Jump to content

SFO Incident


UpperDeck

Recommended Posts

12 hours ago, DEFCON said:

 

"Re... 100 feet... when they landed on the return, they were probably much closer to the traffic on the taxiway than 100 ft... wingtip to wingtip, and at the same altitude"

Not likely at all; I think someone would have to be on the grass to get that close to another ac.

 

 

Ok... probably a bit of an exaggeration on my part to illustrate a point... aircraft are often much closer to each other in quite normal circumstances than we think. We don't actually know the minimum altitude of the approaching aircraft.... 100 ft is the newspaper number at this point.

However, the minimum distance between a runway centerline and the hold short line is 250 ft. An A380 with a 262 foot wingspan would be within 119 ft of that hold line. The minimum distance between taxiway and runway centerline for an airport like SFO is 400 ft, so two design group 6 aircraft (B747-8 or A380) could be within 138 feet of each other if both were right on centerline... a bit of drift and they could easily be close to 100 ft from each other. This is obviously not the case here, but shows how close landing and taxiing aircraft can be to each other. 

Link to comment
Share on other sites

  • Replies 186
  • Created
  • Last Reply
9 hours ago, Zan Vetter said:

When tower called go around the response from AC759 was "...IN the go around." I take that to mean they had already started the procedure.

This is obviously the case (at least based on AvHerald info), as ATC did not direct the go around until it had overflown the taxiway by .25nm... 1500 ft. For the benefit of non-pilots here, normal touchdown is about 1000 ft. So, clearly, go around was initiated well before ATC's intervention.

Link to comment
Share on other sites

10 hours ago, GDR said:

Perhaps the "best guess" explanation of the events. Both the LDA and Rnav approaches for 28R are offset north of the runway centreline. If you delayed the visual segment (left turn for runway centreline) of the LDA approach it might appear that you're lined up for "C" taxiway. Those UA pilots may be buying the next round.

 

IMG_5916.PNG

IMG_5917.PNG

Link to comment
Share on other sites

10 hours ago, GDR said:

Good explanation, however, with RWY 28L closed I don't think that there would be any need for the LDA DME 28R to be used, rather the ILS 28R. Maybe with all the waiting aircraft wait on taxiway C with their taxi lights on could cause some confusion.  Also with the 30Kt NW xwind the Air Canada 320 would have had a good crab giving the pilots on the ground the illusion that the 320 was pointed right at them. I'm not sure if Air Canada has FDM, a go-around probably would have recorded the event.

Link to comment
Share on other sites

Blues.....

It appears that you are adopting the remarks of the consultant interviewed on CTV. Is that correct?

Personally, I had some trouble with his explanation but I readily admit my ignorance.

I will note that the heading of AC759 on the first approach was 3 degrees off the heading on the landing.

Gator....again, acknowledging my ignorance.....1) " in go around" and " going around" seem synonymous. There was very little time elapsed between the initial exchange, United comment and AC acknowledgement; 2) the taxiway is about 1000 feet longer than the runway...is that correct? The waiting aircraft would be some distance down that taxiway, wouldn't they?; 3) by my calculation, 18 seconds elapsed between the inquiry by the pilot about lights and the statement " in go around". At 150 kts., the aircraft would have travelled 4600 ft (+/-) in that time. Am I in the ballpark?

Link to comment
Share on other sites

Upper Deck, Re

38 minutes ago, UpperDeck said:

2) the taxiway is about 1000 feet longer than the runway...is that correct? The waiting aircraft would be some distance down that taxiway, wouldn't they?; 3) by my calculation, 18 seconds elapsed between the inquiry by the pilot about lights and the statement " in go around". At 150 kts., the aircraft would have travelled 4600 ft (+/-) in that time. Am I in the ballpark?

Here's the actual taxi chart of KSFO

IMG_0193.PNG

 

Taxiway and Runway are same length.  Aircraft cross threshold 35-50 ( normally 35 feet) above the end of the runway.  If these pilots actually hadn't started going around before being .25 miles mile past the end of the runway, they would already have been on the ground.

The goaround was absolutley initiated by the AC pilots prior to ATC waking up to the clues being handed to him.

 

Link to comment
Share on other sites

37 minutes ago, UpperDeck said:

Blues.....

It appears that you are adopting the remarks of the consultant interviewed on CTV. Is that correct?

Personally, I had some trouble with his explanation but I readily admit my ignorance.

I will note that the heading of AC759 on the first approach was 3 degrees off the heading on the landing.

Well I do believe his comments about what "may" have occurred makes the most sense at this point. However, he doesn't have access to any more detailed information than anyone else.

And as mentioned on page 4 of this thread perhaps not an LDA but certainly an RNAV. The RNAV (X) approach has a similar path prior to runway alignment.  

I know many pilots make a point of turning off annoying taxi/landing lights when close to an active runway and pointed towards landing traffic. I know I do and only leave enough (wing inspection or opposite side exterior) on to show where we are located. 

My two cents. :)

 

Link to comment
Share on other sites

"The go around was absolutley initiated by the AC pilots prior to ATC waking up to the clues being handed to him". 

