Lufty at SFO

[1201alarm]

PukinDog, thanks, very interesting insight into the controllers "universe".

To me it seems well thought out what is done in case of "own visual separation". Such knowledge might help Lufthansa to consider their SOP's into SFO.

However doesn't change the requirement to not punish Lufthansa in case they do not "maintain own visual separation" and still bring them in.

[Request Orbit]

It does not matter. If they give you 160/4 (or whatever they give), and what should be a 4nm gap to touchdown becomes a 3.7 when you’re on a visual, that’s on you not them. Sure, they’re not regularly sticking a Citation XL 4nm behind an A380 and going “ah, visual, it’ll be grand” but the speeds they’re giving are not aimed at maintaining wake to touchdown on a visual. That is your job.

Generally speaking in a sequence like that the speed is to make sure you don’t slow early and back the sequence up behind, or to give the tower their requested gaps. It is nothing to do with wake from the aircraft ahead. It is your responsibility to say “UNABLE” if that speed won’t allow you to do what you need to do. You can’t ignore the speed given, but the reason it’s re-stated is the US has a published difference from ICAO where an approach clearance cancels previously iterated speed control (as you correctly stated). Pretty much anywhere else in the world the speed isn’t re-stated because it was never cancelled. The speed is for the sequence, not necessarily for the wake, and definitely not on a visual.

Example 1, normal sequence. You are number 1 for the day (ie, no wake ahead), I have a steady sequence spaced behind you. I clear you for the approach at 12nm and give you no speed. The approach clearance cancels speed, you come back to 140 knots at 10nm (entirely legally) and my sequence behind you is screwed. That’s why speed is re-iterated, nothing to do with wake.

Example 2, LVPs (/LVOs..?). You are number 2, number 1 is on a 3.5nm final. I’m doing 12nm gaps for the tower to get departures away. You’re doing an ILS in this case. I turn you to establish 16nm out and clear you for the ILS. I re-iterate a speed of 160/4 with the clearance because if I hadn’t, you’d have potentially (entirely legally) stayed fast or even sped up and eroded the 12, resulting in tower losing a departure. Speed issued, nothing to do with wake.

Vortex Avoidance Procedures (FAA AIM) 7-4-6 a):
“Under certain conditions, airport traffic controllers apply procedures for separating IFR aircraft. If a pilot accepts a clearance to visually follow a preceding aircraft, the pilot accepts responsibility for separation and wake turbulence avoidance. The controllers will also provide to VFR aircraft, with whom they are in communication and which in the tower's opinion may be adversely affected by wake turbulence from a larger aircraft, the position, altitude and direction of flight of larger aircraft followed by the phrase “CAUTION - WAKE TURBULENCE.” After issuing the caution for wake turbulence, the airport traffic controllers generally do not provide additional information to the following aircraft unless the airport traffic controllers know the following aircraft is overtaking the preceding aircraft. WHETHER OR NOT A WARNING OR INFORMATION HAS BEEN GIVEN, HOWEVER, THE PILOT IS EXPECTED TO ADJUST AIRCRAFT OPERATIONS AND FLIGHT PATH AS NECESSARY TO PRECLUDE SERIOUS WAKE ENCOUNTERS. When any doubt exists about maintaining safe separation distances between aircraft during approaches, pilots should ask the control tower for updates on separation distance and aircraft groundspeed.” (The capitalisation is the FAAs, not mine)

Pilot Responsibility 7-4-8:
a) Research and testing have been conducted, in addition to ongoing wake initiatives, in an attempt to mitigate the effects of wake turbulence. Pilots must exercise vigilance in situations where they are responsible for avoiding wake turbulence.

b) Pilots are reminded that in operations conducted behind all aircraft, acceptance of instructions from ATC in the following situations is an acknowledgment that the pilot will ensure safe takeoff and landing intervals and accepts the responsibility for providing wake turbulence separation.

