Hacker News Reader

Comment by noxvilleza | Hacker News Reader

This wasn't some new policy of Lufthansa though, apparently it's their SOP for basically all airports. Outside the US, I'm not sure if visual approaches are all that common for (heavy) aircraft at night. Overall, the reason that SFO wants visual approaches is to increase rate of landing, and the reason that Lufthansa wants ILS is to increase safety -- your phrasing "Lufthasa is the one making this hard on everyone" just seems wrong, having more safety seems totally justified and reasonable here.

I'm actually surprised SFO still allows visual approaches at night after that Air Canada 759 flight nearly landed on the taxiway ~5 years ago (https://www.youtube.com/watch?v=aGQlQFn0euI).

View on HN

Instrument approaches are the norm in Europe, visual approaches are the norm in the US. Visual approaches can and normally do still use ILS when possible; SFO's incidents were when ILS was out of service (Asiana 214) or not engaged (Air Canada 759)

Which is why ATC obliquely asked whether Lufthansa bans visual approaches altogether, or simply requires the use of ILS. If it had been the latter, that's fully compatible with normal SFO operations.

> I'm actually surprised SFO still allows visual approaches at night after that Air Canada 759 flight nearly landed on the taxiway ~5 years ago

The Air Canada incident happened because one of the runways at SFO was closed for maintenance, and after it happened the FAA specifically updated their regulations to require ILS when there's a possibility of runway confusion. There's no reason to think VFR landings at SFO are unsafe in normal conditions.

> There's no reason to think VFR landings at SFO are unsafe in normal conditions.

"No reason" seems like a very strong claim. Is that the reason Lufthansa cites for prohibiting visual separation at night? I mean it would clearly be in their interest to increase throughput as well, so they must think there's a reason. Why do you think that reason is invalid?

I'm surprised a computer assisted landing sequence can't _increase_ the rate of landed planes, though I assume doing so would require subordinating all the inbound aircraft to the local traffic control system.

The problem with having computer separation be less than pilot separation is, what happens if the computers fail and the pilots have to take over? Then they're suddenly in a situation where they're already below whatever the minimum safe pilot separation is, and now you've potentially turned a recoverable failure into an unrecoverable one. If the whole point of having pilots is to be backups for the computers, the computer situation has to have more margin than the equivalent pilot one, even if efficiency is being left on the table.

(The only way around this is to not have pilots as backup, and turn the whole thing over to the computers, but we're not ready or willing to take that step yet)

'go around' but with different registered abort vectors / plans.

Also if any unit within the stack falls out of automatic mode the entire chain gets cleared; if there's sufficient window for the pilots to declare emergency manual landing they can maintain the path as everything else aborts and the failed unit lands.

Considering that only one runway (28R, IIRC) at SFO even winks in direction of supporting autoland, and there's no continental USA support for controller-pilot datalink (only used for oceanic flight), at least outside of test flights (correctme if I'm wrong, not an expert here).

On top of that current CPDLC doesn't seem to be as well integrated to autopilot support other than pilots acking each command, and then I do not think Mode S is mandated (or exact enough) to provide good enough separation.

> reason that Lufthansa wants ILS is to increase safety

Yes because "more technology is more better" right? And forgetting about how to do things by hand (or eye) doesn't pose a risk to anyone neither

A visual approach is no more unsafe than an ILS one in good weather. Sure, ok, the caveat here is "at night" but I don't think the multiple pilots that were doing it at the time were being risky on purpose

(and people are quoting Asiana, but understand that fumbling a visual approach landing is not a thing that should be common)

> Yes because "more technology is more better" right? And forgetting about how to do things by hand (or eye) doesn't pose a risk to anyone neither

That's a very nifty strawman, but that's not my claim (at least not in such a general way). In this case it's become SOP by experts in a field, specifically for safety reasons. Other experts in the field have said "well, we don't think it's necessary - there are downsides such as aircraft landing rate". These groups of experts don't disagree on the fact that it's more safe - they just disagree as to what the optimal combination of (safety, financial value) is appropriate.

(speaking about strawmen) I'm not saying ILS is less safe

I'm saying that have people rely on ILS and never practice visual landings, especially when the conditions are favourable, makes pilots lose skill and confidence on it, making the overall situation less safe

Then people wonder how come other pilots miss a perfect good approach on a sunny day.

> your phrasing "Lufthasa is the one making this hard on everyone" just seems wrong, having more safety seems totally justified and reasonable here.

It's so "justified and reasonable" that nobody else does it. Maybe it's not quite so justified or as reasonable as you think.

> It's so "justified and reasonable" that nobody else does it.

Airlines are a cutthroat business and many will go for profit rather than for safety if they're allowed to. The large American airlines, for what it's worth, are actually loss leaders [1].

[1] https://happyrichadvisor.com/loss-leaders/

> that nobody else does it.

Apparently some Canadian carriers do also have this as a SOP.

Around the time this happened I spoke to some friends who are ATCs (in the US) who all immediately agreed it was a very reasonable request, especially given that the request was made far enough out (so it wasn't like they'd have to quickly scramble to respace the incoming planes correctly in the sequence).

Because Lufthansa left a few hours late they arrived during a super busy arrival window. AFAIK ATC had nearly 30 planes already in the queue for landing with spacing for visual. To get Lufthansa in any sooner they'd need to send updated instructions to a lot of planes to make a gap.

If Lufthansa had arrived two hours earlier or later it wouldn't have been an issue. Indeed they were able to depart OAK around two hours later and land at SFO via IFR with no problems.

SFO handles a lot of traffic for having just two active runways - one of the reasons they constantly operate in parallel. Much like other super busy constrained airports (eg JFK) they have very little room to accommodate special requests.

Ideally Lufthansa would have let ATC know of this need while still a long way out so they could build a bubble in the sequence ahead of time but I don't know if procedures even allow for that.

What's the rationale? Maybe the jetstream reverses and the 2-hour delay leaving MUC gets erased during flight?

According to other posts here it's something the FAA specifically recommends (but stops short of requiring).

Reply on news.ycombinator.com