Asiana flight crash at San Francisco

[nitpicker330]

There is an RNAV GPS Approach available for R28L
They could have programmed the FMC and followed LNAV VNAV right down to the runway if they had wanted to.

Maybe SFO will have to stop giving visual approaches to foreign carriers AND they will have to stop requiring 180kts to 5 nm.

[DaveReidUK]

To say the ILS outage was a contributory factor is nonsense.

In NTSB-speak, the absence of a glideslope, like pretty well any other aspect of the accident, is classed at this stage simply as a "finding" - a fact or conclusion that may or may not have had a bearing on what happened.

By the time the investigation report is published, all findings will have been classified in one of 3 categories:

A. Probable Cause (of which there can, of course, be more than one)
B. Factor (a situation, circumstance or event that contributed to one or more of the causes)
C. Event (anything that isn't a Cause or Factor)

There is no specific sub-category for something that had a bearing on the accident, but really shouldn't have had ...

[Flyer94]

It's been a few years ago when I travelled on the jumpseat with Korean Air. The Captain instructed "his" F/O to demonstrate some avionics, using a pointer to refer to the devices to be explained. Some minutes later the F/O wanted to answer an ATC call, but he was reminded by the Captain's pointer, patting to the F/O's fingers, to first finish explanations before talking to ATC.

Since that day I understood that KAL is light years away from good CRM and that this F/O never would mention any irregularity caused by the captain.

However, I do not understand why the Koreans, after all these years and accidents, are still not able to establish flat hierarchies and a good CRM in order to reduce the number of corresponding accidents.

Regarding the UAL crew: maybe they assumed Asiana struggling with a technical problem and decided not to burden with radio messages. Additionally, everything happens fast, it's hard to capture the unexpected situation, quickly consider a solution and take action - if only a moment of time remains.

[PaperTiger]

Originally Posted by ross_M

Any stories of a bad landing that was salvaged by an alert crew on the ground?

I recall one at ORD when a crew at the hold (UA I think) called "Continental's got no gear!" as a CO 727 approached. The CO (sans wheels) banged the tail but managed to go around.

Originally Posted by nitpicker330

I haven't flown into SFO as crew but looking the sat photo and TV shots where exactly is this "laser" light supposed to have come from when they were at 500' approx 1 .5 nm from the sea wall over the open water?

Originally Posted by ross_M

Do we get many people trying to shine l@sers into cockpits during a bright July midday sun?
...
Besides isn't most of that approach over the sea?

I think it more likely to a have been a flash reflection than a laser, maybe from the holding UA 744. However, the SFO 28s approach is over water but parallel to the shore along which is a footpath and some public parks, so it's possible just not likely IMO.

[draglift]

This forum makes painful reading as those who do not fly commercial aircraft make knee jerk reactions and try to tell professional pilots how they should be flying.

Saying you should not fly an approach without an ILS glidepath is a bit like someone who does not drive observing a car skidding off a road in the rain and trying to legislate that cars should not be allowed to drive in the rain on the basis that the accident would not have happened on a dry road.

Saying that pilots should not fly an approach without an ILS glideslope would simply mean pilots becoming overly dependent on ILS and not able to fly visual approaches, ie it would makes the situation worse.

I spent many years flying 747s to Africa and the Caribbean and rarely saw an ILS. The more approaches you do without an ILS the better pilot you will be.

[legomaniac]

Hasn't anyone listened to the ATC tapes? You can clearly hear "go around" called when Asiana is on short final. That "go around" was called by the UAL flight crew holding short.

[legomaniac]

at 15s "go around"

Asiana 214 KSFO Crash Landing ATC - YouTube

[rottenray]

Being allowed to fly a couple of visual approaches early on in the Sim (I accept I have no idea if that's accurate) and being signed off isn't really proper training for doing it after long flight and being given speed constraints several months later.

I could buy that if this were a low-time pilot. But this pilot was not - he had around 10k total hours and was previously a PIC domestically on A320. At some time during those many hours and sim sessions, the PF must have had to do a visual or two.

I can understand a poor performance (hard, slightly long or slightly short landing) but I can't understand what appears to be a complete loss of memory on how to do a visual.

I'm becoming more and more curious about sterile cockpit in this accident.

[joema]

"There is an RNAV GPS Approach available for R28L
They could have programmed the FMC and followed LNAV VNAV right down to the runway if they had wanted to."

Apparently RNAV GPS with WAAS augmentation was available at SFO and provided an artificial glideslope. The out-of-service GS was planned and the RNAV GPS was provided as an alternative precision approach -- including vertical LPV guidance. See FAA bulletin, section 4: http://www.faa.gov/documentLibrary/m...N_8900.216.pdf

"For about three months in mid-year 2013, the FAA will render the LDA for runway 28L and the ILS for runway 28L out of service (OTS) at San Francisco International Airport (SFO) due to runway construction. The loss of these navigation aids will eliminate the ability for SFO to conduct PRM approaches during simultaneous offset instrument approach (SOIA) operations. The FAA plans to publish RNAV (GPS) PRM procedures prior to this navigation aid shutdown."

RNAV GPS background: http://www.faa.gov/about/office_org/...l_06122012.pdf

So the inop GS need not have affected Asiana. They had equivalent vertical guidance available (ie an artificial glideslope), had they chosen to use it. If they didn't feel comfortable hand-flying, they could have made a fully-automated approach. The equipment was available.

[The Ancient Geek]

Let me make this absolutely clear for the hard of thinking.

ILS is an approach aid intended to allow landings when visibility falls below the minimum standards required for a visual landing.

