1/20/2011 Near Midair - AA B-777 & 2 C-17s

Here's a pictorial of the incident:

NYOceanicNearMiss.png


NTSB Docket
 
I've never been a military pilot, so I'm not personally comfortable with the same level of distance some of you guys might be.

Exactly my point -- pilots are a product of their experience.

IA heavy jet simply doesn't have the same maneuvering capability that smaller aircraft do, particularly in the flight levels. 2,000' lateral separation, unintentionally, could have serious consequences to an aircraft that can't maneuver out of the way quickly.

That is true -- however maneuverability has nothing to do with the IFR and Flight Level bubbles. As was mentioned, those numbers are derived based on the accuracy/capability of the ATC radar/transponder system and the navigation/altitude precision/accuracy of the aircraft that are operating in the national airspace system. So, in the F-15 and T-38, I had to have exactly the same separation that a 747 and A300 have, despite being physically smaller and lots more capable of maneuvering should it be needed.

My point was simply that folks who spend their careers in the protective cocoons of the IFR world build judgment and habit patterns that are reflective of that, and not necessarily of risks that are real.
 
So, why is it that you guys are cool with 1000' of vertical separation, but crapping your diapers if it's 1000' of lateral sep? We operate in a 3 dimensional environment, and airplanes don't know which way is 'up' -- separation is separation.

It's no more of a 'close call' in one dimension than in any other.
 
Because 1000' of lateral deviation as displayed on a CDI is hardly noticable. 1000' of vertical deviation is one full sweep of the long needle on the altimeter.
 
So, why is it that you guys are cool with 1000' of vertical separation, but crapping your diapers if it's 1000' of lateral sep? We operate in a 3 dimensional environment, and airplanes don't know which way is 'up' -- separation is separation.

It's no more of a 'close call' in one dimension than in any other.
Because generally transport aircraft are not traveling several hundred knots in a lateral direction. The closure rate is far greater, as evidenced by the image above. Again... no time to do the actual math right now, but those guys don't have a whole lot of time to react given their positions, closure rate, distance from each other, and speed. If they were 2,000' apart vertically, at the same heading, and converging towards the same altitude, they would have longer to react. 2,000' at the same altitude, on a head-on convergence at their speeds? That can't be more than 10 seconds.
 
Just did a bit of quick math. One aircraft, flying at approximately 250kts toward a fixed object 2,000 feet away would strike that object in about 5 seconds. 5 seconds. This is two aircraft, converging on a fixed point (speeds unknown right now, right? 250kts a fair guess for these two?). I'm sure someone faster than I can figure out the geometry of them being separated by 2,000', and triangulate what each of their distances from the fixed point are based on the angles in the above image. Suffice to say... not a whole lotta time to react.

Were these two guys at the same altitude, changing altitude, off by a few hundred?
 
The story reports that controller mistakes in 2011 were 80% greater than in 2007: "Total controller errors reported by the FAA last year were about 80% higher than in 2007

OK, before anybody freaks out about the 'big' increase in controller errors, consider that the reason 2007 was used as a baseline is that ATC began their ATSAP program in 2008. ATSAP is the equivalent program for pilots in place at most airlines (called ASAP) that allows crews/controllers to self report mistakes, incidents, screw-ups, etc with a wide range of protections that generally keep the employee from facing any repurcussions. In return, the FAA and airline Safety Departments now get thousands of reports and data points that they never would have had access to before allowing them to see trends and areas of vulnerability. NATCA says that since the program began, 60% of their 21,000 controllers have filed at least one ATSAP report. At Delta, we have had over 8300 ASAP reports in the first 6 months of 2012 alone (one of them was mine!)....so of course 'reported' ATC errors have increased since 2007, because controllers now have an avenue to report problems that they would have covered up before.
 
....so of course 'reported' ATC errors have increased since 2007, because controllers now have an avenue to report problems that they would have covered up before.

Bu...bu..buu.but that's not what Joe Six Pack wants to be readin' while he's drinking his breakfast beer!
 
Just did a bit of quick math. One aircraft, flying at approximately 250kts toward a fixed object 2,000 feet away would strike that object in about 5 seconds. 5 seconds. This is two aircraft, converging on a fixed point (speeds unknown right now, right? 250kts a fair guess for these two?). I'm sure someone faster than I can figure out the geometry of them being separated by 2,000', and triangulate what each of their distances from the fixed point are based on the angles in the above image. Suffice to say... not a whole lotta time to react.

Were these two guys at the same altitude, changing altitude, off by a few hundred?

Okay, I'll play -- they were 5 seconds away from impact at their closest point. And? Although their lines cross when plotted on a map, it doesn't mean they were headed on a collision course. It's not like everyone was blindfolded and suddenly at 5 seconds to impact, with the airplanes both headed for exactly the same point in space, the blindfolds came off and they had to do something. The article said they were headed at each other for at least a minute and nobody said or did anything about it.

Again, from my perspective, this just isn't that significant. IMHO, people's "personal minimums" here are just too comfortably large and they get excited over something that isn't really a big deal. In this case, TCAS appears to have at least assisted, although there's no way to know if they would have hit each other if neither of them had acted.

More importantly many hours and minutes prior to that did they have to see the other aircraft and do something about it? There were at least 6 people sitting in front seats that could have seen traffic. There were any number of avionics systems that were there to help, PLUS the controllers.

P1-BH435_MIDAIR_G_20120805192712.jpg


The jets, both at 22,000 feet, barreled directly toward each other for at least a minute without pilots seeing the other aircraft or realizing the extent of the danger.

Check twelve, dudes. Boeing and McD engineers spent a lot of time designing those big clear panels in front and beside you so that you could look through them. There's no priority higher in your duties as a pilot than avoiding hitting the ground, anything attached to it, or anything flying through the air.
 
