Air Canada Near-Miss at SFO

[Ryan]

Landmarks don’t do a bit of good if you’re not paying attention, which was the real cause of QZ214.

 

[jis]

What do you feel makes San Francisco airport especially tricky? What lesson do you feel should have been learned by the last crash?

My understand is the airport approach is over water which of course has a lack of landmarks. One of the factors in the crash of 214. The water approach is why it's tricky.

Asiana Airlines 214 that crash. One of the factors was the aircraft was on visual. Several other issues also helped in that crash.

The recurring issues is the aircraft is on visual. If we invented a radio beacon, a radar, or a guy on the ground providing a signal that could tell the pilot that there approached is good. Well then this mess would not happen nor would of Asiana Airline.

Oh wait we have said systems we just don't use them.

Profits above safety. A winning combination.

maybe you should provide your expert services to FAA. You might save so many lives
In terms of profit before safety, the safest system would be if no one flew anything of course. There is no absolutely safe system. It is always a case of finding the right balance between profit and safety. The reason they use VFR is because that allows them to get better throughput and arguably there is little to suggest that VFR alone has caused any of the problems. SFO just has a few difficult approaches because of various factors, and usuall bigger problems occur when people truly need to fly IFR because of fog and such. Just because they are on VFR does not mean they cannot use glide slopes and other guidance systems. It is just that they don't have to maintain IFR separations. Whether they use such or not is a choice the pilots get to make. The Asiana flight did not have glide slope because, as saxman explained to me, at that point the equipment was being moved to put in place the displaced threshold for the two 28s to reduce the likelihood of an undershooting aircraft falling into the water. All pilots had gotten NOTAMs to that effect and it was also notified as part of the final approach airport condition notification. Incidentally, the displaced threshold probably saved the lives of everyone on those five planes potentially involved this time around, since that caused the 320 to fly over the two planes at the head of the queue on Charlie instead of into them.

There has been some criticism of not just SFO but most Western airports that they turn some aids off, like the Rabbit (allegedly), which IMHO is a silly thing to do, specially when all Eastern airports always keep them on apparently.

Landmarks don’t do a bit of good if you’re not paying attention, which was the real cause of QZ214.

Well, very often, fatigue is a much bigger factor than almost anything else.

 

[Devil's Advocate]

What do you feel makes San Francisco airport especially tricky? What lesson do you feel should have been learned by the last crash?

My understand is the airport approach is over water which of course has a lack of landmarks. One of the factors in the crash of 214. The water approach is why it's tricky. Asiana Airlines 214 that crash. One of the factors was the aircraft was on visual. Several other issues also helped in that crash. The recurring issues is the aircraft is on visual. If we invented a radio beacon, a radar, or a guy on the ground providing a signal that could tell the pilot that there approached is good. Well then this mess would not of happened. Oh wait we have said systems we just don't use them. Profits above safety. A winning combination.

Any competent commercial pilot should be able to manage a visual approach to a major airport with massive lighted runways, a visual glide slope, and active instrument landing system. Yes you're approaching over water, but it's not over the open ocean. There are visual cues all around you night and day. Heavy fog is possible but in that case you'd be following instrument flight rules anyway. The two primary lessons I learned from the OZ 214 crash are that some pilots need better training on how autopilots respond to unusual and contradictory command inputs and that some emergency ground services need to be taught to avoid blindly driving large commercial vehicles into firefighting foam. Watching the footage of the ground services blindly running over the bodies like clueless morons was extremely aggravating to me. If you want SFO to always activate the ALS/LDIN during low light conditions I'd support that proposal.

 

[Hotblack Desiato]

The argument of profits vs. safety might be one of those emotion-jerkers that doesn't stand up to the logic test.

Flying is (even if the Air Canada flight had ended in tragedy) arguably the safest mode of transportation out there. If the "safety" argument is taken too far in too narrow of a window of focus (meaning, just thinking of SFO operations), that reduces capacity at the airport by 50%.

Do the folks on the other half of the flights simply disappear/not travel? Probably not. What happens then is flights into/out of SFO become more scarce, meaning fares rise. Especially likely to be sacrificed are the shorter/regional flights, where the alternative is often driving (for most, rail travel isn't a realistic option yet, and may not be for who knows how many years to come). Now, what's one of the least safe modes of transportation? (Hint, it's the alternative mentioned in the previous sentence.)

