Hexacopter

February 14, 2010 9:45 PM

Presenting the hexacopter fantatsic miniature helicopter with multiple rotors. boy this thing can fly. has camera and gps. single link, wimp.com
posted by marienbad (113 comments total)

Fascinating device. I don't imagine it would be worth trying to scale up larger, though, because it would introduce far more points of failure as compared with a contemporary helicopter design. What happens if one of the rotors busts mid-flight, I wonder?

But still: Coool!
posted by barnacles at 9:47 PM on February 14, 2010 [1 favorite]

want
posted by brundlefly at 9:49 PM on February 14, 2010 [1 favorite]

Wow.

There's something unnerving about the bits where he just puts the controls down and watches it hover around.
posted by Artw at 9:50 PM on February 14, 2010

very cool.
posted by Ratio at 9:50 PM on February 14, 2010

I was prepared to be unimpressed. But it's pretty cool, in a sinister and creepy way. it reminded me of a large robotic insect.
posted by carter at 9:51 PM on February 14, 2010

Wow, that would be pretty great if it didn't sound like a swarm of angry bees. That makes me like it a LOT less.
posted by aubilenon at 9:56 PM on February 14, 2010

Actually, an intelligent large robotic insect. Which is worse, I think.
posted by carter at 9:58 PM on February 14, 2010

The URL at the end leads to this wiki, parts of which are in English. So you can build one yourself, although it looks like most of the schematics and instructions are for a 4 rotor design.

Here's the German wiki page for the 6 rotor design.
posted by jedicus at 10:01 PM on February 14, 2010 [4 favorites]

Its apparent complexity is reminiscent of the V-22 Osprey. Which also probably worked pretty well at a model scale.
posted by M.C. Lo-Carb! at 10:01 PM on February 14, 2010 [1 favorite]

It's as if the human species just wants to make it easy for SkyNet.
posted by Kattullus at 10:04 PM on February 14, 2010 [13 favorites]

I don't imagine it would be worth trying to scale up larger, though,

It wouldn't necessarily have to, if it was developed for UAV use.
posted by carter at 10:08 PM on February 14, 2010

It's nothing like the Osprey, since the rotors here are fixed in a vertical orientation. The tilt-rotor mechanism in the Osprey is what has been causing all the problems.
posted by tss at 10:10 PM on February 14, 2010 [2 favorites]

It wouldn't necessarily have to, if it was developed for UAV use.

Yep, it would make a great assination-bot. It's precise enough to be flow indoors and could blow up/eviscerate a target while they were in the shower.
posted by Burhanistan at 10:12 PM on February 14, 2010

(more re. osprey) Oh, but since your point was about the complexity... well, perhaps. Still, not needing to have crazy fuel linkages or dealing with flight modes somewhere between airplane and helicopter count for a lot. What I would worry about at scale is strange modes of vibration setting themselves up between all the rotors.
posted by tss at 10:14 PM on February 14, 2010

What happens if one of the rotors busts mid-flight, I wonder?

Obviously less of a problem then if the rotor busts in a regular hellicopter.

Its apparent complexity is reminiscent of the V-22 Osprey. Which also probably worked pretty well at a model scale.

Some things are complex, and some are not. Some things work and other things don't, but obviously it's possible to make complex things that also work. It just takes more effort.
posted by delmoi at 10:18 PM on February 14, 2010 [1 favorite]

"This is the new hexacopter. It has six propellers. Now we want to see how it flies."

It flies like a swarm of angry bees.
posted by smcameron at 10:20 PM on February 14, 2010

And LO when the Space Bees descend from the Heavens to bring woe woe WOE unto the inhabiters of the Earth, ye shall know them by THIS sound.
posted by WolfDaddy at 10:22 PM on February 14, 2010

I liked how when the coke bottle was swinging you could hear different sets of propellers going off at different times.

(Also, the Osprey's problems have been fixed (supposedly) and it's currently in full scale production. From Wikipedia:

<sub>The Osprey has provided support in Iraq, racking up some 2,000 flight hours over three months with a mission capable availability rate of 68.1% as of late-January 2008.[39] They are primarily used in Iraq's western Anbar province for routine cargo and troop movements, and also for riskier "aero-scout" missions. General David Petraeus, the top U.S. military commander in Iraq, used one to fly around Iraq on Christmas Day 2007 to visit troops.[40] Then-presidential candidate Barack Obama also flew in Ospreys during his high profile 2008 tour of Iraq.[41]

The only major problem has been obtaining the necessary spare parts to maintain the aircraft.[42] The V-22 had flown 3,000 sorties totaling 5,200 hours in Iraq as of July 2008.[43] USMC leadership expect to deploy MV-22s to Afghanistan in 2009.[42][44] General George J. Trautman, III praised the increased range of the V-22 over the legacy helicopters in Iraq and said that "it turned his battle space from the size of Texas into the size of Rhode Island."[45]
V-22 landing on the USS New York 19 October 2009
...
The MV-22 saw its first offensive combat mission, "Operation Cobra's Anger" on 4 December 2009. Ospreys assisted in inserting 1,000 Marines and 150 Afghan troops into the Now Zad Valley of Helmand Province in southern Afghanistan to disrupt communication and supply lines of the Taliban.[23][48] In January 2010 the MV-22 Osprey is being sent to Haiti as part of Operation Unified Response relief efforts after the earthquake there. This will be the first use the V-22 in a humanitarian mission.[49]</sub>)

