AC Lining Up with a Taxiway SFO...?

[complexintentions]

Of course they did a go-around when instructed. That doesn't mean they didn't either initiate it themselves at the same time, or wouldn't have.

Calling a go-around, or even acknowledging ATC's instruction to do one, is far lower priority than actually doing it. Aviate-Navigate-Communicate, remember?

Personally I think they knew something wasn't right, tried to figure it out while they still had time to do so, and would have simply gone around if ATC hadn't instructed it. But then, I look at a runway before I land on it, even after I've been cleared to land. Crazy, but true.

It frightens me a bit when I read how many people seem to believe that if it wasn't for ATC this would have been a disaster, because that implies that many people also believe that it would be perfectly logical, and quite likely, to land on top of airplanes you could see, because ATC said they weren't there, and if you made a mistake identifying a runway you'd just land because ATC has cleared you to do so...

[Jet Jockey]

I think Rockie and all Air Canada Bus drivers should add RAAS equipment to their Christmas wish list for their A320s but first I hope they get GPS/WASS equipment.

[Cat Driver]

Well it's clear when you listen to the audio that they did a go around when they were instructed to or they would have made a call earlier saying were going around no?

No.

[confusedalot]

I think Rockie and all Air Canada Bus drivers should add RAAS equipment to their Christmas wish list for their A320s but first I hope they get GPS/WASS equipment.

RAAS is nice. WASS is nice. Many operate out of SFO without the nice gadgets with no issues. The only time I had RAAS was in a fancy bizjet, and I never had the luxury of WAAS.

Question; were the lights out on 28L. What was the intensity of the lights on 28R, and on C taxiway. Anybody can fall into the expectation trap, don't kid yourself. Not the first time a crew ended up lining up with a taxiway, and landing on it to boot.

Look how close those runways are, and consider the nighttime eyeball limitations in conjunction with said expectations.

[pelmet]

Just to clear things up, we have no idea of the exact timing of when the go-around was initiated. This incident has the hallmarks of the pilots making an assumption that a taxiway was a runway, yet we see the same sort of assumption fault thing here in this thread discussion. It is quite possible that the go-around was initiated prior to ATC telling them to do so, then duing the go-around, ATC said to do so and the flight responded the way they did after the go-around had been initiated.

If you have come to the conclusion that the go-around was initiated after ATC said to do so, you may be correct or you may be incorrect but are exhibiting the same assumption type characteristics that has lead to many incidents in the past.

[crazyaviator]

It was an OFFSET approach,,,THERE SHOULD have been a briefing to expect the taxiway to line up first ALSO, on the briefing, should there not be a call for ILS lined up? I am NOT buying the apologists saying that it was hard to differentiate between the taxiway and the runway,,,,,

[brooks]

Air Canada jet was 30 metres from colliding with 2 planes in San Francisco: TSB report

The summary said Air Canada Flight 759 had already travelled almost half a kilometre over the taxiway before aborting the landing.

As the Airbus 320 pulled up sharply it flew 30 metres over the first two jets, about 60 metres above the third and about 90 metres over the fourth, the summary said. It then circled and landed safely.

http://www.cbc.ca/news/canada/air-canad ... -1.4204738

[Shady McSly]

It was an OFFSET approach,,,THERE SHOULD have been a briefing to expect the taxiway to line up first ALSO, on the briefing, should there not be a call for ILS lined up? I am NOT buying the apologists saying that it was hard to differentiate between the taxiway and the runway,,,,,

Too many commas, you're fired.

[PositiveRate27]

I love how my tax dollars pay for a news corporation that summarizes a report from the "NTSB" when the report clearly states it was published by the Transportation Safety Board of Canada.

Meh. In such a scandalous story like this who cares about details...

The AC pilots questioned the controller at 0.6nm or roughly 200' AGL. They crossed the closest aircraft 100' vertically. ATC instructed a go around after the aircraft flew 0.25nm or 1500' down the taxiway. Everything else being equal, you'd expect the aircraft to cross the threshold at 50' AGL and be touching down at 1500' (0.25nm). Seeing as they flew over the first two aircraft at 100' AGL, the third at 200' and the 4th at 300' and those aircraft were stacked on the hold short line, I'd say it's a safe assumption that the AC pilots were already on the go when ATC issued that instruction.

Still lots of questions to be answered about why the aircraft flew over the taxiway, but I don't think it's quite as close to two 747's plowing into each other as the oh so reliable NCBC is telling us. (Or is it just CBC? What ever, they're basically the same thing)

[Eric Janson]

Incidentally, at my own company, if ANY pilot on the flight deck (including relief crew) calls for a go-around below 1,000 feet, a go-around is mandatory.

Exactly the same in my Airline - this is the only way to do things imho. Below 1000' is not the time for discussion.

It appears there are still companies that don't have this as SOP.

