-
Our new ticketing site is now live! Using either this or the original site (both powered by TrainSplit) helps support the running of the forum with every ticket purchase! Find out more and ask any questions/give us feedback in this thread!
Inquiry after BA jet stalls over London on approach to Heathrow
- Thread starter Jimini
- Start date
Sponsor Post - registered members do not see these adverts; click here to register, or click here to log in
R
RailUK Forums
gabrielhj07
Established Member
500fpm is a rather more relaxed way of describing that descent.
SargeNpton
Established Member
- Joined
- 19 Nov 2018
- Messages
- 1,706
“It is unheard of for a captain to physically intervene in the cockpit.”
This puzzles me. I always assumed that the captain was the pilot, and would be in charge of the takeoff/landing. Or at the very least correct any error the co-pilot makes as a matter of course.
This puzzles me. I always assumed that the captain was the pilot, and would be in charge of the takeoff/landing. Or at the very least correct any error the co-pilot makes as a matter of course.
Nope, roles are defined for each flight, with a "pilot flying" versus "pilot monitoring"
gabrielhj07
Established Member
The Captain remains in command of the flight but the actual flying can be delegated as he sees fit.
A more explanatory version of events
hard to do a precis on here as you really need the whole article and I don't want to bust copyright
hard to do a precis on here as you really need the whole article and I don't want to bust copyright
What a shockingly sensationalist article - full of over the top language and clearly written by someone with very little understanding of aviation.
It is most certainly not “unheard of” for a captain to intervene in an irregular situation, to point out one of the many flaws with this piece.
Tetchytyke
Veteran Member
From what I've read on pPrune, it seems the aircraft had a slight tech issue and moved into 'alternate law' on the fly-by-wire systems. The First Officer, who was pilot flying, didn't react correctly to that, so the Captain as pilot monitoring intervened.
Not great, hence the AAIB getting involved.
Not great, hence the AAIB getting involved.
It is correct. Alternate law is a degraded flight control mode on Airbus planes.
edwin_m
Veteran Member
It is an Americanism but also the accepted terminology.
It is the term that Airbus have used since the A320 was designed in the 80s. There is normal law, alternate law and direct law.
gabrielhj07
Established Member
The BBC join the race to the bottom for journalism standards:
https://www.bbc.co.uk/news/articles/cn9n2v3lwr0o
Quite a poor article. Vox pop-style interview with an ill-informed passenger that adds basically nothing to our understanding of the incident. Hopefully they'll write something more worthwhile when the AAIB have something to say.
https://www.bbc.co.uk/news/articles/cn9n2v3lwr0o
Quite a poor article. Vox pop-style interview with an ill-informed passenger that adds basically nothing to our understanding of the incident. Hopefully they'll write something more worthwhile when the AAIB have something to say.
edwin_m
Veteran Member
I think it's fair to report how it made people feel. The article does go on to explain more about what actually happened, and there are plenty of more authoritative sources for those that want more detail.
Thanks. It was a little confusing (for me) - when I saw it at first, I was thinking it meant alternating between two rules.
GRALISTAIR
Established Member
Captain Sullenberg uses the line "My plane" and takes over from First Officer Stiles.
Cloud Strife
Established Member
- Joined
- 25 Feb 2014
- Messages
- 2,962
Usually the roles are switched for each leg, although the captain can choose otherwise depending on the situation. Some airports mandate that only the captain flies into them, or for instance, a tired pilot might ask for the other pilot to fly a particular leg. There's no set rule for it, some captains are quite happy to let the first officer do most of the flying so that they can build hours in control, or they might do most of the flying themselves. Or you might have a situation that they're flying to an airport that the FO doesn't know, so the captain will get them to fly to/from there.
There are usually some rules such as requiring the captain to land following an emergency situation, but other than that, it's mostly up to the captain to make a decision. I know one Ryanair captain who is a great teacher, and he always asks the first officers to take on most of the flying so that he can help them build the skills they need to get into the left seat. His view of the matter is that young pilots need to fly, and that an old guy like him doesn't need to do most of the flying duties.
Alternate is correct. Alternative has a different meaning:
Alternate: it means a backup, substitute, secondary option, used when the primary option isn't available
Alternative: a choice between two (or more) equal possibilities.
Alternate law is not a choice, it's a fallback safety state. So, it's not an Americanism, it's just the correct way to describe what it is.
It's one of the best things about the entire Airbus design, but one major issue is that many pilots have no real experience with alternate and direct laws. I've always thought that it should be mandatory for every pilot to fly in direct law at least once a year, just to make sure that they can handle the lack of protections available.
