American Airlines crash

billvon

Valued Senior Member
I wrote this on FB but thought I'd repost it here:

Yesterday I was out flying, practicing to get ready for my next BFR (biennial flight review.) It was a beautiful day for flying, mostly clear with some scattered clouds around 4000 feet. After I landed and came back home I heard about the crash of American Airlines flight 5342, the first crash of an airliner in the US since 2009.

Commercial aviation has become so safe that we often forget about the risks inherent in aviation. But they are always there. Captain A. G. Lamplugh, a pilot in the Royal Air Force during World War 1, once noted that “aviation in itself is not inherently dangerous. But to an even greater degree than the sea, it is terribly unforgiving of any carelessness, incapacity or neglect.” And that has held very true over the years.

When I first started flying 30 years ago we had four major tools to use to maintain separation from other airplanes. The first and most important were our eyes. VFR pilots (like me) fly under those "visual flight rules" which basically mean see and avoid other airplanes. At uncontrolled airports you can fly without a radio and do all your navigation and separation with just your eyes, and for decades after that first flight in North Carolina, that's what all pilots did. And back then there were so few airplanes in the air that it usually worked. But not always. The first midair collision happened at the Milano Circuito Aereo Internazionale airshow in 1910, between a monoplane and a biplane. Both pilots survived, due both to luck and the slow airspeeds that aircraft were capable of at the time.

In 1915 pilots got the second tool that would turn out to be invaluable - radio. With radios, pilots could not only talk to each other, they could talk to the ground. At first this just meant getting weather reports and updates on the runway, but soon ground based controllers were able to talk to pilots and give them advice on where to go and how to remain separated. This is important in part because pilots do not have a 360 degree field of view; parts of the airplane block their sightlines. This is why, for example, collisions between low-wing and high-wing airplanes are more common when the low-wing is above the other aircraft; the top pilot cannot see down past the wing, and the bottom pilot cannot see up through his wing. Now a controller could easily see both and tell them how to remain separated.

The next tool we got was radar. First developed during World War II to find enemy aircraft, it was first used for air traffic control in 1952 in Washington, DC. Now controllers did not have to physically see aircraft - the radar would show them where, and how far away, they were. This was quickly improved through transponders, so the air traffic controller could not only see a blip on his screen, but also had information about the plane's ID and altitude.
The most recent innovation has been ADS-B / ATAS. This is an autonomous system that can be installed in any aircraft to both transmit the location of the aircraft and determine if a close call will occur. ADS-B "out" has been mandatory in the US in most US airspaces since 2020. It's a big advance in pilot traffic awareness, since warnings no longer have to go through air traffic control, and you will immediately see any traffic that's heading towards you if you have the right receiver. Larger aircraft even have the ability to autonomously avoid other aircraft based on information from a similar system called TCAS.

But with all these systems, it still often comes down to what you can see - and even modern air traffic control uses this regularly. As I was flying back to the airport yesterday the tower said "One one bravo, you're number two following a Cherokee at your 11, base to final at 1200 feet, report traffic in sight." Which in plain english means there's a small plane to my left just below my altitude, and I should tell the tower when I see them. I strained to find the tiny dot against all the houses, buildings and cars but could not see anything. Ten seconds later I got another "11B, report traffic in sight." At that point I saw the Cherokee a few miles away, just finishing the turn onto final. "11B has traffic in sight," I reported. They immediately replied with "11B follow the Cherokee, clear to land 24." Which means I could now land on runway 24 without any additional communication.

This whole exchange happened because even though air traffic control can see us both on radar (and likely out the window at this point) the primary method of clearance in VFR is always the pilot's eyes. And once I could see him, the odds of me accidentally turning into him or overtaking him on final were very low - and the controller could turn his attention to other things.

I was thinking about this as I was going over the transcript from air traffic control from yesterday's crash. The relevant sections are these:

15:50 (ATC) PAT25 traffic just south of Wilson Bridge CRJ is 1200 feet for runway 33
17:30 (ATC) PAT25 do you have the CRJ in sight?
17:32 (ATC) PAT25 pass behind the CRJ
18:05 (Unknown) Tower did you see that?

Again in English, at 15 minutes and 50 seconds into the recording, the airport tower (responsible for all traffic in about a 5 mile radius) first tells PAT25 - the helicopter - that they have jet traffic that they might conflict with. We can't hear the helicopter's response (they may have been using a different frequency to respond) but we can infer their replies from what the air traffic controller was saying.

