you know these thinks can be and are handled much more subtle than outright conspiracies or untruths. It is often so that you have to read very carefully between the lines as well as look for the weighing of the actual circumstances in order to figure out what really happened.
In recent years, there have been several highly “political” accident reports which indeed did report the true reasons in the technical part to most extent but ended up with conclusions and recommendations not reflecting those. Concorde is one of those, but there are many more. All I am saying is that if we do want to learn from those things, we need to make our own picture rather than just overreading the conclusions.
I don’t want to derail this thread any further by again delving into the reasons why this accident happened, but Mike Bannisters book is a quite good starting point which includes a lot of inside information I also did not know and I am not sure if “Hutch” was aware of at the time he wrote his conclusions. Yet, they independently reach similar conclusions. Assigning the blame to a foreign airline and the final hole in the cheese simply comes way short. Had that flight been operated as it should have been, they would never have hit that stripe in the first place and even if, it would not have caused the consequences it did. But that is for another discussion and apart from that, has been discussed many times over.
@Mooney_Driver please enlighten me.I don’t see what political conclusions the BEA report contained. It was very factual otherwise it would have been thrown out by the initial investigating judge.
There were three elements which were in the report that others suggested didn’t carry the relevance they should have in the conclusions. But this was not political as they continued that opinion even when challenged.
I also don’t see what Mike Bannister brought to the table that makes it a good starting point. He was not part of the investigation team and would not have been under any other country’s regime including the UKs. He might well have been considered an expert witness or interviewed for television.
There were also several very good investigatory tv documentaries and not just French ones. A canadian tv company did a very good one as did the UKs Channel 4 which AFAIK do not pull back from criticising the establishment.
If you think that the BEA conclusions were not enough, take a look at the court cases. There were many, and judges questioned every conclusion that the BEA had come to against other theories put forward by people like Johnn Hutchinson and Mike Bannister. AFAIK no one was let off the hook politically. Huge compensation payments were ordered to be paid to victims families. I don’t know what the result of the manslaughter charges ended up being.
This is why I said I was sad that some on this forum are setting so much in store by a book written 20 odd years after the event bu someone on the periphery of investigations and yet pay no heed to the results of the judicial enquiries or things that were covered in television documentaries.
maxbc wrote:
I don’t think we can say they went easy or protected the state of things during AF447. The cause for this is stated as : temporary incoherence between measured speeds, AP disconnect and switch to alternate law inappropriate control inputs destabalizing the trajectory no link made between the loss of airspeed and the appropriate procedure late identification of the trajectory deviation by the pilot monitoring and insufficient correction by the pilot flying failure by the crew to identify near-stall, failure to react and exit from the flight envelope lack of stall acknowledgement and consequently lack of recovery actions So, saying they don’t come to conclusions and don’t clearly point what went wrong is a bit of a stretch. In this case the whole system was lacking (mainly in high altitude manual flying, UPRT and CRM), and was improved thereafter.
What’s missing here are two factors that IMHO were crucial:
- stall warning going silent at IAS of 60(IIRC) kts
- no haptic feedback on the sidesticks, IOW the pilots didn’t know what the other one was doing.
I have no idea if these issues were subsequently addressed, perhaps @Dan, who flew AB, can enlighten us.
In the Concorde case, they also initially tried pretty much every trick in the book to deflect the blame from AF.
@172driver who tried every trick in the book to deflect the blame from AF.?
Air France were prosecuted as were Aerospatiale and Continental.
So please give examples of where the BEA did not criticise AF.
In the case of AF447 Air France criticised itself for many factors. Whilst the whole training industry introduced greter UPRT techniques AF changed its whole policy regarding layovers/stopovers on long distance flights as they accepted criticism that fatigue could have played a major role in this accident.
I’m not sure where you got the no chaotic feedback information from. AIUI from the enquiry that the pilot non flying who was in the left seat and more senior to the pilot flying tried to take control but failed. This being due IIRC to him not operating the mechanism to allow him to take control.
As in the Concorde investigation the BEA listed all facts and conclusions. It was not their job to pass opinions on Airbus design, neither do they have that sort of capability. If something had broken or in this case a stall warner that stopped working at 60kts then that should have been included in the report. But would it really have made much difference to the cause of the accident to be included in the conclusions. In other words how relevant was it to the accident?
Let’s start with a clarification… no one flies an Airbus. It tolerates you. I therefore managed the aircraft and it’s associated flight path for slightly more than 8K hours.
172driver wrote:
no idea if these issues were subsequently addressed
No, AFAIK (been retired for 6 1/2 years) nothing was modified on the Airbus models involved in accidents. Stall warning flight phase dependant, no moving throttles, no sidestick feedback.