And I'm not really sure this design is really that "complicated". Presumably all the arms and propellers have the same design, so if you get one right, you get 'em all right)
posted by delmoi at 10:27 PM on February 14, 2010 [1 favorite]

It looks like it would be really fun to fly. If you could scale it up to carry a good size video camera and it had some battery life, it could be a cool mobile camera platform.
posted by doctor_negative at 10:28 PM on February 14, 2010 [1 favorite]

I don't want one. I need one.
posted by maxwelton at 10:29 PM on February 14, 2010

Can it deliver pizza?
posted by cman at 10:31 PM on February 14, 2010 [2 favorites]

It looks like it would be really fun to fly. If you could scale it up to carry a good size video camera and it had some battery life, it could be a cool mobile camera platform.

In controlled spaces, you could just tether the sucker and have full A/C flowing into it. That would allow for less weight, more power, and you could then have your hexatron camera pod flying indefinitely. Obvious limitations would be a big cord whipping around, of course.
posted by Burhanistan at 10:32 PM on February 14, 2010

Throw some of these on the rotor blades to give a freaky image display at night, and you are SET to screw with some UFO spotters.
posted by FatherDagon at 10:37 PM on February 14, 2010 [1 favorite]

Mine would have a speaker. I would use it to greet missionaries and other unwanted visitors to maxwelton acres. It would also have a glowing red "eye" and some Styrofoam weaponry looking stuff glued on.
posted by maxwelton at 10:40 PM on February 14, 2010 [8 favorites]

wow wow wow wow WOW! When it takes off into the air.
posted by water bear at 10:43 PM on February 14, 2010 [1 favorite]

Get this man a government contract, stat. Do we have a Department of Sheer Awesomeness?
posted by Cool Papa Bell at 10:44 PM on February 14, 2010 [2 favorites]

Neat, but doesn't say what the advantages are over a quadcopter. Would be interesting to see what it could do in the hands of someone like this.
posted by Soupisgoodfood at 10:47 PM on February 14, 2010 [6 favorites]

Fucking Germans man... why are they so good at this shit? This thing is BADASS.
posted by nathancaswell at 10:53 PM on February 14, 2010

Would be interesting to see what it could do in the hands of someone like this.

OK, what the heck? How is that thing hovering upside down?
posted by Cool Papa Bell at 10:57 PM on February 14, 2010

Neat, but doesn't say what the advantages are over a quadcopter.

4 is 2 more than 6?
posted by nathancaswell at 10:57 PM on February 14, 2010 [5 favorites]

Wait fuck, I'm drunk... 6 is 2 more than 4? Fuck!!!
posted by nathancaswell at 10:57 PM on February 14, 2010 [7 favorites]

I'm waiting for the day when it'll be possible to slap a stabilized 5D Mk II on something like this. Most of the footage you see shot from helicams are generally aerial surveys - golf-course promotions and the like - but I think the real potential is in a small, stable platform that can manage a continuous over-the-shoulder shot of a mountain biker riding down a hill.

Also, I just finished watching The Collective, so that's where I'm coming from.
posted by awenner at 10:58 PM on February 14, 2010 [1 favorite]

OK, what the heck? How is that thing hovering upside down?

Reverse!
posted by nathancaswell at 10:58 PM on February 14, 2010

glad you all (mostly) liked it. i too thought it was awesome.
posted by marienbad at 10:59 PM on February 14, 2010

Yeah, the raw speed with which this thing accelerates upwards is just sick.

All it needs are lasers. Frickin' lasers.
posted by mosk at 11:01 PM on February 14, 2010

Reverse!

Yes, but if so, how does it instantly flip from full speed in one direction to full speed in another? Without, like, exploding? And staying 5 feet off the ground?
posted by Cool Papa Bell at 11:01 PM on February 14, 2010

4 is 2 more than 6?

This is why I am never going to design an insane flying machine :(
posted by nathancaswell at 11:01 PM on February 14, 2010 [3 favorites]

Or a minimalist approach. You need only have one blade: Maple seed monocopter.
posted by Kronos_to_Earth at 11:02 PM on February 14, 2010 [1 favorite]

If we can get them to fly quieter, I'd have the glowing red eye option and just program a dozen or so to fly in formation, about 5 feet behind people.

"Mommy! They're watching me!!!!"
posted by yeloson at 11:02 PM on February 14, 2010 [1 favorite]

Engineers + cheap microcontrollers + cheap GPS chips + R/C parts + free time = fucking awesome
posted by killdevil at 11:03 PM on February 14, 2010 [4 favorites]

Oh my God, I want this. Either this or the rover I've been daydreaming about making for years.

