If there is misreporting for political reasons, that is unfortunate. How and where drones can be operated is a real concern and will become only more and more of an issue as they become cheaper, better, and more common.
Sigh. I'm a "drone hobbyist" myself, although I mostly refer to the hobby as "RC model flying". Nothing new, because the RC flying hobby has been around for decades.
Part of me applauds the stricter rules. It's just so easy for some knucklehead to buy a quadcopter from China and fly it around the White House or an airport or over someone's backyard. Unfortunately, these people destroy the hobby, giving it a bad name.
On the other hand, the rules we have had for decades (don't fly near airports or restricted zones, don't fly too high, don't fly over crowds, etc.) are perfectly fine and reasonable. We don't need stricter rules and regulations, we need common sense from drone pilots.
Secretly I hope that drones are really a fad, and the interest will fade away. That leaves the hobby for the real enthousiasts.
Yup. If you expect a nebulous and undefined "common sense" to prevail after your hobby has entered its eternal september what you'll end up getting is a severe crackdown when shit inevitably hits the fan.
"Common sense" is mostly social norms (usually starts out as "hard-earned wisdom" until the culture forgets how it was acquired), when there's a huge continuous influx of new participants the "old culture" can't keep up, neither can its "common sense". Doubly so when a hobby lends itself to nefarious uses and mindless fuckups as much as drones do.
But in a lot of cases, isn't this the same as the people who film themselves climbing tall buildings, bridges, etc? For many people, drones will always have the exhibitionist grandeur attached to them. Sure, they know they aren't allowed to fly near Point A, but the sunset is just so perfect, they can already smell the 4000 upvotes they'll get that night posting on Reddit.
A simple licensing process would work here; a written test that verifies that pilots have read and understand the rules (similar to a written driver's license test) would likely be enough.
I want to know more about this drone that was supposedly pestering an aircraft during firefighting operations. 10,000 ft altitude is quite high; if it was a quad-copter type thing, I've got my doubts that it could make it to that altitude at all, much less maintain that altitude for any length of time. I'm going to assume that anything flying higher than ~15,000 ft is a military drone, and the FAA using these particular sightings to cast aspersions at private individuals is dishonest. Including incidents involving drones operated by the police is also pretty sketchy. Media are parroting this crap uncritically; I hope more of them find some gumption and ask a question or two. I can see where drones are potentially a problem, especially considering that they are only getting better/cheaper; but this sham-press release thing from the FAA appears to be scaremongering and it deserves ridicule.
Links to both the FAA's press release and the AMA's response in this AMA blog post.
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
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.
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.
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.
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.
>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 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.
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.
As someone taking off from a runway the other day and having to actually dodge a stupid camera tied to balloons, we don't always report all that stuff. I, for one, welcome the rules. I also understand that not everyone will follow them. Welcome to the world.
That stupid camera tied to balloons wasn't by chance a permitted-in-advance activity was it? We have that at my local airport. A few times per year a prof from a nearby uni sends a balloon up mostly for the benefit of his students. You ought to hear the Saturday morning FBO free-coffee-drinkers club whine about the inconvenience of it. All GA pilots may not always report everything, but there are a few who report everything they see and everything they imagine they saw.
I also welcome the rules. People flying stuff (balloons, RC aircraft, drones, kites, etc.) near airports should be punished. It's just that I don't think even stricter rules are the answer.
I don't buy it. What if the miss wasn't near? Would it be a far miss? An average miss? What exactly is the point at which you go from colliding to a near miss, to any other kind of miss? It's redundant language.
Sure, if you miss by a huge amount it could be a "far miss" or a "distant miss" or similar.
The point where you go from colliding to a near miss is the point where the two things don't touch. There's no exact point where you go from a near miss to a regular sort of miss, since it's not a precision term.
But they don't far miss or distant miss. They weren't aiming for each other. "Near miss" is the only way you would ever refer to planes almost colliding with the word "miss", so the "near" part is redundant. They missed. If you wanted to say they were close, you could say they almost hit, which is not only more accurate, it's not redundant, because you can't take out either word and still express the same idea.
This isn't just to be pedantic; there's lots of redundant or incorrect language out there which Carlin talks about at length.
So "near miss" is redundant if it's an accident, but not redundant if you're actively trying to collide with the other plane? Having different terminology for the two cases seems like a lot more trouble than it's worth. In any case, this doesn't change the wrongness of "A collision is a near miss!"
It's all about reading. e.g "near dead" means "I was almost, but not quite dead."
Therefore "near miss" could be read as "close to, but not quite a miss" and it would be more accurate to say a "near hit" meaning "close to, but not quite a hit". In this context, a "near miss" is a collision - that almost didn't happen and a "near hit" would be a collision that almost happened but didn't.
Edit:
If trying to collide, it would be appropriate to call a collision that almost didn't happen a "near miss".
> there's lots of redundant or incorrect language out there which Carlin talks about at length.
While Carlin -- as comedians often do -- used the manner in which natural language idioms are irregular as the basis for some of his humor, the irregularity of natural language idioms is not incorrectness. It may be aesthetically displeasing to people who have a strong preference for regularity -- a preference which is probably overrepresented in the HN audience -- but that's a different issue.
I'm not concerned with aesthetics. I'm concerned with incorrect language which leads to incorrect thinking. I could talk at length about a single word and how its misinterpretation by a huge subculture leads to bad decision making, but let's just say words are very important.
reply