Pilots Say That the F-16-Cessna Crash Illustrates a Civil-Military Difference
theatlantic.com
theatlantic.com
"Your responsibility is to NOT get into an accident. It doesn't matter what the air traffic controller said, what I told you to do, or what the other guy did. If you collide with another plane - IT IS YOUR FAULT. You are the PIC."
This absolute responsibility is well placed when even a "fender bender" is death for all.
It sounds like the military pilot didn't get taught that concept.
And it is entirely the case.
Unfortunately, AF pilots often confuse the ability to excel in specific very demanding roles as far more general aviation experience.
Also, the USAF Academy is famous for their gliding program. One of the glider pilots who trained at USAFA subsequently flew as US Airlines 1549 and glided into the Hudson.
https://en.wikipedia.org/wiki/Chesley_Sullenberger#Military_...
The AF does promote people learning more general skills in their free time, but fighter jets are very special aircraft and that tracks focus is defiantly on becoming a better fighter pilot not just a well rounded pilot.
While traffic callouts are nice to have and I really appreciate ATC giving me heads up when I'm flying, I still spend the majority of my flying time looking out the windows for other airplanes. Especially before making any turn or other maneuver where I might have limited view, always dip the wings to check the area first.
Unfortunately, I know some pilots who fly as if they've never seen a radio. Used to do a lot of flying out of a semi-uncontrolled field (it's untowered, but right under the 2,000ft shelf of a nearby class C). I would be doing pattern work, turning base and some yahoo would do a straight-in without a single call on the radio. If I hadn't seen him it and turned early it might have ended badly.
Not necessarily expecting you personally to know. But I feel like the incremental cost to the price of a plane should make this a no-brainer. Did I just come up with a ycombinator 2016 idea?
>It monitors the airspace around an aircraft for other aircraft equipped with a corresponding active transponder, independent of air traffic control, and warns pilots of the presence of other transponder-equipped aircraft which may present a threat of mid-air collision (MAC).
> The next step beyond identifying potential collisions is automatically negotiating a mutual avoidance manoeuver between the two (or more) conflicting aircraft.
This works along side the ADS-B broadcasts for generally knowing where other planes are without ground based radar - however that's not going to be fully implemented until 2020 in the US https://en.wikipedia.org/wiki/Automatic_dependent_surveillan...
It only tells you about other FLARM-using aircraft, though. (Some also look for transponder-carrying aircraft).
My cousin is an air controller in a little French airport and she told me that when military jets are coming in, everything is more tense. They fly so fast that she has far less time to react and to redirect other airplanes.
So even if you are on a collision course and know it, you might not have time to avoid the collision.
Drive by a GA airport and look at all the planes; many of them are 30, 40, sometimes 50+ years old. Now, this is not a big deal in the aviation world. Airplanes are pretty simple compared to most modern cars. They are also required to be inspected once a year by a mechanic (an "annual"), engines are replaced or rebuilt after a certain time interval has passed (usually around 2000 hours), extensive documentation is required to be kept, etc.
But there are just a LOT of them out there, so refitting all of them takes time. Unlike the car world where, if you make a change in the law, you can reasonably expect in 10-15 years the majority of cars would be compliant just because people buy new cars. In the aviation world that isn't the case. New airplanes can easily run $200k+, so people hold on to airplanes for decades.
It's also expensive. Anything involving airplanes is hellishly expensive. :)
You're looking at probably about $5k for equipment + installation costs (which must be done by a mechanic, pilots generally aren't allowed to work on their own aircraft [1]). And that only broadcasts your position; a separate ADS-B receiver must also be purchased if you want to actually see other airplanes around you. The cheapest one I can find is $500, and that also requires a tablet and software.
So yeah, we're going to get there. But it's going to take some time.
[0] https://en.wikipedia.org/wiki/Automatic_dependent_surveillan...
[1] Unless you are a mechanic yourself or you are flying a homebuilt experimental airplane that you were the primary builder.
[2] http://www.sportys.com/pilotshop/dual-xgps170-ads-b-receiver...
It's an entirely different matter to take that into an airplane and use it in such a way as that my life depends on the accuracy of the information. Has it been extensively tested in actual airplanes in real-life flying? Does it work with and not interfere with other aircraft systems? Does it work with the software I already use and know well?
Yes, $100 < $500. Now, if I built my own airplane (something I've given some thought to doing), I might be a little more willing to experiment with it. But the majority of pilots are not hardware hackers. They're average, everyday people.
