Lufty at SFO
”EIN12A, report British Airways 737 ahead in sight.”
EIN - ”We have them in sight.”
BAW - “Er, we’re in cloud.”
EIN - “Yes, but we saw you go in there”
/sorry for the drift.
Join Date: Apr 2018
Location: Scotland
Posts: 7
Likes: 0
Received 0 Likes
on
0 Posts
This might sound a bit thick, but surely if you take more on outward leg, and don't use it, you don't have to take on as much for the inbound leg?
Same overall cost surely?
Same overall cost surely?
Confusion
My apologies, but a slightly confused SLF here. (Not a journalist or blogger, but I would say that, wouldn't I)
Could one of you kind people tell me if my understanding here is correct.
The Lufthansa flight in question had a standard operating procedure (aka 'company rule') that they
a) were allowed to fly a visual approach into SFO at night, if weather conditions allowed.
b) were not allowed to conduct an approach into SFO at night where the pilot and first officer were required to maintain visual separation with other traffic,
the two terms seem to be being confused by some, or at least, not necessarily being used in the same context.
The same confusion between the two might have been a factor between whoever was operating the radio on the flight in question and the air traffic controller. At least, that is the impression I get from listening to the RT on the YouTube video. As I am not a professional pilot, I do not have the experience of interpreting the RT, so it might be perfectly clear to others.
If this is an inappropriate question, I apologise for posting it.
Could one of you kind people tell me if my understanding here is correct.
The Lufthansa flight in question had a standard operating procedure (aka 'company rule') that they
a) were allowed to fly a visual approach into SFO at night, if weather conditions allowed.
b) were not allowed to conduct an approach into SFO at night where the pilot and first officer were required to maintain visual separation with other traffic,
the two terms seem to be being confused by some, or at least, not necessarily being used in the same context.
The same confusion between the two might have been a factor between whoever was operating the radio on the flight in question and the air traffic controller. At least, that is the impression I get from listening to the RT on the YouTube video. As I am not a professional pilot, I do not have the experience of interpreting the RT, so it might be perfectly clear to others.
If this is an inappropriate question, I apologise for posting it.
Originally Posted by Nimrod
I understand it costs a we bit more to carry the wee bit extra fuel
Semreh,
There is a review of the incident here: https://theaviationplace.space/2023/...cident-review/
Not everyone posting in this thread will agree with the debatable points, but at least there is supporting text for the analysis.
There is a review of the incident here: https://theaviationplace.space/2023/...cident-review/
Not everyone posting in this thread will agree with the debatable points, but at least there is supporting text for the analysis.
Except that such approaches are NOT published in the FAA AIP (https://www.faa.gov/air_traffic/publ...alifornia.html). Moreover, nothing in the General Remarks regarding the "Visual Flight Procedures".....
Learning from everyday work - The best time to learn
Re "… no incident… " #270, this is a view of safety based on outcome.
Alternatively, considering 'safe' as minimising the risk of harm - future harm, then as argued at #250, night visuals have higher risk than daytime, and similarly visual separation for closely spaced runways more than IFR ILS approaches.
One's view of this depends on judgement, context, experience, knowledge, and the way that these come together in our thinking.
The ATC procedure is an accepted operation (aircraft, operators, ATC, FAA; as a system), however this does not preclude different risk assessments and SOPs to manage risk. The industry encourages this, it is a core aspect of safety management.
The operator's SOP requires ATC adjustment, which in this particular situation and time, resulted in a mismatch in requirements; the need for a (higher risk) night visual approach/sepatation vs a (lower risk) ILS approach. This situation was compounded by minimum fuel - time available for crew and ATC.
This mismatch is an opportunity to learn; not from an after the fact (with hindsight) apportioning of blame, but to reconsider the levels of risk in such scenarios, and the actual or inferred responsibilities.
One way is to conduct a 'Pre-mortem', a debated, justifiable 'what if' before an event:
What are the respective hazards, how best managed.
Is the level of risk minimised - are there better safety options.
Who holds the responsibility for these judgements (everyone); is this responsibility acknowledged, exercised (how), or mistakenly delegated to others.
