Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

AF 447 Search to resume (part2)

Tech Log The very best in practical technical discussion on the web

AF 447 Search to resume (part2)

Old 23rd May 2011, 19:42
  #2201 (permalink)  
 
Join Date: Feb 2008
Location: In the Old Folks' Home
Posts: 420
Received 2 Likes on 1 Post
Sidestick Position

OK465:

Quote:
I have always wondered about the position of the sidestick.
Is that any different than flying a 727 with the left hand on the control wheel and right hand on the throttles?
I have about equal experience in planes with yokes and sticks. I never seemed to have a problem in switching from the left seat to the right seat where yokes were involved. I can specifically remember flying an F-6A (F-4D) with a stick in the morning and an HU-16 (Grumman amphibian with a yoke and throttles in the middle) in the afternoon. However, the idea of a stick in my left hand seems awkward. Of course, you could get used to it. It's all in the training and experience.
Smilin_Ed is offline  
Old 23rd May 2011, 20:10
  #2202 (permalink)  
 
Join Date: Dec 2010
Location: Middle America
Age: 84
Posts: 1,167
Likes: 0
Received 0 Likes on 0 Posts
Time To Impact

On the other AF 447 thread in R&N, there is an interesting Post # 377 by MFgeo. In this post there is a link to an author (Matthew Squair) where he has calculated the time to impact using information provided by the BEA in their Interim Report #1. With additional information provided by the BEA in Interim Report #2 he revised his figures accordingly.

Based on a terminal airspeed derived from the cabin pressure advisory message, the time to impact (tti) can be calculated as being 57 seconds after the cabin vertical speed advisory was generated (2:14:21).
This would result in a time of impact (toi) of 2:15:23.
But, these tti and toi are based upon the assumption of an average terminal vertical velocity in the last 57 seconds of flight that is the same as the average calculated for the preceding flight phase.
As a still flying aircraft would have transmitted the class 2 ADR 2 fault message before impact these two times can be used to revise the time to impact (tti) and correct the terminal velocity (vterm).
He previously calculated the altitude at which the safety valve opened (6,898.5 ft) and he believed that the valve opened at 2:14:26, 5 seconds before the advisory would have triggered and he calculated the terminal velocity associated with both ends of the transmission window as follows.
In the case of (time to impact) timp = 2:15:00:
tti (2:15:00) = 2:14:26 – 2:15:00 = 0:00:34 seconds.
Giving an average terminal velocity of:
vterm (2:15:00) = 6,898.5 ft / 34 sec = 12,173.9 ft / min.
In the case of timp = 2:15:14:
tti (2:15:14) = 2:14:26 – 2:15:14 = 0:00:48 seconds.
Giving an average terminal velocity of:
vterm (2:15:00) = 6,898.5 ft / 48 sec = 8,623.1 ft / min.
The corrected results. Given the transmission window for the ADR 2 fault message the average terminal velocity of AF 447 is in the range 8,623.1 to 12,173.9 ft / min.
His Conclusions. The increase in average vertical velocity from 6,718 ft/min to at a minimum 8,623.1 f/min indicates a sustained vertical acceleration of the aircraft.
This terminal vertical acceleration and high terminal vertical speed when combined with the positive attitude and low bank of the aircraft is strongly indicative of the aircraft having departed into a deep stall and possibly a subsequent flat spin.


If his theory is correct, it might explain the relative closeness of the found debris to the LNP.

Last edited by Jetdriver; 23rd May 2011 at 21:20.
Turbine D is offline  
Old 23rd May 2011, 20:39
  #2203 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
His Conclusions. The increase in average vertical velocity from 6,718 ft/min to at a minimum 8,623.1 f/min indicates a sustained vertical acceleration of the aircraft.
He assumes that the airplane was at FL350 at 2:10:10, which is possible but not an established fact. He then goes on to assume that the rate of descent between 2:10:10 and 2:14:21 was constant, which is highly unlikely. (I recall that an earlier calculation was in error but didn't check whether that has been corrected. The matter was discussed quite some time ago on part 1 of this thread.).