1. Traffic information.
2. Instructions to follow an aircraft; and
3. The acceptance of a visual approach clearance

I’m not posting the entire thing but it goes on to list the techniques it would expect you to use (if in a lighter aircraft fly above the glide path, land longer than the preceding. If in the preceding heavy do not deviate from the nominal glide path etc).

SFO ATC doesn't pass responsibility for wake avoidance spacing to the pilot with the clearance for the approaches or the "maintain visual separation" instructions

That is 100% incorrect. Wake separation on a visual approach is pilot responsibility regardless of any assigned speed.

____________________________________________________________ ___________
************************************************************ ***********

edit 12/10/23:

To avoid detailing the main thread again, I’ll add these extra references here for anyone who happens to stumble across this at a later date.

AIM 7-4-5b) note

A common scenario for a wake encounter is in terminal airspace after accepting clearance for a visual approach behind landing traffic. Pilots must be cognizant of their position relative to the traffic and use all means of vertical guidance to ensure they do not fly below the flight path of the wake generating aircraft.

FAA AIM 5-4-23e) Arrival procedures - Separation responsibilities

If the pilot has the airport in sight but cannot see the aircraft to be followed, ATC may clear the aircraft for a visual approach; however, ATC retains both separation and wake vortex separation responsibility. When visually following a preceding aircraft, acceptance of the visual approach clearance constitutes acceptance of pilot responsibility for maintaining a safe approach interval and adequate wake turbulence separation.

AINOnline - The dark side of visual approaches

Pilots are required to see and avoid other traffic during a visual approach. In addition, pilots are responsible for wake turbulence separation.There is no requirement for ATC to provide wake turbulence separation to an aircraft on a visual approach.

NASA Callback 524 September ‘23

​​​​​​​Differing from an instrument approach procedure in significant ways, responsibilities such as terrain avoidance, traffic separation, wake turbulence, and cloud clearance migrate somewhat from Controller to pilot during a visual approach.

This one is worth a read as the first example is a Challenger 350 that experienced a wake encounter on final at…SFO.

IFALPA briefing: Visual approach considerations in the USA

​​​​​​​For an aircraft operating behind an aircraft that requires wake turbulence separation, the pilot can expect a wake turbulence cautionary advisory from the ATC when the approach clearance is received, however it is impossible for the pilot to accurately assess the position of wake turbulence. There are no tools to predict this phenomena, which is why the wake turbulence separation standards were implemented. By agreeing to reduce those standards (accepting the Visual Approach), the pilot may be increasing the risk of a wake turbulence encounter.

This document is superb, including full coverage of separation/speed/wake responsibilities and requirements with references. It is clear and very well written.
​​​​​​​

[megan]

Originally Posted by Del Prado
I’ve handled many fuel maydays, double figures in just one incident/closure, none of the captains were demoted or sacked.
What sort of strange blame culture do you operate under?

Well, some one understood the thrust of what I inferred.

Originally Posted by 1201alarm
I think what megan was alluding to was to actively provoke a Mayday to force the unwilling controller to do his job. I agree with that, that would be unprofessional.

[Bbtengineer]

Originally Posted by 1201alarm

BS.

The purpose of the airline indutry is to bring people to the place they booked.

Yes, but on aggregate.

Every single one of those passengers wants to arrive infinitely more than they want to arrive at the right place.

Let me try this a different way. Do we want the controller out of their comfort zone?

I hope we can agree that we don’t want that. We don’t actually want a controller finding out on the job where their personal limits are.

So what seems to have happened here is a controller soft pedaling on an important but non-urgent request for LH. One that we would all hope and likely they do to that they could have accommodated, but they clearly felt unable.

What we are left with, in my view, is why is the controller so close to the edge of their envelope that they just cannot countenance accommodating this request more timely. That they will have or care to find a reason not to?

And then by consequence, what would it take to put them back into their comfort zone for when this situation inevitably arises again?

TL;DR what is required to fix it?

[Request Orbit]

Originally Posted by Bbtengineer

Let me try this a different way. Do we want the controller out of their comfort zone?

