AF 447 Thread No. 7
Join Date: Jul 2002
Location: UK
Posts: 3,182
Likes: 0
Received 0 Likes
on
0 Posts
the "anti-Airbus brigade" obsession
and your self assigned mission of defending Airbus at any cost, even those elements that are scientifically indefensible
I'm just here to make sure that press-induced falsehoods about the design, including but not limited to it being "designed to take pilots out of the loop", "the first step in replacing pilots", "designed by managers and computer nerds without pilot input" and "reliant on the French Government covering up the real reasons for accidents" meet with robust and evidence-supported rebuttal. Any or all of which are not only substantively untrue, but also defamatory against the hard work of all the teams that have worked on it over the years, some of whom are no longer with us.
for anyone that is a professional, and wrongly extrapolating faults with certain few Airbus elements, as being applied to the entire Airbus design, is a good indication of the type of alignment you're driven by.
It's not about the manufacturer and it's not about defending modern automated aircraft in the slightest, it's about demanding a degree of factual basis behind the discussion rather than a bunch of tired old cliches that have been doing the rounds since 1988.

Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,688
Likes: 0
Received 0 Likes
on
0 Posts
aside from the pitot tube issues (which are serious, and Airbus/AF have some 'splaining to do there...)
Given -
- the history of the pitot issues in Interim #1,
- the recommendations in the area of certification in Interim #2,
- the existence of UAS procedures in AF and AB manuals,
- that DFDR and CVR have hardly added anything to this issue,

Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,810
Likes: 0
Received 0 Likes
on
0 Posts
Clandestino
Well .. it's not always wastage of valuable information
We must take in account the airbus FBW laws and autopilot
We have indeed a multicrew .. in fact (normal crew manning) 3 members .. 2 humans and the additional crew member .. the AP computer-FBW system entity
In normal law ... it's no wastage (human crew members hands are out of SS)
In other laws (or when a human crew member put hand on SS) it can be a wastage of valuable information
Can anyone provide any verifiable source for this statement? If we were to believe it then right now thousands of FBW Airbi currently airborne are operating with considerable wastage of valuable information in a multi crew operation.
We must take in account the airbus FBW laws and autopilot
We have indeed a multicrew .. in fact (normal crew manning) 3 members .. 2 humans and the additional crew member .. the AP computer-FBW system entity
In normal law ... it's no wastage (human crew members hands are out of SS)
In other laws (or when a human crew member put hand on SS) it can be a wastage of valuable information

Join Date: Jul 2002
Location: UK
Posts: 3,182
Likes: 0
Received 0 Likes
on
0 Posts
In short, I'm not trying to "defend" or "protect" Airbus indiscriminately, I am and always have been simply trying to keep the discussion to the matters at hand and not get sidetracked by the same old kvetching.

Join Date: Jul 2009
Location: The land of the Rising Sun
Posts: 171
Likes: 0
Received 0 Likes
on
0 Posts
The Red Herrings become a shoal. Direct v Indirect input now. Instruments, AirTren, instruments are the key. They are not fooled by panic, tension, stress or any of the other factors which so much mess up human perception. They can be relied on to give an accurate reading of a situation provided you remember your training and cockpit drills. No need to look at the yoke or whatever, look at your instruments. One can be sure that BEA know who was making the inputs - a yoke is not needed for that. As for the flight crew they also knew so where does the yoke help? A yoke is a piece of equipment to transmit inputs to the aircraft it is not a magical cure all to save lives. It performs its function like sidesticks do and some pilots prefer it. This is human nature not everyone likes the same thing. But it is not superior in any way to a sidestick.
Conf iture - what on earth are you talking about?
Conf iture - what on earth are you talking about?
Last edited by Old Carthusian; 10th Nov 2011 at 23:11.