Their reply to the first ATC transmission was likely something like "PAT25 looking for traffic" since at night in a crowded airspace, there's a lot to look at before you find the one bit of moving light that is your traffic.

Two minutes later, ATC asks them again because it's been a while and they are getting closer. ATC is getting more insistent because if they CAN'T see them they can't reasonably avoid them, even if they have ADS-B or a similar system. And if they still could not see them after those two minutes, the next transmission from ATC would have likely been an avoidance instruction, something like "PAT25 turn left heading XX, maintain YY feet" to ensure they were clear.

However PAT25 likely replied that the traffic was in sight, because the next transmission from the controller, two seconds later, is "pass behind the CRJ." (CRJ stands for Canadair Regional Jet, the sort of aircraft American 5342 was.) At this point, once the controller knows that they have the aircraft in sight, they can see and avoid the aircraft themselves.

About 30 seconds later another pilot says "did you see that?" indicating the collision.

So it's way too early to determine exactly what happened and why, but from the ATC recordings it sounds like ATC was relying on the helicopter to see and avoid the CRJ, based on their visual contact with it. And that's not at all unusual for helicopters operating in busy airspaces, since they often operate near airports but below approach/departure corridors.

The politics has already started, unfortunately. Trump has blamed DEI, Barack Obama and Joe Biden respectively for the crash, and there is currently no FAA administrator (or deputy administrator) after Elon Musk forced out the former administrator, Mike Whitaker. Hopefully these impediments will not have too much of an effect om the investigation.
 
  • Like
Reactions: (Q)
The actual cause of the crash, it appears, is the helicopter pilot being told there is traffic nearby, and verifying that he sees it, when it appears he saw the next airline back in the pattern rather than the one almost on top of him.

The problem is that situation shouldn't be allowed to happen that close to the airport when separation was only vertical and not horizontal as well. Usually helicopter traffic is kept away from the active runway and there is usually a marked heliport removed from the runway area.

So the helicopter is never in the same vertical airspace as the airplane. That doesn't appear to be the procedure here. If the procedure is just relying on vertical separate that close to the airport, that policy is going to change after this I'm sure.
 
Apparently, the helicopter was on a 'dark' training mission in which certain electronics are turned off, specifically the one that shows other aircraft your position.
 
The actual cause of the crash, it appears, is the helicopter pilot being told there is traffic nearby, and verifying that he sees it, when it appears he saw the next airline back in the pattern rather than the one almost on top of him.
So not "DEI hiring policies?" That's a start I suppose.
 
Just out of curiosity, do commercial airline pilots all know how to operate, say, a Cessna prop plane? They are very different beasts, of course, but does pilot training for commercial airline type planes necessarily cover operation of these very different types of planes?
 
Just out of curiosity, do commercial airline pilots all know how to operate, say, a Cessna prop plane? They are very different beasts, of course, but does pilot training for commercial airline type planes necessarily cover operation of these very different types of planes?
The flight principals are the same but each type requires specific training and checkouts. I can fly a private (Cessna) basic airplane and I have a private and commercial certificate for small helicopters. I couldn't go and fly a Blackhawk without expensive and extensive checkouts on that type.

Most airline pilots probably started out with Cessna's so they could fly those. Most people who can drive a Corolla could drive an Indy car but not without more training and checkouts. The principals are the same however.

The differences are just in systems and complexity. Most flight training is less about flying and more about "what ifs" and training for emergencies. On the other hand, when I went from airplanes to helicopter training I though at least my prior experience would be a great benefit. It's not really. Each skill needed is different enough that there is little carry-over.

More basic aircraft are generally more hands-on than more complex aircraft. None of it is particularly hard, just different.
 
Last edited:
Most airline pilots probably started out with Cessna's so they could fly those. Most people who can drive a Corolla could drive an Indy car but not without more training and checkouts. The principals are the same however.
As far as comparisons go, I wonder if going from a Cessna to a commercial airline is more like going from a regular automobile to an Indy car or to an 18 wheeler? It seems like there are elements of both in the gap--speed and scale--but I don't how the comparison works with respect to basic operation and handling. I've flown plenty of flight simulators which, no matter how sophisticated, are still a pretty far cry from the panel of an actual plane.
 