I hope we can agree that we don’t want that. We don’t actually want a controller finding out on the job where their personal limits are.

So what seems to have happened here is a controller soft pedaling on an important but non-urgent request for LH. One that we would all hope and likely they do to that they could have accommodated, but they clearly felt unable.

What we are left with, in my view, is why is the controller so close to the edge of their envelope that they just cannot countenance accommodating this request more timely. That they will have or care to find a reason not to?

And then by consequence, what would it take to put them back into their comfort zone for when this situation inevitably arises again?

TL;DR what is required to fix it?

Finally. It all comes down to capacity, both of the controller and the airspace. It’s been well established this was a busy rush of scheduled planes so it will have been forecast on whatever traffic monitoring systems exist there.

Assuming Bbtengineer is on the right lines: Why was this controller put in a position where they had no capacity left to handle anything else? Factors that come into it: airspace, controllers personal capacity, training, expectation, staffing, airspace and procedure design. Was the position further splittable? If it was, were there enough staff to actually split it? Was there pressure on to power through and not split? Would the split actually have helped? What would have happened if there was an actual mayday in the middle of that? The NYT article did a good job of highlighting that US ATC are handling relatively, more planes than ever with fewer staff than ever. Is the airspace evolving to assist that, or are they just trying to cram as many planes into an already creaking system as possible?

After near on 400 comments we’ve gone round and round about what happened, without ever getting any closer to why it happened.

[1201alarm]

Bbtengineer, what you say is fair.

But to arrive at the discussion about the "why", first we needed to establish that the controller should have sorted Lufthansa into the arrival sequence. Already this was disputed by many contributers.

Personally I still have the feeling that is was more a problem of unwillingness by the controllers side and not a problem of unableness. But this is for others to decide, if there even is an investigation.

As we have learned the controller anyways has to variably sequence the arrivals, some need more space, some less. Lufthansa would have just been another one with a bit more spacing.

[Uplinker]

KSFO have built parallel runways that are too close together, (probably owing to lack of space), to allow ILS approaches when both runways are in use simultaneously.

KSFO have decided to reduce safety margins in order to get more traffic on the ground during busy periods.

LH's safety department - or their insurers - have undertaken a risk-assessment and decreed that night visual-separation approaches at KSFO are not quite safe enough, so have banned them.

This LH crew were - or sounded - cocky and unprofessional, and used a swear word, which (understandably) got ATC's backs' up.

ATC allowed their (understandable) annoyance with LH to influence their handling of the situation.

ATC did not issue an EAT.

USA ATC is apparently a difficult place to work right now, owing to controller shortages ? Don't know, but US controllers at large busy airports often sound stressed-out to me.


Not having a go at any one party here; just as I see it.

[1201alarm]

Not having a go at any one party here; just as I see it.

You forgot to see the main point: for 40 minutes KSFO ATC did not make any effort to land an IFR flight at KSFO while other aircraft who arrived later were landed.

[Request Orbit]

Originally Posted by Uplinker

This LH crew were - or sounded - cocky and unprofessional, and used a swear word, which (understandably) got ATC's backs' up

Agree with 1201alarm, in my view they sounded entirely professional up to that transmission, which itself came after 40 minutes of poorly explained and un-updated delay. It wasn’t like they rocked up on frequency with that attitude to begin with, and the lack of updates on the delay was directly linked to that change in attitude. For the next point, that may or may not have led to a change in how ATC handled them, but it was so busy anyway it could have made no difference. We don’t know for sure.

Excellent summation otherwise.

[Uplinker]

Originally Posted by 1201alarm

You forgot to see the main point: for 40 minutes KSFO ATC did not make any effort to land an IFR flight at KSFO while other aircraft who arrived later were landed.

OK, fair enough - but I have mentioned that in a previous post. Here, I was trying to think of the "WHYs" as you asked about earlier, rather than what did or did not happen.

However, I did say that ATC did not issue an EAT, which was part of the problem; If ATC could not reconfigure the approaches and calculate an EAT, then why not ?