My rover will be like ¼ the size of the Sojourner. It will have a solar panel and a video camera, along with a Internet connection through 3G/GSM stuff. Next time I'm visiting my folks, I will drop the rover off near their house in the suburbs. Then I will go home. I will log into my remote control console, and I will drive it around.

I will drive it around at night, and Explore, and then when I'm done with it I will have to move him to a good hiding spot so he'll survive the next day without being stolen. He'll recharge his batteries while he sleeps at day, and every single night I will log in and drive him around. Eventually, he will slowly but surely make the 15 mile trek to return to me. It will be a long, perilous journey. There will be scary moments. He'll need to get onto I5, cross two bridges. Several miles of urban surroundings where there's people around at night. All the way from the top of downtown to where I live at the bottom of downtown. There will be close calls. But when he returns to me, and I have this amazing video and notes I took and I start posting it like a blog in real-time but delayed (obviously I must wait until after the trip is done before I can let people know, or they'll come take him), me and my robot friend will be rich INTERNET SENSATIONS.
posted by floam at 11:08 PM on February 14, 2010 [17 favorites]

The rate of ascent is crazy!
posted by sycophant at 11:09 PM on February 14, 2010

Also, this thing reminds me of the Lunar Landing Research Vehicle for some reason (although that had just a single large aircraft jet engine firing straight down, plus attitude thrusters).
posted by killdevil at 11:11 PM on February 14, 2010 [1 favorite]

Yes, but if so, how does it instantly flip from full speed in one direction to full speed in another?

The motor is always running at full speed, just like a normal heli. The pitch of the blades is all that changes. I think the hexacopter here uses variable speed rotors, with no pitch control, so it will never be as responsive as a normal RC heli can be, but it still seems to perform pretty well.
posted by Soupisgoodfood at 11:11 PM on February 14, 2010

I look at this carrying a 1 liter bottle of Coca Cola (1 Kg ) payload and all I can imagine is it carrying a 1 KG brick of C4. I understand this costs $1500 or so to build.

I am a bit worried.
posted by Poet_Lariat at 11:29 PM on February 14, 2010 [3 favorites]

"I look at this carrying a 1 liter bottle of Coca Cola (1 Kg ) payload and all I can imagine is it carrying a 1 KG brick of C4. I understand this costs $1500 or so to build.

I am a bit worried."

I've wondered why there haven't been more RC aircraft assasination attempts. Seems like a pretty obvious way to go. My guess is that the interesection of people smart enough to do it and the people that actually really want to do it is vanishingly small. With time, the set of people smart enough to do it is going to get larger though, because it's going to take less smarts, and so the overlap will eventually reach into the positive integers.
posted by smcameron at 11:43 PM on February 14, 2010

Why does it look so much more alive than normal one- or two-blade copters? Not just the speed.

Maybe the six rotors trigger some deep insect-reaction.
posted by gottabefunky at 11:47 PM on February 14, 2010 [1 favorite]

Yes, but if so, how does it instantly flip from full speed in one direction to full speed in another? Without, like, exploding? And staying 5 feet off the ground?

Propeller pitch. The good stuff is $120 per gallon.
posted by weapons-grade pandemonium at 11:49 PM on February 14, 2010 [1 favorite]

you can buy designs like this and larger variants already from a california company called draganfly. they were linked on the blue a while back.
posted by krautland at 11:58 PM on February 14, 2010

I don't imagine it would be worth trying to scale up larger, though, because it would introduce far more points of failure as compared with a contemporary helicopter design.

Not exactly... These schemes work with electric motors, and on small scales, because both factors help to make the rate of rotation of the rotors more responsive to control signals. For good responsiveness on huge turbine or piston powered rotors, you need pitch adjustable blades. Now I guess you could argue that it is complexity that keeps you from putting six adjustable pitch rotors on one vehicle, but..
posted by Chuckles at 12:24 AM on February 15, 2010

Yes, but if so, how does it instantly flip from full speed in one direction to full speed in another? Without, like, exploding? And staying 5 feet off the ground?

How do you think helicopters move in general? They do it by changing the angle of their blades. To fly upside down, all they have to do is change the angle even farther.

I look at this carrying a 1 liter bottle of Coca Cola (1 Kg ) payload and all I can imagine is it carrying a 1 KG brick of C4. I understand this costs $1500 or so to build.

I am a bit worried.

Worried about what? There are lots of RC aircraft you can get, many of which can carry payloads.
posted by delmoi at 12:50 AM on February 15, 2010

Would be interesting to see what it could do in the hands of someone like this.

Oh my god, it's been grabbed by an invisible 50-foot toddler! RUN!!!!
posted by teraflop at 1:22 AM on February 15, 2010 [1 favorite]

I've wondered why there haven't been more RC aircraft assasination attempts.

Radio jammers and machine guns make short work of most RC aircraft. AI still isn't good enough to do without the pilot for that sort of thing (fast, evasive flying) and by the time it is, the countermeasures will also be just as sophisticated. If your target is less protected, then you probably don't need to go to such extremes to begin with.
posted by Soupisgoodfood at 1:28 AM on February 15, 2010

I laughed when I saw that it had a camera embedded. I'm sure it just records, but if you could get it to live stream, that would be awesome.
posted by doozer_ex_machina at 2:13 AM on February 15, 2010

Neat, but doesn't say what the advantages are over a quadcopter.