As far as comparisons go, I wonder if going from a Cessna to a commercial airline is more like going from a regular automobile to an Indy car or to an 18 wheeler? It seems like there are elements of both in the gap--speed and scale--but I don't how the comparison works with respect to basic operation and handling. I've flown plenty of flight simulators which, no matter how sophisticated, are still a pretty far cry from the panel of an actual plane.
Yeah, the 18 wheeler would probably be more apt. A Cessna 152 is more responsive than even a 172 and an airliner is going to be lumbering by comparison.

I've never been in a commercial flight simulator and I haven't messed with one on a personal computer for decades but back in the day I played around with MS Flight simulator and it was harder to land than the real thing and the feeling is missing of course and the visual aspect of turning your head, feeling gravity forces on your body etc. is missing.

I know that some people have a lot of hardware and a setup where it's a more realistic experience I'm sure. My thoughts at the time was just that it was a good approximation of how the nav radios worked and it was better than just reading about those things. Actual flying in a landing pattern is about reducing throttle, putting on carb heat, flaps, turning your head as you turn base, feel, etc.

I do remember the one time where landing wasn't too hard was when I just forgot about the computer and treated all techniques and settings as if I was in the actual airplane and I landed it a lot more easily. The real thing was still easier though. Especially back in the day with hitting keys for throttle and flaps and for switching views, it just had no natural feeling at all.

I think the helicopter simulators probably feel more realistic other than hovering.
 
The remarkable safety achieved in aviation continues to astonish people although risks persist in aviation operations as you stated. A simple impairment in pilot vision functions as the deciding factor for entire aviation operations despite technological advancements. Everyone needs to keep a high level of attention because safety depends on continuous awareness for collision prevention. The recent plane crash makes everything clear now. People maintain a political agenda regarding this incident rather than concentrating efforts on identifying the root cause through analysis and subsequent use of acquired knowledge. The investigation needs to deliver answers to explain present safety concerns and implement new protective procedures.
 
The remarkable safety achieved in aviation continues to astonish people although risks persist in aviation operations as you stated. A simple impairment in pilot vision functions as the deciding factor for entire aviation operations despite technological advancements. Everyone needs to keep a high level of attention because safety depends on continuous awareness for collision prevention. The recent plane crash makes everything clear now. People maintain a political agenda regarding this incident rather than concentrating efforts on identifying the root cause through analysis and subsequent use of acquired knowledge. The investigation needs to deliver answers to explain present safety concerns and implement new protective procedures.
Well thank you, the Sirius Cybernetics Corporation, for that extraordinary insight.:biggrin:
 
Unfortunately, impossible to ask the helicopter pilot if -- like a horse with blinders on -- he was looking purely straight ahead at an empty area that was suddenly filled by an airliner.
Yeah, Black Hawk didn't divert even at the last second. Must have been looking out the wrong window.
 
The problem is that situation shouldn't be allowed to happen that close to the airport when separation was only vertical and not horizontal as well. Usually helicopter traffic is kept away from the active runway and there is usually a marked heliport removed from the runway area.
There is a marked helicopter route that goes right under the approach. The map is here:


The helicopter was traveling south, going from Route 1 to Route 4. The altitude restrictions are listed on the map. When you are heading south on Route 1, max is 200 feet when you pass over Memorial Bridge, just to the north of the airport. But when passing over James Creek Marina you are allowed to climb to 300 feet. Once you start on Route 4, you have to descend to 200 feet.

The collision happened just south of James Creek Marina on Route 4 at ~280 feet. Had the helicopter (legally) climbed to 300 feet over James Creek, then started descending after entering Route 4? I am sure that's one of the things they will look at.
 
People maintain a political agenda regarding this incident rather than concentrating efforts on identifying the root cause through analysis and subsequent use of acquired knowledge.
That's not entirely true, is it? I mean "rather" suggests that they're doing one thing instead of the other, but in this case the politics is what Trump is doing, while the investigations are being done by people who actually know what they're doing. Trump will undoubtedly put a spin on the eventual report, and where possible he will ensure that DEI-hires and the DEI-policies have some of the blame.
But my point is that politics is the game people play, not "rather than" the investigations, but while they don't have facts and while they can therefore score political points based on lies, half-truths, guesses, and wishful thinking. Some would hope that the "politics" of such an incident would initially be thoughts for those who lost their lives, and a call for calm while the root causes are investigated. And I'm sure that would have happened under almost every other President of the past 50 years. Just not Trump.