Join Date: Jan 2005
Location: W of 30W
Posts: 1,939
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by DozyWannabe
No sir, it has already been addressed - by people far more worthy of addressing it than me - in several places on the previous threads, hence the reference to the hamster wheel.
No reply in sight ... but put your own thoughts using your own words.
Habsheim ?
AF447 has its own thread.
Habsheim should also ...
Originally Posted by Clandestino
Can anyone provide any verifiable source for this statement? If we were to believe it then right now thousands of FBW Airbi currently airborne are operating with considerable wastage of valuable information in a multi crew operation.
Usually more relevant in take off and landing phases but AF447 demonstrates how it can also be true at FL350.
Originally Posted by OK465
If the SS input did not change the existing dynamic, why would there be a requirement for the autotrim to change even in ALT2?
Both posters may well want to monitor the SD FLT/CNTL page the next time they get a chance to do this.

Join Date: Jan 2005
Location: W of 30W
Posts: 1,939
Likes: 0
Received 0 Likes
on
0 Posts
One can be sure that BEA know who was making the inputs - a yoke is not needed for that. As for the flight crew they also knew so where does the yoke help?

Join Date: Jul 2009
Location: The land of the Rising Sun
Posts: 171
Likes: 0
Received 0 Likes
on
0 Posts
Yes by looking at the instruments. An input produces a certain pattern on your instruments and you can understand what your aircraft is doing. The seated PNF could see the zoom climb from his instruments. He even mentioned this. The captain also could see the instruments when he entered the cockpit.

Join Date: Jun 2009
Location: Germany
Age: 70
Posts: 782
Likes: 0
Received 0 Likes
on
0 Posts
An input produces a certain pattern on your instruments
Confiture
2 PNF knew the PF was making inputs - Did they know which inputs he was making ?
2 PNF knew the PF was making inputs - Did they know which inputs he was making ?
Answer Old Carthusian
Yes by looking at the instruments. An input produces a certain pattern on your instruments and you can understand what your aircraft is doing.
Yes by looking at the instruments. An input produces a certain pattern on your instruments and you can understand what your aircraft is doing.
Stick was full left, aircraft was banked to the right, how do you know by instruments, what kind of inputs are made by the guy in the dark 1 meter away?