The (German) wiki says the advantage is it can carry a heavier payload than the quadro version, but without needing larger props, which would increase risk of injury.
posted by uncle harold at 2:31 AM on February 15, 2010

Ok, without reading any technical info whatsoever, I'm going to assert that this device uses fixed pitch propellers. Furthermore, I claim that it can change the direction of prop rotation extremely quickly since it uses super lightweight propellers and electrically commutated brushless motors. This results in a very low inertia drivetrain with huge torque. Generating reverse thrust is not a problem.

(Basically, there's only one sensible way to make this device using off-the-shelf R/C parts. Variable-pitch rotors ain't it.)
posted by ryanrs at 2:49 AM on February 15, 2010

As I discovered to my astonishment this weekend, Liverpool police are now using UAVs just like this to fight crime.

The future is looking more and more like Half Life every day.
posted by Acey at 4:26 AM on February 15, 2010 [1 favorite]

For a few hundred euros more you can get the Oktokopter.
posted by LVdB at 5:08 AM on February 15, 2010 [1 favorite]

As I discovered to my astonishment this weekend, Liverpool police are now using UAVs just like this to fight crime.

(cough)bullshit(cough)

Considering how much people dislike speed cameras, I wonder how long it will be before there are dogfights between the police UAVs and those flown by "domestic terrorists".
posted by Enron Hubbard at 5:22 AM on February 15, 2010

So, this thing has GPS and a compass so that you can pin it in place and it stays there. That's great. But how quick is the 'wobble' on the GPS signal? If it's going to drift around in a 1m (or is it 5m?) cube over 24 hours that's fine (assuming the battery life is sane) but if it drifts over a 20 minute period, that's going to make for some interesting automated accidents.

Not that it matters, because I want this more than I want my car, computer and phone combined. Anyone want a second hand Android...?
posted by sodium lights the horizon at 5:23 AM on February 15, 2010

If it's going to drift around in a 1m (or is it 5m?) cube over 24 hours that's fine (assuming the battery life is sane)

Sorry to disappoint, but this thing's battery life is probably measured in minutes. Double digits if you're lucky.
posted by ryanrs at 5:45 AM on February 15, 2010

I kept expecting it to suddenly zoom forward and slice the guy's neck open, then come after the cameraman.

Also, the Osprey's problems have been fixed

Yeah, we've heard that before.
posted by mediareport at 6:08 AM on February 15, 2010 [1 favorite]

This is very cool, and a great experiment in applied control theory. I'm especially impressed with how the closed loop dynamics are resistant to the huge dynamical change when the liter of coke is attached, nice design.

But this thing is in some variant of hovering flight all the time, which uses an incredible amount of energy. A "normal" helo can cruise much like an aircraft by gaining forward speed and using the rotor disk for life sort of like a wing (the "sort of" requiring a lot more explanation, but variable pitch is needed). I don't see how you could do this easily/efficiently with this kind of design, so don't expect to see a human-scale one, or one of any size with more than a few minutes of range.

Yeah, I want one.
posted by Ella Fynoe at 6:09 AM on February 15, 2010

Much as ryanrs has already said (#$(#&$ preview)
posted by Ella Fynoe at 6:09 AM on February 15, 2010

I. Want. One.

NOW!
posted by wierdo at 6:11 AM on February 15, 2010

I have nothing to add, except that if you skip to 1:52 without seeing the first part of the video, because of where the helicopter is in the sky it ends up looking like this enormous futuristic space craft is landing right behind the dude with the controls. I got really freaked out for a second. When he put out his hand to touch the rails I was like, "Oh, thank God."
posted by en forme de poire at 6:50 AM on February 15, 2010

Aieee! I just flashed to a scene of several of these things chasing me through the woods, shooting laser beams at me!
posted by Ron Thanagar at 6:50 AM on February 15, 2010

Needs a bunch of tentacles and antennae hanging off the bottom.
posted by sneebler at 7:07 AM on February 15, 2010

Sorry to disappoint, but this thing's battery life is probably measured in minutes. Double digits if you're lucky.

Ah, no... I meant that if the GPS drift is slow enough that it takes 24h to wobble around inside a 1m cube, then any helicopter with a sensible battery life will have no problems, since the zero point won't have moved by more than a few centimetres during it's flight time.

If GPS drift is fast enough that the zero point could have moved by 5 meters in the 10 minutes the thing is aloft, it could make for some interesting problems when it tries to return to its starting point.
posted by sodium lights the horizon at 7:10 AM on February 15, 2010

The engineer in me thinks this is TOTALLY FUCKING RAD; the all-things-hexapod-fearer in me is expecting a hidden hatch to pop open and a score of facehugger paratroopers to burst forth squealing in hungry rage.