Check twelve, dudes. Boeing and McD engineers spent a lot of time designing those big clear panels in front and beside you so that you could look through them. There's no priority higher in your duties as a pilot than avoiding hitting the ground, anything attached to it, or anything flying through the air.

Now there is something we can both agree on. That and TCAS doing its job.

The article doesn't mention it, but did the aircraft come within 2,000' of each other while they were in the process of avoidance? Because that would be a different story.

Not having ever flown transport-level aircraft, I'd be curious to know, at the speeds and convergence they were traveling, how soon would the pilots realistically been able to see each other?

(And just so you know, Hacker15e, I'm not arguing any points your making... I'm honestly trying to learn and understand why you think that two transport-sized aircraft on a converging path 2,000' apart is no biggee).
 
(And just so you know, Hacker15e, I'm not arguing any points your making... I'm honestly trying to learn and understand why you think that two transport-sized aircraft on a converging path 2,000' apart is no biggee).

Guess Air Refueling with transport sized aircraft could resemble this situation some times...

While I do understand the FAA minimums as they pertain to system capabilities, I do agree with Hacker that "personal minimums" can be greatly exaggerated. I think that many of them are rarely based on any sort of factual knowledge and are instead based more on a "gut feeling" about what someone thinks might be right.
 
The article said they were headed at each other for at least a minute and nobody said or did anything about it.
Because the two flight (elements) were on different frequencies and/or under control of different controllers.

Again, from my perspective, this just isn't that significant. IMHO, people's "personal minimums" here are just too comfortably large and they get excited over something that isn't really a big deal.
Your perspective is one from which you've been trained for close quarters maneuvering. Mine, and a lot of others here, is not. This discussion is ongoing on another board, and not ONE of the contributing members is saying "quit being a baby and up your personal minimums..."

It's not unlike the difference between 91 and 135 regarding factored landing distances. I'm perfectly comfortable flying a plane into a runway which is just barely longer than what the AFM shows I need to stop. On the other hand, someone that's flown 135 and used nothing but the 60% rule for as long as they remember may not be comfortable operating in my world. The fact of the matter is that rules are rules, and I'm not going to go into a 135 operation and say "You pansies. Up your minimums, because if I can handle it, you can too."

In this case, TCAS appears to have at least assisted, although there's no way to know if they would have hit each other if neither of them had acted.
Lets not forget the C-17s were in TA only and weren't taking part in the resolution advisory calculations. From what I've read, the AA plane actually had a reversal of commands; first a "descend now", then an "increase descent", and finally a "climb now". There was actually ZERO vertical separation when it was all said and done. Why that happened, no one really knows yet.

More importantly many hours and minutes prior to that did they have to see the other aircraft and do something about it?
It was night. I sure as hell can't tell whether or not someone is at my altitude or on a converging course at night. And how do we know they weren't in a cloud layer?

There were any number of avionics systems that were there to help, PLUS the controllers.
Again, only one plane would have been playing the TCAS game, and the two flights were on separate frequencies.

Check twelve, dudes. Boeing and McD engineers spent a lot of time designing those big clear panels in front and beside you so that you could look through them. There's no priority higher in your duties as a pilot than avoiding hitting the ground, anything attached to it, or anything flying through the air.

See all of the above...
 
I should also add that I'm usually first to blow off the sensationalist nature of the modern media, but, this was pretty serious. According to a ZBW controller that is contributing to the discussion on the other board that I mentioned, and has seen the FALCON replay and listened to the tapes, "it was ugly".
 
So, why is it that you guys are cool with 1000' of vertical separation, but crapping your diapers if it's 1000' of lateral sep? We operate in a 3 dimensional environment, and airplanes don't know which way is 'up' -- separation is separation.

It's no more of a 'close call' in one dimension than in any other.

Yes, yes it is. With targets merged and 1,000' vertical separation, the controller still has positive separation. In this case, the controller had no way to determine whether or not the aircraft were separated.

You going and flying tactical jets with proper formation training, and most importantly, a thorough briefing, is not the same as a heavy 777 climbing through the flight levels with an unintentional near miss.

And yes...the TCAS worked. Mostly. The TCAS RA on the 777 commanded a descent, then an increased rate of descent, then a reversal to a climb. THUG11, the flight of two, had their transponders set to TA only and couldn't receive RAs; there was no "negotiation" between the systems of each aircraft. The 777 was on its own for avoiding the C17s.
 
Random aside, if any of you guys get a chance to hop into a heavy jet sim at some point, have the instructor load you up at max takeoff weight, climb into the flight levels, and see how much maneuver margin there really is before the buffet boundaries. Several seconds of visual contact is simply not enough time to click everything off and maneuver to avoid.
 
Well... I'm glad there is some validation that I'm not crazy. As far as I'm concerned, 2,000' lateral separation on a converging flight path at the same altitude is a really BFD for just about every type of aircraft out there. Like I posted above... at 250kts, 2,000 feet is only 5 seconds away. Assuming both of these heavies were going 250, there was precious little time to get out of each others' way.
 
Well... I'm glad there is some validation that I'm not crazy. As far as I'm concerned, 2,000' lateral separation on a converging flight path at the same altitude is a really BFD for just about every type of aircraft out there. Like I posted above... at 250kts, 2,000 feet is only 5 seconds away. Assuming both of these heavies were going 250, there was precious little time to get out of each others' way.

The 777 was more than likely doing 330+ KIAS in the climb.
 
Yeesh. Just out of curiosity, I'm going to do the full math on this with what little we know (at least we have a fairly accurate convergence angle and good guesstimates on speed) when I get home. Based on some really rudimentary stuff, this looks like it could have been a gnat's ass from being a giant fireball.
 
Back
Top