I will happily pay $500 for a tested piece of hardware that is guaranteed to provide accurate information and work in my airplane over something I homebrewed using instructions I found on the Internet. My safety is worth the extra $400.
The imaging sensors are available for limited civilian use.
Sure, and the same could be said of the ATC. But the ATC never has direct control of an aircraft, unlike the pilot.
ATC are humans too, just like pilots, so mistakes sometimes happen. ATC has a larger picture, but pilots are on the scene up close. The important thing is that we work together to keep everyone safe.
There has only been one crash[1] where TCAS was used on both aircraft, and functioning. One of the aircraft followed the controller, ignored the TCAS resolution and a mid-air collision occurred.
[0] https://en.wikipedia.org/wiki/Traffic_collision_avoidance_sy...
[1] https://en.wikipedia.org/wiki/%C3%9Cberlingen_mid-air_collis...
One afternoon I was departing east, climbing and preparing to turn crosswind at 1000 AGL. I had announced my departure like usual, but otherwise the channel was quiet. I was maybe 750 AGL when I noticed a tiny sparkle directly in front of me (fortuitously, the sun was at my back). After observing it for a couple of moments and realizing that it didn't appear to be moving relative to my position, I re-announced my departure and angled to the right (SOP in head-on convergence is everyone turn right, immediately). Sure enough, about 20 seconds later the inbound aircraft zipped past on my left. Hard to estimate distance in the air, but way less than a mile and equal altitude. Fucker never said a word, just landed straight in, with the wind, completely ignoring the pattern. Blows my mind to this day.
Yokel: "What traffic?"
ATC: "The traffic you just overflew."
(As a Cessna goes over me about 50 ft above me.)
And mind you this is with a ATC trying to order traffic.
If course, you're still responsible for following instructions, but you can't necessarily blame an IMC mid-air on the pilots.
Although it's not surprising that military pilots would not be sharp on FAA regs, it does raise the question, "so how does civil and military aviation share the same airspace?"
From accidents seen around the world, you can't safely mix them. Civil pilots are "solely responsible" and military pilots are "flying under orders" and "flying a mission", which is why they defer to controllers.
A few folks in the article suggest 'slamming' the plane into a turn when ATC uses 'immediately'. That's really not a great idea in a variety of situations. Remember: aviate, navigate, communicate. AOPA has a great article this week on a situation where ATC screwed up and cleared a pilot for landing with another plane on final. He turned onto a tight base ready for short final and ATC told him to turn quickly. He executed a 60 degree bank/climb at 300 ft AGL and stall/spun the plane and put a 4 foot smoking hole in the ground. http://www.aopa.org/AOPA-Live.aspx?watch={384817B3-70C4-4147...
I'd add that I've been in the pattern at small airports around Colorado Springs with Air Academy pilots (and flown through the MOA's in the area) and they've always been polite and helpful. Although it sounds like the culture issue described comes up with military pilots on IFR plans around VFR civilians.
Second that, "Fate is the Hunter" [0] by Ernest Gann [1] describes the precarious life of a US career airline pilot (including descriptions of flying DC2's, DC3's & DC4's, flight seniority and luck) between the 30's to the 50's. The book is dedicated to a hundred or so commercial pilots not as fortunate as EKG.
"Although we can never be absolutely certain,
we now believe the Eastern Airline crash at
Bainbridge was caused by unporting. Do you
know what that is?"
I confessed that I had never heard of it.
[Unporting: could be caused by a missing
elevator hinge bolt, like that missing on
Gann's DC-4, and cause an uncontrollable
nose-dive, if the plane were flown at certain
speeds and center of gravity loadings.]
"Did you slow down when you first noticed
the vibration? You did not because you had
no fear of it. But if you had been the
nervous type, if you slowed down, the center
of gravity would have changed. That would have
been quite enough to complete the process of
unporting which had partially begun."
One interesting chapter I do remember is Ch16, "A Pretender" where an untrained pilot takes command of an aircraft only to be found out by very poor co-ordination. Gann describes how nobody checked his credentials prior to employment of commercial flying.
[0] https://en.wikipedia.org/wiki/Fate_Is_the_Hunter
Midairs happen all the time, at least a few a year. Usually this is in dynamic situations (BFM, etc) or joins, especially at night. Usually between fighters and seldom with civilians.