I'm especially impressed with how the closed loop dynamics are resistant to the huge dynamical change when the liter of coke is attached

I may or may not have gnashed my teeth in a jealous rage upon viewing that scene.
posted by elizardbits at 7:21 AM on February 15, 2010

Quadcopters are sexy, and I've been daydreaming about building one from Mikrokopter or other schematics for years. The design trades off mechanical complexity for software complexity. Rather than swashplates and tail-rotors (mechanical bits that are expensive to build and maintain, and prone to crash damage), solid-state rate gyros and accelerometers meter power to counter-rotating blades -- the motor's gearbox is the most complex moving part, and some models use direct drive. I wonder what benefits, aside from "because we can," the extra pair of rotors imparts. I guess the useful load increases somewhat, but the complexity of the control algorithms also scales up.
posted by Alterscape at 7:45 AM on February 15, 2010

It is failsafe up to a point; see this page for what happens when one of the rotors gets fouled. It's the third video down with the banner hanging from the copter.
posted by warbaby at 7:53 AM on February 15, 2010

Quaduino -- arduino based quadcopter project.
posted by stp123 at 7:57 AM on February 15, 2010

Sure, it looks and sounds scary, but all you need is the crowbar and some good timing.
posted by jquinby at 8:12 AM on February 15, 2010 [2 favorites]

So hexapodia is the key insight?
posted by ROU_Xenophobe at 8:23 AM on February 15, 2010 [1 favorite]

Couldn't get the feature video to work, but YouTube to the rescue.
posted by hangashore at 9:03 AM on February 15, 2010

Anyone want a second hand Android...?

Do second hand Androids dream of second hand sheep?
posted by cjorgensen at 9:10 AM on February 15, 2010 [2 favorites]

Anyone want a second hand Android...?

Let's call the reseller market for those "Second Handroid" and see if it catches on.
posted by Burhanistan at 9:16 AM on February 15, 2010

Wow, that would be pretty great if it didn't sound like a swarm of angry bees.

I'm not comfortable with the idea of this being silent.
posted by The Whelk at 9:17 AM on February 15, 2010 [1 favorite]

I was going to make a post about the Parrot AR.Drone once they finally get it priced and released. I've wanted a flying device that could record video for a long time. The AR.Drone allows you to switch from a down angle view to forward looking, but doesn't have the ability to record.

It also can't fly over a house. I have been amazed at the advances made by the dragan people mentioned above, but they remain out of my price range for one capable of carrying a deccent camera.

This thing is way cool. I can't imagine them being mass produced cheaply though (and by cheap I mean sub $1,000).

I thought it would have been so cool to have it have the ability to drop it's payload, but only because I wanted to see that Coke bottle plummet to the Earth and explode!
posted by cjorgensen at 9:25 AM on February 15, 2010 [1 favorite]

if the GPS drift is slow enough that it takes 24h to wobble around inside a 1m

A consumer GPS with WAAS will often get ±1m in ideal conditions. But the spec doesn't promise that kind of accuracy. Officially, WAAS only needs to be within ±7.6m, and only 95% of the time. Most of this error is due to varying ionospheric delays. The conditions change over the course of hours. Given the short battery life, you can assume zero drift but sort of crap initial accuracy.

It would be nice if several helicopters operating in the same airspace had very precise relative positioning. Then you could do neat synchronized acrobatics and such. Dense flocking would be very cool. But I suspect that is not possible. There are significant errors that will vary greatly among nearby vehicles (like multipath).

Of course, if you're willing to buy professional gear and set up a small portable base transmitter, centimeter precision can be yours. There are plenty of industrial GPS systems that can be hacked to guide a UAV (assuming you can't get your hands on a real UAV guidance system).
posted by ryanrs at 9:26 AM on February 15, 2010

I'm not comfortable with the idea of this being silent.

Production versions will sound like this.
posted by Burhanistan at 9:27 AM on February 15, 2010

So how long until Google combines a fleet of automatic hexacopters with Google Maps/Earth/Liquid Galaxy in a final bid for total world omniscience?
posted by Western Infidels at 10:14 AM on February 15, 2010

Ok, without reading any technical info whatsoever, I'm going to assert that this device uses fixed pitch propellers. Furthermore, I claim that it can change the direction of prop rotation extremely quickly since it uses super lightweight propellers and electrically commutated brushless motors. This results in a very low inertia drivetrain with huge torque. Generating reverse thrust is not a problem.

I doubt this thing reverses the direction of the props. That would still put a huge load on the motors. Both the motors and the speed controllers are usually designed to go only one way.


(Basically, there's only one sensible way to make this device using off-the-shelf R/C parts. Variable-pitch rotors ain't it.)

Give me 6 tail booms with tail rotor assemblies and I'll have a hexacopter for you with variable pitch props.
posted by Soupisgoodfood at 10:37 AM on February 15, 2010

I was going to make a post about the Parrot AR.Drone once they finally get it priced and released

Heh, me too. :)

The AR.Drone allows you to switch from a down angle view to forward looking, but doesn't have the ability to record.
As I understand it, the AR Drone generates its own wi-fi network and will have an open API. In which case, it should be a matter of getting domeone to quickly write an app for a regular OS that would grab and record the Drone's video stream.
posted by the cydonian at 10:56 AM on February 15, 2010

That would still put a huge load on the motors.