I think clues were being handed out from several sources, however, how do you know exactly when the A320 commenced the GA? 

Link to comment
Share on other sites

Blue...part of my problem...just one of many...is that I can read but not necessarily understand. You reference Rnav but I read that ALPA doesn't recommend Rnav on a visual approach. And then I am confronted by my ignorance.

I remember my experience in the cockpit landing at YVR and YYZ in a 320 at night and the runway lights...the center line....jumped out at you. And I recognize that and try to imagine to what extent the view was different to these pilots coming into SFO after hundreds ( or thousands) of landings.

Accepting the landing was rejected by the pilot prior to the ATC directive, there was very little separation in time and the rejection was late in terms of sequence.

Link to comment
Share on other sites

12 minutes ago, UpperDeck said:

Blue...part of my problem...just one of many...is that I can read but not necessarily understand. You reference Rnav but I read that ALPA doesn't recommend Rnav on a visual approach. And then I am confronted by my ignorance.

I remember my experience in the cockpit landing at YVR and YYZ in a 320 at night and the runway lights...the center line....jumped out at you. And I recognize that and try to imagine to what extent the view was different to these pilots coming into SFO after hundreds ( or thousands) of landings.

Accepting the landing was rejected by the pilot prior to the ATC directive, there was very little separation in time and the rejection was late in terms of sequence.

Good point about how CL lights can appear at night. Especially in CAVOK conditions. 

I've noticed lately that many airports are slowly converting runway and taxiway lights to LED. Very white (runway edge and CL) and a very different look to them. Interesting to see in low vis conditions. 

Link to comment
Share on other sites

BD,  the initial reports as well as AVHerald all report the crew was told to go around .25 mile past the runway end.  This is the point the were TOLD to go around.  One mile equals 5280 feet.  One quarter of a mile is 1320 ft.  The touchdown zone on any runway is the first 3000 feet, or one third of its length, whichever is shorter.  AC SOPs say plane is on ground by 1500' or you're going around.

At 1/4 mile down the runway, these pilots are flaring and at idle, which meant they crashed.  As they did not crash, they must have been going around before ATC chimed in at the threshold -.25mile point.  Aviation-navigate-communicate.

As Flight Aware has removed the actual 4 minutes from the FAF to the first MAP point, it's hard to know exactly what happened.  The .25 past end of runway goaround initiation smells pretty fishy though, and I know you are experienced enough to realize that.

The good news is, the investigators have all the actual data, and will release a report of what really happened.  Each post above may have snippets of the truth, or simply show armchair WAGs which only contributed to sullying reputations, not actual reconstructive fact.

Link to comment
Share on other sites

24 minutes ago, acsidestick said:

The good news is, the investigators have all the actual data, and will release a report of what really happened.  Each post above may have snippets of the truth, or simply show armchair WAGs which only contributed to sullying reputations, not actual reconstructive fact.

I think the numbers mentioned regarding the airplane's position relative to the T/D zone can likely be dismissed as inaccurate. Some do not make any sense  at all. It's more than likely the AC crew saw what they needed to at the same time as the tower's instructions to abort and GA. 

I don't think anyone is trying to sully this crew or it's airline. I believe most people in this crazy airline business (and on this forum) worth their weight are generally good problem solvers and only want to find the cause. However, I still think TC needs to refine crew duty days for Canadian pilots further because while these 21st century jets don't get tired the people flying them still do. 

Link to comment
Share on other sites

Agreed, the problem is that an aviation "expert" weighed in on CNN early on and presented these numbers, which don't make sense.  Actual facts are going to have an uphill battle against innpaccurate sound bites made for money.

Big questions regarding fatigue will be identified, and the airport itself may have to answer some big question, re turning off the ILS signal and why, where was the 15 foot tall X showing the closed runway, why were offset visuals still in effect if the left side was closed etc.  It would seem if 28 l was closed, there is no reason a straight in ILS couldn't have been up and running.

Link to comment
Share on other sites

Lots of misquotes and the like.

Does anyone have the wind data?

In this sort of situation, would the miss normally be hand flown, or conducted by the aircraft auto flight system?

 

Link to comment
Share on other sites

11 hours ago, blues deville said:

Well I do believe his comments about what "may" have occurred makes the most sense at this point. However, he doesn't have access to any more detailed information than anyone else.

And as mentioned on page 4 of this thread perhaps not an LDA but certainly an RNAV. The RNAV (X) approach has a similar path prior to runway alignment.  

I know many pilots make a point of turning off annoying taxi/landing lights when close to an active runway and pointed towards landing traffic. I know I do and only leave enough (wing inspection or opposite side exterior) on to show where we are located. 

My two cents. :)

 

Blues, are we sure that this aircraft is indeed equipped with GPS? The 320 in Halifax was not equipped with GPS, therefore, it was unable to do any kind or RNAV approach.

Link to comment
Share on other sites

3 hours ago, thor said:

Blues, are we sure that this aircraft is indeed equipped with GPS? The 320 in Halifax was not equipped with GPS, therefore, it was unable to do any kind or RNAV approach.

This is true and I have no knowledge of how this particular A320 was equipped. It would be interesting to know what was set up in the MCDU for that late night SFO approach. 

After re-reading some of the ATC exchanges, I'm also wondering if the other calls/comments to tower from aircraft on the taxiway help to create a situation that didn't really exist? Because of the 28L closure, aircraft were sent around the north side taxi route and sitting in a place not normally used for departing traffic. So they were seeing arrivals from a different perspective and at night with only the aircraft exterior lights of the inbound aircraft. 

Link to comment
Share on other sites

 

Well, that will teach me to go flying.  more than two pages to get caught up on.

That said, the extrapolation in the press, and elsewhere, based on partial data continues to cause nothing but problems all around - unless you are a lawyer I suppose.

The questions here about what we don't know are important ones.  What was the aircraft's navigational accuracy, what was the wind doing? How much crab did the aircraft have on approach?  Was it enough to convince the crew they were tracking correctly?   What did the crew actually see in terms of lighting effects when they transitioned to the visual segment?  How did they interpret the aircraft lights once tower told them they were alone on 28R?  Would there have been a different conclusion had tower told them there were aircraft holding for departure?

airbrake, interesting link.  My reading of that article is that it is an attempt to maintain momentum to serve another agenda.  The NTSB has not, from what I read in the article, said it was even worse than originally imagined.  Indeed, the comments look identical to their initial reports.

I have seen a number of lateral plots in the press, lots of differences and no shortage of conclusions.  I don't know the sources of these plots or whether any error checking has been done.  I suspect the NTSB will be cleaning up the data.  The terminal radar may or may not have the resolution to  differentiate between an offset approach to the TDZ and an alignment on the taxiway..  Unlikely given the reaction of the controller to the (presumably United) crew call, 'where is he going?'

A couple of broader observations.

I find the comments about pilots distancing themselves from an event like this odd.  That's never been my experience in all of the time I have worked with and around pilots and aviation.  This thread is just another example of how driven airframe drivers are to understand a threat that might lie in wait for us on our next trip.  Perhaps the black humour and commentary that aviators often use gets misunderstood.  I honestly don't know how such a mis-impression could be formed. 

That said, that drive to understand and assimilate information has in the past resulted in pilots changing their behaviour, based on rumours about what happened to a peer, and themselves going on to suffer an event due to a threat they themselves introduced, thinking it would help.  Speculation is not risk free, that's why the process is as rigid as it is.

Likewise, the quick-hit phrases like 'fatigue' and 'pilot error' are easily spoken but hard to use constructively.   Hearing that 'pilot error' caused anything is like saying that 100% of serial killers were exposed to milk as a child.  It's not that pilots believe themselves perfect, it is just the opposite.  Pilots, being human, commit errors on a regular basis on nearly every flight.  That is the reason for all of the layers of safeguards, to create a robust mitigation to the inevitable.   The fact is, the vast majority of flights proceed normally despite the errors that occur in the flight deck, ATC, loading, maintenance, etc, etc.  The entire discipline of human factors seeks to understand the complex nature of errors, and sometimes asks specifically why an error,  that on any other day would have resulted in no consequence,  today was part of an incident sequence.

So, did this crew perform perfectly?  Unlikely, and perfection was not expected.  'The devil is in the details' was never truer than following an event like this.  All we can do is ask questions, knowing that some of them may require data that is lost to the process. 

At the centre of all of this is a crew and those who support them.  They are being pilloried in the press.  Keep a thought for them if you can.

Vs

 

Link to comment
Share on other sites

Regardless of direction, a cross wind at the critical altitudes during the approach may have created a flight path resembling a flat 'U'. An approach shaped like that would have resulted in a constantly changing pilot perspective of the landing zone. Darkness, the lights of holding aircraft, some of which may have been in motion, changing horizontal & vertical perspectives, fatigue and other factors may have created an illusion that made it look like an aircraft was on the active; the crew query suggests unusual optics were causal in the incident.

 

 

 

Link to comment
Share on other sites

VSPLAT is back so I'll step away from this thread as an exercise in discretion--- for which I am NOT noted.

Air Canada has a "crisis group" which was activated.

This was a very unfortunate incident that affects careers. I don't know what changes will be implemented as a result but I genuinely sympathize with those involved.

However, a very, very severe incident was narrowly averted. For that, I am most grateful.

I personally believe that the exchanges on this thread are of value and in a small way, contribute to understanding....even if you're a lawyer. LOL.

Link to comment
Share on other sites

UpperDeck. the choice to stay or go is of course yours.  I agree with you that this event has its share of bad luck and good for all concerned.

Things here can get heated.  The price of caring about a topic perhaps.  Having read much of what you post, I suspect you would rather the heat than apathy.  If so, then perhaps on that we can agree. 

Vs

Link to comment
Share on other sites

I firmly believe that even in the total absence of warnings given by ATC or anyone else, this AC crew would have eventually initiated a last minute (and, yes, rather hair-raising) goaround, rather than settling their bird into a lineup of waiting aircraft.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.




×
×
  • Create New...