is ~250uS for computation and switching time. Now,
power FETs switch in the 2-5us range, so no problem
there. CPU clock speeds mean nano seconds per cycle, so
no problem there. |
|
| |
//That's going to put a much greater load on the motors,
and therefore increase power consumption, no?// |
|
| |
err... working that out is above my pay grade. |
|
| |
And it's below mine. Who can we ask? |
|
| |
//tuck in for that small window// good point. All the blades could be briefly stopped in the inward-pointing position for ballistic entry of even smaller windows. |
|
| |
Or rotate through perhaps 30º, then feathering for the return. |
|
| |
//That's going to put a much greater load on the motors,
and therefore increase power consumption, no?// |
|
| |
Yes, but maybe even worse is that the motor will need to
be much more powerful (larger and heavier). Normally
the prop is accelerated to full speed over many
revolutions. To have the slightest hope of this working,
you'd need your prop to be super light weight, but it will
also need to be stronger than a normal prop to handle
the rapid acceleration. |
|
| |
//rotate through perhaps 30º, then feathering for the
return.// |
|
| |
Hmm, if you had sophisticated blade angle control, hard
to do with something weighing a few grams and spinning
that fast, but you briefly change the angle of two flexible
blades. One gets a coarser pitch and flexes up, the other
briefly reverses and flexes down, missing each other. You
could do this mechanically with a cam, to occur at the
same place in every rotation... that is over the drone
body where any prop downdraft is wasted anyway. |
|
| |
// Or rotate through perhaps 30º, then feathering for
the
return. // |
|
| |
That's an oscillatory motion that's not coupled to a rotary
or
other continuous motion (apart from the overall motion
of
air), meaning high power consumption. |
|
| |
The oscillatory component of [bs0]'s idea will still result
in higher power consumption, but not much higher than a
constant-speed rotor, because most of the thrust comes
from the constant-speed component. |
|
| |
[scad mientist], this oscillatory rotary motion
superimposed on continuous rotary motion has been
done, for a slightly different helicopter rotor
purpose: [link]. Acceleration from zero to full speed
within one rotation shouldn't be necessary; just sync the
blades the same as during full-speed operation, right? |
|
| |
[notexactly] That MAV is pretty cool, I wonder if they could
do the same thing with single bladed props and an increase
in RPM to compensate for thrust loss? Folding the blades
back would make it extremely small.. for [pocmloc]'s
ballistic entry point. Would be an extremely capable way of
getting a grenade exactly where someone doesn't want it. |
|
| |