[Rockie]

Incidentally, at my own company, if ANY pilot on the flight deck (including relief crew) calls for a go-around below 1,000 feet, a go-around is mandatory.

Exactly the same in my Airline - this is the only way to do things imho. Below 1000' is not the time for discussion.

It appears there are still companies that don't have this as SOP.

Exactly the same at Air Canada as it has been since before I got there. Wrong assumption Eric.

I should also point out that it isn't necessary for a crew to realize an approach is unstabilized for it to actually be unstabilized, as this one obviously was. When any pilot in the cockpit recognizes it they are required to call it and a go-around is mandatory.

Just like you guys amazingly enough.

[Rockie]

Although it is not something we normally do it is possible to manually enter an ILS frequency/course and display it when flying some other type of procedure.

The FMS Bridge Visual is coded in the database, but since it is a visual approach flown only in VFR conditions identifying the runway is not something that should be a problem so no one would think to manually enter the ILS as a backup. Perhaps that's something that could be considered in the future but that involves other considerations than just the obvious ones and is way above my pay grade.

Based on the large number of aircraft that have landed on wrong runways, taxiways, and even the wrong airports over the years, you might want to re-evaluate that idea that no one would think to use the ILS as a back-up. It was one of the first things I thought of when I heard about this incident.

There are too many illusions and possibilities for misidentification as we see over and over again. When the system being used is not accurate enough to guarantlee alignment of the aircraft with the runway, manually tuning an available ILS as a backup, if reasonably easy to do, is an obvious simple thing to do if it does not cause interference with other required indications, especially at night.

One shouldn't need to wait for others to consider it in the future when we have the tools to do certain thing ourselves in the present.

You aren't the first person to have that idea Pelmet. It is something I and my colleagues here do routinely on every other airplane if for no other reason than having displayed vertical guidance to the runway. However things are rarely that straight forward on a bus because an action here often has unintended reactions there that don't occur on conventional airplanes. For example flying in/out of an airfield not contained in the database is a non-event in most airplanes - not so with the bus where some reading is required before you do it.

Even seemingly obvious things need to be carefully thought out on the bus. That particular backup just isn't a regular part of the Airbus routine, but perhaps now it will be looked at among other mitigation strategies.

[av8ts]

The AC pilots questioned the controller at 0.6nm or roughly 200' AGL. They crossed the closest aircraft 100' vertically. ATC instructed a go around after the aircraft flew 0.25nm or 1500' down the taxiway. Everything else being equal, you'd expect the aircraft to cross the threshold at 50' AGL and be touching down at 1500' (0.25nm). Seeing as they flew over the first two aircraft at 100' AGL, the third at 200' and the 4th at 300' and those aircraft were stacked on the hold short line, I'd say it's a safe assumption that the AC pilots were already on the go when ATC issued that instruction.

Still lots of questions to be answered about why the aircraft flew over the taxiway, but I don't think it's quite as close to two 747's plowing into each other as the oh so reliable NCBC is telling us. (Or is it just CBC? What ever, they're basically the same thing)

While this makes sense to us. Lots of people on here will ignore this and pretend they would have landed on those planes if not for ATC

[rxl]

What is going on with all the incidents on the 320 fleet? Is it possible this related to the PML and hiring of crews that otherwise perhaps wouldn't have been?

This is one of the most ignorant and stupid comments I've seen on this forum in a long time.
Hopefully a Jimmy version 3 will think just a little bit before pushing the submit button.

[skittishfox]

I may be very wrong, but there's an interesting animation posted by NBC, described as constructed by the flight data.

From what I can tell, Air Canada crew was actually trying to land between a taxiway and a runway, seemingly misreading the taxiway midline illumination for the right side of landing strip, and that they aligned with taxiway (as shown on many overhead animations) only after they already initiated the ascent:

http://www.nbclosangeles.com/news/calif ... 63353.html

[Maritimer]

While this makes sense to us. Lots of people on here will ignore this and pretend they would have landed on those planes if not for ATC

Maybe they would have, they had doubts, called ATC and were told everything was fine. Maybe they thought it was.

I'm not saying they would have btw

[Eric Janson]

Exactly the same at Air Canada as it has been since before I got there. Wrong assumption Eric.

I should also point out that it isn't necessary for a crew to realize an approach is unstabilized for it to actually be unstabilized, as this one obviously was. When any pilot in the cockpit recognizes it they are required to call it and a go-around is mandatory.

Just like you guys amazingly enough.

My comment was a general one - not aimed specifically at Air Canada. I should have phrased it better.

I'm not familiar with air Canada SOPs hence my original question.

Good to hear that we're all doing things the same way.

[CpnCrunch]

I love how my tax dollars pay for a news corporation that summarizes a report from the "NTSB" when the report clearly states it was published by the Transportation Safety Board of Canada.

Meh. In such a scandalous story like this who cares about details...

