D L X wrote:

The NTSB said it has security-camera video of the late-night incident and will release it in the coming months.

https://www.yahoo.com/news/investigator ... 54239.html

From the article:

"Also Monday, California Insurance Commissioner Dave Jones, who was a passenger aboard the flight, released a letter to Air Canada in which he requested that the airline keep passengers informed of the results of any investigation. He said passengers were not told what was happening, and instead, the pilot "made a nonchalant announcement that he had to go around due to traffic at the airport.""

I guess it would have been better if the pilot had said:

"Uh....ladies and gentleman, I'm sorry but we are having to go around and try to land again. Uh....we....uh, almost landed on top of some other planes and.....gosh, my hands are still shaking.....uh, we're going to, uh, go around and try this again....man, my hands are really shaking....anyhow, ya, so sorry about...uh, sorry about the, uh, near disaster, but uh, you, uh...are in good hands and we'll....uh...well, let's just try this again.....uh ,thank you..."

I'd take what they actually said in a heartbeat.

I don't know if this has already been mentioned, but I heard from at least one source that 28L was not in use, and potentially the lights were.

Maybe, when told to land 28R, AC saw 28R, believed it to be 28L, and therefore lined up right of that? And possibly if the the planes on the taxiway had their taxi lights on, that could give a faintly similar look to the lights on the runway (again, considering 28L is dark).

MidEx216 wrote:

I don't know if this has already been mentioned, but I heard from at least one source that 28L was not in use, and potentially the lights were.

Maybe, when told to land 28R, AC saw 28R, believed it to be 28L, and therefore lined up right of that? And possibly if the the planes on the taxiway had their taxi lights on, that could give a faintly similar look to the lights on the runway (again, considering 28L is dark).

I'm guessing you haven't read the whole thread, but this has been a pretty common hypothesis in general.

DIRECTFLT wrote:

Remind me not fly through SFO...

Nice intelligent well thought out reply.

The RAAS system has a feature that will give an aural if the airplane is approaching a taxiway. However, most airplanes don't have RAAS and most who do don't have the Taxiway approach feature activated. This is ripe for a new technology be developed to prevent it though.

PlanesNTrains wrote:

I'm guessing you haven't read the whole thread, but this has been a pretty common hypothesis in general.

To say the least...

Adding to this hypothesis, is there a chance the airframes obstructed the green centerline lights for the CA crew? There were three pretty big ones there in line and UA1 at the hold short, possibly shading a number of the green lights with the tail. Instead of the green centerline lights, the CA crew saw white logo lights and some taxi lights... all white! At a 3° glideslope, a tail 20 meters high would cover almost 400 meters of lights, and there were 4 aircraft there to cover them.

/Fredrik

FredrikHAD wrote:

PlanesNTrains wrote:

I'm guessing you haven't read the whole thread, but this has been a pretty common hypothesis in general.

To say the least...

Adding to this hypothesis, is there a chance the airframes obstructed the green centerline lights for the CA crew? There were three pretty big ones there in line and UA1 at the hold short, possibly shading a number of the green lights with the tail. Instead of the green centerline lights, the CA crew saw white logo lights and some taxi lights... all white! At a 3° glideslope, a tail 20 meters high would cover almost 400 meters of lights, and there were 4 aircraft there to cover them.

/Fredrik

What about the approach lights? Even if the pilot sees the taxiway and assumes it's 28R, how does the presence of just one set of approach lights, to the left of his approach, not raise any alarm bells? Is this assumed to just be a fatigued brain fart?

IF THIS WAS ALREADY POSTED...SORRY

nice animation in READ MORE section:
https://www.yahoo.com/news/m/11f10ddf-4 ... plane.html

per that video, 51 feet of clearance at the lowest point ... how much less is there above the top of the vertical stabilizer?

DIRECTFLT wrote:

Remind me not fly through SFO...