What might be improved.
Learning from everyday work - The best time to learn.
https://www.dropbox.com/scl/fi/3jt2g...cvfvz7u1l&dl=0
'Death hides in normal work'
'What is happening when nothing bad is happening'
'Complacency is a logical outcome of a stable system'
Alternatively, considering 'safe' as minimising the risk of harm - future harm, then as argued at #250, night visuals have higher risk than daytime, and similarly visual separation for closely spaced runways more than IFR ILS approaches.
One's view of this depends on judgement, context, experience, knowledge, and the way that these come together in our thinking.
The ATC procedure is an accepted operation (aircraft, operators, ATC, FAA; as a system), however this does not preclude different risk assessments and SOPs to manage risk. The industry encourages this, it is a core aspect of safety management.
The operator's SOP requires ATC adjustment, which in this particular situation and time, resulted in a mismatch in requirements; the need for a (higher risk) night visual approach/sepatation vs a (lower risk) ILS approach. This situation was compounded by minimum fuel - time available for crew and ATC.
This mismatch is an opportunity to learn; not from an after the fact (with hindsight) apportioning of blame, but to reconsider the levels of risk in such scenarios, and the actual or inferred responsibilities.
One way is to conduct a 'Pre-mortem', a debated, justifiable 'what if' before an event:
What are the respective hazards, how best managed.
Is the level of risk minimised - are there better safety options.
Who holds the responsibility for these judgements (everyone); is this responsibility acknowledged, exercised (how), or mistakenly delegated to others.
What might be improved.
Learning from everyday work - The best time to learn.
https://www.dropbox.com/scl/fi/3jt2g...cvfvz7u1l&dl=0
'Death hides in normal work'
'What is happening when nothing bad is happening'
'Complacency is a logical outcome of a stable system'
How absurd to think that a pilot of one plane who can see a well lit airplane a few miles away would want to take some responsibility for not bumping into that plane.
”That’s ATC’s job” seems to be the cry here, yet many would object to ATC “flying the plane for them”.
I get the feeling most here have never flown this approach. Nobody’s measuring the spacing with a micrometer. “See that plane over there? Don’t hit it. Cleared for the visual”. That’s all there is to it. No fancy aerobatics involved. Turn off the automation and fly the thing like you flew a 172 and enjoy the view. It’s great fun.
”That’s ATC’s job” seems to be the cry here, yet many would object to ATC “flying the plane for them”.
I get the feeling most here have never flown this approach. Nobody’s measuring the spacing with a micrometer. “See that plane over there? Don’t hit it. Cleared for the visual”. That’s all there is to it. No fancy aerobatics involved. Turn off the automation and fly the thing like you flew a 172 and enjoy the view. It’s great fun.
This is an established commercial passenger airline, whose safety department has done a risk assessment and decreed that visual approaches at KSFO - where mark one eyeball is required to ensure and maintain separation during parallel approaches - are not allowed during the hours of darkness.
It is not that the LH pilots are incapable of visual approaches; their company has told them it is not allowed. It might well be an insurance issue, for all we know.
Semreh,
There is a review of the inciedent here: https://theaviationplace.space/2023/...cident-review/
Not everyone posting in this thread will agree with the debatable points, but at least there is supporting text for the analysis.
There is a review of the inciedent here: https://theaviationplace.space/2023/...cident-review/
Not everyone posting in this thread will agree with the debatable points, but at least there is supporting text for the analysis.
I'd better bug out and keep reading quietly.
Originally Posted by MarcK
Both have localizer guidance in the final segment.
The TT does (it uses the LLZ the whole way by the looks) but the QB clearly isn't depicted like that; it's shown as an offset the whole way in. And the tracks the aircraft actually fly on the QB bear no resemblance to the QB chart either: they look like they head in to the bridge then do a 15° right turn onto the CL at around 5nm. The QBchart is misleading/wrong.
.
Join Date: Jul 2013
Location: Within AM radio broadcast range of downtown Chicago
Age: 71
Posts: 851
Received 0 Likes
on
0 Posts
1. Thanks sftypee for posting the link to a summary of the incident.
2. Admit I started watching the "second" video which has been referred to by many posters; also admit it became too tedious to continue watching to end. But regardless, its content (as has been noted) isn't necessarily official.