The AC pilots questioned the controller at 0.6nm or roughly 200' AGL. They crossed the closest aircraft 100' vertically. ATC instructed a go around after the aircraft flew 0.25nm or 1500' down the taxiway. Everything else being equal, you'd expect the aircraft to cross the threshold at 50' AGL and be touching down at 1500' (0.25nm). Seeing as they flew over the first two aircraft at 100' AGL, the third at 200' and the 4th at 300' and those aircraft were stacked on the hold short line, I'd say it's a safe assumption that the AC pilots were already on the go when ATC issued that instruction.

Still lots of questions to be answered about why the aircraft flew over the taxiway, but I don't think it's quite as close to two 747's plowing into each other as the oh so reliable NCBC is telling us. (Or is it just CBC? What ever, they're basically the same thing)

According to the Mercury News, they only got down to 175ft AGL, which is about 100ft above the plane on the taxiway. So if they were actually at 200ft at the 0.6nm point, they presumably initiated the go-around very shortly after.

[crj_705]

I think Rockie and all Air Canada Bus drivers should add RAAS equipment to their Christmas wish list for their A320s but first I hope they get GPS/WASS equipment.

I have seen quite a few people as of late state something about WASS...it is WAAS if anyone cares to become educated!

http://www8.garmin.com/aboutGPS/waas.html

[Eric Janson]

The NTSB has released a preliminary report.

https://www.ntsb.gov/_layouts/ntsb.avia ... 1141&key=1

[cmadude]

Considering this flight was last one for the boys/girls with an easy laid back approach. Bet they started early in day to get to Toronto then fly across country. I'd say fatigued is a BIG factor on this one.

[Cat Driver]

I'd say fatigued is a BIG factor on this one.

That is a real possibility, however it then brings up the question of why were they flying that fatigued?

[goingnowherefast]

Because fatigue is sneaky and you often don't notice it until it's there. At 6pm, you are feeling fine for a 3 hour sector. Then you bounce around in turbulence for a bit, then it gets dark, other guy starts yawning.... Combine that with all the previously discussed factors and you get to practice a real go around.

[Rockie]

Exactly. One minute you think you're doing fine and ten minutes later there's an elephant sitting on your shoulders. Stupid question Cat, and you should know better given your two centuries in aviation.

But that's not in any way suggesting this was the cause or contributed.

[pelmet]

Fatigue is frequently used as an excuse. It is near impossible to do long haul and avoid fatigue. No shortage of long haul carriers go into SFO. Aside from one crew from a country that has written off a large number of wide body jets around the world, I haven't heard of too many issues at this airport. Misidentifying a runway has happened plenty of times though over the years in a variety of locations.

Just landed at a triple parallel runway airport tonight with plenty of parallel taxiways with the green centreline lighting. Including taxiways, there are 8 parallel lighted strips. Cleared for the visual....ILS tuning made sure that there was no problem. It's that easy on some aircraft. But can it be done on an Airbus....

From another forum, it appears that this guys has some experience on type and knows the possibilities....

"In the A320 the display of ILS GS and LOC symbology on the PFD is incompatible with using the autoflight system to fly a non-precison approach. The ILS may be hard-tuned but the crew will not see the data unless they select the LS pushbutton or switch the Nav Display over to ROSE LS mode. If they do press the LS pushbutton after loading an approach with vertical guidance, they will get a flashing amber V/DEV message on the PFD to highlight the incompatible selection. I do not have experience of any 'FMS coded' visual approaches (RNAV visuals in other parlance?) but I expect that the FMGS behaviour is the same.

In the case of a classic visual approach, with just the runway selected in the FMGS, the ILS will have auto-tuned (the FMGS will know that it is an ILS runway) and the data can be displayed on the PFD without any advisory messages; the selections are compatible. However, by coding the visual approach trajectory into the FMGS it has effectively been transformed into a non-precison approach, and the selection of supporting ILS information is not as straightforward.

Bearing in mind the above information, just try writing a procedure to turn what may sometimes be a valid technique into a SOP. You would have to decide at what point it would be safe for the crew to start making fundamental changes to their EFIS set-up in order to discard the FMGS guidance and switch to ILS data. You would also need to consider if your procedure is robust enough to cope with the different cases of a non-ILS runway, ILS not available and a full-up ILS.

I also think it is worth pointing-out that hard-tuning the ILS requires manual entry of data from a different approach and introduces the chance of incorrect data entry. What if the FMS Quiet Bridge approach to 28R had been selected, but going to a different plate to obtain the ILS ident resulted in the tuning of the ILS for 28L? The technique is valid, but it is not a completely threat/error free."


Bottom line.....if you don't have the ILS back-up, be a bit paranoid. Brief the lights expected to be seen. The different types of approach lighting, perhaps confirmation with tower on what is illuminated, a reminder of the colours, etc. Similar to a low vis non-precision approach using an NDB or VOR. I landed on the wrong parallel runway once in low vis with a crosswind NDB approach. It was in the sim but it can easily happen.