Most fighter-to-fighter mid-air collisions occur during dogfighting practice, or during formation join-up. In both these situations, the radar might not show the other jet. Again the radar has to be in "dogfight" mode to have any chance of finding the other jet in close. The pilot would not be manually adjusting the radar scan position during a a visual join-up.
Military aircraft typically trasmit to air traffic controllers in a frequency range between 225-380 MHz, while civilan aircraft transmit to air traffic controllers in a frequency range between 118-136 MHz.
The air traffic controllers simulcast their transmissions betweeen both frequency bands to make sure their instructions reach both military and civilan pilots, however the transmissions back to the controllers cannnot be heard between civilan and military pilots.
This is a prime example of the "Civil-Military" difference outlined in the article.
* Note: Most Military pilots have the ability to transmit on both VHF (118-136 MHz) and UHF (225-380 MHz) however it is up to pilots discretion.
Basically the most dangerous type of people across any situation.
The amazing things is that most surgeons are like this -- the traits described are a near archetype for surgeons -- and yet outcomes for surgery are pretty damn good, and rapidly getting better for many types of surgery.
I would say that "good" is relative to what can be reasonably expected. If most surgeons fit that description, then we may not have an independent way to establish our expectations. That is, maybe we think it's "good" because we don't know better.
Hospital error is a leading cause of death in the U.S., though I don't know how much of that is due to surgery.
Also, pilots must sustain a good enough performance for hours, while surgeons must sustain a great performance for minutes. Those are very different requirements, and may quite well fit different personalities.
[1] http://www.who.int/patientsafety/safesurgery/ss_checklist/en...
I can't say that I agree, and a lot of people whose surgery didn't turn our well aren't around to complain.
We did software for Hospitals here in Europe. They told us that confidence is one of the most important things for surgeons but arrogance is very bad. We had lots of data on our hands.
Also, while surgeons are well considered here in Europe, they are not the "Ferrari at the door" US types.
One of the big problems is that people can see the errors on others, but not in oneself. So our job was basically organizing things automatically to minimize errors, using lots of psychology, like making sure other professionals could see the errors of the first, recording things and so on.
We told them that it was for University kids to learn. While true, the main idea is controlling the surgeon by other surgeons. It works flawlessly.
"infallibility" could be good for making people confident and no doubt. At the same time is very bad if applied with authority if the surgeon is bad.
This training was all kicked off by a British Airways captain whose wife died on the operating table due to the surgeons ignoring the less senior people in the operating theatre when they raised the alarm and offered solutions. More power to him.
Our instincts and the lessons we learned in childhood are often wrong, especially in environments very different from our childhood, historical, and evolutionary environments, like aviation.
That doesn't sound like adequate vertical separation to me. Why not ask for a 500 foot climb instead of giving a vector?
I suspect this crash is the result of a gap in FAA regulations. I would suggest that no pilot should be doing IFR training alone while around aircraft not flying IFR. But I can see why the rule doesn't exist: it is an extraordinarily rare situation that, probably, nobody thought needed covering.
This is one of the reasons that American's and Brits go to Goose Bay for low-level training. Canada has more empty sky than either of them combined.
And if the pilot is simulating instrument conditions, there is a rule: FAR 91.109 "No person may operate a civil aircraft in simulated instrument flight unless - The other control seat is occupied by a safety pilot who possesses at least a private pilot certificate with category and class ratings appropriate to the aircraft being flown".
If not simulating IMC, it's the pilot's responsibility to see and avoid per 91.113(b)
Civil aircraft means aircraft other than public aircraft.
Public aircraft means any of the following aircraft ...
(1) An aircraft used only for the United States Government ...
(2) An aircraft owned or operated by the armed forces ...
(3) An aircraft owned or operated by the National Guard ...
FAR 91 doesn't apply to military aircraft, although military rules require they maintain a visual scan outside the aircraft.
It's perfectly normal for a training, proficiency, or currency flight to be operated under IFR, even single pilot.
1. All all the facts are NOT in, yet judgment has been reached and mudslingers are throwing.
2. By the NTSB preliminary report, the entire time from the first pilot warning (1100:18) to midair (1100:52) was 34 seconds. Barely half a minute for life and death decisions to be made. All of which fell on ATC and the F-16 pilot with the civilian not in radio contact.
