Hacker News new | more | comments | ask | show | jobs | submit login
Drones Flying Near Newark Liberty Airport Temporarily Halt Flights (nytimes.com)
49 points by jbegley 31 days ago | hide | past | web | favorite | 62 comments

FYI, here's a video of a drone striking an airplane wing in a test environment:


The main idea not highlighted by this video: Aviation fuel tanks are frequently held in the wings, and a drone’s charged lithium batteries, when deformed, will arc, burn violently and often explode.

I'm not sure what your intention was behind posting this video with minimal commentary, but, given the context, it reads like fearmongering to me.

there are now more registered drones than manned aircraft in the US, and they have been accessible to consumers for years now. there has yet to be a single fatal collision with a drone. on the other hand, there were 5-10 (graph is hard to read)[0] fatal collisions (and over 200 fatal accidents) in general aviation just in 2016.

[0] https://www.ntsb.gov/investigations/data/Pages/AviationDataS...

Can any air people comment on whether that collision would have been fatal to the aircraft? I'm sure it possibly could be, but at what likelihood? Personally I'd be more concerned about a propeller or wind screen strike, but these guys know more about it than me.

There's just no good reason to be flying near an airport. On the other hand, if you're five miles away from the nearest airport, the odds of a collision (even if you break the rules r.e. flying in LOS or under a certain ceiling) are just astronomically low. The sky is an enormous place and the odds of two things occupying the same space in it without trying to are once in a century if that. At the current level of density, at any rate.

Another thing that strikes me, and probably won't make me popular with pilots on this board: dozens or hundreds of GA pilots get themselves killed every year. Most of the time, this is due to pilot error. Drones represent some amount of increased risk to GA pilots. But, at the current level of drone market penetration, it seems empirically tiny compared to the risks GA pilots already faced voluntarily. I don't know what to do with this observation, except perhaps downgrade my personal concerns about the risks of drone <-> plane incidents.

> The sky is an enormous place and the odds of two things occupying the same space in it without trying to are once in a century if that.

Not near an airport. Planes takeoff and land in extremely predictable patterns. And at about the same height as drones typically operate.

I think you're thinking of a meteorite hitting an airplane, which yes is probably extremely rare.

No to mention that birds strike 13k planes a year in the US alone.


> I think you're thinking of a meteorite hitting an airplane, which yes is probably extremely rare.

Not as rare as you might think: for one airplane yes, but there are many planes and many meteorites. There’s a 5% chance a plane, somewhere, gets hit in the next 20 years: http://blogs.discovermagazine.com/cosmicvariance/2009/06/04/...

I wouldn't call "five miles away from the nearest airport" to be near an airport.

Five miles off the end of the runway a landing plane is only ~1,500ft above the ground. That's higher than the legal ceiling for drones, but it's certainly well within their technical capabilities.

As a pilot, I would. Most of the bug smashers I rent will be flying around 90-120 nautical miles per hour, so 5 miles out puts you 2-3 minutes from the field. Within one minute or so you better be setting up for landings, usually dropping to 1000' AGL thereabouts. A fair bit of traffic in a 5 mile zone when towered.

The sky is an enormous space... and then people decide to all go to the same place, like some airport's runway. It's quite easy for some idiot drone operator to fly into an approach path, maybe just to take cool photos of landing planes, but still putting everyone at risk nevertheless.

Just think about what happened with laser pointers: the sky is enormous, and yet many idiots decide to aim them at planes.

Right, I'm speaking only of a drone operator who is not seeking to be near other aircraft. For such operators, the risk of hitting something is very near zero. Surely lower than the odds of getting struck by lightning by an order of magnitude or more.

Wait, something happened with laser pointers?

I can't answer a fatality question, it seems out of scope. Anyway, big sky theory does not apply to class B airspace. And also, just go to flightaware and look at incoming flights to EWR which are going right over TEB. And the drone will not show up on either primary or secondary radar, and there's no radio contact with the drone operator, so there's no possible way to maintain the require positive separation guarantee of class B airspace.

>The sky is an enormous place and the odds of two things occupying the same space in it without trying to are once in a century if that.

Bullshit, and also it's a stupid statement. Just do some basic research about how many midair collisions there have been. A huge reason why we have positive control airspace and instrument flight rules is because of midair collisions.

It seems like this is a cheap and easy way to cause a whole lot of disruption. We're going to see more and more of this.

At some point Congress or some state legislature is going to pass restrictive legislation regulating drone ownership and use much more strictly than is the case today. It might require manufacturers to include backdoors to enable law enforcement to take control of a drone. It will be portrayed as a safety measure, perhaps even as an antiterrorism measure.