Control stick inputs, instruments, laws
Once again, I have to go with Retired.
If I am not the pilot in control, then simply looking at the instruments does not provide the best feedback required to maintain control or get the beast going where you wish. It is true that having a mechanically coupled wheel/stick would help the "other guy" understand what was happening, but the jet involved did not have that feature. Retired pointed out that the "other guy" did not know exactly what the control inputs were and finally advised to use some rudder to pick up the low wing.
I realize that I do not have 20,000 hours as some here. But I prolly had as many takeoffs and landings with my measly 4,000 hours, none of which had a co-pilot to help/monitor. I also had to solo pilots from a "chase" aircraft on their first hop, as we had no family models. So not feeling their control inputs was very familiar to me. 'nuff of that.
Rely on the instruments? Hell, the tragedy started when the instruments became unreliable!! BEAM ME UP! So would you trust the speed and air data after the A/P disconnected and then the PF states we're in alternate law? Then we have stall warnings and other chimes.
Those that have not flown FBW still need to realize that even in the Airbus "direct" law, the computers still modify the control surface deflections and deflection rates. To wit from FCOM manual I have:
- "In all configurations elevator deflection varies as a function of aircraft C.G."
- Roll rate commands are limited in ALT 2 depending upon gains and configuration". 20 - 25 deg/sec compared to 15 deg/sec in "normal". So roll is not a direct movement of the ailerons/spoilers WRT to stick inputs.
Finally, this old curmudgeon is not anti-Airbus. I crowed about the A320 with partners of mine on our first flight in one of them. Asked them to look at the ailerons and spoilers moving to keep bank angle constant even in mild turbulence. Small movements not possible with the older systems. I gained a lot of respect for the GD folks that did the Viper system after a few minutes flying the beast. I gained more after my leading edge flap failure ( see my bio on my profile).
I simply feel that there are too many protections that are different between all the reversion laws. For example, why the change in roll rates? Why keep changing things when air data is deemed unreliable by HAL ( had to get that in for Doze, heh heh) versus switching to a simple standby gains vale or all the gains and such? And the beat goes on...
And then there's the issue, a big one, of crew training for worst case scenarios, and that may be a major recommendation of the accident board, IMHO.
Gotta go, and I only have a basic B.S. in the academic arena, but prolly a PhD in the real world school.
If I am not the pilot in control, then simply looking at the instruments does not provide the best feedback required to maintain control or get the beast going where you wish. It is true that having a mechanically coupled wheel/stick would help the "other guy" understand what was happening, but the jet involved did not have that feature. Retired pointed out that the "other guy" did not know exactly what the control inputs were and finally advised to use some rudder to pick up the low wing.
I realize that I do not have 20,000 hours as some here. But I prolly had as many takeoffs and landings with my measly 4,000 hours, none of which had a co-pilot to help/monitor. I also had to solo pilots from a "chase" aircraft on their first hop, as we had no family models. So not feeling their control inputs was very familiar to me. 'nuff of that.
Rely on the instruments? Hell, the tragedy started when the instruments became unreliable!! BEAM ME UP! So would you trust the speed and air data after the A/P disconnected and then the PF states we're in alternate law? Then we have stall warnings and other chimes.
Those that have not flown FBW still need to realize that even in the Airbus "direct" law, the computers still modify the control surface deflections and deflection rates. To wit from FCOM manual I have:
- "In all configurations elevator deflection varies as a function of aircraft C.G."
- Roll rate commands are limited in ALT 2 depending upon gains and configuration". 20 - 25 deg/sec compared to 15 deg/sec in "normal". So roll is not a direct movement of the ailerons/spoilers WRT to stick inputs.
Finally, this old curmudgeon is not anti-Airbus. I crowed about the A320 with partners of mine on our first flight in one of them. Asked them to look at the ailerons and spoilers moving to keep bank angle constant even in mild turbulence. Small movements not possible with the older systems. I gained a lot of respect for the GD folks that did the Viper system after a few minutes flying the beast. I gained more after my leading edge flap failure ( see my bio on my profile).
I simply feel that there are too many protections that are different between all the reversion laws. For example, why the change in roll rates? Why keep changing things when air data is deemed unreliable by HAL ( had to get that in for Doze, heh heh) versus switching to a simple standby gains vale or all the gains and such? And the beat goes on...
And then there's the issue, a big one, of crew training for worst case scenarios, and that may be a major recommendation of the accident board, IMHO.
Gotta go, and I only have a basic B.S. in the academic arena, but prolly a PhD in the real world school.

Join Date: Jul 2002
Location: UK
Posts: 3,182
Likes: 0
Received 0 Likes
on
0 Posts
@Franzl, CONF:
The precise position of the stick is considerably less important than being able to work out that the stick is not where it should be. Once that is ascertained, then the only logical recourse is "I have control".
This isn't a trainer, where the junior pilot follows through his senior's movements - this is line flying, and if your colleague is not successfully controlling the aircraft into a recovery, then they have no business continuing to do so.
Also, CONF - if you use the Search function - my username plus the keyword "yoke", you have instant access to everything I've ever said on the matter, including conversations with you. I'm not going to be your monkey by rehashing it again.
The precise position of the stick is considerably less important than being able to work out that the stick is not where it should be. Once that is ascertained, then the only logical recourse is "I have control".
This isn't a trainer, where the junior pilot follows through his senior's movements - this is line flying, and if your colleague is not successfully controlling the aircraft into a recovery, then they have no business continuing to do so.
Also, CONF - if you use the Search function - my username plus the keyword "yoke", you have instant access to everything I've ever said on the matter, including conversations with you. I'm not going to be your monkey by rehashing it again.