Last edited by HazelNuts39; 23rd May 2011 at 20:56.
HazelNuts39 is offline  
Old 23rd May 2011, 21:02
  #2204 (permalink)  
 
Join Date: Jul 2009
Location: Planet Earth
Posts: 91
Likes: 0
Received 0 Likes on 0 Posts
3. So far, so good. But (and there is always a but), this will inevitably lead to an increasing and insidious disconnect between the pilot and the aircraft. Manual flying skills have likely degraded somewhat. Pilots now may have less situational awareness than in years gone by. There is perhaps a tendency for them to not fully understand all the systems - they know how to operate them under normal conditions sure, but is it possible for them to really understand what is happening when a serious problem suddenly appears out of routine. I accept that all this is a generalization and a gross over-simplification. But even so, is there some truth in this?
PM is a euphemism. And whats the PF doing anyway?

This raises interesting questions about experience. What does it mean to have 10,000 hrs of experience if 7,500 or more of that time was spent staring at the computer flying the a/c. The reality of the modern flight deck may be closer to Pilot Staring and Pilot Doing Nothing. Talk about ennui.
CogSim is offline  
Old 23rd May 2011, 21:26
  #2205 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
Deep stall and flat spin to the end, model

This terminal vertical acceleration and high terminal vertical speed when combined with the positive attitude and low bank of the aircraft is strongly indicative of the aircraft having departed into a deep stall and possibly a subsequent flat spin.
Just to remember: The conditions (mech. damages) of the debris recovered floating seems consistent with a flat spin "heading to RH" (CW as viewing from top)

Exception: Galley "good conditions".


Edited: Please don´t consider what is in blue:


Question 1): Supposing a flat spin til the end. Typically, how many degrees per second? If tail cone separated in flight, was at low height as showed by "concentration" in seabed.
Question 2): It´s possible (spinning to RH) hit the water slightly banking left? And if the VS detached before impact? LH would drop further?

Last edited by Jetdriver; 24th May 2011 at 06:09.
RR_NDB is offline  
Old 23rd May 2011, 21:35
  #2206 (permalink)  
 
Join Date: Aug 2009
Location: Germany
Age: 67
Posts: 1,777
Likes: 0
Received 0 Likes on 0 Posts
Cool

Hi,

This raises interesting questions about experience. What does it mean to have 10,000 hrs of experience if 7,500 or more of that time was spent staring at the computer flying the a/c. The reality of the modern flight deck may be closer to Pilot Staring and Pilot Doing Nothing. Talk about ennui
Always difficult to describe(evaluate) an experience comparing to a number of hours or age ...
A pilot can fly 10,000 hours very well (in full manual) and have little experience other than uneventful flight ... if no exeptional events (non-routine) occurs
In my job .. in one month (due exeptional events) I acquired more experience than I had been able to accumulate in 12 years ....
jcjeant is offline  
Old 23rd May 2011, 21:39
  #2207 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,131
Received 320 Likes on 204 Posts
Just to remember: The conditions (mech. damages) of the debris recovered floating seems consistent with a flat spin "heading to RH" (CCW as viewing from top)
I think you said that backwards, or I understood it backwards.

If the aircraft is spinning CCW (and the aircraft is not inverted), its heading keeps changing to the left (being south of the equator would not change that ). Put simply, if we mark the nose of the aircraft each time it passes through 360/North, it's heading will decrease during the next rotation until it arrives again at 360/North, and repeats the decrease on the next rotation.

If you spin/rotate CW, as viewed from above, (I have in my minds eye a fast moving, old fashioned compass card), as we pass through 360/North on each rotation, your heading will increase during the rotation. That's more of a right hand turn.

Exception: Galley "good conditions".
Wasn't that explained earlier as the airframe absorbing most/more of the impact? (Think "crumple zone" in a car collision).
Question 1): Supposing a flat spin in the end. Typically, how many degrees per second? If tail cone separated in flight, was at low height as showed by "concentration" in seabed.
It might be that Airbus never collected those data points in testing.
Question 2): It´s possible spinning to RH hit the water banking left? And if the VS detaches before impact? LH wing can fall?
"Flat spin" will not by itself rule out a bit of roll along the longitudinal axis during the spin. The amount depends on aircraft type. The type I am familiar with had some (small) nose roll/oscillations in an erect spin. (Grateful I've never been in a flat spin, I understand it's a horror show.)

It also depends on the coupling between vertical stab and wings, which again would depend on a particular aircraft's unique characteristics.

I'll leave to those who are familiar with big jet/passenger jet spin and upset characteristics to explain in detail.
Lonewolf_50 is offline  
Old 23rd May 2011, 21:53
  #2208 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
Spinning CW!

Lonewolf_50,

I´m sorry, CW!

It´s always good to put redundancy in the phrases.

OK, about roll oscillations, thanks. I asked because if she lost VS+assembly, some seconds before impact could "bank" left after losing it. Intuitive? It seems and could increase your "roll oscillation" factor.

Last edited by Jetdriver; 24th May 2011 at 06:06.
RR_NDB is offline  
Old 23rd May 2011, 22:00
  #2209 (permalink)  
JPK
 
Join Date: Jul 2009
Location: houston
Posts: 4
Likes: 0
Received 0 Likes on 0 Posts
Wall Street Journal, latest leak:

Preliminary Findings Suggest Pilot Error in Air France Crash - WSJ.com
JPK is offline  
Old 23rd May 2011, 22:13
  #2210 (permalink)  
 
Join Date: Jul 2002
Location: UK
Posts: 3,093
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by RR_NDB
OK, about roll oscillations, thanks. I asked because if she lost VS+assembly, some seconds before impact could "bank" left after losing it. Intuitive? It seems and could increase your "roll oscillation" factor.
The problem is that such a scenario contradicts all the physical evidence we have so far.
DozyWannabe is offline  
Old 23rd May 2011, 22:22
  #2211 (permalink)  
 
Join Date: Sep 2010
Location: by the seaside
Age: 74
Posts: 558
Received 17 Likes on 13 Posts
JD-EE
forget about the HF, when I had the dubious pleasure of using it one could go several hours without getting through.
Crossing the ITCZ especially in the middle of the night the important bit was finding our way through the cells.
Secondary was transmitting on VHF to let the other aircraft know where we were - same as over Africa and Asia.

We had the luxury of a double crew and occasionally we had the second captain watching as we transited the active zone. There was never ever an occasion when the working skipper was off the flight deck during the critical phase.
blind pew is online now  
Old 23rd May 2011, 22:25
  #2212 (permalink)  
 
Join Date: Jun 2009
Location: Oxford, England
Posts: 297
Received 0 Likes on 0 Posts
RR NDB, #2168
Would be useful to the thread to make a short briefing on "Finite State Machines", for me a fascinating issue.
Ok, will will give it a shot and apologies to those who find this
oversimplified or even patronising:

Long before the days of software programmed microcomputers, there was a
need to be able to manage fixed sequences of events or 'states'. For
example, repetitive industrial processes.

For a trivial and incomplete example, consider the industrial bottle
filling line below. The state column defines valid states, while the
description defines actions, or what happens for each of those states.
For each state, an action is performed and a decision made as to which
state to move (transition) to next.

State Action / Transition
----- -------------------
0 Move next bottle to fill position
1 Open valve to fill bottle
2 If bottle not full, hold at state 2, else, goto state 3
3 close valve
4 if more bottles, goto step 0, else goto step 5
5 end, process complete

The key things to note here is that each defined state is associated with
an action and that only defined states are valid. Any other condition,
such as a broken or jammed bottle, is defined as an error. Errors may in
fact be handled by more states, after which the normal state sequence
would be restarted, but let's keep it simple. In the old days, such a
plant would have been implemented using electromechanical relays, pneumatics
and motors. To summarise then: The states, the action at each state and
transitions between states are designed to "encapsulate" the system functionality completely.

