You are at netAirspace : Forum : Air and Space Forums : Civil Aviation

Air Canada SFO Near Miss

All about Airlines and Airliners.
 

IFEMaster (Project Dark Overlord & Founding Member) 11 Jul 17, 16:07Post
http://www.bbc.com/news/world-us-canada-40571913

http://avherald.com/h?article=4ab79f58&opt=0

How does this even happen? 28R has CAT-III ILS. Night conditions but good visibility. The video below shows 28R approach at night; there's no way taxiway C can be confused with 28R. So...how on earth?

"Great spirits have always found violent opposition from mediocre minds." - Albert Einstein

Amazon.com

First Class members don't see ads. Upgrade!
 
vikkyvik 11 Jul 17, 16:55Post
IFEMaster wrote:How does this even happen? 28R has CAT-III ILS. Night conditions but good visibility. The video below shows 28R approach at night; there's no way taxiway C can be confused with 28R. So...how on earth?


I imagine that if visibility was good, they were doing a visual approach.

In the AvHerald comments, someone said that 28L was closed. If so, I can sort of imagine a pilot mistaking 28R for 28L, and taxiway C for 28R.

There are of course still cues that it's a taxiway, not a runway, but you know, when you think you see one thing, your mind can reinforce that.

Also, it might have been difficult to see the taxiway lights well due to the landing lights of the taxiing airplanes.
ANCFlyer (netAirspace ATC & Founding Member) 12 Jul 17, 13:52Post
I can't remember exactly who (Cathay or China Cargo) departed off ANC on Taxiway R (adjacent to Runway 33). That was prior to extension of the Taxiway to the end off 33/15.

Left an impression of the mains in the snow berm at the end of the taxiway. Tower called them a couple times during the take off roll, but it was too late.

Incidentally, there's a 100' cliff at the end of 33/R into open water (Point Waranzoff, Cook Inlet). That could have been really ugly.
Retired, retired again, and lastly retired again. I'm finally done.
halls120 (Plank Owner) 13 Jul 17, 10:50Post
vikkyvik wrote:I imagine that if visibility was good, they were doing a visual approach.

In the AvHerald comments, someone said that 28L was closed. If so, I can sort of imagine a pilot mistaking 28R for 28L, and taxiway C for 28R.

There are of course still cues that it's a taxiway, not a runway, but you know, when you think you see one thing, your mind can reinforce that.

Also, it might have been difficult to see the taxiway lights well due to the landing lights of the taxiing airplanes.


Sorry, but that makes no sense.

Image

I don't care if 28L was dark, how could they possibly think that 28R was 28L and that the taxiway was 28R???
ShanwickOceanic (netAirspace FAA) 13 Jul 17, 11:23Post
De-saturate that some, turn the 28L approach lighting off, flatten it out to a more realistic approach (still too high, I think), and put some traffic on the taxiway, and you get this five-minutes-in-photoshop view:

sfo.jpg
sfo.jpg (554.31 KiB) Viewed 976 times

Now come at it from the right:

https://resources.globalair.com/dtpp/gl ... IS28LR.PDF
Let him that hath understanding count the number of the beast:
For it is the number of a man; and its number is One hundred threescore and twelve.
halls120 (Plank Owner) 13 Jul 17, 13:55Post
ShanwickOceanic wrote:De-saturate that some, turn the 28L approach lighting off, flatten it out to a more realistic approach (still too high, I think), and put some traffic on the taxiway, and you get this five-minutes-in-photoshop view:

sfo.jpg

Now come at it from the right:

https://resources.globalair.com/dtpp/gl ... IS28LR.PDF


Looking at 28R, I see a runway. To the right, I see green lights, which screams "taxiway." to the right of the green lights, I see the midfield hangar complex.

The pilots screwed up, plain and simple.
miamiair (netAirspace FAA) 13 Jul 17, 14:22Post
You're seeing the green lights which is just an indicator of a center line. Taxiway boundaries are blue. Runways are white. Taxiways and runways have center line lights.

They almost screwed the pooch. Human error. One of the safety nets caught it, or they got lucky.
And let's get one thing straight. There's a big difference between a pilot and an aviator. One is a technician; the other is an artist in love with flight. — E. B. Jeppesen
halls120 (Plank Owner) 13 Jul 17, 14:35Post
miamiair wrote:You're seeing the green lights which is just an indicator of a center line. Taxiway boundaries are blue. Runways are white. Taxiways and runways have center line lights.

They almost screwed the pooch. Human error. One of the safety nets caught it, or they got lucky.


I get it that the taxiway center line lights are green. But right next to the midfield hangar complex? And no approach lights in the bay leading up to the landing areas?

Unless this was the pilot's first nighttime approach to SFO, that simply makes no sense to think the taxiway is 28R.
Zak (netAirspace FAA) 13 Jul 17, 14:54Post
The problem of the human mind is that it often makes you believe that you do see something you totally expect to see.

These pilots were probably not aware of 28L being closed, and absolutely expected to see 2 parallel active runways. Might be because they flew into SFO the first time, or maybe they have flown there many times already.

Either way, they wanted to see 2 parallel runways, so they ignored all apparent visual clues that what they mistook for 28R was indeed a taxiway.

This is a typical human error, but pilots are trained especially to avoid such errors.

Which is why I agree with halls:
halls120 wrote:The pilots screwed up, plain and simple.
None are more hopelessly enslaved than those who falsely believe they are free.
(Johann Wolfgang von Goethe)
Click Click D'oh (Photo Quality Screener & Founding Member) 13 Jul 17, 15:14Post
miamiair wrote: Taxiways and runways have center line lights.


Well yeah, but the runways center line lights are white. The only other green lights are the runway end lights.

Someone posted a much better looks at a night time 28R short final:

Image
We sleep peacefully in our beds at night because rough men stand ready to do violence on our behalf
ShanwickOceanic (netAirspace FAA) 13 Jul 17, 15:21Post
halls120 wrote:The pilots screwed up, plain and simple.

I don't disagree with the "screwed up" part, but the "plain and simple" part...

Yes it's a visual approach, but I hear all the time about how professional pilots would dial in the ILS anyway and monitor it. Was this not done, and if not, why not? ILS u/s, just forgot, not SOP at Air Canada?

As I recall, the Delta who actually landed on a taxiway at ATL was faced with a very similar picture of a night-time approach with the far runway out of use, so this isn't a new way to screw up. That means, of course, that pilots should be aware of the possibility, but it also means it's a known trap that shouldn't be set if at all possible.

Zak wrote:The problem of the human mind is that it often makes you believe that you do see something you totally expect to see.

"Confirmation bias", I think they call that. You approach off-centre, see two runway-looking things, and by the time you're close enough to make out that the centreline of your "runway" is green you're too stuck in the belief it's a runway to notice. Even seeing other aircraft on it has you questioning why there's someone on your runway, not whether you're aligned with the wrong strip of tarmac.

Yeah, they screwed up.

How they got into that mess in the first place, and how others can be prevented from doing the same, that's fascinating.
Let him that hath understanding count the number of the beast:
For it is the number of a man; and its number is One hundred threescore and twelve.
vikkyvik 13 Jul 17, 15:57Post
halls120 wrote:Sorry, but that makes no sense.


Maybe it doesn't, but guess what? It happened. There's a reason why it happened. And simply saying "the pilots screwed up" without trying to explain why is just allowing it to happen again.

halls120 wrote:I don't care if 28L was dark, how could they possibly think that 28R was 28L and that the taxiway was 28R???


If there's no way it could possibly happen, then how did it happen?

The mere fact that it did happen against all reasoning, leads me to think there might be some valuable lessons to he learned.

I don't think anyone would say the pilots didn't screw up.
miamiair (netAirspace FAA) 13 Jul 17, 17:08Post
Zak wrote:These pilots were probably not aware of 28L being closed, and absolutely expected to see 2 parallel active runways. Might be because they flew into SFO the first time, or maybe they have flown there many times already.


Something for the report to dig up, but the ATIS usually provides that tidbit, and if it was a planned shutdown, it would have been in the NOTAMs.
And let's get one thing straight. There's a big difference between a pilot and an aviator. One is a technician; the other is an artist in love with flight. — E. B. Jeppesen
ShanwickOceanic (netAirspace FAA) 13 Jul 17, 19:24Post
vikkyvik wrote:The mere fact that it did happen against all reasoning, leads me to think there might be some valuable lessons to he learned.

{check} This.

vikkyvik wrote:I don't think anyone would say the pilots didn't screw up.

{check} And this.

Mark my words: If some day we see someone landing on an occupied taxiway in similar circumstances, people will look back at this incident, the DL incident and countless others and ask why nobody heeded the warnings. Wrong runway, not even a runway, wrong airport altogether are all events that have happened in the recent past. Everyone from Southwest to the Dreamlifter has had a go. Sooner or later, the holes in the cheese will line up.

With my code-monkey hat on, I'll throw out the "couldn't you just" that Stefan loves bugging me with:

Wrong-runway, wrong-airport and not-a-runway have one thing in common: The extended flight path when stabilised on final does not go through the threshold of the runway that the aircraft and pilots expect to land on. That should be at least detectable in software, with current technology. The aircraft knows where it is and where it's going accurately enough for systems like Brake To Vacate to be viable. If it knows what runway has been used in the landing calculations, and knows that the flight path shows no sign of crossing the threshold of that runway, then it should be capable of knowing that someone lined up for the wrong bit of concrete, even on a visual approach. Is there anything like that out there already? My google-fu is failing to find it.

There is a patent for doing a similar job with wrong-runway take-offs: https://www.google.com/patents/US6614397
Let him that hath understanding count the number of the beast:
For it is the number of a man; and its number is One hundred threescore and twelve.
miamiair (netAirspace FAA) 13 Jul 17, 21:28Post


And let's get one thing straight. There's a big difference between a pilot and an aviator. One is a technician; the other is an artist in love with flight. — E. B. Jeppesen
MrGavin 02 Aug 17, 22:44Post
More of what you know, but from today.

https://www.bloomberg.com/news/articles ... ium=social
Click Click D'oh (Photo Quality Screener & Founding Member) 02 Aug 17, 23:10Post
The NTSB wasn’t able to obtain the Air Canada plane’s cockpit recorder. The device was overwritten during subsequent flights, the NTSB said.


Air Canada might want to rethink their policy on CVR retention if they don't already do it when you almost wipe out a couple widebodies on a taxiway.
We sleep peacefully in our beds at night because rough men stand ready to do violence on our behalf

Amazon.com

First Class members don't see ads. Upgrade!
 
 

Who is online
Users browsing this forum: No registered users and 0 guests

LEFT

RIGHT
CONTENT