Join Date: Jul 2009
Location: Not far from a big Lake
Age: 81
Posts: 1,461
Likes: 0
Received 0 Likes
on
0 Posts
O.C.
Only partially correct I think. Up until the stall, the instruments did tell the story of what the inputs were.
After the stall, the aircraft began to act perversely in roll and pitch, and at that point, the control inputs were masked by the aircraft's gyrations.
This must have been immensely confusing to both the PF and PNF.
The Red Herrings become a shoal. Direct v Indirect input now. Instruments, AirTren, instruments are the key. They are not fooled by panic, tension, stress or any of the other factors which so much mess up human perception. They can be relied on to give an accurate reading of a situation provided you remember your training and cockpit drills. No need to look at the yoke or whatever, look at your instruments. One can be sure that BEA know who was making the inputs - a yoke is not needed for that. As for the flight crew they also knew so where does the yoke help? A yoke is a piece of equipment to transmit inputs to the aircraft it is not a magical cure all to save lives. It performs its function like sidesticks do and some pilots prefer it. This is human nature not everyone likes the same thing. But it is not superior in any way to a sidestick.
After the stall, the aircraft began to act perversely in roll and pitch, and at that point, the control inputs were masked by the aircraft's gyrations.
This must have been immensely confusing to both the PF and PNF.


Join Date: Jan 2005
Location: W of 30W
Posts: 1,939
Likes: 0
Received 0 Likes
on
0 Posts
Also, CONF - if you use the Search function - my username plus the keyword "yoke", you have instant access to everything I've ever said on the matter, including conversations with you. I'm not going to be your monkey by rehashing it again.

Join Date: Jul 2009
Location: The land of the Rising Sun
Posts: 171
Likes: 0
Received 0 Likes
on
0 Posts
Retired F4
Both the PF and PNF have displays with the same instrumentation on them. Your artificial horizon will tell you if the plane is climbing or banking. You derive your rate of climb (or descent) or descent from your altimeter. Your bank rate from your rate of turn indicator. These are reliable and accurate instruments. They are not affected by UAS. Anyone who doesn't rely on their instruments ends up in a dangerous situation.
Both the PF and PNF have displays with the same instrumentation on them. Your artificial horizon will tell you if the plane is climbing or banking. You derive your rate of climb (or descent) or descent from your altimeter. Your bank rate from your rate of turn indicator. These are reliable and accurate instruments. They are not affected by UAS. Anyone who doesn't rely on their instruments ends up in a dangerous situation.

Join Date: Jul 2011
Location: Northern Hemisphere
Posts: 195
Likes: 0
Received 0 Likes
on
0 Posts
I'm just here to make sure that press-induced falsehoods about the design, including but not limited to it being "designed to take pilots out of the loop", "the first step in replacing pilots", "designed by managers and computer nerds without pilot input" and "reliant on the French Government covering up the real reasons for accidents" meet with robust and evidence-supported rebuttal. Any or all of which are not only substantively untrue, but also defamatory against the hard work of all the teams that have worked on it over the years, some of whom are no longer with us.
In this case, aside from the pitot tube issues (which are serious, and Airbus/AF have some 'splaining to do there...), it's looking increasingly likely that there were no hardware (or indeed software) faults involved in the crash of AF447.
It's not about the manufacturer ... it's about demanding a degree of factual basis behind the discussion
It does not really matter if it's caused by a lack of comprehension, a lack of technical expertise, or ill intentioned tactics. The predominant appearance of spreading a smoke screen over a few problems is counter productive not only for those very few problems, but the entire Airbus technology, as it reflects badly on the many things that are good.