CAT C airfields can be Captains only landings, depends on the operators SOPS which are developed through operator specific risk assessments which can be fluid and, if operating mixed fleets, be fleet specific.
What, you mean like recurrent training..? Like sn LPC/OPC which happens every 6-12 months?
That's not what I was taught.
For example. We would say "I took an alternative route because of roadworks" not "I took an alternate route because of roadworks"
Two options are an alternative. You can alternate between two options.
Anyway, i don't want to drag the subject off topic
Last edited:
Tetchytyke
Veteran Member
I think that's rather a harsh interpretation.
It does sound like the passenger is describing the TOGA rather than the preceding issues which led to the TOGA. But TOGAs and rejected takeoffs are pretty scary things, even for more experienced fliers.
They do do LPCs in the simulator which should include this but, yes, as we saw with Air France 447 it can be a real issue.
The AvHerald article linked above suggests it was a stall warning.
A stall at that kind of altitude might be difficult to recover from, depending on the circumstances. I think we'd know about it if that had happened.
A TOGA is a thrust setting. Take-Off/Go-Around. What they’re presumably referring to is a Go-Around.
The 447 accident wasn’t a lack of understanding of Alternate law, it was a case of dual input leading to inadequate stall recovery (Captain was trying to unload the wings by pushing forward on the column whilst the FO was simultaneously trying to pull up -dual input sums the inputs so the stall recovery technique employed by the captain was effectively cancelled out) The initial cause was a frozen pitot tube resulting in spurious airspeed information. All of which was adequately trained for prior to 2009. However stall recovery practice prior to this event was power through by maintaining pitch up and selecting TOGA to minimise altitude loss. An arguably counter-initiative action when considering the physics of a stall, but it’s believed the old-school preferred the idea of not pointing the aircraft at the ground. This is one change that was made following the accident.
Startle factor again too, this is what resulted in the FO putting in incorrect control inputs. However in my APS training (which I completed on Airbus) it was drilled into us that during the stall recovery (and any upset recovery) the first call is ‘I have control!’ whilst simultaneously pressing firmly on the take-over push button. After holding this for 40 seconds the system latches on to that control column. The Captain on 447 did not do this, he assumed the FO was not inputting on his sidestick. As the takeover PB was a known but not necessarily formally trained this first call and action was added to standard recovery by Airbus as a result of AF447.
Indeed it was AF447 that contributed to the requirement to complete Upset Prevention and Recovery Training (UPRT) for all low hour pilots prior to the first type rating. It also forms part of the LPC/OPC and other recurrent training. I was one of the ‘lucky’ ones to complete after 2020 and therefore had to complete Advanced UPRT which is completed in aircraft (sims, even full motion ones, cannot replicate the sensations of an upset - least that’s the theory of why it must be completed on aircraft) and so I was doing all sorts of unusual manoeuvres in an Extra 200. All good fun, but for anyone who’s done quite a bit of instrument flying, seat of the pants has to go out of the window somewhat. The paradox of AF447 I guess.
In the case of the BA flight, looks like Normal Law dropped out as a result of ADR failure. This removed various protections like Alpha Prot, Alpha Floor etc. As a stall is the exceedance of the critical angle of attack (not a set speed as is typically believed) then it’s possible that the ADR failure included AOA sensors which may have been issuing spurious data. The aircraft may not have stalled in actuality. The AAIB will take this seriously though as LOC-I (loss of control in-flight) is a serious threat.
This is why I cautioned on the (thankfully now temporarily closed) AI171 thread that catastrophic accidents are almost always systemic — not the result of one neat, isolated root cause. AF447 is the classic example of the Swiss Cheese model in action: multiple small failures, none fatal on their own, lining up at the same moment. Crucially, as I have explained, the learning points were identified and acted upon.
Incidentally it took 3 years from the AF447 until the investigators published their findings. This lead to a similar rag tag of armchair experts identifying what had occurred simply by looking at press snippets. Striking similarities to AI171!
Last edited:
Terry Tibbs
Member
AF 447 was predominantly a failure to recognise what was actually happening to the aircraft and what the correct response should be. The dual input was, sadly, a failure to clearly state who was actually flying the aircraft. The captain was on his break and it was he, when he entered the cockpit, who recognised that the aircraft was in a terrible stall but by then the situation was completely unrecoverable.
The 40 second control disconnect press is irrelevant in this case as the aircraft was never going to be recovered within 40 seconds. UPRT usually does not involve using the control disconnect in my experience.