3. ATC made the traffic call. 8 seconds later ATC gives the ‘turn south’ call. The F-16 pilot asks ‘confirm 2 miles’. 8 seconds later ATC says ‘if traffic not in sight, turn immediately 180.’ The F-16 pilot complies and begins his turn south. 18 seconds later they collide.
4. From 2 miles to collisions in 34 seconds coincides with just under 4 miles a minute closure or 240kts. Cesna at 60 and F-16 at 180 seems reasonable.
Could the F-16 driver have overbanked and started his turn quicker by a few seconds? Maybe, but there was only 16 seconds and the pilot collided in a turn, so if he’d delayed a few more seconds the collision would have been a miss.
Also, how responsible is a civilian pilot for climbing through an instrument approach near a military airfield without radio contact?
It’s a tragedy. Military aviation could probably ditch Tacan approaches, but they are good training. Perhaps any that are close to civilian fields need to be given the axe.
Yet to blame this only on the military pilot and military pilots in general is a complete disgrace. An accident is like swiss cheese, with multiple holes. Errors with people, processes, and things all help cause it.
If only one thing would have changed (if the turn started earlier or later, if ATC picked up the radar hit earlier, if the F-16 pilot got a visual quicker, if the F-16 had a radar lock, if the Cesna was in radio contact, if the Cesna stayed below and didn’t climb through an instrument approach) the crash wouldn’t have happened.
Aviation is dangerous, especially if you aren’t in a bus. Even a bus driver is mildly dangerous.[1] General Aviation and Fighter Aviation are especially dangerous. While I’m all for lessons learned, this salacious, click-bait, profiteering-off-a-tragedy, poised-as-a-question trash doesn’t help.
[1] http://qz.com/410585/garbage-collectors-are-more-likely-to-d...
Other than responsible for not seeing a fast moving jet rear-end him, not very responsible. Even if he was under radio control and his head on a swivel, it would have have been near impossible to see the jet coming from behind and left at that speed. He wasn't "near a military airfield" - he had taken off from a general aviation airport just north of Charleston's International airport (CHS), also shared with military, however FAA ATC and rules prevail in this airspace. There is no "military airfield" difference around CHS. The F-16 was on a practice approach to CHS, just as anyone could do a practice approach to CHS by requesting to do so from ATC. One major contributing issue to this was the controller descending the F-16 to such a low altitude so far out (1600' at 34 miles out) [1]. This is well below any slide glope and frankly, as a pilot who flies IFR and no-radio-VFR, I would never expect to see a jet on an approach that low even 15 miles out.
The incident did not occur in the military operation areas (MOA) north of CHS.
[1] http://www.ntsb.gov/_layouts/ntsb.aviation/brief.aspx?ev_id=...
If you fly GA, I’d recommend you know where nearby airports are (and thirty miles isn’t that far for jets), where instrument approaches are, where MOAs are, where low level routes are, etc. I’d also recommend using flight following. You can say that these things aren’t your responsibility, but I don’t think that’s a good attitude.
Also, I’m not sure how you interpret that the Cesna was rear ended. Reading the NTSB report it sounds like the Cesna was heading southeast (135) and climbing and the F-16 was heading west (265) and turning south (180). Seems closer to head-on or perpendicular with the Cesna climbing into the F-16’s altitude and the F-16 in a belly up turn with no visibility.
Regarding the Tacan, I don’t see any mention that the F-16 was below altitude or off course. So just because you wouldn’t expect a plane there, doesn’t mean that one won’t be there. Again, maybe this is a bad Tacan with nearby civilian airfields and should be axed, but it doesn’t seem any rules were broken.
We don’t know the results of the investigation, yet all the blame is being put on the F-16 driver. It sounds like small failures from all involved resulted in a tragic outcome. There may be lessons learned from this incident that can help prevent another in the future, but placing blame without the facts and pointing fingers in the media isn’t how we figure them out.
Why delay doing exactly what you're told to do, twice?
1. Turn left if you don't have that traffic in sight, does not mean ask questions to confirm the distance, it means turn left if you don't have that traffic in sight. This delay is a minimum of 8 seconds difference in whether this accident happens.
2. Turn left immediately means do it now, not beginning to turn over 18 seconds... I've put a Cessna in a 60˚ 2G bank inside of 5 seconds many times (it is an unusual attitude, done for training purposes), and this guy is in a fighter jet. He can do better than that.