Comments like this bring up an important issue. That is SFO has been dam lucky in it's history of aviation. It has barley averted a number of near huge accidents and those that have happened have had minimal life loss. ATC at SFO are fantastic, especially given what they have to work with. However, when you look at the history dating back from a JAL that crashed on landing, a PAN-AM 747 that almost went bye-bye, multiple UAL 747 engine flame-outs where the plane was almost crashed into the side of the mountain/hill on take off, Asian and Eva pilots who can't seem to land a Jumbo on clear day, and Air Canada which by the grace of god just skimmed off from most likely the greatest aviation accident of all. SFO has seemed to get bye over the years, but with all the congestion, runway issues, fog and other environmental issues and booming growth. Is the airport able to handle all this safely now and into the future? I know SFO is not the only airport in the world to have such issues and is pressed with limited options for growing but it leads to my next point which is

In the Early 2000's when the last runway proposal was put up for vote, It seemed as though the pitch was "Growth" and economic development and environmental mitigation. Granted all those still apply, but what about Safety!! I think this may have been a wake up call for the Bay Area. A new runway scheme wouldn't solve all the problems but It could help to spread out the risks and lessen many of the issues which are rapidly coming to light!

NTSB has released an update on AC 759 and has also released image of the A320 narrowly missing UA 1, a 787-9

The captain was the pilot flying ACA759, and the first officer was the pilot monitoring. Both pilots held Canadian airline transport pilot certificates.


Runway 28L was closed to accommodate construction; its approach and runway lights were turned off, and a 20.5-ft-wide lighted flashing X (runway closure marker) was placed at the threshold. Runway and approach lighting for runway 28R were on and set to default settings, which included a 2,400-foot approach lighting system, a precision approach path indicator, touchdown zone lights (white), runway centerline lights (white at the approach end), runway threshold lights (green), and runway edge lights (white at the approach end).

The incident pilots advanced the thrust levers when the airplane was about 85 feet above ground level. Flight data recorder data indicate the airplane was over the taxiway at this time. About 2.5 seconds after advancing the thrust levers, the minimum altitude recorded on the FDR was 59 feet above ground level.


Both pilots said, in post-incident interviews, 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.

Minimum recorded altitude of 59ft and the height of the 787-9 and the top of the tail is 55ft 10"....

https://www.ntsb.gov/news/press-release ... 70802.aspx

reading this packet from the NTSB and seeing the additional pictures really show how close disaster came to knocking on their doors. https://ntsb.gov/investigations/Pages/DCA17IA148.aspx

That picture is hair-raising

stealth777 wrote:

reading this packet from the NTSB and seeing the additional pictures really show how close disaster came to knocking on their doors. https://ntsb.gov/investigations/Pages/DCA17IA148.aspx

Here is when the aircraft began to climb away from the airport so people can just see it in the thread.

I can't remember what was stated in the thread now, but this does show that they began to go around before the tower instructed them to, though after the UA pilot commented on their position. It's probably safe to assume that if the UA pilot hadn't said anything, the PAL pilot might not have turned on his landing lights (may not have been aware - not sure at this point) and the AC plane would have continued it's decent. Obviously in the final report it'll show timelines of cockpit recordings and we'll have a better idea of what the AC pilots were thinking.

Also, it states that the lighted X was on a the end of the runway 28L - I thought there was a question mark earlier in the thread. I'll have to reread it a bit later when I have more time.

The report says that the CVR was overwritten and thus not available.

airtechy wrote:

The report says that the CVR was overwritten and thus not available.

Yikes.

They thought the lights on the runway were abnormal but thought nothing of the lack of approach lights?

"Minimum recorded altitude of 59ft and the height of the 787-9 and the top of the tail is 55ft 10"...."

When margins are so small, it gets interesting to discuss where and how those measurements are made. An altitude of 59 ft, is that the bottom of the fuselage, center, top, cockpit or wherever that particular sensor happens to be located? Are landing gears and engine nacelles included in that reading? We have 3 ft 2" left (if my imperial measurements are correct), and the MLG extends more than that below the wing (but not below the fuselage).

/Fredrik

airtechy wrote:

The report says that the CVR was overwritten and thus not available.

How does this happen? Did the pilots tamper with the tapes? A go around and 2nd landing including taxi to gate at midnight can't be more than 40 mins time. Those images tell enough of the story but clearly the pilots were distracted and or fatigued and it'd be interesting to hear their conversations.

Holy $%#{, but that looks way to close.

The NTSB reports that PAL 115 turned on its landing lights within the 3 seconds before AC overflew UA1. It looks as if AC started the go around as they were over UA1, ie, a few seconds after the PAL lights went on, and took 2.5 seconds to develop enough thrust to stop the descent ... a few seconds more, and they may not have cleared UA 863 / UA. 1118 (the warning from ATC would have been too late).

We may never know if AC would have started the go around in time but for the action by PAL, but maybe PAL crew should get the medal for quick thinking!

Boeingphan wrote:

airtechy wrote:

The report says that the CVR was overwritten and thus not available.

How does this happen? Did the pilots tamper with the tapes? A go around and 2nd landing including taxi to gate at midnight can't be more than 40 mins time. Those images tell enough of the story but clearly the pilots were distracted and or fatigued and it'd be interesting to hear their conversations.

The only resource I have available is flightaware, but if you look at the flight track log, you can see somewhere between 23:54 and 23:57, the go-around was initiated. Then, landing at 00:11 and arrival at the gate at 00:17. So, already that's 20 minutes. Then, how long before the aircraft powers down? Unless I'm mistaken, the CVR is always on, as long as the plane is powered.

Speechless!

YYZYYT wrote:

Holy $%#{, but that looks way to close.

The NTSB reports that PAL 115 turned on its landing lights within the 3 seconds before AC overflew UA1. It looks as if AC started the go around as they were over UA1, ie, a few seconds after the PAL lights went on, and took 2.5 seconds to develop enough thrust to stop the descent ... a few seconds more, and they may not have cleared UA 863 / UA. 1118 (the warning from ATC would have been too late).

We may never know if AC would have started the go around in time but for the action by PAL, but maybe PAL crew should get the medal for quick thinking!

I'd like to the the GPWS would have been blaring as they were flying over UA 1.

Yes and the plane flew for a number of hours after THIS landing. It's 2 hours of recording time.

GF

Re: The cockpit voice recorder. I assume that someone in authority did not recognize the importance of preserving the recording. That would have probably involved turning it off until the recording media or the recorder itself could be replaced.

I seem to remember another instance where the crew themselves saved the recording record at the gate. I'm somewhat surprised that there is not a established procedure for doing this.....maybe there is.

ikolkyo wrote:

YYZYYT wrote:

Holy $%#{, but that looks way to close.

The NTSB reports that PAL 115 turned on its landing lights within the 3 seconds before AC overflew UA1. It looks as if AC started the go around as they were over UA1, ie, a few seconds after the PAL lights went on, and took 2.5 seconds to develop enough thrust to stop the descent ... a few seconds more, and they may not have cleared UA 863 / UA. 1118 (the warning from ATC would have been too late).

We may never know if AC would have started the go around in time but for the action by PAL, but maybe PAL crew should get the medal for quick thinking!

I'd like to the the GPWS would have been blaring as they were flying over UA 1.

No it wouldn't. They were in proper landing configuration and within the calculated "landing tunnel". None of the GPWS alerts would have occurred.

It would appear AC would have missed the first two aircraft and hit the third and forth. I don't even want to think of the number of deaths that would be with just the three planes, let alone all five .

aerolimani wrote:

Boeingphan wrote:

airtechy wrote:

The report says that the CVR was overwritten and thus not available.

How does this happen? Did the pilots tamper with the tapes? A go around and 2nd landing including taxi to gate at midnight can't be more than 40 mins time. Those images tell enough of the story but clearly the pilots were distracted and or fatigued and it'd be interesting to hear their conversations.

The only resource I have available is flightaware, but if you look at the flight track log, you can see somewhere between 23:54 and 23:57, the go-around was initiated. Then, landing at 00:11 and arrival at the gate at 00:17. So, already that's 20 minutes. Then, how long before the aircraft powers down? Unless I'm mistaken, the CVR is always on, as long as the plane is powered.

God Dammitt !! It's the 21st Century !!! When any commercial aircraft has a near miss, then the CVR leading up to it ---- there should be an easy way to "Save" that data from being overwritten!!!! Can't be that hard to program:!:

From the Guardian website (UK newspaper)...

https://www.theguardian.com/business/20 ... hing#img-2

Article includes a graphic of the airport with the aircraft involved, and an image of the incident showing the AC aircraft approaching the occupied taxiway, according to the report, just 59 feet above the taxiway.

