American Airlines crash at Reagan international 29 Jan 2025

[OldSolduer]

I’ve seen a few things on TV and experts weighing in.
We can never ignore the fact as humans we can become tunnel visioned during periods of stress - like landing a jet plane.

 

[YZT580]

…and if you miss something outside because you’re focusing inside on the scope? Where does second guessing stop? The controller gave conditions for continuing forward and the pilot positively acknowledged. That is not at all an unreasonable scenario. The impact came 5-6s after that communications, the very same 5s you say above is all it takes to make sure things are good. The two aircraft were 900’ apart when the pilot confirmed visual on the aircraft.

You are correct. It is a reasonable scenario but unfortunately it is the wrong one, not only in this case but in many others as well even when there is no report to write up. VFR is see and be seen but in all complex environments it come with limitations. Controllers must compensate by providing additional safeguards alongside the rulebook and you should never give up one form of separation until another form is assured to your satisfaction. An example would be two a/c converging to the downwind on STAR tracks. One is descending to 10,000 the second to 9,000 as per the star which builds in a buffer as a matter of safety. You will often hear the controller clear both to the same altitude sacrificing the altitude safety to eliminate having to make a second transmission. Should there be a conflict his plan is to alter one of their headings. Works well, saves time until the rare occasion when the controller is distracted. ATC has encouraged slackness in the guise of expediting traffic.
there are very few controllers around who still focus out the window. Some focus almost entirely on the scope. The better ones alternate between the two as it is almost impossible to gauge distance between a/c and closing headings visually yet looking out the window gives you better situation awareness. Regardless of the pilot's comments, the predicted track would have indicated the intercept. The altitude readout would have shown conflict. The controller ignored rule number 1.

 

[Good2Golf]

You are correct. It is a reasonable scenario but unfortunately it is the wrong one, not only in this case but in many others as well even when there is no report to write up. VFR is see and be seen but in all complex environments it come with limitations. Controllers must compensate by providing additional safeguards alongside the rulebook and you should never give up one form of separation until another form is assured to your satisfaction.

And so deliberately deviate from MANOPS?

An example would be two a/c converging to the downwind on STAR tracks. One is descending to 10,000 the second to 9,000 as per the star which builds in a buffer as a matter of safety. You will often hear the controller clear both to the same altitude sacrificing the altitude safety to eliminate having to make a second transmission.

Haven’t heard that to be honest, but that’s a deviation by the controller. Frankly, that control should not be issued as you describe.

Should there be a conflict his plan is to alter one of their headings. Works well, saves time until the rare occasion when the controller is distracted. ATC has encouraged slackness in the guise of expediting traffic.
there are very few controllers around who still focus out the window. Some focus almost entirely on the scope. The better ones alternate between the two as it is almost impossible to gauge distance between a/c and closing headings visually yet looking out the window gives you better situation awareness.

So what dwell time between scope and outside should be used?

Regardless of the pilot's comments, the predicted track would have indicated the intercept. The altitude readout would have shown conflict. The controller ignored rule number 1.

I see, so controller is more accountable for actions not taken the pilots for actions taken?

 

[YZT580]

And so deliberately deviate from MANOPS?

not a deviation. Pearson for example has/had a tower radar position with a job description including vectors that were binding in the zone. MANOPS provides your basic thou shalt and shalt not processes that are supplemented by unit procedures

Haven’t heard that to be honest, but that’s a deviation by the controller. Frankly, that control should not be issued as you describe. It is indeed a deviation but events like that occur frequently. Controllers amend the profiles all the time to expedite traffic. How often have you heard the expression SID is cancelled, followed by a change in heading, altitude or speed? Altering a STAR is no different. In the case I described the controller would say that his secondary plan is to alter the heading if required while the safest course would be to wait until longitudinal separation was established and then issue the same altitude.