[1201alarm]

Uplinker

[Bbtengineer]

Originally Posted by PukinDog

I'm well-aware of who has responsibility for what when accepting a Visual Approach and what my own is under FAA rules. Instrument flying 101. Thanks though , I've been aware of them for about 40 years or so. Not sure if this this is your first foray into them because...,

I hate to break it to you, but I'm not incorrect, let alone glaringly, because (from your own source) https://www.faa.gov/air_traffic/publ...section_7.html.

The generic Visual approach rule you referenced assumes the Visual approach clearance is received with no accompanying or further speed restriction issued by ATC. Speed-control restrictions are, as already mentioned, commonly re-stated/issued at SFO at the time Clearance for these 28 Visuals are issued. This restatement of a restriction is no different than for any Instrument approach when ATC desires or needs to retain positive control of spacing since issuance of any approach Clearance, Instrument or Visual, automatically cancels any previously assigned speed restriction.

Ignoring the effect that restating the speed restriction has on the Visual approach clearance when issued is not unlike ignoring that there is such a thing as a conditional clearance. For the purpose of clearing up your misunderstanding, it's not an unimportant detail.

Because details matter, ATC can and does restate/issue a speed assignment with the Visual approach clearance and it essentially becomes part of that Visual approach clearance, so the generic rule you reference does not invalidate the restated/issued speed restriction (as it does a previously-assigned one). And, for the same reason ATC restates a speed restriction for an Instrument approach Clearance when they want to retain positive control for spacing, ATC does the same when issuing a Visual approach clearance; to retain positive control of spacing.

The only difference between a Visual and Instrument approach as far as Speed control and wake is what happens within 5 miles of the runway. But the subject is creating space to plug LH into the stream and the gap ATC needs to create for wake, not something that happens within 5 miles on any approach.

But, since you keep making your assumptions and hate my Manifestos, I'll assume you enjoy your own sources, so here's an excerpt from yours in the link above. Note, especially, points #9 and #10..

So yes, ATC does require one to fly a speed so restates the restriction when issuing the Clearance in order to retain positive control for required or desired spacing. SFO does it, as per the reasons in bold, and they are not alone in doing so. They also restate speed restrictions during Visual approaches at JFK, DCA, LGA...pretty much anywhere Charted Visual approaches are conducted and for the same reason. This shouldn't come as a shock to anyone except those who refuse to believe it's not a highly unsafe, free-for-all up there with pilots jockeying for position all over the sky just because "Visual" appeared in the Approach Procedure. No, it's pretty organised, controlled, and happens hundreds of time every day where Charted Visual Procedure are in use.

On the issue of "maintain visual separation" again, excerpts from your source... https://www.faa.gov/air_traffic/publ...section_4.html

The "maintain visual separation" instruction for pilot-applied visual separation is necessary if ATC wants to have aircraft positioned 750' apart side-by-side in pairs tracking the charted Visual inbound courses. While tracking those Visual inbound courses the aircraft targets on Controller's display with touch/merge, so pilot-applied visual separation becomes necessary in order to conduct them in that manner.

If ATC decides to alternately stagger aircraft with sufficient in-trail spacing during the Visuals (which does happen during off-peak hours), there is no need for the pilots to accept and confirm they will "maintain visual separation" when positive speed control is retained and ATC provides the approved separation.

The "maintain visual separation" instruction has nothing to do with the pilot jockeying his own speed to manage the in-trail spacing between himself and the aircraft miles ahead already spaced for wake turbulence by ATC. The Controller, by restating a speed restriction issued in conjunction with the Visual Clearance, is already providing-for and maintaining the wake turbulence spacing under positive control for as long as the positive speed control is valid, That's how they do it at SFO during the paired Visuals. Please don't keep repeating how something theoretically works, or can't work, until you've done sone ride-alongs in the cockpit to see how they work the Procedures.

