Peter wrote:
However W-N, Cobalt and myself are all electronics designers so it’s easy to say these things…
Peter I don’t really understand your point. Systems knowledge doesn’t help you too much here, I think it would be very obvious what the nature of the failure was when the two AI’s contradicted, even with little systems knowledge. The challenge would have been, which one, if either, can I trust?
The decreasing altitude and increasing airspeed doesn’t help you at all, because unless you can get the wings level, you can’t correct the situation anyway.
I’m actually quite shocked that aircraft at this level of certification are not required to have triple redundant IRU/AHRS. Two systems is effectively no redundancy at all since you have no way to know which is at fault (unlike, say, two engines of which one has obviously failed).
To expect the pilot to manually introduce the third standby instrument to do his own fault isolation, and do so in a few seconds under the real-world conditions that what_next describe above, is totally unrealistic.
I think my systems/electronics/avionics knowledge is comprehensive but I would not be at all confident of reacting to and dealing with this issue, if it caught me off guard out of nowhere and the aircraft was already in an unusual attitude, and if I had only a few seconds to recover before it was too late.
Peter wrote:
Perhaps the airframe broke up due to aerodynamic forces, which would happen within maybe 10-20 seconds after pitching down at 60+ degrees.
It did not. Everything was in one piece until impact. Also, there were no indications that the pitch-down was ever near 60° degrees.
Yes, and I also think many pilots don’t understand aircraft systems properly. In light GA if you don’t know the prop is screwed on the end of the crankshaft it doesn’t really matter (I have met many pilots who thought that a variable pitch prop is actually a variable gearbox between the engine and the prop!). If you don’t have an autopilot then you don’t need to know how one works (and many pilots who have an autopilot don’t know how it works and either avoid using it completely or just press AP+ALT). But if you are flying something big with nice complex systems then you do need to know a whole load more. But you are still the same human, who at one point started with a PPL, then sat a load of bollox JAA ATPL exams (which cover 1970s FOG gyros which never really existed as described in the exam paper, etc) and then you had a TR course and then an LPC every 6 months. So you get stuff like AF447 where they had no idea that the attitude data is inertially derived so the +15deg indication was probably accurate, while the airspeed indication is obviously a lot more vulnerable. I reckon the AF447 pilots had no idea where the data came from… it was all just a big video game. And maybe in this case it was similar. But then I am “lucky” having had several autopilot failures, some quite subtle, including a pitch servo failure in the final stages of an IAP which resulted in a -1000fpm VS within seconds… These people probably didn’t see their “video game” ever go wrong before. However W-N, Cobalt and myself are all electronics designers so it’s easy to say these things…
To answer Peter’s earlier question – why did they not sppt the discrepancy between attitude displayed and airspeed / altitude trend?
After all, that is all the integrity checking I usually have…
I have just one simple explanation – which has nothing to do with glass vs classic.
If this sort of thing happened to them when hand flying, I am sure the flying pilot would have noticed – starting to push, seeing airspeed increase and altitude starting to drop while the AI goes in the opposite way than expected… that is clearly nonsense.
But if with the autopilot flying, the pilot is out of the direct control loop (so wouldn’t notice the obvious discrepancy between control applied and attitude displayed) and, if not currently scanning, confronted with several instruments telling different things without knowing which one started the whooe thing and what the cause was.
This is the one downside of automation – the situations when it goes wrong are getting rarer and rarer, but the cases when the stuff gets handed back to the pilot to sort out the mess are getting harder and harder, and with the pilot less involved and less current in actually flyng it becames even harder…
Here is the link for the report:
http://www.havkom.se/en/investigations/civil-luftfart/olycka-i-lappland-med-flygplanet-se-dux-av-typen-canadair-crj-200
lenthamen wrote:
I’m surprised that there is no integrity check on the AI of a CRJ-200…
There was an integrity check. It was removed from the screens for “declutter” reasons when the indicated pitch attitudes exceeded predefined values. IIRC it is one of the conclusions of the accident report that these integrity flags should always be displayed.
RobertL18C wrote:
wonder if the mandatory upset recovery courses will address this type of situation,
We have had these since AF447. But never with contradicting indications between pilot flying and pilot monitoring. I guess they will introduce those as consequence of this accident.
Peter wrote:
Perhaps the airframe broke up due to aerodynamic forces, which would happen within maybe 10-20 seconds after pitching down at 60+ degrees.
According to the report the aircraft hit the ground in one piece.
NB: After AF447 the mantra went: Watch attitude and power, disregard everything else. This pilot did exactly that. The mantra needs to be changed I guess.
I wonder if the mandatory upset recovery courses will address this type of situation, in theory they are supposed to do so as they are aimed at multi crew.
EASA CPL also will require upset recovery training in 2018 – after fifty years it will be interesting to see instructors whose focus has been on teaching stabilised approaches etc having to undertake spin training in earnest.
I’m surprised that there is no integrity check on the AI of a CRJ-200…
How come the pilots did not spot the increasing airspeed, decreasing altitude, etc?
Maybe glass cockpits make that less obvious, assuming the main PFD attitude has failed.
Perhaps the airframe broke up due to aerodynamic forces, which would happen within maybe 10-20 seconds after pitching down at 60+ degrees.
LeSving wrote:
The pilot’s job is to fly the plane, not to be a observer,…
In a modern highly automated airliner the pilot’s job really is that of an observer, like it or not. SOPs of a typical airline call for “positive rate – gear up” and “autopilot – engage” before passing 1000ft AGL. Upon landing, the last action before touchdown is “autopilot – disengage” at minimum or even below, if not autoland. Even in our business operation, on every single flight I have to call out “NON STANDARD: manual flight to FL 100” (above that we are not supposed to fly manually at all) if I want to do any flying myself. And this will be getting even worse in the future. Maybe rightly, because a computer would have compared those three AIs and reacted to the indication of the two correct ones, not the single at fault.