So what dwell time between scope and outside should be used? As required. What has happened is that controllers have become overly reliant on tools to cover up bad control techniques. Plus it is really easy to sit in the high chair looking down on the scope as lord and master of all you survey. When you are driving, how much attention do you pay to your mirrors when reversing and how much are you looking at the backup display? We tend to focus on the technology rather than paying attention out the window. The display focuses on a specific point looking over your shoulder or at the mirrors will help you see the lady with the shopping cart converging from the side.


I see, so controller is more accountable for actions not taken the pilots for actions taken? Exactly

 

[mariomike]

Sadly, Trump laid the blame on Biden and Obama DEI hiring policies. That is completely beyond the Pale, but in keeping with the actions of a sociopath with Center of Attention Deficit Disorder. Remains were still being recovered while he politicized a tragedy.

The DEI angle was discussed with the Transportation Secretary,

 

[KevinB]

More data comes out of the flight paths, specifically altitude.

Remains of all 67 victims of the midair collision near DC recovered as NTSB probes altitude data

Investigators said they need to get more information from the still-submerged Black Hawk to verify the data.

apnews.com

Plane was at 325’ and descending.
Helicopter was reported at 300’ but rounded to nearest 100’, it is visually below the plane at impact.
- Helicopter was not to exceed 200’

For the pilots out there - where is the altitude reading taken from ? Bottom, top, or middle of the hull?

Now even if the Blackhawk was at (or below) 200’ that doesn’t seem to offer much of a safety margin ~100’ of elevation.

 

[SeaKingTacco]

If on Radar Altimeter- bottom of the hull (where the antenna is). If barometric altimeter, about mid hull (not that any aircraft manufacturer will warranty you that kind of accuracy.

 

[KevinB]

Caveat, I’m just a has been GIB:

But as I understand it while the Blackhawk is 16’10” tall that due to rotor span that min safety is 200’ due to rotors (formation flying). While I’ve been in some that have done tighter landings (simply due to available space) that was operational, and it appeared that we had an over watch bird (either another MH-60, or a MH-6) helping deconflict the landings (and give support fires).

Even if the Blackhawk was a 200’ the AE jet at 325’ would be well within a danger space.

Now I’m not anyway conversant as to what ATC does, but to me it seems that the 60 should have been held at a specific location and told to transition the zone after the jet had passed?

 

[Good2Golf]

Plane was at 325’ and descending.
Helicopter was reported at 300’ but rounded to nearest 100’, it is visually below the plane at impact.
- Helicopter was not to exceed 200’

For the pilots out there - where is the altitude reading taken from ? Bottom, top, or middle of the hull?

Now even if the Blackhawk was at (or below) 200’ that doesn’t seem to offer much of a safety margin ~100’ of elevation.

Caveat, I’m just a has been GIB:

But as I understand it while the Blackhawk is 16’10” tall that due to rotor span that min safety is 200’ due to rotors (formation flying). While I’ve been in some that have done tighter landings (simply due to available space) that was operational, and it appeared that we had an over watch bird (either another MH-60, or a MH-6) helping deconflict the landings (and give support fires).

Even if the Blackhawk was a 200’ the AE jet at 325’ would be well within a danger space.

Now I’m not anyway conversant as to what ATC does, but to me it seems that the 60 should have been held at a specific location and told to transition the zone after the jet had passed?

Where is the altitude taken? Talking barometric it’s taken from the aircraft’s pitot-static system.

On a Black Hawk, there are two pitot-static tubes on the upper forward fuselage (est. 8’ above ground/4’ below main mast height), either side of the forward fairing:



For altitude reporting, ADS-B Out (automatic dependent surveillance - broadcast) transmits uncorrected pressure altitude, ie. The barometric pressure altitude for a standard 29.92”hg or 1013mbar (at mean sea level) air mass. It does so in 25’ increments@, hence why when your flying on a plane and watching the flight data in the entertainment system, you’ll see your altitude change every 25’.