Join Date: Jul 2011
Location: Northern Hemisphere
Posts: 195
Likes: 0
Received 0 Likes
on
0 Posts
No.
It's not Direct or Indirect Input.
It is perhaps too abstract, and a terminology that is not part of your profession.
It is Direct versus Indirect Information Transfer.
If B is interested in information about A, it can have a Direct information exchange from A to B, which is not altered by any intermediate agent.
An Indirect information exchange involves one or more intermediate agents - let's say C, D, E, F, G. Each intermediate agents transform, translate, or process the information as part of the transfer. C, D, E, F, and G transforming, translating, or processing, is introducing distortions, and errors. Consequently, when the information arrives to B from A, via C, D, E, F, and G, it can be distorted or altered significantly.
No.
The KEY is the visual contact, the visual perception.
In case of a blinding attack, the instruments are not good at anything.
The panic, tension, stress, are affecting the visual perception.
Instruments can be inaccurate, can be fooled. They show data collected by sensors, processed by sensor data processing units, and as such they can be fooled by malfunction of sensors, by margin of errors in the processing algorithms, etc... This was the case with the AF447 out of specs AOA.
See also Machinbird's post.
You're missing the point, you're missing the difference between Direct and Indirect Information Transfer.
The direct visual contact of the PNF with the PF's SS, yoke, or whatever other control mechanism is in the cockpit, is a Direct Information Transfer - from A (SS of PF) to B (PNF).
The reading of instruments, is Indirect Information Transfer - from A to B, via C, D, E, F, G etc.... , because, the instruments don't show directly the PF actions on the SS, but rather the effects of the PF's SS actions onto the deflective surfaces, along with effects onto the A/C attitude, A/C speed, etc,... (which combine other controls, like throttle, trim, etc...) Information about those effects is collected by the A/C sensors, each sensor with its margin of errors, and passed to the sensor data processing units, and so on, (with each sensor data processing algorithm having its own margin of errors).....
Finally, the PNF reading the instruments need to make a mental translation of the instrument readings into PF SS actions. For accuracy,
the PNF need to exclude the effects of the PF actions on Throttle, or trim, or others... That is in fact impossible....
As you can see - if you're willing to - the Indirect Information Transfer is not only so convoluted, that by now, we've probably forgot that the PNF was just interested in the PF's actions onto the SS,.... but it is introducing the effects of other controls than the SS, which is impossible to determine only from a reading of the instruments......
It's not Direct or Indirect Input.
It is perhaps too abstract, and a terminology that is not part of your profession.
It is Direct versus Indirect Information Transfer.
If B is interested in information about A, it can have a Direct information exchange from A to B, which is not altered by any intermediate agent.
An Indirect information exchange involves one or more intermediate agents - let's say C, D, E, F, G. Each intermediate agents transform, translate, or process the information as part of the transfer. C, D, E, F, and G transforming, translating, or processing, is introducing distortions, and errors. Consequently, when the information arrives to B from A, via C, D, E, F, and G, it can be distorted or altered significantly.
The KEY is the visual contact, the visual perception.
In case of a blinding attack, the instruments are not good at anything.
The panic, tension, stress, are affecting the visual perception.
Instruments can be inaccurate, can be fooled. They show data collected by sensors, processed by sensor data processing units, and as such they can be fooled by malfunction of sensors, by margin of errors in the processing algorithms, etc... This was the case with the AF447 out of specs AOA.
See also Machinbird's post.
They can be relied on to give an accurate reading of a situation provided you remember your training and cockpit drills. No need to look at the yoke or whatever, look at your instruments.
The direct visual contact of the PNF with the PF's SS, yoke, or whatever other control mechanism is in the cockpit, is a Direct Information Transfer - from A (SS of PF) to B (PNF).
The reading of instruments, is Indirect Information Transfer - from A to B, via C, D, E, F, G etc.... , because, the instruments don't show directly the PF actions on the SS, but rather the effects of the PF's SS actions onto the deflective surfaces, along with effects onto the A/C attitude, A/C speed, etc,... (which combine other controls, like throttle, trim, etc...) Information about those effects is collected by the A/C sensors, each sensor with its margin of errors, and passed to the sensor data processing units, and so on, (with each sensor data processing algorithm having its own margin of errors).....
Finally, the PNF reading the instruments need to make a mental translation of the instrument readings into PF SS actions. For accuracy,
the PNF need to exclude the effects of the PF actions on Throttle, or trim, or others... That is in fact impossible....
As you can see - if you're willing to - the Indirect Information Transfer is not only so convoluted, that by now, we've probably forgot that the PNF was just interested in the PF's actions onto the SS,.... but it is introducing the effects of other controls than the SS, which is impossible to determine only from a reading of the instruments......

Last edited by airtren; 11th Nov 2011 at 03:44.

CONF:
With the SS full forward, I would expect the elevators to be full leading edge up for a simulator full nose down command, and be displayed as such on the SD.
One then would in general expect the nose of the simulator to follow this command. Why it didn't in the 330 and evidently did in the 320 is what's at issue.
As far as SS commanded proportional 'G' and the THS, what was the 'G' indication in the Flight Data block during the full nose down SS command prior to manual re-trim?
With the SS full forward, I would expect the elevators to be full leading edge up for a simulator full nose down command, and be displayed as such on the SD.
One then would in general expect the nose of the simulator to follow this command. Why it didn't in the 330 and evidently did in the 320 is what's at issue.
As far as SS commanded proportional 'G' and the THS, what was the 'G' indication in the Flight Data block during the full nose down SS command prior to manual re-trim?


Join Date: Jul 2009
Location: The land of the Rising Sun
Posts: 171
Likes: 0
Received 0 Likes
on
0 Posts
I am afraid you misunderstand the situation - it is not about information transfer it is about information interpretation. How you judge the information you receive is the important factor. A yoke does not add any advantage because the interpreter of the position may well differ in where he sees the position of the yoke and the threat or otherwise it represents. If you think of visual perception as reliable I would point out that an instrument is a much more likely to be read reliably than the position of a yoke.
Now I assume you are not a pilot otherwise I would not need to explain this. At night without any visual references it is easy to become disoriented. This is where the aircraft instruments come into play. Running off separate reliable mechanical or electronic inputs they give an accurate and measured stream of information which a trained pilot can use to fly an aircraft safely. If one instrument does prove to be unreliable then there is enough redundancy in the others to enable the pilot to fly the aircraft smoothly. It is important to remember that the instruments are not an indirect transfer but a reliable indicator of the state of the aircraft at the time. Over many years instruments have been refined so that they are more accurate than human perception.
Of course to use your instruments requires training and cockpit discipline. The mode of input for control directives is actually irrelevant and comes down to preference. Your direct v indirect information transfer is one of the shoal of red herrings. It falls at the hurdle of interpretation. There is clear evidence that the PNF was aware what was happening with the aircraft. How was this - he couldn't see the sidestick but he could see the instruments and they gave him the information he needed. His actions and later perceptions were not sufficient but this does not alter the fact that he knew there was something wrong very rapidly. The PF does not seem to have looked at his instruments at all and this is a clear training failure for which the airline must also be held responsible.
You also owe Dozy an apology - of the non-pilots he is perhaps the one who has understood the dynamics of the accident clearly. It was human generated in a situation which was clearly survivable if the crew had been adequate.
Now I assume you are not a pilot otherwise I would not need to explain this. At night without any visual references it is easy to become disoriented. This is where the aircraft instruments come into play. Running off separate reliable mechanical or electronic inputs they give an accurate and measured stream of information which a trained pilot can use to fly an aircraft safely. If one instrument does prove to be unreliable then there is enough redundancy in the others to enable the pilot to fly the aircraft smoothly. It is important to remember that the instruments are not an indirect transfer but a reliable indicator of the state of the aircraft at the time. Over many years instruments have been refined so that they are more accurate than human perception.
Of course to use your instruments requires training and cockpit discipline. The mode of input for control directives is actually irrelevant and comes down to preference. Your direct v indirect information transfer is one of the shoal of red herrings. It falls at the hurdle of interpretation. There is clear evidence that the PNF was aware what was happening with the aircraft. How was this - he couldn't see the sidestick but he could see the instruments and they gave him the information he needed. His actions and later perceptions were not sufficient but this does not alter the fact that he knew there was something wrong very rapidly. The PF does not seem to have looked at his instruments at all and this is a clear training failure for which the airline must also be held responsible.
You also owe Dozy an apology - of the non-pilots he is perhaps the one who has understood the dynamics of the accident clearly. It was human generated in a situation which was clearly survivable if the crew had been adequate.
