Hacker News Reader

There is some confusion in this particular case stemming from the difference between altitude above sea level and above ground level.

The UAV in question was spotted at ~10,000ft above sea level, but it was only about 800-900ft above the local ground level at the time.

There is no confusion. I assumed 10,000 ft above sea level. Starting at 9,000 ft doesn't help much. The air is quite thin at 10,000 ft; many small general aviation aircraft have difficulty reaching that altitude. Here is a YT video showing a few minutes of someone's drone flight allegedly at that altitude. It doesn't appear to me that sustained flight for any reasonable duration is possible with that particular drone, nor do I expect any/many other drones to perform much better.

https://www.youtube.com/watch?v=WnnlAS5Zzfg

Apologies, your phrasing "make it to that altitude" made me think otherwise!

Multi-rotors using BLDC motors cannot really be compared to small general aviation aircraft with infernal combustion engines.

"Reasonable duration" is a nebulous phrase. Some of the airframe configurations we use here couldn't manage more than about 10 minutes at 100ft, whereas others could quite happily carry a gopro at 10,000ft for 10 minutes (assuming it didn't have to climb from sea level!)

re: my phrasing, I did assume that the drone pilot in question started from a much lower (unknown to me) altitude.

>Multi-rotors using BLDC motors cannot really be

It's got nothing to do with the type of motor or fuel.

>"Reasonably duration" is a nebulous phrase. So

So is "quite happily carry". I'll admit that it appears that it's not impossible for such a flight, but I don't agree that any drone is "happily carrying" anything at that altitude.

Well I gave a payload (gopro, 350-400grams with gimbal) and a duration (10 minutes), I'm not sure what else to say! The aircraft was completely controllable with no memorable handling differences apart from a very slightly reduced response to high thrust demands, but as we fly our platforms for stability rather than speed it wasn't something that we really noticed.

The thing is, at 10,000' you have to move a lot more air in order to maintain the thrust necessary to stay at altitude. This additional RPM requires more current and that cuts into your battery life. It would be an interesting comparison to run a number of drones at 10K' to see how they held up.

One of my colleagues has done some investigations into the effect of density on multirotor performance before.

I don't have time to run the model for this specific case, but from a similar case with a lower air density of 0.71kg/m^2 (40c at 4km altitude) a simple quad uses ~20% more power in a stationary hover than at sea level.

Interestingly at high flight speeds (15-20m/s) you could actually start to see power savings at this altitude, as the parasitic drag of the vehicle body is reduced.

There was a post on reddit/r/multicopter a while ago of prop thrust at different air pressures with the same current. It's pretty much exactly what you're asking for.

https://i.imgur.com/Z5toIVm.jpg

Assuming air pressure at 10000 ft is ~69kpa, and sea level is ~100kpa, there really doesnt appear to be much of a difference, except for medium-sized props, which had very interesting thrust outputs- gaining efficiency at low rpm and loosing a lot at high rpm.

Anyway, point being is a 250 class quad could pretty easily fly at 10000 feet. Not particularly well, but considering 1:4 (or even 1:8) thrust ratios arent uncommon with 250 class quads they could still pull it off.

Now, idk about bigger quads like the phantom/iris, their props have quite different geometries which probably act quite differently.

At 10000ft, pressure is around 30kpa. Using the data you provided in the link and assuming a quad weighs any were around 600-800grams with camera, the multi rotor will be flighting at near 100% power. Sustained flight time would be very limited.

> At 10000ft, pressure is around 30kpa.

Do you mind if I ask where you are getting that pressure value from?

The ICAO Standard Atmosphere gives a value of 70kPa for 10,000ft.

>lot more air in order to maintain the thrust necessary to stay at altitude.

Not really... you need to move the same MASS of air, just more air VOLUME. Bigger props solve that problem, at the expense of moving a large prop (greater moment of inertia, slower acceleration).

Sustaining at higher altitudes requires a higher prop speed, but a lower prop torque. The difference in consumption is by drag, friction, electrical losses, and such.

Flight at that altitude is mostly problematic for GA aircraft because of the engines. Less air into the engine means less power produced means climbing is harder.

Most small drones these days are electric, so that doesn't happen. It's a bit harder to generate lift from the props in thinner air like that, but you also experience less drag, which balances that out to a large extent.

I fly standard quadcopters above 10,000ft ASL regularly. Soon I will be on a trip where I will be taking off from aronud 20,000ft. Increase prop size or pitch to compensate and you don't lose any thrust, but you do lose some efficiency, esp. if you go for higher pitch. On Erebus (~ 0.65 atm) I had about half my sea level battery life fully loaded. https://www.youtube.com/watch?v=gMh8kM4o6tk

I don't doubt a quad copter could hit 10k feet, but I do doubt many people would go that high. The only way to control that would be via a remote telemetry and/or video link or via strict autonomous programming. A standard 250 class quad could likely hit that 3km in 1.5-3 minutes under full power. I know from my time build and flying that a flight time for a quad hovered (hah) around 15-20 minutes on average. That leaves 9-12 minutes of loiter time as max. Then you need to factor in the power cost of stabilizing the aircraft at the height. The wind will be shoving it around so the FC will need to be constantly fighting for position hold.