They can take it.


Both the motors and the speed controllers are usually designed to go only one way.

These aren't standard motors. They're brushless DC motors. They can be reversed by activating the phases in reverse sequence. This is purely a matter of timing, so it's essentially free, cost-wise.

That said, the helicopter control algorithm may not need reverse thrust. But if it did, it wouldn't be a problem.
posted by ryanrs at 10:58 AM on February 15, 2010

I had thought it must be gas powered, since it seemed to keep flying so long. Batteries weigh a lot! But I like electric, because I hate the sound of those little gas engines. But this thing lifts more weight than I expect, from my (disappointing) experience with toy RC aircraft.
posted by Goofyy at 11:05 AM on February 15, 2010

I thought it would have been so cool to have it have the ability to drop it's payload, but only because I wanted to see that Coke bottle plummet to the Earth and explode!

On the other hand if he'd put a hook under it and a loop on top of the coke bottle, millions of boomers would see their childhood Vertibird fantasies fulfilled.
posted by George_Spiggott at 11:11 AM on February 15, 2010

See, this is the kind of living-in-the-future shit I can get behind. Give me a version with an HD camera on the relatively cheap, and I'll give you an absolute boom in independent/ low budget/ porn movies suddenly doing really awesome crane style-shots.

Or, alternatively, make it chrome, add in some spinning razors and the desire to give chase, and you have the single greatest home security device ever invented.

Better yet, combine both these ideas and you have the next big reality show for Fox!
posted by quin at 11:20 AM on February 15, 2010

Checking out that Draganfly Krautland linked above, I'm wondering how these aircraft are performing when there is any wind. To head out on an extra clear day all loaded down with gear, then get stopped by ordinary amounts of wind, would get stale. Except when it works, of course.
posted by Goofyy at 12:00 PM on February 15, 2010

Looks like the Draganfly is like $15,000. Anybody have an idea what a hex-or-quadra-copter can be built for?
posted by fake at 12:11 PM on February 15, 2010

Germany looks so cool from the sky.
All neat and organized.

You'd never guess it was harboring a squadron of laser-wielding, sentient, R/C copters....
posted by madajb at 12:27 PM on February 15, 2010 [1 favorite]

Once they solve that damn energy density problem, we're all fucked

Hexabot is watching you masterbuate....
posted by PissOnYourParade at 12:31 PM on February 15, 2010

Add object avoidance, and you have the workings of a very cool autonomous eye-in-the-sky. The object avoidance doesn't have to be very sophisticated: if the copter is human-launched, it only needs to look out for moving objects, and safe landing zones below (in case of power failure).


Or a minimalist approach. You need only have one blade : Maple seed monocopter.
Kronos_to_Earth, now my head is filled with ideas of synchronized camera shots, so you can choose to transmit either a single-angle video with one frame per rotation, or a 360-deg view (perhaps limiting resolution) of the world (sync-stabilized, of course, so the screen display image on the ground didn't spin).


Sorry to disappoint, but this thing's battery life is probably measured in minutes. Double digits if you're lucky.
ryanrs, that video was over 11 minutes long, with no landings. Guaranteed minimum of double-digits.

Plus, a 1-kilo-plus payload makes for a helluva spare battery. Still probably not going to be aloft for several hours, but its current limitations (no pun intended) are much higher than you've guessed.
posted by IAmBroom at 1:34 PM on February 15, 2010

From the designer's website, it uses:
Motorization
* 8 * Robbe Roxxy 2827-34 [MOTOR]
* 8 * Propeller: 10" EPP
* Lipo 4S/3300mAh (flat make) [BATTERIES]

Data from a vendor site for the battery (apparently only one is used):
Hyperion G3 VX 14.8V 4S 3300mah 35C/65C Lipo
HP-LG335-3300-4S
Voltage (V) 14.8
Capacity (mAh) 3300
Dimensions L x W x H mm 138 x 46 x 32.5
Weight grams / oz 361g / 12.7oz

So, it could probably carry 3 extra batteries, at a payload of 1.083 kg. That gives it a fly-time of at least 33 minutes, including the huge bursts of energy used for the two "elevator mode" climbs. Hovering only - maybe 45 min to an hour? But even that is a bottom limit - we don't know how depleted the copter batteries were after 11 min of flight.
posted by IAmBroom at 1:45 PM on February 15, 2010

I wonder if one could use a gas motor to drive a generator to drive the props. Increased range, lower weight (increased load capacity), reduced complexity.
posted by five fresh fish at 1:46 PM on February 15, 2010

stp123 noted an Arduino version above - Quadrino. Roughly prices out at $500 for parts. Less with a little scrounging.
posted by warbaby at 2:16 PM on February 15, 2010

FatherDragon: Already done!
posted by Duug at 2:40 PM on February 15, 2010

I wonder if one could use a gas motor to drive a generator to drive the props. Increased range, lower weight (increased load capacity), reduced complexity.