For those that do still fly, it means going into alternate airports such as San Jose and Oakland. However, if their destination is San Francisco, it means additional travel to get there, most likely to be done by road.

So, if we reduce the capacity of SFO in the name of "safety," we'd actually be making travel around the area less safe.

 

[Just-Thinking-51]

Time for a white paper. If we going to design a air transport system for the ground up. Pretty sure we get something better, and safer.

To be risking people's lives, just for a dollar or two.

Once we used Freon in our refrigerators, then we discovered the damage it cause. The industry fought tooth and nail to keep using it. The government outlawed it. Industry found a cheaper and non-damaging solution to replace it.

 

[SarahZ]

It seem to maximize the airport capacity we are making trades. Safety vs Profit. By using the visual approach we can land more planes. The lost of one or two out of the hundreds of thousands is worth it.

We do not lose 1-2 planes out of every 100,000.

The U.S. averages 26,527 passenger flights every day**. By your math, we'd "lose" 1-2 planes every four days. If we increase the number to 900,000, that's still 1-2 planes every 33 days. That's not happening. At all. Anywhere.

** Source: Federal Aviation Administration, 2017 - https://www.faa.gov/air_traffic/by_the_numbers/

 

[Ryan]

He didn’t say how many hundreds of thousands he was talking about.

Maybe he meant “one or two out of the thousands of hundreds of thousands”.

Or he’s just making stuff up.

 

[jis]

OK. I think he is off bringing in all kinds of irrelevant stuff to this incident. So at least I am done responding to that sub thread. I will continue with anyone interested in discussing the specifics of this incident.

 

[Just-Thinking-51]

It seem to maximize the airport capacity we are making trades. Safety vs Profit. By using the visual approach we can land more planes. The lost of one or two out of the hundreds of thousands is worth it.

We do not lose 1-2 planes out of every 100,000.

The U.S. averages 26,527 passenger flights every day**. By your math, we'd "lose" 1-2 planes every four days. If we increase the number to 900,000, that's still 1-2 planes every 33 days. That's not happening. At all. Anywhere.

** Source: Federal Aviation Administration, 2017 - https://www.faa.gov/air_traffic/by_the_numbers/

Was not try to use a exact number just point out that a safety system was not in use. Someone is make a business decision to maximize the aircraft per hour. At some point you have a failure. If the cost worth it?

Not to me.

 

[Hotblack Desiato]

It seem to maximize the airport capacity we are making trades. Safety vs Profit. By using the visual approach we can land more planes. The lost of one or two out of the hundreds of thousands is worth it.

We do not lose 1-2 planes out of every 100,000.

The U.S. averages 26,527 passenger flights every day**. By your math, we'd "lose" 1-2 planes every four days. If we increase the number to 900,000, that's still 1-2 planes every 33 days. That's not happening. At all. Anywhere.

** Source: Federal Aviation Administration, 2017 - https://www.faa.gov/air_traffic/by_the_numbers/

Was not try to use a exact number just point out that a safety system was not in use. Someone is make a business decision to maximize the aircraft per hour. At some point you have a failure. If the cost worth it?

Not to me.

Then drive, where you're significantly more likely to die in an accident. But as I pointed out above, reducing the capacity of the safest mode of transportation, in the name of safety, just puts more people onto a less-safe mode of transportation.

 

[saxman]

This incident had nothing to do with safety vs. profit whatsoever. It happen to be a clear night which is the norm for 90% of time just about anywhere. Within the last 12 months, I've flown 454 approaches but only shot 28 instrument approaches. Of those 28, very few were with ceilings below 1000 feet.

All the normal safety systems were in place and available. My airline requires use of backup navaids during a visual approach at night and briefing it as an instrument approach. We can still shoot a visual approach at night. We just have to have a backup in place, which is usually an ILS. It really doesn't change much anyway other than being slightly closer to the guy in front, but thats not what caused this incident.

I'm assuming the AC guys had the ILS and approach loaded into the aircraft. I've never seen or heard of someone not doing this, day or night, clear or foggy. I'm not on an Airbus but being about 300 feet off to the side, the needle would be offset a little bit but not a huge about to cause concern. As they approached the taxiway, the needle would move slightly farther away, but by this time they are short final, and the runway/taxiway is easily in view and the pilots are simply fixated on it, probably not noticing how far the needle is off. You might be thinking that how could they not notice the needle being off?! Well it's hard to explain but it's super easy for even the most experienced. 28R does have a PAPI, but it is on the left side of the runway. PAPI's can be on either side of the runway, so perhaps they were thinking that was for 28L.