This is why we can't have nice things.

You could almost replace "drone" with "gun" in your comment and save for the "take control of a gun" line, it would still make sense. You could get an air port to shutdown by shooting a gun yet the consequnces of doing so are severe enough that it is in ineffective means of causing disrucption. The key difference is that a drone can go places a person with a gun cannot so I suspect that methods of finding drone operators will improve and become more widely deployed.

Not to mention that you can reconstruct where a bullet was shot from, but a drone could be operated from anywhere. Bullets are also loud when they go off, so localization becomes easier.

We’ve been living with guns for centuries, we’ve been living with drones for decades, and only the last decade saw them in the hands of many people. I suspect the current “Wild West” of drones is a temporary phenomenon.

I'm not sure about this (the regulation). At least it doesn't seem to solve anything. It is fairly easy to make a drone. If you're making it then there's no registration or back door that could help with this.

The biggest concern is what if some organization (let's say ISIS) organized a drone disruption "attack". With a few thousand dollars you could disrupt TONS of airports (by keeping constant drone coverage over just a few large airports). There's only 36 class B airports (37 if you include Hawaii). You could even be smart about it and just shutting down a much smaller subset of that and you'd have HUGE disruptions. At least millions lost to the economy, if not billions, due to the inability for these airports to operate (and the loss due to passengers not being able to reach their destinations). We're talking a (bs number) lower bound of 5 drones and possibly shutting down the East coast. That's not an expensive attack. Let's even take an estimate to ball part what it would cost to shut down all class B airports in the (contiguous) US. We'll say 5 drones per airport at $1k each (upper bounding). That's still under $200k (not including operative costs). That's too cheap NOT to do. And I'm sure you can think of a million ways to extrapolate this kind of attack to make it FAR worse for not much more money.

This isn't a regulation problem. This is a technical one and it needs to be solved fast.

What needs to be figured out is some way to disable the drones. And then to automate that process for drones that violate these air spaces. I honestly can't see any other good solution.

> It seems like this is a cheap and easy way to cause a whole lot of disruption. We're going to see more and more of this.

Pretty sure there are consequences for violating FAA regulations, so I doubt there will be a lot of intentional aircraft disruption by drone operators.

> It might require manufacturers to include backdoors to enable law enforcement to take control of a drone. It will be portrayed as a safety measure, perhaps even as an antiterrorism measure.

Congress is too lazy to figure out this language. They'll just opt for harsh penalties. Minimum 5 year jail terms, say.

Causes about the same disruption as a hoax call from a burner phone... for about 90 times the cost.

I’m not sure how we could ever gather statistics but I’m really curious if reported drone sightings could just be other objects mistaken for drones. To really see an object clearly such as a consumer drone like a DJI Phantom you’ll have to be very close and at that range it’s going to zip by extremely fast.

The problem is that one guy genuinely sees a drone and after that everybody else thinks they see a drone that isn't there.

That's the genius of this form of disruption: fly a drone past an airport just once, and people in the airport will report spurious drone sightings for hours afterwards. Flights will be delayed and much money will be spent tracking down things that don't exist.

The sun sets in NJ at 5:03 pm tonight, and backtracking through the article, the disruption started around 5:30, probably reported just a few minutes before ATC redirected traffic.

Likely they saw the lights of the drone flying around. It's a clear sky tonight, so they would have been able to see the drone for several miles.

It’s hard enough to pick out the lights of an aircraft 5 nm away in the sea of lights surrounding the NYC metro area. I highly doubt they were able to identify a drone by its lights from up to ‘several miles’ away.

Let’s assume they did though. All of the drones I’ve owned have red and green lights. Aircraft typically have red and green lights in addition to white lights facing aft and typically a red beacon and strobes. What’s so different about these lights that they’re able to distinguish them as those of a drone? Isn’t it possible it was a helicopter? A GA aircraft flying VFR? What about the red, green, white lights from a boat?

In my experience there have been many times I’ve seen something strange in the sky where my mind wants me to think it was something but in the end I really cannot be sure what it was. There’s been multiple times something has gone whizzing by at cruise or in the terminal area and for all I know it could have been a bird, a drone, a balloon, a kite, etc. Given the report was at night and it’s at an altitude that a drone shouldn’t normally be operating it seems questionable to me whether what they saw was actually a drone. With the recent media attention drones have had around airports I suspect there will be an uptick in reported sightings that likely are not drones at all.