Yeah, not that last one.
posted by ryanrs at 7:09 PM on February 15, 2010

Yes, that last one. Driving six rotors with a single motor would make for an incredibly complicated mechanical linkage, especially if the rotors are going to have independent speed control. Driving a generator is child's play by comparison.
posted by five fresh fish at 7:35 PM on February 15, 2010

I suppose you're saying it's more complicated than a battery-only design. Well, duh. But in terms of a design that could have a much, much longer air time, I think it might be on the right track.

Say, I wonder how difficult it would be to do that with six micro jet engines, instead of the rotors. The elevator test would be too quick to see...
posted by five fresh fish at 7:41 PM on February 15, 2010

Kronos_to_Earth, now my head is filled with ideas of synchronized camera shots, so you can choose to transmit either a single-angle video with one frame per rotation, or a 360-deg view (perhaps limiting resolution) of the world (sync-stabilized, of course, so the screen display image on the ground didn't spin).

IAmBroom, check out the video in this blog post by Jack Crossfire.
posted by jjwiseman at 7:49 PM on February 15, 2010

They can take it.

If you use big enough motors. And yeah, most brushless motors don't have set timing -- my mistake. The speed controller could still be a problem. If you have one with equal reverse power, I'm still not sure the rapid reversing all the time would be that good for it, although you could design one for that purpose, I guess.


Driving six rotors with a single motor would make for an incredibly complicated mechanical linkage, especially if the rotors are going to have independent speed control.

If you wanted to power it with an IC engine, you could link them all up and use variable pitch props. The mechanics wouldn't be that complex. You could just have the four rotors unless you really needed the advantage of the smaller props.


Say, I wonder how difficult it would be to do that with six micro jet engines, instead of the rotors. The elevator test would be too quick to see...

Jet engines would give you a higher top speed, but less acceleration and responsiveness than a propeller. They wouldn't have the same "bite". Would sound pretty neat, though.
posted by Soupisgoodfood at 10:05 PM on February 15, 2010

If you use big enough motors.

You keep insinuating the motors will need to be overbuilt to withstand reversing. I disagree. The only motor components that could possibly be affected are the bearings. But R/C aircraft have super low duty cycles and a propeller is a very soft, very gentle, very low-speed load. And since the motor is brushless, the bearings don't get covered in abrasive dust. This helicopter's motor bearings should be invincible.

Reversing won't affect the windings since they remain stationary. I suppose the torque reversals could damage a gearbox, but this design doesn't use one.


The speed controller could still be a problem. If you have one with equal reverse power, I'm still not sure the rapid reversing all the time would be that good for it, although you could design one for that purpose, I guess.

The speed controller uses the same drive transistors for forward and reverse. The six MOSFETs are arranged in three half H-bridges, one for each phase.


Sorry for not letting this go. But an outrunner brushless motor directly driving a prop is such a clean, elegant design: no wear parts, no gears, no linkages, lightweight, low speed, and super efficient. It is an engineering ideal. I demand you acknowledge its awesomeness.
posted by ryanrs at 8:05 AM on February 16, 2010 [1 favorite]

I suppose you're saying it's more complicated than a battery-only design.

Yeah. And R/C glow engines are disgusting little fuckers. Unreliable, noisy, temperamental, nasty plume of bluish smoke. Yuck.

Much nicer would be a small-scale turbine engine with an integrated alternator on the main shaft. You should be able to eliminate nearly all the plumbing if you use foil bearings and a gaseous fuel (butane?). No oil, no coolant, no pumps. You really want to get rid of all the liquids. Liquids suck.
posted by ryanrs at 8:50 AM on February 16, 2010

What's a "foil bearing"?

I think I remember prototypes of butane-powered batteries floating around the net a few years ago, about the same time all those lithium batteries were catching fire. That's be a super-sweet setup: no moving parts, with hours of powers.
posted by five fresh fish at 12:00 PM on February 16, 2010

Foil bearing.
posted by ryanrs at 12:13 PM on February 16, 2010

You keep insinuating the motors will need to be overbuilt to withstand reversing. I disagree. The only motor components that could possibly be affected are the bearings.

High loads = heat = damage to the magnets if they are overheated for long enough.


The speed controller uses the same drive transistors for forward and reverse. The six MOSFETs are arranged in three half H-bridges, one for each phase.

I didn't know that. All the older, bushed ESCs I've seen have separate MOSFETs for reverse, and all the modern brushless ESCs I've encountered have been of the forward-only variety. Although, I'm still not convinced that the sudden reversing won't put a huge load on them. But if the equipment can handle it and you have the power capacity of spare, I guess there's not a problem, depending on how big your props are, of course.


I demand you acknowledge its awesomeness.

I do. I also like to keep in mind limitations of each technology.


You should be able to eliminate nearly all the plumbing if you use foil bearings and a gaseous fuel (butane?). No oil, no coolant, no pumps. You really want to get rid of all the liquids. Liquids suck.