What ATC is unable to do is see or provide for is lateral separation between the closely-spaced (750') paired aircraft, abeam each other with a slight stagger. If pilots are flying the Charted Visuals correctly, established on the inbound courses, the lateral separation will be there because that's how the Visuals are constructed. One is straight in, the other is Offset. The pilot only need to visually ensure and maintain the separation that flying to procedure establishes. Again, nobody has gets a clearance to eyeball and fly formation with anyone. The clearance is to fly a Charted Visual Procedure that has electronic lateral guidance and fly them at the ATC-assigned speeds until within 5 miles.

The excerpt directs a Controller "Do not permit an aircraft to overtake another aircraft when wake turbulence separation is required". The Controller can only prevent what he/she can clearly determine on the display, and the only way to prevent an aircraft from overtaking is through speed control. The Controller can see and prevent this for the pairs in-trail of other pairs, but the controller cannot see that the slight stagger between the aircraft that make up each pair, where pilot applied visual separation is, is being maintained. If a high-side aircraft in a pair overtakes the other and pulls ahead far enough, wake can become an issue. Therefore, the instruction.

Sure, yes, but this is getting lost in the weeds.

The controller failed to provide a published service to LH.

What you are describing is a laundry list of reasons why LH could have accepted that failure on the part of the controller to deliver the published service, and deal with it anyway.

What you are not tackling is why the controller could not provide a service that was published and therefore reasonably expected.

[galaxy flyer]

The reason is… traffic. KSFO has an inbound flow of traffic that won’t work unless the pilots accept visual separation. Well, it’ll work with except the delays and holding will go back to SLC Center. Without visual separation, it the acceptance rate less than half. Theres lots of published services that cannot be offered. 01L/R have published approaches, try getting it when the 28s are for landing, 01s departures.

[Bbtengineer]

Originally Posted by galaxy flyer

The reason is… traffic. KSFO has an inbound flow of traffic that won’t work unless the pilots accept visual separation. Well, it’ll work with except the delays and holding will go back to SLC Center. Without visual separation, it the acceptance rate less than half. Theres lots of published services that cannot be offered. 01L/R have published approaches, try getting it when the 28s are for landing, 01s departures.

I can see why this thread is going round in circles.

It’s one aircraft. They didn’t request the other runway.

Why is it reasonable for you to extrapolate to all aircraft and halve the acceptance rate?

Why is it reasonable for you to compare it with a runway switch that they didn’t ask for?

[Request Orbit]

In the SFO perspective response video, they said they treat it as though it’s asked for an opposite end approach. Why they’d choose to treat it like that I have absolutely no idea, and is why a lot of this make no sense at all.

[ATC Watcher]

Originally Posted by galaxy flyer

The reason is… traffic. KSFO has an inbound flow of traffic that won’t work unless the pilots accept visual separation. .

And that is indeed the root of the problem : Declared capacity. For economic reasons the acceptance rate into SFO has been increased to such an extend that to make it work you had to bend the normal rules. Now if someone asks for the rules to be applied the numbers can\t be reached , The issue here is that ATC, a safety service provider has been turned into an economic enabler, the numbers being more important than the separation standards . , and with time ,perhaps with the help of their local management, some of the local controllers were led to believe that it was their duty to provide numbers instead of applying the basic safety service they were initially trained for .
The rest are just details.

[51bravo]

Originally Posted by Request Orbit

In the SFO perspective response video, they said they treat it as though it’s asked for an opposite end approach. Why they’d choose to treat it like that I have absolutely no idea, and is why a lot of this make no sense at all.

Why? Maybe a YT thing. Exaggerating things out of proportion.

Just as academic excercise, if someone would be so kind: What a gap would be needed to be created in the 28 inbound stream if LH indeed requested an opposide 10 approach?

[safetypee]

ATC, #396;

Bottle that and put it on the top shelf … no, send it to the FAA via a safety report.

[Bergerie1]

ATC, You are absolutely right, and I speak as one who was involved in the Eurocontrol PRC comparisons of US ATC vv European ATC.

[Uplinker]

+ 1 ATC Watcher, very good summing up

This incident has highlighted a practice that is not primarily safety driven.