Understanding of what Alternate Law looks like on the instruments and how this manifests itself within the protections of the flight envelope is just as important as knowing what Alternate Law actually is.
Airbus have addressed the issues surrounding the rare occurrence when instruments give different readings from their respective sources. On their latest aircraft, the FADECs (Full Authority Digital Engine Control) are used also as back up to provide accurate readings.
I would caution against reading anything on aviation websites and jumping to conclusions with statements like ‘The First Officer, who was pilot flying, didn't react correctly to that, so the Captain as pilot monitoring intervened.’ It doesn’t even make any sense. What was the incorrect reaction to - the change in law? A stall warning? A stall? What was the correct reaction?
Pug above seems to generally talk a lot of sense across all these aviation threads. Aviation is a very complex and nuanced industry and Pug has said on many occasions about the Swiss Cheese Model. Aviation incidents and accidents are nearly always multi faceted with factors the lay person frequent cannot understand or find difficult to believe. Human Factors will nearly always play a role also.
The 40 second control disconnect press is irrelevant in this case as the aircraft was never going to be recovered within 40 seconds. UPRT usually does not involve using the control disconnect in my experience.
Understanding of what Alternate Law looks like on the instruments and how this manifests itself within the protections of the flight envelope is just as important as knowing what Alternate Law actually is.
Airbus have addressed the issues surrounding the rare occurrence when instruments give different readings from their respective sources. On their latest aircraft, the FADECs (Full Authority Digital Engine Control) are used also as back up to provide accurate readings.
I would caution against reading anything on aviation websites and jumping to conclusions with statements like ‘The First Officer, who was pilot flying, didn't react correctly to that, so the Captain as pilot monitoring intervened.’ It doesn’t even make any sense. What was the incorrect reaction to - the change in law? A stall warning? A stall? What was the correct reaction?
Pug above seems to generally talk a lot of sense across all these aviation threads. Aviation is a very complex and nuanced industry and Pug has said on many occasions about the Swiss Cheese Model. Aviation incidents and accidents are nearly always multi faceted with factors the lay person frequent cannot understand or find difficult to believe. Human Factors will nearly always play a role also.
Thank you for your kind words. Just a point of clarification though, I’ve revisited the AF447 event and a few points to note. It was in fact a First Officer in the left hand seat, although he was trying to push forward on the stick he only pressed and released the take-over push button, for this reason the dual input alert was triggering consistently in the flight deck. Pressing and holding the PB locks out the other stick input in one second, it’s 40 seconds only to be fully latched. By the time the captain arrived in the flight deck the aircraft was in a deep stall. Any chance of recovery would have required a decision to manually move the trim wheel forward (the stabiliser was trimmed 13 degrees nose up by this point, in alternate law the aircraft was no longer trimming for 1g and instead was trimming for the inputs of the guy in the right have seat) within around 10-15 seconds of the captain entering the flight deck. Adding startle factor and you could perhaps argue there was no chance.
Further clarification. I do not operate the Airbus on the line. UPRT is subtlety different on the type due to the differing philosophy. During my APS MCC (on the A320 series FNPT II) we were trained on generic Airbus SOPs but based heavily on the big orange lot. The standard stall recovery (and recovery from unusual attitudes) was from memory announcing ‘I HAVE CONTROL’ with simultaneous press and hold of the take-over push button - disconnecting autopilot and autothrottle. Thereafter you were expected to announce each step as in ‘PUSH’, await the cessation of the stall warning, ‘THRUST’ etc etc.. in there somewhere was making sure the speed brake was stowed and consider adding a stage of flap if below FL200.
I remember this (perhaps badly) because it was my first practical experience in how the pitch power couple can induce a secondary stall if you’re too heavy handed with the thrust increase. I actually find the Boeing aircraft inherently more intuitive in this scenario. Perhaps that’s just me.
Ultimately though I concur. It’s easy to sit on the ground, whether you’ve only ever flown as a passenger, have a PPL and enjoy flying around in a Cessna, or have 20,000 hours flying airliners, to look at things in a very linear fashion. Particularly if you are, as I am right now, sat in the garden with a cold one browsing the internet. Aviation media, and the press in general, tend to like to promote a narrative. I suppose it’s human nature to say ‘I wouldn’t do it like that’, even highly experienced investigators can easily get sucked in to that train of thought. However there is rarely one simple, single point of failure in catastrophic events. The human almost always has a part to play, and I’ve found with experience that mode confusion, confirmation bias and startle factor are often a significant contributory factor in aviation accidents and incidents.
Last edited: