Near miss with 5 airliners waiting for T/O on taxiway "C" in SFO!
Join Date: Apr 2017
Location: England
Posts: 75
Likes: 0
Received 0 Likes
on
0 Posts
It wouldn't be a big deal. Those jets on the ground would stay there, and those in the air would land in an orderly fashion. There are places where passenger jets operate in and out of airports that don't have any controllers at all. Despite what they might like you to think, we don't really need them 
.
.On the night in question, there was a long queue of flights lined up for landing. If everything went smoothly, flights may indeed have landed in 'an orderly fashion'. But things don't always go smoothly. Assume that the single controller had, say, become unconscious. It one flight had failed to clear the runway in time (perhaps not knowing which exit to take, lacking tower instructions), or a vehicle encroached on the runway, there would be no way for the following flight to know that and execute a go-around. How long would it take for the 'spare' controller to become aware of his/her colleague's absence? How long to run upstairs and assimilate the situation?
The single-controller situation seems to me to be a risk. That might be acceptable in a small, remote airport with few flights. But in SFO? With ground and airport full of a/c and vehicles? A major international airport should not be allowed to take that kind of risk.
Join Date: Jul 2002
Location: Confusio Helvetica
Posts: 311
Likes: 0
Received 0 Likes
on
0 Posts
Ok, so they can't produce a picture of what it looked like out the window, so we can't know how red the X was or how runway like the taxiway seemed (wasn't it a smaller runway for a few years?)
But single controller was a factor, just not for the obvious reasons.
There's no way you can expect him to spot the misalignment in time, given the angle he has on the 28s.
But, for a relatively calm part of the day, what were all those a/c doing on the taxiway? The lead A/C had been waiting at the bar for thirty minutes.
When the ATCOs made the determination to go to a single controller, they did so based on the number of strips they had.I wonder if they took into account the evening closure of 28R? So, with one controller on ground and tower, Line Up And Wait goes into the break room. With all arrivals on 28R, the heavies have very few gaps to get out in, and no-LUAW means those gaps need to be bigger. So the heavies pile up on the taxiway, because the controllers didn't fully account for the impact of construction, as the AC bus lines up on them, for similar reasons.
ATC has changed their procedures since.
But single controller was a factor, just not for the obvious reasons.
There's no way you can expect him to spot the misalignment in time, given the angle he has on the 28s.
But, for a relatively calm part of the day, what were all those a/c doing on the taxiway? The lead A/C had been waiting at the bar for thirty minutes.
When the ATCOs made the determination to go to a single controller, they did so based on the number of strips they had.I wonder if they took into account the evening closure of 28R? So, with one controller on ground and tower, Line Up And Wait goes into the break room. With all arrivals on 28R, the heavies have very few gaps to get out in, and no-LUAW means those gaps need to be bigger. So the heavies pile up on the taxiway, because the controllers didn't fully account for the impact of construction, as the AC bus lines up on them, for similar reasons.
ATC has changed their procedures since.
In non-ATC airports, pilots are aware and operate accordingly. This is quite different from a major international airport where flight crew know that there are many a/c in the same airspace and expect a highly-managed environment.
On the night in question, there was a long queue of flights lined up for landing. If everything went smoothly, flights may indeed have landed in 'an orderly fashion'. But things don't always go smoothly. Assume that the single controller had, say, become unconscious. It one flight had failed to clear the runway in time (perhaps not knowing which exit to take, lacking tower instructions), or a vehicle encroached on the runway, there would be no way for the following flight to know that and execute a go-around. How long would it take for the 'spare' controller to become aware of his/her colleague's absence? How long to run upstairs and assimilate the situation?
The single-controller situation seems to me to be a risk. That might be acceptable in a small, remote airport with few flights. But in SFO? With ground and airport full of a/c and vehicles? A major international airport should not be allowed to take that kind of risk.
On the night in question, there was a long queue of flights lined up for landing. If everything went smoothly, flights may indeed have landed in 'an orderly fashion'. But things don't always go smoothly. Assume that the single controller had, say, become unconscious. It one flight had failed to clear the runway in time (perhaps not knowing which exit to take, lacking tower instructions), or a vehicle encroached on the runway, there would be no way for the following flight to know that and execute a go-around. How long would it take for the 'spare' controller to become aware of his/her colleague's absence? How long to run upstairs and assimilate the situation?
The single-controller situation seems to me to be a risk. That might be acceptable in a small, remote airport with few flights. But in SFO? With ground and airport full of a/c and vehicles? A major international airport should not be allowed to take that kind of risk.
Join Date: Apr 2017
Location: England
Posts: 75
Likes: 0
Received 0 Likes
on
0 Posts
Well, it happens all the time. The tower controller doesn’t have a lot to do when it comes to landing aircraft, just clear them to land, they’ve already been sequenced by approach. Remember the controller is busy because he’s dealing with ground vehicles, aircraft wanting an airways clearance, aircraft wanting to taxi and the aircraft on final, if he dies the only aircraft that can’t just wait around are the couple on final. As for vacating the runway, you think we land without briefing an expected exit? We just get off by the most efficient exit, we don’t need someone to hold our hand every step of the way. Is operating somewhere like SFO with no controller a good idea? Of course not, but if there is one controller and he has a heart attack, the aircraft on finals will land with no problem, and once they realize no ones talking to them they’ll be letting approach know pretty quick. If the preceding aircraft is still on the runway then the following just goes around, same with vehicles on the runway. This incident notwithstanding, it is normally easy enough to see if the runway is a, a runway, and b, clear.
b) And you can see everything with absolute clarity at night?
The whole point of aviation safety measures is to minimise the possibility of system failure. Belt and braces and belt and braces. There's a line where there's no real safety gain for extra safety measures. It probably wouldn't increase safety much to have ten controllers in the tower at 23.30. It's probably pointless to have five. But one controller? To go with a single point of failure on the basis that the chances of failure are low and only a few a/c would be affected?
This is not how aviation safety is supposed to work.
Is it just SFO? What other safety measures are shaved to the bone for reasons of cost-cutting, hubris or machismo, that this trusting, fare-paying passenger doesn't know about?
Join Date: Dec 2013
Location: Dallas
Posts: 108
Likes: 0
Received 0 Likes
on
0 Posts
You may therefore have missed the following:
May I thus venture to suggest that the actual situation as seen through the cockpit window on that particular night may not have tallied with the "sim program screencap" -- 28L not dark, and red-X drowned out by construction floodlights perhaps?
May I thus venture to suggest that the actual situation as seen through the cockpit window on that particular night may not have tallied with the "sim program screencap" -- 28L not dark, and red-X drowned out by construction floodlights perhaps?
The closed runway was the subject of a NOTAM and was mentioned in the ATIS broadcast. In spite of this, the crew made no mention of that fact during the approach briefing. While some look for reasons to exonerate the crew from blame, it seems to me they ignored several opportunities to recognize the approach required a bit more attention than they gave it.
wingview
"SFO should have better equipment and more staff on duty. This was almost a disaster but everybody seems to forget Asiana which would have been prevented with the right radar. To have a blanc spot on very very short final is not acceptable for an airport like SFO with runways used for parallel landings and take off."
The AC crew were the only ones to get "confused" that night, it was severe clear weather.
The Asiana crew totally screwed up, blaming SFO for not having the "right radar" is totally pointing at the wrong problem.
And it is "blank", blanc is French for white.... maybe you are trying to "white-wash" the poor performance of those 2 crews...
f
"SFO should have better equipment and more staff on duty. This was almost a disaster but everybody seems to forget Asiana which would have been prevented with the right radar. To have a blanc spot on very very short final is not acceptable for an airport like SFO with runways used for parallel landings and take off."
The AC crew were the only ones to get "confused" that night, it was severe clear weather.
The Asiana crew totally screwed up, blaming SFO for not having the "right radar" is totally pointing at the wrong problem.
And it is "blank", blanc is French for white.... maybe you are trying to "white-wash" the poor performance of those 2 crews...
f
Yeah, that's what "notwithstanding" means. In spite of the current incident, it is normally pretty easy to tell if a runway is a runway and if it is clear at night.
No, but it's normally pretty good. Look, I'm a night freight pilot, I do a lot of it, often into largish airports that have only one controller. If the controller falls over dead, it would be of little consequence. We'd all go-around, change to approach, talk to him for a bit while they sorted things out then go and land.
You just don't seem to understand that not only are the chances of the single controller falling over very slim, but the consequences of a single controller falling over are minimal. It just wouldn't matter, therefore there is no need to go totally belt and braces over it.
b) And you can see everything with absolute clarity at night?
The whole point of aviation safety measures is to minimise the possibility of system failure. Belt and braces and belt and braces. There's a line where there's no real safety gain for extra safety measures. It probably wouldn't increase safety much to have ten controllers in the tower at 23.30. It's probably pointless to have five. But one controller? To go with a single point of failure on the basis that the chances of failure are low and only a few a/c would be affected?
This is not how aviation safety is supposed to work.
Is it just SFO? What other safety measures are shaved to the bone for reasons of cost-cutting, hubris or machismo, that this trusting, fare-paying passenger doesn't know about?
This is not how aviation safety is supposed to work.
Is it just SFO? What other safety measures are shaved to the bone for reasons of cost-cutting, hubris or machismo, that this trusting, fare-paying passenger doesn't know about?
Oh indeed, but it seems to me that over those five months the nature of the construction and the number of lights used and/or their position and/or direction may well have changed, possibly rendering that assumption invalid.
And we do have the evidence of the previous flight who said that on that particular night they had found the construction lights far too bright and as a result had found locating the runway difficult.
I'll freely confess that -- just as you haven't had time to read the newest part of this thread -- I haven't had time to go through the whole docket yet, but according to this recent post, they DID read the NOTAM:
I can feel for the crew. They tried to do everything right (saw the NOTAM, noted the risks of an oh-dark-thirty arrival on their own human performance) - but in the end, they still "lost the picture" in a critical phase of the flight, were slow to respond to their own growing doubts, and nearly produced a major catastrophe.
(EDIT: I've just realised that there's a difference between seeing the NOTAM at the start of the flight, and (not) mentioning it during the approach briefing. Apologies if this applies.)
Oh I have no wish to exonerate the crew. From my humble position it seems clear to me that they screwed up pretty badly as mentioned above, not least leaving it far too late to go-around. However it does now seem that they were not entirely to blame due to various issues at SFO, and it seems to me best that those are understood by all concerned.
Join Date: Apr 2017
Location: England
Posts: 75
Likes: 0
Received 0 Likes
on
0 Posts
Oh indeed, but it seems to me that over those five months the nature of the construction and the number of lights used and/or their position and/or direction may well have changed, possibly rendering that assumption invalid.
And we do have the evidence of the previous flight who said that on that particular night they had found the construction lights far too bright and as a result had found locating the runway difficult.
I'll freely confess that -- just as you haven't had time to read the newest part of this thread -- I haven't had time to go through the whole docket yet, but according to this recent post, they DID read the NOTAM:
Oh I have no wish to exonerate the crew. From my humble position it seems clear to me that they screwed up pretty badly as mentioned above, not least leaving it far too late to go-around. However it does now seem that they were not entirely to blame due to various issues at SFO, and it's best that those are understood by all concerned.
And we do have the evidence of the previous flight who said that on that particular night they had found the construction lights far too bright and as a result had found locating the runway difficult.
I'll freely confess that -- just as you haven't had time to read the newest part of this thread -- I haven't had time to go through the whole docket yet, but according to this recent post, they DID read the NOTAM:
Oh I have no wish to exonerate the crew. From my humble position it seems clear to me that they screwed up pretty badly as mentioned above, not least leaving it far too late to go-around. However it does now seem that they were not entirely to blame due to various issues at SFO, and it's best that those are understood by all concerned.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
Oh I have no wish to exonerate the crew. From my humble position it seems clear to me that they screwed up pretty badly as mentioned above, not least leaving it far too late to go-around. However it does now seem that they were not entirely to blame due to various issues at SFO, and it seems to me best that those are understood by all concerned.
Given the sometimes very nasty denunciation of pax (perhaps selfish, perhaps panicked) who endanger the lives of others by dallying to get luggage during an emergency evacuation -- a subject dear to the hearts of some pilots on pprune who sometimes suggest prison or lifetime flying bans for those miscreants -- why is the no similar outcry (in the interests of safety) for the pilots who fail to secure the CVR after a screw-up.
So excoriating pax for taking luggage and thus endangering lives is because they broke a regulation not because of the safety implications and therefore pilots who aren't bound by a regulation shouldn't worry about preserving CVR for reasons of furthering aviation safety?
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
There are no Canadian rules that stipulate that a CVR or DFDR must be saved after an event like this.
There are no company (Air Canada) rules that stipulate that a CVR or DFDR must be saved after an event like this.
Let us not forget that this was simply a missed approach followed by a normal landing by those two pilots (in their minds) regardless of what we now know to be the truth in how close to a disaster this could have been.
Guest
Join Date: Apr 2009
Location: On the Beach
Posts: 3,336
Likes: 0
Received 0 Likes
on
0 Posts
When ACA759 was about one-tenth of a mile on final, the airplane looked “extremely strange,” regarding its proximity to the aircraft on taxiway C, and the taxiway itself. It was then Mr. Delucchi made the decision to send ACA759 around. There was no indication that ACA759 was in the wrong place until the aircraft was on short final. About that time, he heard a second transmission on the frequency, which he assumed was from one of the three United Airlines pilots holding short of runway 28R on taxiway C, state “he’s lined up on the taxiway.” Mr. Delucchi assumed it was a United Airlines pilot because three of the four aircraft holding short of runway 28R were United Airlines, and one was Philippine Airlines. The pilot making the transmission did not have a foreign accent. After the second transmission, “he’s lined up on the taxiway,” he directed ACA759 to go around, assigned a runway heading, and instructed the pilot to contact NCT. He noted the callsign, type aircraft, and time for documentation purposes and directed ACA759 to contact NCT. He took a minute to relax and then started departing aircraft from runway 28R. ACA759 landed shortly thereafter without incident.
After ACA759 landed, the pilot asked for and was provided the telephone number for the tower’s unrecorded line. Mr. Delucchi was going to give him the number for the tower anyway. The pilot called a bit later and was concerned about the possible seriousness of the go around, and he sounded “shaken up. During the telephone conversation he was primarily attempting to calm the pilot down. They discussed the process of what happened. It was not a lengthy conversation, because he was still working ground traffic. Mr. Delucchi had calmed down between the time of the incident and the phone call with the pilot; he had not realized how close the aircraft had gotten to each other. Mr. Delucchi was slightly more concerned with how the pilot was taking it and did not want to “freak the guy out” since he seemed shaken.
After ACA759 landed, the pilot asked for and was provided the telephone number for the tower’s unrecorded line. Mr. Delucchi was going to give him the number for the tower anyway. The pilot called a bit later and was concerned about the possible seriousness of the go around, and he sounded “shaken up. During the telephone conversation he was primarily attempting to calm the pilot down. They discussed the process of what happened. It was not a lengthy conversation, because he was still working ground traffic. Mr. Delucchi had calmed down between the time of the incident and the phone call with the pilot; he had not realized how close the aircraft had gotten to each other. Mr. Delucchi was slightly more concerned with how the pilot was taking it and did not want to “freak the guy out” since he seemed shaken.
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
Let us not forget that this was simply a missed approach followed by a normal landing
Would that also be the case if the GA was due to a microburst GA; or a misjudged/cowboy approach that need up at 500' & 30kts too fast and only half flap; or an attempt to land at the wrong airfield...or...?
Are you saying that all GA's are just normal manoeuvres and the reason has no significance?
About this specific approach I have always been curious what they were using for vertical guidance. PF had gone manual flight some miles out. One would normally do this, especially at night, when one was good confident visual with the landing runway. From photos the only PAPI's were those associated with 28R. One is used to the PAPI's being closely adjacent to the runway and touchdown lights. The PAPI's for the correct 28R were displaced a significant distance from the centreline. If they were following those PAPI's the picture should have looked very iffy from a long way out. Only the crew will know.
Would that also be the case if the GA was due to a microburst GA; or a misjudged/cowboy approach that need up at 500' & 30kts too fast and only half flap; or an attempt to land at the wrong airfield...or...?
Are you saying that all GA's are just normal manoeuvres and the reason has no significance?
About this specific approach I have always been curious what they were using for vertical guidance. PF had gone manual flight some miles out. One would normally do this, especially at night, when one was good confident visual with the landing runway. From photos the only PAPI's were those associated with 28R. One is used to the PAPI's being closely adjacent to the runway and touchdown lights. The PAPI's for the correct 28R were displaced a significant distance from the centreline. If they were following those PAPI's the picture should have looked very iffy from a long way out. Only the crew will know.
The controller told AC to go around after the crew had already initiated the procedure so it is irrelevant. It doesn’t matter how bad the approach was flown or how close they came to other aircrafts on the taxiway in so far as the CVR and DFDR are concerned.
My two points here are not about the Go Around itself but a reply about why some of you think the CVR and DFDR should have been preserved.