I'm not sure butane would work very well. I think it's what they use to start most model turbines these days, but not run them. Liquids are great. They have a high energy density and don't require a pressurised container.
posted by Soupisgoodfood at 3:19 PM on February 16, 2010

R/C jets can't start themselves. They need to be (1) spun up and (2) hot before they can run on kerosene. The first part is easy. The alternator winding can be used as a powerful starter motor to spin up the turbine. The second part, heating, is not necessary if you use a gaseous fuel.

I suggested butane because it is a common, clean burning gas that can be easily stored as a liquid under pressure. A liquid butane tank can be pretty lightweight since it doesn't hold very high pressures. At 50°C, the pressure in the fuel tank will be about 60 psi. An aluminum soda can would be uncomfortably thin, but something like a spray paint can would be fine.

Although liquid butane has a lower energy density than kerosene, it's specific energy is slightly higher:

Energy density:
34.5 MJ/liter kerosene
27.5 MJ/liter l.butane

Specific energy:
43.1 MJ/kg kerosene
45.8 MJ/kg l.butane

So butane wins on weight, but not on volume (although it is quite close). In practice, butane's small weight savings will be wiped out by its metal tank, but the overall difference in vehicle weight will be very small.


Other design questions:
1) How efficient is the engine at part load?
2) How much useful thrust from the exhaust?
3) Does slewing the shaft speed spin the vehicle? Props might not be able to counter the rotation.
4) Since we're dreaming, it would be cool to have thrust vectoring.
posted by ryanrs at 7:16 PM on February 16, 2010

Outdoor camping stores stock a wide variety of LPG/butane/etc cartridges; as does the "beauty aids" section with its cartridges for hair curlers. The cartridge source, connectors, and weight is a solved problem.

Using a gas turbine design, powered by burning LPG/etc is a kickass idea. I'll bet that between electronic ignition and a simple regulator, it can even be so sophisticated as to not produce (much) more electricity than is needed by the entire 'copter. The batteries could be much smaller, as they'd serve only as a "buffer" against the generator cycling.

Those foil bearings are a brilliant idea. Spiffy stuff abounds in the flight world.
posted by five fresh fish at 7:58 PM on February 16, 2010

R/C jets can't start themselves.

They can. FADEC is pretty common now. No more sitting at the field with a tank of propane and a leaf-blower. The turbines have a starter motor and a small butane or propane tank, all hooked up to a computer, everything on-board. All you have to do it throttle up.


Although liquid butane has a lower energy density than kerosene, it's specific energy is slightly higher:

I don't know the details, but I don't think it's as simple as that. Otherwise you'd get better performance with it in cars. And surely someone would be using it if it was good enough for RC aircraft. I suspect it might have something to do with vaporisation and fuel to air ratios, perhaps.


1) How efficient is the engine at part load?
2) How much useful thrust from the exhaust?

I know that RC jets planes (not turbofans) can go pretty fast, but they go through a lot of fuel. For use in helis, I think they often run a similar turbine just at a lower RPM, and the performance is still pretty good while giving you more flight time than an equivalent electric (at least last time I checked).


Since we're dreaming, it would be cool to have thrust vectoring.

Quite a few electric planes with partial thrust vectoring (usually just the elevator/pitch). There is a turbine with trust vectoring, but the video doesn't have much about the details.
posted by Soupisgoodfood at 10:12 PM on February 16, 2010

Capstone MicroTurbine makes that thing I was talking about. The turbine generator I described is a shipping commercial product. They're mostly used as emergency generators for large commercial sites. But they also have a smaller 30kW version for hybrid-electric buses. You'd have to scale it down for your UAV.


For use in helis, I think they often run a similar turbine just at a lower RPM

Generally they take the jet engine and pipe the exhaust through a second turbine. The second turbine is on a separate shaft and turns at a slower speed. Its turbine blades have a different geometry so it spins slower than the main turbine, but generates more torque.

By decoupling the main turbine and the second turbine, you can optimize the shaft speeds of the compressor and gearbox. The compressor needs high rpm, so it gets connected to the main turbine (shared shaft). The gearbox wants a lower rpm to minimize friction losses and reduce vibration, so it pulls power of the secondary turbine.

The two turbine system also improves responsiveness, I think. When more power is needed, the compressor shaft can quickly spool up to peak power while the second turbine lags behind. With a single turbine, the whole drivetrain needs to be accelerated just to get the engine into its power region.

But none of that matters if your load is an alternator. If you need to spool up the turbine quickly, the electronic controller just stops pulling current for a moment, thereby diverting all power to accelerating the shaft. It could even use a burst of battery power to spool up the shaft electrically. Like the alternator in a Prius, it can move energy in either direction.

Damn, that would be cool. I want to hear a microturbine spool up from 80k to 200k RPM in 500 ms. That should be possible. The alternator is designed to harness the jet's full power. So it should be able to dump that much energy back into the shaft. In fact, since this is only for a short burst, you could probably overdrive the alternator 3:1. When you add that electric power to the jet power, you should be able to cut spool up times by at least 75% (an underestimate because electric power is constant whereas jet power is much lower at beginning of spool up).
posted by ryanrs at 11:09 PM on February 16, 2010

(Air) Robot Makes “Arrest”
posted by Artw at 11:42 AM on February 17, 2010