There is no requirement for any airport to have an ILS, if there is no ILS and the weather falls below minimums then flights will be diverted to an alternate airport. This costs the airport in lost revenue so fitting an expensive ILS system is a commercial decision based purely on financial advantage.

There is no requirement for a crew to use an ILS unless the visibility falls below visible minimums, in which case it is mandatory to use the ILS or an equivalent system.

There are thousands of airports around the world with no ILS and big jets fly into them every day with no problems - this is perfectly normal and what pilots do to earn their living.

In other words - ILS is IRRELEVANT unless the weather is bad.

[Groundbased]

Course_profile

"ILS dependency isn't a problem when everywhere has one"

But it is when he one you want to use suddenly goes u/s.

[Lonewolf_50]

joe, thanks for that fact.

draglift: thank you. While you may be right, (my gut and experience says that you are) I don't think that a lot of airline company management groups agree with you. That's the nut that has to be cracked, don't you think?

[Keylime]

As they say in the infomercials, "But wait, there's more". Five days before the Asiana accident, a KAL B747-400 on approach to runway 28 in SFO descended to ~70' AGL before pulling up and getting back on a proper descent profile for landing(rather than go-around as per SOP's). The SFO tower has a low altitude alerting system. There should be a record of this deviation. Makes for interesting commentary with reference to some of the posts here concerning visual approach proficiency.

[givemewings]

Spandex, I realise it doesn't always work like it does on paper, however emergency services are more than just firefighters or medics. Response planning will have someone allocated to round up and direct the pax , maybe not right away but definitely once the rescue efforts are underway.

[Lonewolf_50]

Causes and effects aside, and mitigation, the NTSB has been releasing a lot more info a lot faster than I am used to seeing after a fatal accident.

Regarding information release, ALPA's justifiable complaint with the NTSB in this case, and the relationship between FOQA and flight safety, I recommend a look at posts by PJ2 in the thread discussing another very controversial wreck (AF 447) which has elements of systemic overreilance on automation as a factor.

The post I linked to is an explanation to a question by a novice. It may be worth going a few posts up to see the conversation start and then digest PJ2's explanation in the linked post.

For all the complaints I have seen in this thread about the FAA, US airports, and ATC in large cities, I'd ask our foreign colleagues to consider how important to safety culture the FAA's approach has been to flight data from accidents, legal protections, and more. (The US military has a similar accident data protection system in place)

[Keylime]

Courseprofile:

How did aviation get here? Over-reliance on automation?

You are correct. Another aspect of this is; Training is a necessary evil as far as the bean counters are concerned. It is another cost to the operation they have to endure. Therefore, much effort is spent trying to make the training "footprint" as small as possible. Less time away from the line is less money. We spend time teaching how to type in the box, not how to manage the automation. If we trained as we should an initial/transition course would be 1-2 weeks longer. Imagine the cost. Training managers bonuses are based on how much they are under budget. And so on and so on. Don't expect this to change any time soon. Just another part of this whole puzzle. The question is not necessarily over reliance on automation but maybe misunderstanding of how to use the automation.

[Lonewolf_50]

Also - it wouldn't matter about bad weather if we did use ILS/Autoland everywhere.

Course_profile, consider the KISS principle, if you please.

Then consider the implication to both flight safety and traffic flow if every time an ILS or G/S went U/S (or a bit flaky) a runway was shut down in VFR or weather other than " at minimums."

You'd be keeping more airplanes airborne longer in the volume of air around an airfield for no good reason. So doing will decrease safety margins and increase the opportunity for error or mishap.

I don't think you've thought that idea through very well.

EDIT: Ancient Geek, thank you sir, well said.
EDIT 2: Course Profile, point on experience taken, and agreed based on my own experiences as well. Gravity and physics don't give a hoot how many hours you (or I) have.

[737 Elpiloto]

I wonder, both pilots flew Airbus before, if the change in automation level has something to do with the events...?

Also, I am convinced that a pilot should be able to fly visual approaches.
It's the one time we can be complete pilots again and handfly the aircraft as we use to do...

I would think that if you really to much on automation, you get stuff like this when they have to handfly the aircraft..

Just my idea....

[Javadreaming]

Course... Based on your logic... All planes should be operated with 7 pilots on board because that would be safer right? Your obvious lack of understanding of aviation means you should not post - but everyone has an opinion right! It is your logic that actually makes aviation less safe. If we said every approach has to be flown using an ILS ( by the way most approaches don't use an ILS, only larger airports have these...usually) what would that mean if the plane failed to capture the ILS?

Many airline accountants think like you which has lead to a reduction in pilot training (cause we don't ned to fly visual approaches right) so training does not focus on this... Well we reap what we sow right...or something like that

[rottenray]

I always figured that was because he was part of the furniture, how does a second tourist look at a guy with 25 years experience and go 'Mate, what's going on?'.

... which brings us right back to the discussion of "gradient," in particular, those situations where some are bestowed with unconditional respect based upon characteristics which aren't necessarily relevant.

A pool of fuel I'd rather not reignite.

Course, I can at least understand your approach here: Put in all the aids possible so that the least-skilled (and those having bad days) are as safe as the most-skilled.

I would agree, if we were talking about something everyone did - like drive to work or to here or there. Certainly, it's in everyone's interest to make lanes wide, curves gentle, et cetera.

But in the case of this accident, I'd rather see things go the other direction. If it proves that 214 crashed because of too little hand flying / visual approach experience, then the industry needs to work at providing more hand flying and visual approach experience.

Because, in the grand scheme of things, an ILS OTS at a 2-mile-long runway on a beautiful sunny day is rather near the very low end of bad things that can happen during the course of a long haul flight.