The important point is that all aircraft are transmitting the same uncorrected (or perhaps more correctly, ‘standardized’ to 29.92/1013) pressure altitude, so the reported altitudes of all aircraft would be comparable (less minor variations of reporting, based on each aircraft’s pitot-static system.

 

[YZT580]

More data comes out of the flight paths, specifically altitude.

Remains of all 67 victims of the midair collision near DC recovered as NTSB probes altitude data

Investigators said they need to get more information from the still-submerged Black Hawk to verify the data.

apnews.com

Plane was at 325’ and descending.
Helicopter was reported at 300’ but rounded to nearest 100’, it is visually below the plane at impact.
- Helicopter was not to exceed 200’

For the pilots out there - where is the altitude reading taken from ? Bottom, top, or middle of the hull?

Now even if the Blackhawk was at (or below) 200’ that doesn’t seem to offer much of a safety margin ~100’ of elevation.

it isn't and in addition, if American was circling visually he could easily have been significantly lower. Honestly, if I were writing the procedure I would have the helicopter track crossing over the threshold of the landing runways at no lower than 750. your comment re: to hold and cross behind is valid but with the amount of helicopter traffic in the zone isn't a viable solution. With two positions the radar guy would have to coordinated each crossing.

 

[Good2Golf]

The aviator-navigate-communicate challenge here is that a circling approach to 33 definitely overflies HELO4. As far as I can track things, AA5342 was originally lined up for 01, I can’t recall if it was IFR or already cleared visual (ie. Cancelled their IFR clearance and continued VFR for the circling approach to 33).

HELO4 really is set up to work with 01 arrivals and 19 departures. 33 arrival or 15 departure tracks definitely cross HELO4. When I flew the HELO4 down to KDAA, there wasn’t any traffic landing at DCA, so I can’t comment on what should happen, other than to surmise it would have been similar to the direction ATC have PAT25, to maintain visual separation and proceed behind the CRJ.

As a ‘semi’-reference (for context, not trying to recreate the respective approach paths), attached is the RNAV/GPS approach to DCA RWY33. This is an IFR approach and isn’t what was being flown by AA5342 that might, but it gives some idea about route and altitude and space.

For an IFR approach, the aircraft would follow the routing via GPS ( via KATRN, NADSE and IDTEK waypoints.) descending to the MDA(minimum descent altitude) of 680’ (corrected pressure altitude) until such time as the aircraft could see the runway environment and descend safely. If it could do so, it would fly a descent path roughly equating to a 3.0-3.1° glide path. To work backwards with altitudes, crossing the threshold crossing height (the TCH on the plate) of RWY33 at 37-40’ (depending if the pilots followed the GPS-driven flight director or followed the visual glide slope indicator (VGSI), back out 1/2mi and a 3° descent path would be up at 189-192’ altitude (37/40 + 6080’mi*1/2mi*3°/60°mi) and 1mi back would be up at 341-344’ (about where the accident occurred).

 

[YZT580]

The aviator-navigate-communicate challenge here is that a circling approach to 33 definitely overflies HELO4. As far as I can track things, AA5342 was originally lined up for 01, I can’t recall if it was IFR or already cleared visual (ie. Cancelled their IFR clearance and continued VFR for the circling approach to 33).

HELO4 really is set up to work with 01 arrivals and 19 departures. 33 arrival or 15 departure tracks definitely cross HELO4. When I flew the HELO4 down to KDAA, there wasn’t any traffic landing at DCA, so I can’t comment on what should happen, other than to surmise it would have been similar to the direction ATC have PAT25, to maintain visual separation and proceed behind the CRJ.

As a ‘semi’-reference (for context, not trying to recreate the respective approach paths), attached is the RNAV/GPS approach to DCA RWY33. This is an IFR approach and isn’t what was being flown by AA5342 that might, but it gives some idea about route and altitude and space.