It's pretty obvious fatigue was a major factor in this. They were coming from Toronto and landing at midnight. Assuming they were based there, it was 3 AM body clock time. Who knows what their day had been earlier? Ever try driving when tired? I did a 4 day trip a couple months that was all West Coast flying, even though I'm ORD based. It was the same trip twice in a row and it was all late afternoon shows. Weather was bad in Seattle so were delayed from the get go. On leg 3 of 4, PDX-SEA, we had to divert back to PDX because of low visibility and not enough fuel. We left PDX again with lots of fuel and again had to wait for the visibility to come up just enough to attempt an approach into SEA. We needed 1200 feet of visibility. We got in at 2 AM. We were starting to feel the fatigue. It was quitting time. We were still schedule to go to YVR (Vancouver). We had to say no, and the flight cancelled and we had to deplane passed 70 pissed off passenger that had been waiting for hours. (Sorry folks) Got to bed at 3 AM, which was 5 AM, my body clock time. The point is that fatigue is dangerous and has been a factor any many accidents. Luckily, now a-days we are allowed to call in "fatigued" without punitive measures. But it's difficult to assess.

For those keeping track, I have 81 takeoffs and 77 landing from SFO.

 

[Bob Dylan]

Thanks Chris! Good Info from one who knows!

 

[jis]

I was hoping Chris would chime in with informed opinions based on his personal experience. Thanks Chris.

 

[SarahZ]

As they approached the taxiway, the needle would move slightly farther away, but by this time they are short final, and the runway/taxiway is easily in view and the pilots are simply fixated on it, probably not noticing how far the needle is off. You might be thinking that how could they not notice the needle being off?! Well it's hard to explain but it's super easy for even the most experienced.

I'm not sure where the needle is located, but the way you described it makes me think of driving on a local interstate late at night.

You've driven down this road countless times, and since the exit is only a mile away, you're not fixating on your speedometer.

Therefore, it's hard to tell if you're going 70 mph or 80 mph, as everything seems fairly normal and the needle isn't so far off that you'd notice it in your peripheral.

Is my analogy even close?

 

[jis]

Preliminary report from the NTSB providing an accurate timeline with photographic evidence to back it up.

https://ntsb.gov/investigations/Pages/DCA17IA148.aspx

Additional facts - the closed runway was marked with a flashing red X at its threshold which for some reason the AC crew missed completely. Perhaps a case of ignoring all evidence contrary to the theory that you have come to believe, and carry on regardless.

Looks like UA 1 was already on the taxiway connecting Taxiway C to 28R, but its tail was still sticking out onto Taxiway C. As surmised before it was the UA 1 crew that commented about the AC being lined up with the taxiway. By the time that comment ended AC was already voerflying UA 1.

The second plane in queue on Taxiway C - PAL 115 - turned on its landing lights.

AC crew initiated a go around while overflying PAL 115, the second plane in queue for takeoff, before being commanded to do so by ATC.

An additional tidbit: "[SIZE=9pt]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. "[/SIZE]

 

[XHRTSP]

I fly my 74 into SFO pretty regularily, almost always at night, and I can assure you it's nothing especially hard or tricky.

I'd go more detail but I'm on layover at HNL (poor me) with just my iPad and it's a pain to do long post, so I'll just second what Saxman said.

Sarah,

Not a good anology. Imagine a car had two drivers and two steering wheels and two dashboards, and one od the drivers had to monitor what the other was doing when they were steering, including specifically the speedomoter. And nit only that, the speedometer needle changed colors and flashed at you when you were even slighhtly off parameters.

 

[jis]

Thanks for the additional info. Very helpful for filling in a few gaps in our understanding.

 

[SarahZ]

Sarah,

Not a good anology. Imagine a car had two drivers and two steering wheels and two dashboards, and one od the drivers had to monitor what the other was doing when they were steering, including specifically the speedomoter. And nit only that, the speedometer needle changed colors and flashed at you when you were even slighhtly off parameters.

Thank you! That does help.

 

[Devil's Advocate]

Here's an update from the NTSB via the Mercury News for those who were interested.

Federal aviation investigators released stunning video footage and data Wednesday showing the near-disastrous landing of an Air Canada flight at San Francisco International Airport came as close as 5 feet from striking a Philippine Airlines jet lined up on a taxiway last July. The National Transportation Safety Board found the crew felt fatigued during the flight, that the first officer was twice rejected in his application for promotion, and that another pilot landing at SFO that night complained about too-bright construction lights that made it difficult to find the proper runway. Aviation experts have said the close call could have led to one of the worst aviation disasters in history with the fully loaded planes carrying upward of 1,000 passengers and crew.

Media Link: https://www.mercurynews.com/2018/05/02/video-new-shocking-sfo-footage-shows-just-how-close-air-canada-plane-came-to-landing-on-four-aircraft/

NTSB Docket: https://go.usa.gov/xQ8Mp

 

[Bob Dylan]

As terrible as this would have been, really it's just an everyday occurrence @ the overcrowded skies around the Major Airports!

I'm a "Retired" Million Mile Flyer, but I can remember many scarey incidents through the years, both as an Airline Passenger and as the Pilot of Small Planes using Major Airports including your own San Antonio International!

 

[jis]

As terrible as this would have been, really it's just an everyday occurrence @ the overcrowded skies around the Major Airports!

I don't think attempted landing on a taxiway that has three planes on it is a common occurrence at all.

 

[Bob Dylan]

As terrible as this would have been, really it's just an everyday occurrence @ the overcrowded skies around the Major Airports!

I don't think attempted landing on a taxiway that has three planes on it is a common occurrence at all.

Not an every day occurence, but more common than you might think jis,especially among Private Aircraft!
I know of instances where Airliners landed @ the wrong Airport, and even on Highways ( Houston Bush was involved in this when first opened since it was known as "The Black Hole of Calcuta" when planes landed on I45 several times!( my Uncle, the Retired TWA Captain, can tell some real Hair curling stories!)

 

[the_traveler]

I was on a DL flight in 94 from IAH to SLC, flying at 30K. My flight went DIRECTLY thru another jet contrail! They jet was crossing 90° from us.

It was so close that I could see the heat glowing at the back of the other engines!

 

[jis]

As terrible as this would have been, really it's just an everyday occurrence @ the overcrowded skies around the Major Airports!

I don't think attempted landing on a taxiway that has three planes on it is a common occurrence at all.

Not an every day occurence, but more common than you might think jis,especially among Private Aircraft!
I know of instances where Airliners landed @ the wrong Airport, and even on Highways ( Houston Bush was involved in this when first opened since it was known as "The Black Hole of Calcuta" when planes landed on I45 several times!( my Uncle, the Retired TWA Captain, can tell some real Hair curling stories!)

I did not say they don't happen. But they are quite uncommon. Probably more people get injured falling off a ladder by a wide margin than two planes come close enough for it to become a reportable incident. I have a good friend who worked for several years at the Newark Airport Tower. Stories, sure, there are many. But reportable incidents according to him are far fewer than hair raising stories. He claims that it is so because the system works most of the time. when a pilot forgets to deploy his landing gear before landing someone else notices and gets him straightened out before anything bad happens, and similar stuff.

 

[cpotisch]

As terrible as this would have been, really it's just an everyday occurrence @ the overcrowded skies around the Major Airports!

I don't think attempted landing on a taxiway that has three planes on it is a common occurrence at all.

Not an every day occurence, but more common than you might think jis,especially among Private Aircraft!I know of instances where Airliners landed @ the wrong Airport, and even on Highways ( Houston Bush was involved in this when first opened since it was known as "The Black Hole of Calcuta" when planes landed on I45 several times!( my Uncle, the Retired TWA Captain, can tell some real Hair curling stories!)

He claims that it is so because the system works most of the time. when a pilot forgets to deploy his landing gear before landing someone else notices and gets him straightened out before anything bad happens, and similar stuff.

As someone who has about 8,000 flight hours on a consumer grade but very realistic flight simulator, I will say that if you get too low with gear up, the cockpit gives you a multitude of warnings and alerts that are pretty much impossible to miss. On older retractable gear aircraft, the alerts were often pretty ambiguous and often didnt effectively communicate or specify that the gear was still up. Therefore, it was not uncommon for pilots to misunderstand the alerts, fail to address the problem, and crash land with their landing gear up. So now, airliners have clearer and more specific alerts to make it very clear to pilots that the issue is that landing gear, and from that gear up landings have become for the most part a thing of the past.