DJI drones have software features to prevent abuse: https://www.dji.com/flysafe

If these are indeed drones causing disruption then they might not be off-the shelf video drones.

I remember seeing a talk in which the DJI (or maybe a different manufacturer) geofencing was just a plain text file of coordinates that anyone could alter.

I can't comment on that, but from the outside, they look to have a pretty complex implementation according to their website.

But then again I never would've assumed that Sony would store passwords in plaintext until they got hacked...

I understand approaching this with an abundance of caution, but the drones were reportedly 17 miles away near a different airport, so this looks like quite an overreaction.

Anecdote: I was flying during instrument training at night and took a laser strike from a small town about 30nm north of a class bravo airport (very large airline hub). When I reported it to approach, they basically closed all the northbound arrival/departure routes and started diverting traffic to the east and west so nobody would overfly the offender. Also, I got to watch a LOT of cop cars light up a minute or two later...

Pardon my ignorance, but what's a laser strike?

An attempt to blind aviators with a laser pointer

I’ve never understood this - aiming a laser at a vessel moving at over a hundred miles an hour many thousands of feet away seems incredibly harmless, unless you have one gigantic, powerful laser...

I've been hit twice by a ground-based laser. Both times the flashes were bright enough to be disruptive to my adaption to night vision and IMO reduced safety of flight temporarily. I called both into ATC. On one instance, I had the time to hang around and report the location quite accurately and saw the beginning of ground response of police (probably 15+ minutes, though). I wasn't able to stick around and learn the outcome.

An EMS helicopter was lased a few months ago resulting in (I think temporary) injury to the crew and a delayed response to an EMS transport mission.

Portable lasers nowadays are very much powerful enough to temporarily disrupt a pilot's night vision. They also light up all the micro-scratches in the windshield, making it difficult to see out. It's extremely dangerous, which is why perpetrators get found and arrested quickly. And the penalties are quite steep.

When the laser hits the cockpit glass, the whole panel lights up.

Example: https://www.youtube.com/watch?v=VshPGuKoa2Q

That's probably the mildest case I've seen. When it happened to me the guy was about half the distance and it lit up the whole cockpit.

3500 feet is 3100 feet higher than drone regulations allow. That's the real problem here. If that was in the direct flight path of an airport even 17 miles away this is wildly reckless behavior.

Hanlon's razor suggests to me it could be a flyaway (which of course doesn't make it less dangerous to aircraft, just not intentionally reckless).

I don't think so. Most fly aways in my experience are horizontal not vertical. A quick check of YouTube will show plenty of reckless behavior and my small drone is more than capable of flying at a distance of 3500 feet away from me.

Hanlons razor says this was somebody trying to get that sweet picture or see how far they can push their toy.

Yeah. Sometimes people like to use laser pointers at helicopters and planes. It carries up to five years in prison, and they do get caught, prosecuted and the feds do aggressively pursue maximum prison sentences, whether a pilot is blinded by it or not.

And I put letting your toy fly into terminal area airspace in the same category. I don't care if it's intentional or accident. Control your shit, or go to prison.

Or wind caught it at a higher altitude. As wind is prone to do.

The drone was reported over Teterboro at around 3,500ft. One of the ILS approaches into Newark[1] has Teterboro as an initial fix, with a minimum crossing altitude of 3,000ft. That puts the drone exactly where aircraft inbound on that approach will be.

[1]: https://aeronav.faa.gov/d-tpp/1901/00285IL22L.PDF

Unless you have an incredibly deep understanding of air traffic control protocols, I don't think you can make this judgement.

Consider that drones may not show up on radar, restricted air space around airports can change at a moment's notice, they cannot be avoided through visual identification, airplanes need a long distance to drop/gain altitude when landing/taking off, airplanes need vertical buffer when sharing airspace with other airplanes, and that much aerial maneuvering and control happens at the speed of humans talking in technical jargon to other humans.

Even if no planned flight needs the airspace, the drone may be in buffer airspace that would be used for emergency complications.

The drone pilot was grossly irresponsible to fly where they did, and should be in jail.

Did he violate any specific rules?

In case this would have been drone flight falling under 14CFR part 107, at least these come to mind.