For an IFR approach, the aircraft would follow the routing via GPS ( via KATRN, NADSE and IDTEK waypoints.) descending to the MDA(minimum descent altitude) of 680’ (corrected pressure altitude) until such time as the aircraft could see the runway environment and descend safely. If it could do so, it would fly a descent path roughly equating to a 3.0-3.1° glide path. To work backwards with altitudes, crossing the threshold crossing height (the TCH on the plate) of RWY33 at 37-40’ (depending if the pilots followed the GPS-driven flight director or followed the visual glide slope indicator (VGSI), back out 1/2mi and a 3° descent path would be up at 189-192’ altitude (37/40 + 6080’mi*1/2mi*3°/60°mi) and 1mi back would be up at 341-344’ (about where the accident occurred).

View attachment 91041

With a visual to 33 I would actually have expected AAL to be dropping down as much as he was able, seeing as it is a significantly shorter runway which would cause further chances of confliction

 

[Good2Golf]

With a visual to 33 I would actually have expected AAL to be dropping down as much as he was able, seeing as it is a significantly shorter runway which would cause further chances of confliction

I’m not sure about that, there are additional restrictions for noise abatement etc. since final on 33 directly overflies the PMQs of Joint Base Anacosta-Bolling.

 

[SeaKingTacco]

IIRC the ATC tapes, AA5342 was still on an IFR approach and did not cancel to go visual (you would be an idiot to cancel and go visual at that airport, at night).

 

[Good2Golf]

IIRC the ATC tapes, AA5342 was still on an IFR approach and did not cancel to go visual (you would be an idiot to cancel and go visual at that airport, at night).

I don’t know. I heard them accept the visual approach to 33 immediately prior to ATC revising their clearance to land 33. They could only do that visually, otherwise they’d need to conduct a missed approach and re-sequence for an IF approach, the RNAV (GPS) 33. I don’t know enough details of FAA ATC MANOPS to say whether a visual approach can be conducted while still on an active IFR clearance. That couldn’t happen in Canada because an IF approach clearance also has the Missed Approach tied to it. At DCA, the MA for 33 isn’t the same as the MA for 01, so if they couldn’t make the landing on 33, they are essentially without a valid Missed Approach. I don’t know…

An interesting video gives a bit more detail on altitudes, and also includes a near miss with another PAT UH-60 the day prior, where a United Embraer 175 coming from the North to land 19, actually conducted a go around because of receiving and complying with a TCAS resolution advisory resulting from that UH-60 climbing up to at least 700’ nearing the Embraer’s track.

 

[Mick]

I don’t know. I heard them accept the visual approach to 33 immediately prior to ATC revising their clearance to land 33. They could only do that visually, otherwise they’d need to conduct a missed approach and re-sequence for an IF approach, the RNAV (GPS) 33. I don’t know enough details of FAA ATC MANOPS to say whether a visual approach can be conducted while still on an active IFR clearance. That couldn’t happen in Canada because an IF approach clearance also has the Missed Approach tied to it. At DCA, the MA for 33 isn’t the same as the MA for 01, so if they couldn’t make the landing on 33, they are essentially without a valid Missed Approach. I don’t know…

An interesting video gives a bit more detail on altitudes, and also includes a near miss with another PAT UH-60 the day prior, where a United Embraer 175 coming from the North to land 19, actually conducted a go around because of receiving and complying with a TCAS resolution advisory resulting from that UH-60 climbing up to at least 700’ nearing the Embraer’s track.

I haven't been able to find any recordings of AA5432's original approach clearance for runway 01, but just for clarification, being re-cleared for the visual 33 would not require the crew to cancel IFR, nor would that be common practice for a crew to do.

Once a visual approach has been cleared, the aircraft is still on its IFR flight plan, but is authorized to proceed visually to the landing runway. IFR has not been cancelled, despite the crew taking responsibility for visual separation from any preceding acft they are directed to follow, and navigating to the final approach course etc.

Once the visual approach is cleared, as you pointed out, there is no associated MA procedure. If a go-around is required, tower will give an appropriate heading / altitude instruction before handing off to terminal controllers.