I'd really love to see the math on the probability of a tiny quad intersecting the flight path of a firefighting aircraft at 3km altitude though. They are small and that's a LOT of open space. I can barely spot one against the open sky at a few hundred feet up. If these guys spotted one they must have amazing superhuman vision or amazing luck to have come within a very close distance.

Personally I put most of the sightings into the mass-hysteria camp. I know there are idiots out there flying in controlled space, but I also think things are being blown way out of proportion.

Edit: I found a site that did the speed/climd rate calculations. My numbers are bit off. The DJI, a common platform, could hit 3km in about 5.5 minutes and a 250 racing quad could do it in about 5. With about 20 minutes of battery power that leaves maybe 5-10 minutes of loiter time.

DJI Phantom 2 Vision+: flight speed of 17 m/s (61 km/h) and a maximum climbing rate of 9 m/s

250 FPV Racing Quad: maximum flight speed of 18.5 m/s (67 km/h) and a maximum climbing rate of 10 m/s

source: http://bpmsg.com/flight-speed-of-a-quadcopter/

10,000 ft has to be above or near the "service ceiling" of what a quad copter is even capable of. As altitude increases, air density decreases. The copter will have to work very hard, and will have less control force available. I think you're over-estimating the available power of a quad copter.

I think it's a certainty that one would/should have a video system for operating without line-of-sight. Video systems are common in high-end equipment.

>Personally I put most of the sightings into the mass-hysteria camp. I know there are idiots out there flying in controlled space, but I also think things are being blown way out of proportion.

Yeah. Agreed.

>>I found a site that did the speed/climd rate calculations. My numbers are bit off. The DJI, a common platform, could hit 3km in about 5.5 minutes and a 250 racing quad could do it in about 5. With about 20 minutes of battery power that leaves maybe 5-10 minutes of loiter time.

10,000 ft seems right at the edge of what the things can even do. They aren't going to be stable or easy to fly/land like a fixed wing aircraft would.

edit: forgot part of my post.

>I'd really love to see the math on the probability of a tiny quad intersecting the flight path of a firefighting aircraft at 3km altitude though. They are small and that's a LOT of open space.

"Big Sky" theory doesn't give me as much comfort as it ought to.

>I can barely spot one against the open sky at a few hundred feet up. If these guys spotted one they must have amazing superhuman vision or amazing luck to have come within a very close distance.

This. I have a difficult time spotting full-sized aircraft in the air, and I have pretty good vision.

> 10,000 ft has to be above or near the "service ceiling" of what a quad copter is even capable of. As altitude increases, air density decreases.

Its "only" a 25% reduction in density, you can still get a fair amount of flying done at that altitude. My colleagues have been to Ecuador twice for field work and had no issues flying quads in Quito (9,300ft), no need to change control gains or anything like that. So I know its perfectly possible.

250mm quads fly 3-7mins and going straight up is not hovering, it takes a lot of power. I fear you dont know what you're talking about.

Basically you can go 3k up then stay a minute and free fall down.

Bigger quads can fly longer - albeit going up always takes most energy and many big quads also need energy to go down / can't free fall without crash

Don't know about you, but I regularly got 15 minute flight times on my quads, so yes, I do know what I'm talking about.

Dilated time bubble i guess!

This one hits the 12,000ft mark at around 8:30 minutes: https://www.youtube.com/watch?v=LKARaE51SPo

Thing is, I seriously doubt many people fly their quadcopters that high. It's not much fun (not more fun than, say, 500ft), you don't have line of sight, and the wind will blow it away so if you don't track where you're going you're likely to lose your craft.

Plus at that range, we're talking some serious long-range equipment. Ready-to-fly consumer drones max out at approx. 1 mile range. If you buy this stuff, you probably understand (you should!) what you're doing.

It's possible alright, but not very likely.

Forget 10,000 ft above sea level. The article quotes one incident as occurring at 51,000 ft. 51,000 ft is a crazy high altitude.

Exactly what does the FAA think hobby kits include now? A complete knock down kit for a RQ-4 Global Hawk?

Agreed. FL 510 is right out. Or else I'm expecting to hear a joke about a drone radioing in to ATC to request a groundspeed check any minute.

edit: corrected post, see reply below.

This is FL510

You can get ballons to fly that high but not much anything else at an affordable price.

my mistake, thanks for the correction.

I don't even think quadcopters could produce 20% thrust at that height... which is pretty bad as consumer quads have something like a 2:1 thrust/weight ratio, maybe 3:1. High quality racing quads can get 8:1, but those have tiny batteries.