- 107.41: over Tererboro airport and did not appear to have contacted ATC (air traffic control) and/or obtained waiver to operate in class B-E airspace

- 107.51: They flew above 400 feet

- 107.31: did not maintain VLOS with their drone (Visual Line of Sight)

- 107.43: interfered with operation of airport

- 107.39: possibly flew UAS over people

- 107.51: possibly within 500 feet vertically or 2000 feet horizontally from clouds

- 107.37: possibly (b) No person may operate a small unmanned aircraft so close to another aircraft as to create a collision hazard

Under CFR 14 part 107 you could request an waiver from FAA for a specific mission to fly above 400ft or outside VLOS or within different class airspace’s etc. 107.21 states also: In an in-flight emergency requiring immediate action, the remote pilot in command may deviate from any rule of this part to the extent necessary to meet that emergency. I suspect they had not.

Summary of FAApart 107: https://www.faa.gov/uas/media/Part_107_Summary.pdf

Note: in the us you can also fly under section 336 (model aircrafts) or under section 333 (pilot licensed remote pilots). But without checking the rules, I think differences would be minimal (e.g. ability obtain waivers, technicalities for ops near airports - e.g. 5 mile rule instead of airspace considerations).

Yes, he violated the specific rule that limits drone activity to below 400 feet. It was flying at 3,500 feet.

The operator is either dangerously ignorant, or reckless.

well, while it is a violation, i think they paid so much attention to it because the Teterboro is in the direction of the main runway at EWR. Anyway, looks like an airspace full of fun:


I think the key phrase from the article was

"The drone was spotted about 3,500 feet over Teterboro Airport"

Drone flying at 3500 feet definitely seems like it could pose a risk to flights out of Liberty 17 miles away


Isn't that still WELL above what you're legally allowed to fly a drone even with a 107?

This is the VFR Terminal Area chart for the airports in question, Teterboro is due north of EWR,


and an object near TEB at 3500' is inside the Class B airspace which requires positive control (not merely radar contact, not merely radio contact, but you must have a clearance or you cannot enter class B). TEB is also within 12 miles of EWR, and that's inside the departure/arrival path for EWR's main north south runways. Aircraft inside class B have a guarantee of positive separation by ATC. Everyone has to follow very specific rules in that airspace. And any aircraft that is not under positive control is a threat to all other aircraft.

Anyway, it is definitely not an overreaction. Also with the government shutdown, I expect there's next to no one available from FAA or NTSB to do a thorough investigation, most inspectors are not categorized as essential.

For a different perspective, 17 miles is only 5 minutes for a 747 at comfortable landing speed. At normal approach speed, this is much less.

What's more, the two airports are 17 miles apart only if you look at driving distance. It's more like 12 miles as the crow (or an airplane) flies.

Instead of trying to take out the drone, someone should take inspiration from that story [1] about lidar ruining some cameras (but not eyeballs) and start targeting the drone cameras.

[1] https://news.ycombinator.com/item?id=18886283

The drone did not halt the flights. Administrators halted the flight out of fears over a drone.

What I wonder is whether those fears are proportionate. Are drones any worse than a flock of birds?

Edit: video from a sibling comment suggests the answer is "yes" for the case of a collision with a wings's leading edge.

> Are drones any worse than a flock of birds?

Yes. A flock of birds has incentive and the means to get out of the way. Drones don't react to their surroundings.

Drones are also hard and do more damage than a bird strike.

I'm skeptical that birds have a means to dodge something moving at hundreds of miles per hour. If anything, a human controlling a drone from a distance has more perspective / ability to dodge a plane than a bird does.

Now whether we should actually trust said human to do so... that's a different question, as is the potential damage vs birdstrike.

I've nearly hit birds flying (and actually hit one but at night). They're actually not bad at avoiding aircraft going 100+ mph. They'll tuck their wings and dive if they feel they're in danger so typical training is to pull to avoid a strike (and lower the chance it impacts the windscreen). Obviously not all birds are great at this, particularly the larger ones but they do seek to avoid aircraft generally.

That video (linked by everdev) makes it look like the bird was more damaging to the plane than the same-sized drone was (it tore a bigger hole). But the tester says the drone strike was more of a problem because it penetrated the wing and damaged internal structures (hydraulics? electronics? fuel lines? Structural components holding the whole wing together?), and that for safety, drones should be made more like birds.

My understanding is that planes mostly continue to fly even if you change the shape of the wing (see all the examples of planes flying upside down or with one wing). Probably the answer is "it depends", especially if we are talking about a flock of birds compared to a single drone.

Don't forget the lithium batteries, which could cause serious damage if they entered the wing fuel tanks and decomposed.

Why does this keep being asked? YES! drones are a lot more damaging than birds: http://aviationweek.com/bca/faa-study-drones-more-dangerous-...

actual report: http://www.assureuas.org/projects/deliverables/sUASAirborneC...

Applications are open for YC Summer 2019

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact