not sure why there seems so much confusion over whether ATC called for GA first or not

in the play back of the ATC tape the United flight on taxiway C end waiting for his take off clearly calls (at 15.5 seconds after AC759 queries some lights on 'his runaway') ''wheres this guy goin'' then 3.5 seconds later United calls again ''he's on the taxiway''
(this is now about 20 seconds after AC759 first questions the lights he sees ahead)

then 4.5 seconds later ATC calls for AC759 to go around,
and AC759 immediately replies (now 29 seconds after his first query of lights ahead) ''IN the go around'' - the word IN and his tone implying I am already doing this (supposition on my part)


that implies to me he was already overshooting - if he was not already then I think he would have hit the aircraft waiting - that's my tuppence and I heard that the first day of the released tape which is on Page one of this thread.

in the 29 seconds of tape recording (and 29 seconds seems quite a long time to me) from AC759 first query of lights ahead to him calling 'in the go around'

how far do you fly and descend in that time ?

Just to drift away from the circular discusssion a little, I wonder what was the effect on those a/c, waiting innocently on the taxiway, of the thrust pouring out of the rear end of rapid full umph GA. I'd expect 'storm force a lot' buffeting those beasts, and scaring the poo out of the pax who could see & hear nothing. Well, up to that point anyway. I guess they heard it. If they did get a ground based ride in a washing machine I wonder what the capatins said by way of explanation.

Rat 5,

That is a very good point, United looked to be about 45 to the flightpath, so the pax got a real good view, the other acs, especially #2, had a very, very good view of an ac less than 50 feet above them, heading to the windscreen...damn, that must have been quite the experience.
Funny how it is always the lowest common denominator that exposes the weaknesses in a system.

Many discussions in the ATM circles on this with the move to automation of towers and remote towers. Especially how worthless ADSB is on final.

Looks good in Aviation Leak (Week). As you suggest, not nearly ready for prime time.

We (as in the U.S.) haven't really solved the problem of a 737 landing atop a commuter carrier at LAX, or, for that matter, assuring the runway is clear at an LAX when I am making a CAT III autoland in RVR 600 conditions.

I have been looking at multiple 'tower' setups from different suppliers in the last few days, I just wonder how this has gotten so much traction, while some of the basics have lanquished.
ADSB mandate is going to be delayed yet again in the US, so while that is not really a viable method of tracking in the terminal environment, there is no other solution being looked at.

Night visual= ILS 4miles, END OFF!
 

Hi All.
Been away , doing some semi longhaul with fuelstop and a 180. Been without my prune code and just taken in all the latest ANTSB analysis . ( A= armchair).
Also did the sim and reviewed the different GPS based approaches we can and can not do.

Considering some companies do not let you do a visual at night and in some country it is prohibited, I think it is fair to propose that if a visual is executed, the ILS must be tuned idented and briefed.I dont much care how that is done in a stoneage non GPS A320, just do it!


So for the 4 time from me, to avoid making the biggest airline accident EVER:
A KISS at night, Eh!
The last S beeing the most important!

On the contrary, the problem has been solved in different ways several times. I have worked with one that was a distributed radar systems that could show and track individual pedestrians and the display was set up like a full airport simulator allowing the ground controller to zoom a point of presence anywhere on the airport and aircraft and vehicles were shown as 3D models of the aircraft/vehicle being picked up and tracked by the radar.
But as it would never happen that an aircraft could land on an occupied runway or taxiway :rolleyes: - it wasn't seen as something necessary :ugh: . Out of some of that research came Aerobahn by SAAB/SENSIS

Aerobahn Now Available for iPad and Android Tablets

Ian, how can you say that the problem has been solved, when you also say that it was not necessary to include provisions for tracking the ac to prevent landing on the taxiway? The information provided by the FAA's Surface Movement Event Service showed the ac was lined up with the taxiway, along with the altitudes, yet this system provided no alert, nor did the systems it feeds data to, ASDEX or in this case, ASSC? How can one provide runway conflict warning when incoming ac are not tracked?

Care to guess who provides these systems? (and the one at SFO?)
U.S. Federal Aviation Administration selects Saab Sensis for Airport Surface Surveillance Capability Program

The Saab/Sensis system does not appear to have provisions for tracking which runway or taxiway the ac is landing, and alert/prevent. As far as I can tell, the Sensis system is really for ground ops, not a combination. How can one provide runway conflict warning when incoming ac are not tracked?


The system at SFO was supposed to be tracking inbound ac to 5nm with option for 20nm, yet it failed to alert the controller. For all we know, the ac could have been lined up on 28L instead of the taxi, what is, or even, is there an alert algorithm?

It was the prototype, and is supposed to be deployed at other airports in the US. Again, it is good that we had a lowest common denominator (airline with history of landing on taxiways) test of the system so that it is fixed and not deployed throughout the network.

It is interesting because NAVCANADA has an ATM ground track management system for tower ops that does track inbound aircraft when the threshold is obscured from tower.

NTSB Issues Investigative Update on San Francisco Airport Near Miss

Summary of new / confirmed details:

• The National Transportation Safety Board (NTSB) was notified of the incident on Sunday, July 9 [two days after the incident]
• The NTSB investigator-in-charge has formed the following groups: Air Traffic Control (ATC), Operational Factors, Human Performance, Airports, and Flight Data Recorders (FDR)
• The incident airplane’s cockpit voice recorder had been overwritten, so NTSB investigators did not have that data.
• The captain was the pilot flying ACA759, and the first officer was the pilot monitoring. Both pilots held Canadian airline transport pilot certificates.
• The captain had over 20,000 total flight hours, of which about 4,797 hours were as captain in Airbus A320‑series airplanes. The first officer had about 10,000 total flight hours, of which over 2,300 hours were in Airbus A320-series airplanes.
• At 2349 PDT (7 minutes before the incident), all positions in the ATC tower (controller-in-charge local control, local control assist, ground control, flight data, and clearance delivery) were combined at the local control position.
• As ACA759 approached SFO, at 2355:52 PDT, the airplane flew too far right of course to be observed by the local controller’s ASDE-X/ASSC and was not visible on the ASDE-X/ASSC display for about 12 seconds.
• At 2355:56 PDT, when ACA759 was about 0.3 mile from the landing threshold, the local controller confirmed and recleared ACA759 to land on runway 28R.
• The flight crew of the first airplane in queue on taxiway C (UAL1) transmitted statements regarding ACA759, one of which mentioned the alignment of ACA759 with the taxiway while ACA759 was on short final (see figures 2 and 3). The flight crew of the second airplane in queue on taxiway C switched on their airplane’s landing lights as the incident airplane approached.
• The incident pilots advanced the thrust levers when the airplane was about 85 ft agl. FDR data indicate that the airplane was over the taxiway at this time, approaching the vicinity of taxiway W.
• At 2356:04 PDT, ACA759 reappeared on the local controller’s ASDE-X/ASSC display as it passed over the first airplane positioned on taxiway C.
• About 2.5 seconds after advancing the thrust levers, the minimum altitude recorded on the FDR was 59 ft agl.
• At 2356:10 PDT, the local controller directed ACA759 to go around. The airplane had already begun to climb at this point
• In postincident interviews, both incident pilots stated that, during their first approach, they believed the lighted runway on their left was 28L and that they were lined up for 28R. They also stated that they did not recall seeing aircraft on taxiway C but that something did not look right to them.

https://ntsb.gov/investigations/Publ...ia148-fig3.jpg
NTSB narrative: Figure 3 shows UAL1’s transmission at 2356:04 and ACA759’s position as it overflies the first airplane waiting on the taxiway; note that the second airplane has turned on its landing lights.

Links:
https://www.ntsb.gov/news/press-rele...r20170802.aspx
https://ntsb.gov/investigations/Pages/DCA17IA148.aspx
NTSB finds 'blind spot' in SFO radar after Air Canada event

As ACA759 approached SFO, at 2355:52 PDT, the airplane flew too far right of course to be observed by the local controller’s ASDE-X/ASSC and was not visible on the ASDE-X/ASSC display for about 12 seconds.
As stated before, they could have been trying to land on 28L and the system would not have known, or it appears alerts to the issue.

[/I]The incident pilots advanced the thrust levers when the airplane was about 85 ft agl. FDR data indicate that the airplane was over the taxiway at this time, approaching the vicinity of taxiway W.[/I]
What time were the levers advanced?

[/I]At 2356:04 PDT, ACA759 reappeared on the local controller’s ASDE-X/ASSC display as it passed over the first airplane positioned on taxiway C.[/I]
Because now you have an aircraft on the taxiway so the system finds it?!?!?

[/I]About 2.5 seconds after advancing the thrust levers, the minimum altitude recorded on the FDR was 59 ft agl.[/I]
Again, a time is missing... 26 feet momentary descent! PAL was an A343 with a tail height of 56 feet...

3 feet from disaster?

So much for the afterglow...

CVR lost.
 

The CVR is overwritten and the crew can not recall overflying 4 aircraft!
WOW!

The problem - tracking aircraft, on the ground and close in on approach to identify which runway (taxiway) they were lined up for etc., has been solved it can be done and it has been demonstrated several times with live aircraft on a live airport. Nevertheless, the people who decide what goes operational thought that this was not a problem that needed a solution - even though it would have been ideal for so called virtual control towers. A version without some sensors and with different plan view displays rather than 3D virtual reality was what went ahead. So technically solutions are available and probably significantly cheaper than the cost of a crash into a line of widebodies - the problem is with bean counting. You persuade the beancounters and the system could be rolled out.

How is it that only UAL 1 said anything? There's some fair distance between them and the PAL A340 ... surely they too must have noticed something amiss? I mean a massive jet heading directly at you is not what you expect waiting on a taxiway!

Do we know if CVR recovery is possible? Or being attempted?

At least that finally settles the debate about which came first: the crew's decision to go around or the controller's instruction to do so.

@ DaveReidUK...

Does it matter if the pilots applied power for the go around first or that the tower told them to go around afterwards?

In the grand scheme of things it doesn't really matter because they were going to land on Taxiway C if it wasn't for the actions of UAL 1's and PAL 115...

To me this statement is the most important clue as to why a major accident was avoided...

"The flight crew of the first airplane in queue on taxiway C (UAL1) transmitted statements regarding ACA759, one of which mentioned the alignment of ACA759 with the taxiway while ACA759 was on short final (see figures 2 and 3). The flight crew of the second airplane in queue (PAL 115) on taxiway C switched on their airplane’s landing lights as the incident airplane approached."

Without the actions of both these aircrafts, especially PAL 115's landing lights being turned on, it's almost 100% sure that AC759 would have landed on Taxiway C and collided with PAL 115 and perhaps the other 2 aircrafts following it on the taxiway.

The experience level of both pilots is high enough to conclude inexperience is not a factor... I would really like to know what was the major factor that created this incident in the first place... Unbelievable.

CVRs are intended for accident investigation, usually when there is a hull loss.

In this case, the crew is alive and well and probably subject to "endless" interrogation. And, the FDR is obviously intact. And, all the data from many points on the airport are intact.

The UAL captain was undoubtedly interviewed.

Personally I believe the UAL 1 call to tower was perhaps at best a first wake up call to the AC crew but they still continued in their landing phase past UAL 1 because go around trust was only applies at 85 feet AGL (past UAL 1).

What I believe saved the day was PAL 115 turning on their landing lights. They must have been scared sh!less at the approaching aircraft!

I'm sure they interviewed the PAL 115 crew too... They had the best view in the house!

it notes they moved throttles forward at taxiway W. not beginning the climb for 2.5 seconds (until after PAL...)

Yes, certainly seems UAL 1 woke them up. In the image, not only does PAL have landing lights on, so does the next ac inline.

That is a very frightening read.

And, some pilot stuff, those may be taxi lights on one or both of the planes in line, not the much brighter landing lights. Also, does the last aircraft in line have the logo lights off perhaps?

Either way, it is customary to have those taxi and landing lights off while you are parked with the brakes set waiting in the queue for a night departure. The taxi lights would normally be on while the aircraft is moving, back off when you stop. Also, if there is an aircraft on short final, most folks turn off the taxi lights even if moving to avoid blinding the landing crew. Until it looks like they are landing on top of you that is. :eek:

Yep, act surprised. ;)

As we discussed earlier here:

Wow!

Good job this was all just "breathless nonsense" and "certainly wasn't anything close to a disaster", isn't it?

Dave - indeed but IMHO
if anyone had listened to the ATC tape on page 1 of this thread and read that of my post just above on this matter I personally had no doubt from the tape that AC759 was already IN the go around when ATC called for him to do it - so most fortunate for all that the AC crew realised they were in the wrong place eventually and got out of there (just)

one may read my post here
http://www.pprune.org/rumours-news/5...ml#post9849003

Rog

Quite so, as indeed I said three weeks ago:

actually i think i do recall reading your post and thought someone else was on the same page as me (and probably the NTSB were as well)

the AC759 pilots' tone implied to me 'We are on it' and 'we are already IN the GA'

any investigator should pick that up and pick it to pieces for clarification

when this first happened, some of my ac friends on fb clearly thought this was being overblown by the media. clearly it was not. This was way too close, somebody needs to figure out why.

The pilot may have been in the GA, but not by much, as the GA did not begin until after the ac passed UAL 1.

Time to use radio while missing parked ac by 3 feet.....

actually, this is different from other events as the press has not rooted out every passenger to detail their harrowing near death experience.

CVR lost!
 

It is a shame the CVR can not be recovered as it is essential to understand what happened.
The crew has conveniently developed amnesia.
Thinking that 28L was still active when it was on the NOTAM and ATIS Q is to lame.
Not noticing ANY aircraft on the taxiway after passing Uniteds wopping big tail litt with logolight broadside AND the second aircraft suddenly lighting up their aiming point for touchdown , and then claim they saw no aircraft.
That is called selective memory , at best!

Another detail:
There was only one TWR controller, as per procedure. Did he let ground and Clerance Delivery still operate on Ground and then switch you to TWR frequency only when holding short of 28R? One man two frequencies.
This is done some places to avoid congestion on frequency . Is this the case here. That would explain why the second aircraft did not scream blue murder on the frequency. ( Mind you the report would have mentioned if he did call out on ground frqz I suppose )

I suspect had he not put on the landing lights, he would have been dead.
If on Twr , why did he not call out?

Situational awareness is the key , perhaps?
One to land , you are nr two holding short. Relax a bit, all is 100% safe!
To process and comprehend the unlikely picture that someone is about to land on you on your taxiway needs a bit of time and reality checks.
Especially if they did not pay attention to AC last query.

If they just clued in when United piped up and then froze in his landinglights and decided to return the favor and light up their own , then man. was it close.

Funny thing AC pilots claim they did not see this.

Again, we are going to learn a lot from this, I shure have so fare, but it would be great to have the CVR to confirm how it came so close with such an experienced crew.

In postincident interviews, both incident pilots stated that, during their first approach, they believed the lighted runway on their left was 28L and that they were lined up for 28R. They also stated that they did not recall seeing aircraft on taxiway C but that something did not look right to them.


Much more detail here in the NTSB link: https://www.ntsb.gov/investigations/...CA17IA148.aspx

It is commonplace for the pilots in aircraft lined up for takeoff on a taxiway parallel to the active runway to douse their landing lights to reduce glare for the oncoming landing aircraft. Fortunately, that did not seem to be the case here, or did crews on the taxiway sense what was happening and then turn on landing lights?

Jet Jockey


Not sure watching an aircraft trying to land where your parked is the Best View in my books

From the NTSB update linked by underfire:

Well, I do hope you realize I was being extremely sarcastic in my comment.

Quote:
The flight crew of the first airplane in queue on taxiway C (UAL1) transmitted statements regarding ACA759, one of which mentioned the alignment of ACA759 with the taxiway while ACA759 was on short final (see figures 2 and 3). The flight crew of the second airplane in queue on taxiway C switched on their airplane’s landing lights as the incident airplane approached.


Exactly... PAL 115 most likely by their action saved the day.

re CVR lost: has crew conveniently developed amnesia.?

well no, but this whole incident (very serious as it was, almost a Tenerife x 2) I reckon will see findings to be mainly down to Circadian low effects and v.v time zones rosters commuting etc etc (IF we are ever told and shared this info)
This is causing fatigue, mind fixations, - ie: I'M VERY TIRED

maybe the CVR was wiped or simply got forgotten to be saved as implied by my above comment(s)

just my 2p

Why didn't they bail?
 

Why didn't the first two planes bail off the taxiway? I can't imagine I'd just sit there and let someone land on my nose. What are your driver's instincts here? When Bob Bragg's captain saw KLM he firewalled it sideways and saved at least some of his passengers' lives. That no one even seemed to get excited, much less take some action, is as strange as the errant line-up.

-drl

the front a/c United was angled holding at about 30-40 degrees ready to line up to take off (not facing AC 759 straight on) - his tail was in jeopardy to be hit but not the rest of him by the look of the photos

the 2nd a/c was PAL an A340 who was actually the first a/c to be head on to him (AC759) and frankly in the few seconds he had where would he have gone? the sea was on one side and the live runway the other?
it seem PAL did have his landing lights on so that's all he could do maybe - ghastly scenario avoided by a few seconds and feet

the United guy at the holding point certainly needs to be bought a few beers for speaking up (twice) his voice was quite excited BTW
listen to the recordings

Fatigue and/or Circadian effect is perhaps the only logical explanation for this near disaster.

I'm sure the authorities will look into this when writing their report.

The NTSB is treating it like an accident. The lack of the CVR is not going to let the crew have a pass; not even close.

I am almost certain JJA4 -
I'm not a safety ''expert'' but since commencing my career in civil aviation in 1971 all air safety its causes and findings immensely interest and fascinate me and I made it part of my work - I worked for BMA and knew the Kegworth skipper well.

and perhaps I had a lean towards investigative work who knows.....
hey ho but i was no good at maths or physics for a degree lol