BoeingGuy wrote:

ikolkyo wrote:

YYZYYT wrote:

Holy $%#{, but that looks way to close.

The NTSB reports that PAL 115 turned on its landing lights within the 3 seconds before AC overflew UA1. It looks as if AC started the go around as they were over UA1, ie, a few seconds after the PAL lights went on, and took 2.5 seconds to develop enough thrust to stop the descent ... a few seconds more, and they may not have cleared UA 863 / UA. 1118 (the warning from ATC would have been too late).

We may never know if AC would have started the go around in time but for the action by PAL, but maybe PAL crew should get the medal for quick thinking!

I'd like to the the GPWS would have been blaring as they were flying over UA 1.

No it wouldn't. They were in proper landing configuration and within the calculated "landing tunnel". None of the GPWS alerts would have occurred.

Totally slipped my mind that being in a landing config cancels out the GPWS, this incident was literally few seconds from a full scale disaster.

WOW!! Scary to see those CCTV images and read more of the detail in the report... really brings home how close this was!!

Hard to understand how this happened... very lucky

iahcsr wrote:

It would appear AC would have missed the first two aircraft and hit the third and forth. I don't even want to think of the number of deaths that would be with just the three planes, let alone all five .

The report states that ACA759 was approaching the intersection of C and W. Thus it would have hit the A340 at least with landing speed, most likely even more since they would have initiated a go around at some point. Scary. They are lucky they flew a smaller jet and the engines spooled up so fast.

ucdtim17 wrote:

They thought the lights on the runway were abnormal but thought nothing of the lack of approach lights?

With one runway closed, and with the assumption or expectation of 2 parallel runways, it seems highly possible that the pilots mistook the lights from the other planes on C as runway lights and proceeded to land. They may had even assumed that the first 2 planes were approach lights and the other 2 or 3 planes were the runway lights.

Adoption of Forward Looking IR (Enhanced Flight Visibility Systems, EFVS in FAA speak) and Synthetic Vision Systems (SVS) would go a long way to eliminating night visual problems. The EFVS view, especially in a HUD, would have shown the line up planes as four large signatures and the lighting clearer, assuming it hadn't been converted to LED. The SVS would have shown the runway off line-up to the left in a daylight view and the taxiway in front of them. SVS is good enough to show you taxiing across a runway and the view on the PFD or MFD is identical to the outside view.

GF

ikolkyo wrote:

BoeingGuy wrote:

ikolkyo wrote:

I'd like to the the GPWS would have been blaring as they were flying over UA 1.

No it wouldn't. They were in proper landing configuration and within the calculated "landing tunnel". None of the GPWS alerts would have occurred.

Totally slipped my mind that being in a landing config cancels out the GPWS, this incident was literally few seconds from a full scale disaster.

Landing configuration prevents the radio altimeter based GPWS alerts. Being n the calculated runway tunnel prevents the look-ahead alerts. As they should.

DIRECTFLT wrote:

aerolimani wrote:

Boeingphan wrote:

God Dammitt !! It's the 21st Century !!! When any commercial aircraft has a near miss, then the CVR leading up to it ---- there should be an easy way to "Save" that data from being overwritten!!!! Can't be that hard to program:!:

CVRs are deliberately built to overwrite themselves, retaining only the last two hours of recording. This dates back to agreements with pilot unions when recorders were first introduced. The unions accepted the use to the investigation of a cockpit recording after an accident but did not want management to be able to routinely sift through cockpit recordings for say, evidence in a diciplinary matter that wasn't safety related.

What is remarkable in this incident is the AC A320 nightstopped after landing from the go-around. The CVR would not have been overwritten until the aircraft flew out the following morning. No one thought to secure the evidence that night - I guess it was only sometime after that the seriousness of what had happened was appreciated.

Yes, the aircraft in question flew out the next morning from SFO to YUL as scheduled.