As to what the cause is, I can imagine the flight-tower people did what they were expected to do. But if so then I am surprised there was no process (followed) to ensure that they were confirming the same a/c that the tower was warning them about. Just to say that they see "it" doesn't mean that "it" is what the tower are referring to.
 
Apparently, the helicopter was on a 'dark' training mission in which certain electronics are turned off, specifically the one that shows other aircraft your position.
There are two systems that allow that. One is TCAS which is what all large aircraft now use; it provides both TA's (advisories) and RA's (tells you what to do) to avoid collisions. The second is ADS-B out, which all aircraft in this sort of airspace now have to use. There's no real operational reason to turn either off no matter what sort of training you are doing, since they do not affect normal navigation or aircraft control. But sometimes military vehicles get waivers for operation without those systems, based on the time it takes to upgrade military aircraft.

It should be noted that there is a third system, the transponder, which plays a role in TCAS - but its primary function is to provide ATC information about the aircraft's altitude. Both aircraft had this operating.

That leads to another issue that the NTSB will be looking into. Radar returns, based on Mode C and Mode S transponder returns, showed the helicopter at 200 feet, and the CRJ at 300 feet. Normally these systems are accurate to +/- 25 feet. One of the first things that ATC tells an aircraft when they enter a new airspace is something like "American 128, radar contact, cleared to LOGGI at 3000 feet, altimeter 29.92." That altimeter setting is critical. Altimeters use air pressure to determine altitude, and thus the altimeter has to be set to the reference air pressure at the airfield. (Barometric pressure, of course, changes with the weather.) If the altimeter is set incorrectly it could easily be several hundred feet off, and the pilot might think he is at the correct altitude when in fact he's 100 feet or more off.
 
Last edited:
As far as comparisons go, I wonder if going from a Cessna to a commercial airline is more like going from a regular automobile to an Indy car or to an 18 wheeler? It seems like there are elements of both in the gap--speed and scale--but I don't how the comparison works with respect to basic operation and handling. I've flown plenty of flight simulators which, no matter how sophisticated, are still a pretty far cry from the panel of an actual plane.
It goes both ways. I've flown level-4 737 simulators (full motion, full vision, 100% accurate avionics etc) and there are some things that are easier, some things that are way harder. With smaller aircraft you have to be light on the controls because they are very responsive. The instructor in the 737, by comparison, encouraged me to "stir the pot" on final - use full control deflections to keep wings level and airspeed at the reference.

But in a small aircraft it's not hard to be "in front" of the airplane. On final often you get to a point where you're just chugging along, waiting to get to the runway, with lots of time to watch airspeed, look for traffic, adjust for crosswinds etc. In the larger aircraft you are going at least 2x, if not 3x, faster on final - and things happen much, much faster. For me it took a lot of concentration to not get behind the aircraft. And when that happens you start reacting instead of controlling.
 
That's not entirely true, is it? I mean "rather" suggests that they're doing one thing instead of the other, but in this case the politics is what Trump is doing, while the investigations are being done by people who actually know what they're doing.
Agreed. I am most concerned that Trump will issue an executive order instructing the NTSB to determine that it was DEI that caused the crash. He has already issued a similar executive order claiming that DEI caused the crash.
 
However, they both rely on the altimeter setting. One of the first things that ATC tells an aircraft when they enter a new airspace is something like "American 128, radar contact, cleared to LOGGI at 3000 feet, altimeter 29.92." That altimeter setting is critical. Altimeters use air pressure to determine altitude, and thus the altimeter has to be set to the reference air pressure at the airfield. (Barometric pressure, of course, changes with the weather.) If the altimeter is set incorrectly it could easily be several hundred feet off.
I always wondered how aircraft could use barometric pressure as an altitude metric. Always seemed error-prone to me.

But of course, the airport will be setting their baseline barometric pressure when they enter the airspace.
 
I always wondered how aircraft could use barometric pressure as an altitude metric. Always seemed error-prone to me.
ADS-B attempts to solve this by sending both barometric pressure altitude and GPS altitude, but I don't know which is used by ATC displays, or indeed if they use ADS-B over Mode C/Mode S at all. MOST Mode C systems send uncorrected (i.e. unadjusted) altitude, and it's corrected by ATC computers based on the measured pressure at the field.

But any system that uses barometric pressure is subject to some error, since many things affect air pressure as read by an aircraft vs on the ground.
 
Back
Top