#1 - Again, there is no Canadian rule of law or Air Canada directive that state that the CVR and DFDR must be preserved after an incident like this. If there were such directives, to preserve both contents everytime there was a GA, wouldn’t you be grounding the aircraft and perhaps crews at every incident until the CVR and DFDR are read and a report concluded?
#2 - Whether justified or not, I am convinced those two pilots never thought about the CVR and DFDR because in “their minds” they had just accomplished a GA followed by a normal landing even though they might have known it was a close call.
Join Date: Jul 2013
Location: Pittsburgh
Posts: 129
Likes: 0
Received 0 Likes
on
0 Posts
SLF here - since GA can occur for all sorts of reasons , is it not normal or even required to describe the cause of the GA? As mentioned it could be something outside the crew's control such as weather....in this case it was a mis-alignment resulting in near-miss.
So GA could be fairly benign vs not so.
Is there no reqmt to describe the cause? The docket seems to say AC took this very seriously indeed once they got the details.
So GA could be fairly benign vs not so.
Is there no reqmt to describe the cause? The docket seems to say AC took this very seriously indeed once they got the details.
Join Date: Jun 2001
Location: Rockytop, Tennessee, USA
Posts: 5,898
Likes: 0
Received 1 Like
on
1 Post
Good question. In years past many carriers would require a report to be submitted to the airline if a go-around was done. As a result, some folks would go to extremes trying to salvage a bad approach in part to avoid some paperwork. Old-timers here on PPRuNe would sometimes brag that they never had to do a go-around not ordered by the tower.
It was realized that the failure to go-around in an attempt to 'get it on the ground' no matter what had caused avoidable accidents. The no-fault go-around became policy whereby no paperwork or explanation was required for the missed approach.
Even if no report is required (I don't know if that is the case at Air Canada), the tower will usually ask the reason for the go-around. 'We didn't meet our stable approach criteria' is normally a good answer. If you say you encountered windshear on final, you'd better make sure you filed the required w/s report. I know all this sounds like CYA but it is the reality of the modern closely monitored cockpit work environment.
Now, about that missing CVR...
As I speculated earlier on this thread:
The AC crew probably has a policy to pull the Cockpit Voice Recorder circuit breaker and make a logbook entry for maintenance to remove the CVR after a 'reportable' incident. Did they? I wouldn't be surprised if they 'forgot' to do this based on some other incidents of this type.
Canada is a little behind the times on CVR legislation by their own admission. The CVR has been added to the shutdown checklist at many places in the U.S. after a rash of 'accidental' failures to pull the CB after an incident.
The tower controller says that the Air Canada captain was shaken up when he called the tower from the gate. The claim that he didn't know that the incident was serious seems unlikely to me.
Anybody from the Great White North know if Air Canada has any guidance in their books about preserving the CVR after a significant event?
Join Date: Jun 2000
Location: last time I looked I was still here.
Posts: 4,507
Likes: 0
Received 0 Likes
on
0 Posts
Again, there is no Canadian rule of law or Air Canada directive that state that the CVR and DFDR must be preserved after an incident like this. If there were such directives, to preserve both contents everytime there was a GA, wouldn’t you be grounding the aircraft and perhaps crews at every incident until the CVR and DFDR are read and a report concluded?
I think you are taking things to extremes, where that was never intended. This incident was akin to a very very near armies. This was not a normal GA. It's a damn good thing they made one, but because it was made so late and during hat we believe to be a period of severe doubt, means it was not a normal GA in the accepted sense of the manoeuvre. They were within a whisker of causing 00's of deaths. The root cause of this could very likely go back to the approach briefing, or lack of, and then into possible disquieting utterances during the approach. All useful information into preventing it happening again. This was close to the most major crash and loss of life in an aviation accident in USA ever. It was not 'just an incident'.
I think you are taking things to extremes, where that was never intended. This incident was akin to a very very near armies. This was not a normal GA. It's a damn good thing they made one, but because it was made so late and during hat we believe to be a period of severe doubt, means it was not a normal GA in the accepted sense of the manoeuvre. They were within a whisker of causing 00's of deaths. The root cause of this could very likely go back to the approach briefing, or lack of, and then into possible disquieting utterances during the approach. All useful information into preventing it happening again. This was close to the most major crash and loss of life in an aviation accident in USA ever. It was not 'just an incident'.