Quoting the NTSB Preliminary Update Report for this incident, “runway 28L was closed, approach and runway lights were off, a large flashing X was active at the runway threshold…runway closure had been NOTAMed, ATIS also included the information that runway 28L was closed and its lightings were out of service. Runway 28R and associated approach lights, including 2400 feet approach lightings, runway edge and center line lighting, runway threshold lighting, runway threshold lighting and PAPIs, were set at default intensity. Taxiway C also was set at default lighting including the green taxiway center line lighting and blue edge lighting.”

What other cues and data did they need? And of course 28R alignment should have been apparent on their ND as well if they were flying the 28R FMS Bridge Visual approach properly.

Others can talk about circadian rhythms, lack of sleep, flying at too late in the day, etc. ad infinitum (but just look at that Tower photo at 23:56:04 while they were directly over Txwy. C) but this incident was seconds and feet (not hundreds of feet) away from becoming the worst aviation accident in US history. This IMO should be this flight crews last landing, and it would probably be best that they “pursue other interests”, as this situation oozes of incompetence and recklessness…just seconds and feet away from a horrific disaster.

cat3appr50 wrote:

This IMO should be this flight crews last landing, and it would probably be best that they “pursue other interests”...

The NTSB report says each crew member stated he thought that they were lined up with 28R. I doubt if they put their heads together before the approach and said, "Hey, let's both be stupid at the same time. That'll give 'em a laugh." Much more useful to determine why two competent individuals could make such a potentially deadly mistake.

After viewing the released images, I feel a bit uneasy given narrowly a serious tragedy was avoided. This will be a very interesting investigation to follow, I have no idea how an experienced let alone ANY professional flight crew made those mistakes. Fatigue must've been a factor and the absence of the CVR isn't going to help either.

hivue wrote:

cat3appr50 wrote:

This IMO should be this flight crews last landing, and it would probably be best that they “pursue other interests”...

The NTSB report says each crew member stated he thought that they were lined up with 28R. I doubt if they put their heads together before the approach and said, "Hey, let's both be stupid at the same time. That'll give 'em a laugh." Much more useful to determine why two competent individuals could make such a potentially deadly mistake.

Exactly. If anything these two pilots will be the last to make this mistake again going forward so removing them does nothing to increase safety.

PlanesNTrains wrote:

D L X wrote:

The NTSB said it has security-camera video of the late-night incident and will release it in the coming months.

https://www.yahoo.com/news/investigator ... 54239.html

From the article:

"Also Monday, California Insurance Commissioner Dave Jones, who was a passenger aboard the flight, released a letter to Air Canada in which he requested that the airline keep passengers informed of the results of any investigation. He said passengers were not told what was happening, and instead, the pilot "made a nonchalant announcement that he had to go around due to traffic at the airport.""

I guess it would have been better if the pilot had said:

"Uh....ladies and gentleman, I'm sorry but we are having to go around and try to land again. Uh....we....uh, almost landed on top of some other planes and.....gosh, my hands are still shaking.....uh, we're going to, uh, go around and try this again....man, my hands are really shaking....anyhow, ya, so sorry about...uh, sorry about the, uh, near disaster, but uh, you, uh...are in good hands and we'll....uh...well, let's just try this again.....uh ,thank you..."

I'd take what they actually said in a heartbeat.

How about just "We have to do a go around". Then again, humans are really really good at making excuses.

California Insurance Commissioner Dave Jones, who was a passenger aboard the flight, [...] said passengers were not told what was happening, and instead, the pilot "made a nonchalant announcement that he had to go around due to traffic at the airport.

Well duh, what honestly did he expect?


.


Tugg

CriticalPoint wrote:

...
See that's the problem there IS something that tells the pilots they are not lined up with the runway.....
.runway centerline lights are white, touchdown zone lights are white, runway edge lights are white (until you get to the end then they are amber and red), REILs are green, ALS is white and red

Taxiways have a green centerline and blue edge lighting.......

You can't fix every problem, especially with regulation, all we can do is try to mitigate human error through training. As I said before if it wasn't for human error the aviation industry world wide would be almost perfect. In this instance a human made an error and several humans corrected the error and the system worked as it should.

Just a civilian here but this post in particular (and others of the same tone) really bother me, esp as someone who has flown SFO-YYZ & back; the system most definitely did not work as it should.

I think the point being made was that in 2017 something that ACTIVELY informs the pilots whether something is amiss on approach, such as whether they are lined up with the runway or the taxiway, is certainly technically possible - the only reason it doesn't exist is because people haven't acted on it. Heck, I rented a volvo last month and it flashed warning lights when I tried to change lanes (there was a car in my blind spot); it even steered me back into the middle of the lane on the highway when I strayed too close to the shoulder. People can discuss how much "nudging" or informing is appropriate, but it's mind-boggling to think that something like can't be/isn't assisted with current technology.

To the last statement, no question - training needs to be part of the solution. But suggesting that "all we can do is try to mitigate human error through training" is simply wrong thinking. We should use every tool we can find - training, technology assists, heck praying if it proves to make a difference - to ensure that this kind of thing NEVER happens.

trgreg wrote:

CriticalPoint wrote:

...
See that's the problem there IS something that tells the pilots they are not lined up with the runway.....
.runway centerline lights are white, touchdown zone lights are white, runway edge lights are white (until you get to the end then they are amber and red), REILs are green, ALS is white and red

Taxiways have a green centerline and blue edge lighting.......

You can't fix every problem, especially with regulation, all we can do is try to mitigate human error through training. As I said before if it wasn't for human error the aviation industry world wide would be almost perfect. In this instance a human made an error and several humans corrected the error and the system worked as it should.

Just a civilian here but this post in particular (and others of the same tone) really bother me, esp as someone who has flown SFO-YYZ & back; the system most definitely did not work as it should.

I think the point being made was that in 2017 something that ACTIVELY informs the pilots whether something is amiss on approach, such as whether they are lined up with the runway or the taxiway, is certainly technically possible - the only reason it doesn't exist is because people haven't acted on it. Heck, I rented a volvo last month and it flashed warning lights when I tried to change lanes (there was a car in my blind spot); it even steered me back into the middle of the lane on the highway when I strayed too close to the shoulder. People can discuss how much "nudging" or informing is appropriate, but it's mind-boggling to think that something like can't be/isn't assisted with current technology.

To the last statement, no question - training needs to be part of the solution. But suggesting that "all we can do is try to mitigate human error through training" is simply wrong thinking. We should use every tool we can find - training, technology assists, heck praying if it proves to make a difference - to ensure that this kind of thing NEVER happens.

Something that is often forgotten is that it's not uncommon for airlines to fly their investments for 30 years. They do, of course, get some upgrades over the years, but there's only so much one can do. Think about home computers from 30 years ago, and how relatively little they were capable of. I know "you can never put a price on safety," but the cost of adding technology like this to every aircraft out there would definitely have an impact at the consumer level.

I am an aviation novice. But I have been landing at SFO as a passenger since the 1960s. In fact, I recall that my first flight was almost 50 years ago on a 727 bound to LAX. I have approached the airport at night and couldn't help but notice the long string of lights that extend what appears to be a couple thousand feet into the bay in front of the runway. The taxiways don't have the same set of lights. So if he was lined up with the taxiway, shouldn't the pilot have noticed those lights in the bay were slightly to his left? Be kind if I am asking a stupid question. I am just a passenger.

CriticalPoint wrote:

And bottom line is if the pilots had not turned off the autopilot this would not have happened. I'm a strong proponent of manually flying the aircraft to keep your skills sharp. So when I say you can't get rid of human error I'm dead serious. The Auto pilot would have prevented this but the crew didn't use it. It wasn't malicious it just happened.

Nothing will change because this is not an epidemic, it was a single close call that fortunately didn't kill anybody.

I can't disagree with the premise of manually flying the aircraft but perhaps some pilots have become a bit too casual about manual flight. It's really a shame the CVR was overwritten. It would have been instructive to have heard how detailed the approach briefing was. Perhaps in the end it's more a case of trying to ensure flight crews take that extra bit of care when flying manually in non normal situations. Flying manually is fine but it's the crew's responsibility to ensure that they don't discard all the safeguards that normally keep their aircraft from danger.

"And bottom line is if the pilots had not turned off the autopilot this would not have happened."

On what basis is that statement made? The approach is visual, not auto land and, has been posted here, making all approaches autoland would send delays in/out of SFO soaring.

GF