I fly quite frequently to a variety of controlled US airports, where visual approaches are the norm (and preferred) and have never cancelled IFR. Go-arounds (almost always due to traffic) have occurred as I describe above.

That said, it blows my mind that an aircraft on short final wouldn't be cause for ATC to keep positive control of PAT25, despite their (approved) request for visual separation.

 

[Good2Golf]

@Mick, thanks for that. I’ve never had a changed IF approach without ATC issuing a new MA procedure with the revised approach clearance…although it was usually “climb runway heading to XXXX’, expect radar vectors.”

I agree with you that there was/is a definite issue with what was allowed to happen. I’ve flown Route 4 once at night, and it is definitely supposed to be on the East side of the Potomac…we stayed well below until South of National Harbor and the Woodrow Wilson bridge, before even thinking of getting anywhere near 200’, let alone crossing over the Potomac to get back into KDAA’s airspace.

 

[Mick]

@Mick, thanks for that. I’ve never had a changed IF approach without ATC issuing a new MA procedure with the revised approach clearance…although it was usually “climb runway heading to XXXX’, expect radar vectors.”

I agree with you that there was/is a definite issue with what was allowed to happen. I’ve flown Route 4 once at night, and it is definitely supposed to be on the East side of the Potomac…we stayed well below until South of National Harbor and the Woodrow Wilson bridge, before even thinking of getting anywhere near 200’, let alone crossing over the Potomac to get back into KDAA’s airspace.

From what I have read, it sounds like the change to 33 was initiated after the approach to 01 had commenced. Reprogramming and re-briefing a new approach would definitely require discontinuing the approach to 01, and probably being vectored to set up for approach to 33.

A late visual approach to 33 would be busy, but wouldn't require as much time to reprogram.

Had the weather / visibility been worse, this accident probably wouldn't have happened, as the visual wouldn't have been requested / offered IMHO.

 

[YZT580]

I don’t know. I heard them accept the visual approach to 33 immediately prior to ATC revising their clearance to land 33. They could only do that visually, otherwise they’d need to conduct a missed approach and re-sequence for an IF approach, the RNAV (GPS) 33. I don’t know enough details of FAA ATC MANOPS to say whether a visual approach can be conducted while still on an active IFR clearance. That couldn’t happen in Canada because an IF approach clearance also has the Missed Approach tied to it. At DCA, the MA for 33 isn’t the same as the MA for 01, so if they couldn’t make the landing on 33, they are essentially without a valid Missed Approach. I don’t know…

An interesting video gives a bit more detail on altitudes, and also includes a near miss with another PAT UH-60 the day prior, where a United Embraer 175 coming from the North to land 19, actually conducted a go around because of receiving and complying with a TCAS resolution advisory resulting from that UH-60 climbing up to at least 700’ nearing the Embraer’s track.

A visual approach is an IFR approach done with reference to the ground. You do not have to cancel. The weather limits are to remain clear of cloud and in ground contact and yes, G2G you can do it in Canada and at any other facility that authorises visuals

 

[Good2Golf]

A visual approach is an IFR approach done with reference to the ground. You do not have to cancel. The weather limits are to remain clear of cloud and in ground contact and yes, G2G you can do it in Canada and at any other facility that authorises visuals

Tracking YZT, in Canada, but I wasn’t sure about the US, other than I remember while conducting an IF approach to LGA in a helo, the controller asked if I wanted a visual approach (min’s just supported an acceptance, but I wasn’t thrilled with having my MA cancelled in the airspace), so I asked what the intended MA directions would be, and he said it wouldn’t be a problem, runway heading to XXX then vectors, so I accepted the visual. In the DCA case, I think the hair on the back of my neck would be standing up as the helo, knowing the pinch point at BOLLING, and having done the daytime routing on 4 earlier and knowing there are no obstacles between Hains Point and the Woodrow Wilson Bridge, I’d be ‘comfortably below’ 200’.