And this is my major gripe against the BEA and other investigating bodies. For mostly protective reasons, no one has ever questioned the philosophy of the Airbus design as such, in particular regarding:
AF447 is a very good example. Reading the report one thinks the icing pitots led to invalid speed indications followed by a inappropriate crew reaction. Yes, this is certainly correct.
But there is more to it, the philosophy of the design. Why does a loss of valid airspeed indication lead to the A/P switching itself off? Because it has been integrated into the architecture, or philosophy of the design. Had the A/P stayed on, as it was the case on older and other designs, we probably would not even mention that doomed callsign here…
As I mentioned before Dan BEA, has no expertise to pass comment or criticise aircraft design, not overtly anyway.
If an engine fails and they can call in an engineer from Rolls Royce or Pratt and Whitney to investigate. It’s a physical investigation and can be monitored.
But who has the expertise to investigate faulty aircraft design. If you call in the designer of the aircraft, they are hardly likely to admit that they have got it wrong.
The BEA investigators can mention concerns but they can only be opinions and as such cannot be put forward to a judicial enquiry as fact. The investigating judge will simply ask “what qualification do you have to make this criticism of aircraft design” meanwhile the aircraft manufacturer will be saying that the BEA don’t know what they are talking about.
Fortunately, in the case of AF447 the enquiry demanded by the families of victims experts eg Airbus pilots were able to be questioned and asked what would they do if this or that happened. The designers and manufacturer could then be asked if they agreed or disagreed. Simulators could also be used. Sadly, for full fact finding in the gap between accident and court case the Airbus pilots called as experts had learnt from the errors and had been better trained on Airbus systems and the design elements were no longer an issue. Airbus were/are being investigated on lack of historic training. It is still possible they will be liable for paying some compensation. But whether they will have to make or have made changes to their system design, I don’t think has been adjudicated yet. But I have not kept up with the the most recent developments. These things often take so long that it disappears from the media until something else drops up.
Dan wrote:
No, AFAIK (been retired for 6 1/2 years) nothing was modified on the Airbus models involved in accidents.
As far as I have been informed, what did change is training. One of the critical points in AF447 was that the FO at the controls at the time of the incident used an “unreliable airspeed” pitch value which was intended for low level flight as nobody had thought of this happening at cruise levels, where the sensible reaction would have been to keep the current pitch and power and analyse the situation, rather than to revert to some standard pitch vs power setting which was not adequate for the conditions. Also I understand that a lot of recovery technique and other human flying bits have been introduced into upset recovery training, which was a dire necessity.
On the hardware side, it is always difficult to say what has been done and what not as Airbus is not too communicative about them. I fully agree with all your points though. While the Airbusses are quite fun to fly in normal law, they can become very demanding in any abnormal situation. I think one very obvious example was the total overkill in warnings and cautions experienced by e.g. QF32 and some others in similar conditions or the incidents involving the A330 over Australia, which were quite close calls, both of them.
Dan wrote:
Why does a loss of valid airspeed indication lead to the A/P switching itself off? Because it has been integrated into the architecture, or philosophy of the design. Had the A/P stayed on, as it was the case on older and other designs, we probably would not even mention that doomed callsign here…
Exactly a perfect example of how that integration can turn into a nightmare the moment you go into non-standard operation on the “bus”.
Nevertheless, it is pretty obvious that also Boeing and others have their weak points. I was quite aghast to learn that the 737-800 does not have a proper EICAM to this day which makes analysing malfunctions “interesting” and a throw back to the 1960ties, as opposed to all other Boeings who do, 757/767/747-400up, 777 and 787.
Going by memory here, the accident of Concorde at Paris had several major factors.
Most of those occurred before the take off run. Some of them were clear no-go’s.
Most or all of those facts are found in one way or the other in the report, but figures differ and so do opinions on the effects. Myself, I tend to believe Mike Bannister’s and Greg Faith, both of whom were involved in the British side of the investigation.
Since AF447, simulators have been refined to model high altitude stall characteristics, and in the case of Airbus types, recovery from high altitude stalls in alternate law is now a scenario used in training .The crew of AF447 had not received training in this scenario, or stall recovery in alternate law.
The longstanding Part 121 philosophy, which has now been adopted universally, is for the crew to always disconnect automatics (A/P and A/T), and to ignore FD commands in the event of an upset event. Arguably Part 121 is also clearer on who takes command, and how passage of command is undertaken (three way call out).
The French authorities went to great pains to recover the data recorders, and also to work with the major manufacturers to improve training after the event.