3. Tomorrow, Thursday Nov. 30, the Aviation Subcommittee of the House of Representatives Transportation and Infrastructure Committee will hold a hearing in Rayburn on Capitol Hill. It will be yet another effort to break the legislative logjam - hearing subject matter title: Importance of Passing A Comprehensive Long-Term FAA Reauthorization Bill.
4. One of the four scheduled witnesses to testify in the hearing is the president of NATCA, National Air Traffic Controllers Association.
Could there be some questions for this witness, and perhaps the others as well, derived from this incident? Yes, I think there are.
Q: What was the main or primary cause for the controller not providing accurate estimates of delay time? (Rather obviously, with endemic or even systemic problems in controller staffing, it appears important to try to know, was the breakdown of communications about delay *only* less than ideal performance by that controller, likely stemming from workload? Or was it, conceivably, temperament flaring? Is there another explanation possible? ... I mean beyond pejorative ones.
Q: What has FAA done so far, if anything, to examine the incident? Doesn't DLH deserve some recourse for a diversion that in reality - or so it seems to just an SLF/atty guest here - did not need to occur, such recourse in the form of assurance it doesn't happen again? (Obviously it had become necessary in fact)
Q: Relatedly, what about providing for better advance notice of operating limitations, that is, both by FAA and by a carrier about its SOPs that impose such limits?
I'm resisting the perhaps sarcastic temptation to ask in any detail, when SWIM is talked about ....System-Wide Information Management - problems like those leading to this incident occuring will be digitized out of existence,.....right?
2. Admit I started watching the "second" video which has been referred to by many posters; also admit it became too tedious to continue watching to end. But regardless, its content (as has been noted) isn't necessarily official.
3. Tomorrow, Thursday Nov. 30, the Aviation Subcommittee of the House of Representatives Transportation and Infrastructure Committee will hold a hearing in Rayburn on Capitol Hill. It will be yet another effort to break the legislative logjam - hearing subject matter title: Importance of Passing A Comprehensive Long-Term FAA Reauthorization Bill.
4. One of the four scheduled witnesses to testify in the hearing is the president of NATCA, National Air Traffic Controllers Association.
Could there be some questions for this witness, and perhaps the others as well, derived from this incident? Yes, I think there are.
Q: What was the main or primary cause for the controller not providing accurate estimates of delay time? (Rather obviously, with endemic or even systemic problems in controller staffing, it appears important to try to know, was the breakdown of communications about delay *only* less than ideal performance by that controller, likely stemming from workload? Or was it, conceivably, temperament flaring? Is there another explanation possible? ... I mean beyond pejorative ones.
Q: What has FAA done so far, if anything, to examine the incident? Doesn't DLH deserve some recourse for a diversion that in reality - or so it seems to just an SLF/atty guest here - did not need to occur, such recourse in the form of assurance it doesn't happen again? (Obviously it had become necessary in fact)
Q: Relatedly, what about providing for better advance notice of operating limitations, that is, both by FAA and by a carrier about its SOPs that impose such limits?
I'm resisting the perhaps sarcastic temptation to ask in any detail, when SWIM is talked about ....System-Wide Information Management - problems like those leading to this incident occuring will be digitized out of existence,.....right?
I’ve no clue about any of those things because after about 300 comments we haven’t had anything at all constructive or useful added from that angle, just a load of conjecture that DLH should have done a visual approach - which they could have done apparently! Just not a parallel, paired approach where they’d be responsible for separation.
The second, rebuttal video didn’t offer any of those mitigations you mentioned, just it was busy and that DLH was treated as though it had asked for an opposite end approach - something that would incur a significant delay for everyone - despite the fact it’s request would only mean a single gap created in the same way a go-around would.
The second, rebuttal video didn’t offer any of those mitigations you mentioned, just it was busy and that DLH was treated as though it had asked for an opposite end approach - something that would incur a significant delay for everyone - despite the fact it’s request would only mean a single gap created in the same way a go-around would.
According to the NAS Airport demand charts, 17:00 and 19:00 are peak times, and on this evening there were over 40 arrivals built into the sequence that would have been impacted in order to accommodate LH458, with further aircraft being held in the adjacent airspace waiting for their inbound slots.
Given the congestion, you strangely place little importance on the lateness of LH's notification to ATC of their restriction, which occurred just 16 miles NW of the airport over BDEGA, and dismiss any impact it would have on their delay. To be clear, LH was originally sequenced for 28R in saturated airspace with no indication of a delay. On the BDEGA Arrival there are multiple, depicted holding patterns yet LH wasn't stuck in one. There was a gap already created for LH on the QB 28R to slide into, yet they couldn't. (that was a wasted gap, man, and someone out there in a hold could've used it).
BDEGA, where they notified ATC, is just 7 miles/2 minutes before CORKK (9 miles from the airport), where they were to fly a 100 heading that feeds them onto a right downwind for QB 28R. But despite chatting with NORCAL for quite some time no matter which transition they used, and Oakland Center before that, LH advised ATC of their restriction only in response to the CORKK heading clearance, just 2 minutes before the turn.
It's only at that point the Controller knows he's not going vector LH for an ILS to 28R (perhaps that's what you believe he could've done to only "create 1 space"). The QB Visual is named "Quiet" for a reason and, not coincidently, there's an offset compared to, say, the 28R Loc which isn't. With the subsequent "can't maintain visual separation" the Controller also knows that during either ILS 28 there will be no pairing LH with Visual approach aircraft on the other runway because both pilots of a pair need to confirm they can "maintain...". (so there goes another gap someone in a hold could've used). Even if you did push him onto 28R you'd have to break off the existing aircraft LH would've be paired side-by-side with that's flying Tip Toe Visual for 28L. So, off for an 28L ILS LH was sent because the TT visual is predicated on the 28L Loc anyway.
Now overhead SFO, to plug LH in somewhere else takes adjusting both streams from the east, not just the stream for 28L. Both have been built and metered so aircraft wind up side-by-side in pairs on the Visuals. Because LH can't "maintain" and therefore can not have an aircraft paired-up off his right wing, one way or another someone in the already-built stream for 28R has to disappear to create that gap for LH (another one wasted). Then there's the 1 you talked about, the gap LH itself needs in the 28L stream with the proper spacing behind. To preserve the pairs that create the gaps for departures, however, the in-trail spacing must also be applied to 28R aircraft paired to LH's own trailer.
Assuming LH picks up ATIS (broadcast as "Simultaneous Charted Visual Procedures in Use") like everyone else in order to set-up and brief an approach long before, does 16 miles out seem like an overly brilliant time to first advise ATC unable to accept any of the approaches being broadcast on ATIS, because it doesn't seem even a little bit brilliant to me. What approach had LH set-up and briefed?.....No telling, but I suppose an ILS because that's what they required when they finally, belatedly, let the world know.
If LH had advised ATC earlier than 2 minutes before the 28R downwind turn over CORKK....say, way out on the Arrival when they got the ATIS...they would've probably found out they weren't going to get an ILS for 28R, the Controllers could've been cooking up some Plan B spacing required for an ILS 28L, and the gap slated for them on 28R they squandered could've been used by some Visual-accepting aircraft stuck in a hold.
Undoubtedly, some of those holding could accept the Visual and to "Maintain". If they weren't just more victims of a petty, spiteful and/or incompetent ATC, perhaps they were being held because a few others before LH showed up, also waited until 16 miles out to inform they needed ILSs, which jammed up the works so completely simple "1 gap" solutions just weren't possible.. Whatever the reason, it's 100% certain they had a limit on their endurance, so LH certainly would't be special in that respect.
Maybe it’s just worth recapping the time and conditions when this occurred. The ATIS at 0356Z gives FEW005, SCT007. Hardly CAVOK is it? Not sure I’d be able to see either the runway or another aircraft with that cloud reported.
One aspect not discussed much is human factors relating to fatigue. You have a long haul crew who have probably operated around 11-12 hours and for their body clock it’s around 05:00. The LH pilot sounds a little tetchy so perhaps that’s one reason and the other is the lack of accurate holding times which is the responsibility of ATC to provide. Also, I suspect that pilots who have never flown long haul do not appreciate that the skills you acquire when flying frequently are likely eroded especially when fatigued. We should be aiming for the highest level of safety that can be achieved. Not “well most pilots can do this so why can’t you?” Even the best operators can have their off days.
To read some of the comments you would think the LH crew requested some weird arcane procedure that meant they were to be treated like Air Force 1! Apparently vectors to an ILS, like you would expect anywhere else, is just too difficult. I’m not a frequent visitor to SFO but over the years I’ve had normally had vectors to an ILS. Sometimes a little sporting and generally ATC have been ok. They always ask “report when you’re visual with xxx aircraft” but when unable, because neither of us could see it then it hasn’t been an issue. Sorry PukinDog I have sometimes used “we’ve got ‘em on TCAS”. Mea culpa.
This is definitely not a rant against SFO ATC in general but I think this particular controller should have done better accommodating a reasonable request by the pilot. Lastly, SFO is a major international airport so it should have the flexibility to allow foreign operators to fly within a SOP that is quite reasonable. If it accepts aircraft on an IFR flight plan it should allow an instrument approach at the end of it.
One aspect not discussed much is human factors relating to fatigue. You have a long haul crew who have probably operated around 11-12 hours and for their body clock it’s around 05:00. The LH pilot sounds a little tetchy so perhaps that’s one reason and the other is the lack of accurate holding times which is the responsibility of ATC to provide. Also, I suspect that pilots who have never flown long haul do not appreciate that the skills you acquire when flying frequently are likely eroded especially when fatigued. We should be aiming for the highest level of safety that can be achieved. Not “well most pilots can do this so why can’t you?” Even the best operators can have their off days.
To read some of the comments you would think the LH crew requested some weird arcane procedure that meant they were to be treated like Air Force 1! Apparently vectors to an ILS, like you would expect anywhere else, is just too difficult. I’m not a frequent visitor to SFO but over the years I’ve had normally had vectors to an ILS. Sometimes a little sporting and generally ATC have been ok. They always ask “report when you’re visual with xxx aircraft” but when unable, because neither of us could see it then it hasn’t been an issue. Sorry PukinDog I have sometimes used “we’ve got ‘em on TCAS”. Mea culpa.
This is definitely not a rant against SFO ATC in general but I think this particular controller should have done better accommodating a reasonable request by the pilot. Lastly, SFO is a major international airport so it should have the flexibility to allow foreign operators to fly within a SOP that is quite reasonable. If it accepts aircraft on an IFR flight plan it should allow an instrument approach at the end of it.
These are the airport's preferred approaches when the weather is VMC, so they've set them up pretty well for use with the special weather mins and remote reporting station. That, combined with the electronic guidance that must be tuned and tracked that's depicted on the Procedures' Charts which, if followed as cleared, establishes and maintains the separations between the paired aircraft, ATC speed control for in-trail spacing and hi/low vertical separation. Compared to most, there's actually very little that's "visual" about them.
The LH guys didn't have a choice, which is too bad because they fly these SFO Visuals all the time during daylight when weather permits. They're very familiar with them. So familiar that the greatest risk probably isn't proficiency, but complacency.
Mea Culpas can't be fined retroactively so no worries.
I've never met a pilot who wasn't guilty of something, myself included, and not sure I'd even want to.
The second, rebuttal video doesn't need to mention mitigations like other traffic, other airports using the airspace, noise constraints, etc... it's common sense unless you believe the Controllers are 1) incompetent or 2) in the habit of punishing aircraft for spite or 3) both. Another clue; the Controller told LH to "expect an extended delay" and, not surprisingly, the info taken from the link in safetypee's Post 308 confirms what common sense told us:
In other words, they were saturated, with streams from the East built and paired-up, other aircraft for SFO holding awaiting their gaps to be created, and LH's extended delay was exactly what the Controller told him to expect. To me, it sure doesn't look like LH was singled-out for a spite-fuelled spanking
In other words, they were saturated, with streams from the East built and paired-up, other aircraft for SFO holding awaiting their gaps to be created, and LH's extended delay was exactly what the Controller told him to expect. To me, it sure doesn't look like LH was singled-out for a spite-fuelled spanking
"If they request something else we should accomodate them to the best of our ability, but they may have to take a delay. They view it the same as if you flew in requesting an opposite direction landing in 10 when the active runway is 28". The underlined part is baffling. It's the air traffic equivalent of deciding you need to buy 4 new tyres because your low pressure warning came on for a just single one, which only needed pumping up anyway. Probably not a great like-for-like approximation, but it would be like resorting straight to the total electrical failure checklist because your transponder failed. A runway change is a fairly significant event that will delay multiple aircraft. A single opposite-end arrival will cause huge delays. I'm not sure why they would elect to treat it in that manner. Why make so much more work for yourself?
Another comment made in that video was that the gap behind would "have to be 5-8nm", which is again, utter rubbish. If you're applying wake separation to a light behind it, yeah you'll need 8 miles. But you're delaying it - no-one has said the DLH should have received no delay whatsoever - so why on Earth would you then choose to stick it in ahead of a light?! Secondly, even if that DID apply, as far as I can tell when you use visual separation, wake responsibility also shifts to the pilot? (Not how it would work here, it would be on the controller all the way to touchdown, regardless of the type of approach you fly). As long as the aircraft behind is happy to be visual, all you need is 3nm radar separation for standard separation to be applied to the DLH. Why would you choose to put an 8nm gap behind it? As long as it's a minimum 3nm - which being a heavy doesn't seem too far off - then as long as the aircraft behind is happy to apply visual separation, you aren't actually losing anything behind it.
Honestly, if you were to tell me the SFO perspective video wasn't a controller at all I wouldn't be surprised. The language is odd "They view it as...", why they and not we? The only provenance for it is "a guy I know from Facebook". I still find it hard to believe any practicing radar controller would seriously describe the application of headings and speeds as "adding risk".
Given the congestion, you strangely place little importance on the lateness of LH's notification to ATC of their restriction, which occurred just 16 miles NW of the airport over BDEGA, and dismiss any impact it would have on their delay [...] There was a gap already created for LH on the QB 28R to slide into, yet they couldn't. (that was a wasted gap, man, and someone out there in a hold could've used it).
If it's their peak (which again, I don't dispute in the slightest) and aircraft are holding, and you're packing planes into every gap you have, there will be another plane you can fit in with 40nm notice. If you can't, that is some exceptionally inflexible procedure design. Which leads to...
Now overhead SFO, to plug LH in somewhere else takes adjusting both streams from the east, not just the stream for 28L. Both have been built and metered so aircraft wind up side-by-side in pairs on the Visuals. Because LH can't "maintain" and therefore can not have an aircraft paired-up off his right wing, one way or another someone in the already-built stream for 28R has to disappear to create that gap for LH (another one wasted).
Pukindog, some very long posts on here. I’ve just got a couple of questions for you.
One of ATC’s primary roles is to issue accurate onward clearance/approach times or at least an accurate estimate of the delay (and an update to that as soon as possible after it’s changed) in order for pilots to adequately plan fuel.
Do you think SFO fulfilled that duty to DLH?
There are various ATCOs on here adamant that an extra 2/3 miles on the spacing was enough to accommodate DLHs requirements. Why do you think that would cause significant delays and ground stops for aircraft outside the sector? Or is the estimate of 2/3 miles wrong?
One of ATC’s primary roles is to issue accurate onward clearance/approach times or at least an accurate estimate of the delay (and an update to that as soon as possible after it’s changed) in order for pilots to adequately plan fuel.
Do you think SFO fulfilled that duty to DLH?
There are various ATCOs on here adamant that an extra 2/3 miles on the spacing was enough to accommodate DLHs requirements. Why do you think that would cause significant delays and ground stops for aircraft outside the sector? Or is the estimate of 2/3 miles wrong?