From this transcript, it's a much easier case proving negligence on the part of the F-16 pilot, than defending it. About the best defense the F-16 pilot I can come up with is that he probably had no chance to see and avoid because of the nose cone of the F-16 obstructing the view of the slightly lower Cessna; and the whole see and avoid principle is probably statistically bullcrap, and all flights should be under positive control. But unfortunately that's not the system we have, and I don't understand the cause for these two delays and the seemingly slow turn when the word "immediately" was used by ATC, which has a very explicit, formal meaning in ICAO terms which is when instant action is required by a pilot.
Doesn't the US have MARSA (Military Accepts Responsibility for Separation of Aircraft) rules that military flights can occasionally use, and which point they are then fully responsible, but otherwise I thought they'd still have to generally follow ATC directions.
Yes:
I was talking more about deciding which flight level to use and general legal responsibility.
Unless the F-16 pilot turned on his radar or could see other aircraft, how does he know exactly what other aircraft there are. Admittedly, ATC only generally knows squarked aircraft, but this is generally a good picture of what's happening, and will probably be more complete a picture than what the F-16 has access to under non-military conditions (AWACS, datalink, etc).
My guess is this is going to end up in court.
Also, you're basically making the same point as the article, except that (so the article conjectures) the military pilot didn't think he needed to look out for other aircraft because ATC was doing it for him.
? Captain A. G. Lamplugh
But I would like to point out that see-and-avoid doesn't really work in general, and particularly not in the case of two small aircraft with wildly different performance characteristics. I suspect that, even if the F-16 was on a practice instrument approach, the bottom of its performance envelope is significantly higher than the top of the Cessna 150's. (According to the 'pedia, maximum speed is 109kt and Vne is 140kt.)
See-and-avoid has been the first and more-or-less primary anti-collision system in aviation use, but if you've ever tried to cross a highway with high-speed oncoming traffic, you've probably already learned that there are limits to the human visual system. It's just not built to do that sort of thing. No doubt keeping your head outside the cockpit does prevent a lot of crashes, but expecting it to have prevented any crash, much less this one, is too much of a reach.
Keep in mind that most animal's visual systems are pretty good at picking up movement, but one of the signs that you are on a collision course with another vehicle is that it isn't moving in you visual field.
From [1]: "[The University of Washington’s Autonomous Flight System Lab] came up with a novel way to [modeled the UAS collision risk], reasoning that the known behavior (and random collisions) of gas molecules could serve as a way to predict collision probabilities between VFR aircraft, including UAS. (PDF)[3] To test the hypothesis, they compared the gas-molecule data to actual midair collision data and found a remarkably good fit. This allows two conclusions: manned airplanes under VFR run into each other at about the same random rate as gas molecules do and that’s pretty strong statistical evidence that see and avoid doesn’t work. (Because of a variance between the gas data and actual midairs, the paper concluded that see and avoid might work 14 percent of the time, but a bit better during the past few years. Possibly for a reason.)"
By the way, [1] is an entertaining post with some information, while [2] is a paper from 'stralia called, "Limitations of the See-and-Avoid Principle".
[1] http://www.avweb.com/blogs/insider/Why-See-and-Avoid-Doesnt-...
[2] http://www.skybrary.aero/bookshelf/books/259.pdf
[3] https://www.aa.washington.edu/research/afsl/publications/lum...
Ps. Ok, I do want to say something about the difference between military and civilian pilots. Back in 2004, six soldiers and a general died when a Blackhawk hit a television transmission tower guy wire while in heavy fog.[4] The tower was 1800 feet tall; apparently the helicopter hit the wire about half way up. It wasn't a training mission; the helicopter was "headed to Red River to check maintenance on equipment". The first question from me, the most holster-sniffing of non-pilots, was, "What was he doing flying below 1000 feet in heavy fog?"
[4] http://www.chron.com/news/houston-texas/article/Seven-soldie...
This^. I was right-seat in a King Air and heard Center call traffic for us at 1/4 mile, looked once, nothing, turned head in case I might have lost my bearing, looked again and there it was. I felt I could almost read the numbers and ID on the aircraft, that's an exaggeration, but it was close. If there hadn't also been some altitude separation, we could've been in a lot worse shape because we were under the other A/C's nose. Neither myself nor the pilot could see it before it was already very close, and it closed too fast for us to do much because we didn't have much energy (climbing spiral with skydivers on board).
This is how two planes collided 37,000 feet above the Amazon rain forest.