Finite state machines as we now understand the term were first (and are
still used) to implement complex digital logic functions in hardware.
If you like, programmed solutions for systems whare the the sequence of
data being processed and it's methods never change. They were widely used in
telecoms to decode and translate data streams between exchanges and
subscribers. Because of the speed limitations of early microcomputers,
such work could only be done in hardware, but later processors allow
this work to be done in software. In a way, it's still really all software
though, even when done in hardware

For a more cutrrent example, we have to digress a little to communications
protocols. You can think of a protocol as an agreed language between one
or more systems that need to talk to each other. Such a language will be
transported in messages between one system and another. For a
trivial example, the wire between your home computer and printer carries a
defined message protocol to allow the two to talk to each other.

Messages are typically split into sections that describe, for example,
the start a new message (ie: phone off hook), control information, (ie, what
the message means) the data itself (ie: engine speed value), provide
error detection and signal the end of the message (phone hangup). Messages
are typically sent sequentially in time. That is, the overall message
is sent sequentially, one section after another.

Because of the agreed protocol and if the receiving node is keeping track
of parts already received, it knows which part of the message is coming next.
It can thus decode what each part of the message means without ambiguity.
So how do we keep track of where we are in the message?. We know that the
message has several sections and we know from the protocol spec the order in
which they should be received. We can thus define states that correspond to
each section, the actions that need to be performed and the transition to
the state that processes the next section of the message.

From a software engineering point of view, state machine based design
is a powerfull technique that allows a problem to to be broken down into
a number of small, well defined steps and can simply the generation of
demonstrably deterministic systems. Arguably the most socially significant
example is the internet, where the tcp/ip network stack is a state machine
based design.

This is a long post and hope it's not overdone. There's loads of info on
the web about state machine design that should fill in the gaps, or can
post a bit more if needed...
syseng68k is offline  
Old 23rd May 2011, 22:44
  #2213 (permalink)  
Thread Starter
 
Join Date: Jun 2009
Location: florida
Age: 81
Posts: 1,609
Received 52 Likes on 15 Posts
the "rest" of the story - F-16 FBW mods

Salute!

To avoid further dispersions as to why a few military would dare to tread amongst this august group of heavy pilots and experts, I'll conclude the story about our development of the first operational FBW system AFTER it was deployed and flown for a few years. My goal all along has been to provide some perspective concerning new concepts and their implementation with a first-person account.
++++++++++++++

Once we discovered that we could enter a deep stall and had no way to recover, things got serious.

- We transferred fuel forward shortly after takeoff. About 1200 pounds if memory serves. This helped our c.g., hurt our pitch rate, and led to some unintended consequences after landing if we forgot to "re-balance" on the way home.

- We added the pitch override switch to bypass angle of attack limits used by the computer..... well, almost. The switch was not active unless our AoA was above 29 degrees or so. Considering the deep stall was at 50 - 60 degrees AoA, no problem.

So we could control the horizontal tail surfaces "directly" in proportion to stick inputs.

- Developed a bigger tail. So we added a few square feet to the stabilators. This moved the aero center of pressure slightly aft, but most importantly, it gave the computers more control surface area to work with. Also gave we dumb pilots better control at slow speeds. We then quit transferring fuel, as the procedure was a kludge and had some unintended effects upon pumps and after-landing characteristics.

- We added a new control law called Category III. Named after our external loadout, which was impressive compared to the basic air-to-air configuration. The new law limited AoA and roll rate. This helped to avoid entering a departure induced by the inertia and aero effects of all that crapola hanging on an otherwise pretty jet.

Bottomline: Our engineers and USAF admitted that we all could have done better anticipating "upset" conditions or outright stupid moves. The system worked exactly as it was designed. Sound like a PR release or rumor from last week?

The flight test folks and GD came up with a great mod and it has served us well for 30 years.

we now return to our regularly scheduled hypothesis cacaphony.


Old fart with no airline experience
Familiar with many planes
Extremely familiar with early operational FBW systems
Extremely concerned with flying in a safe, well-engineered airliner
A.A.G.G.
gums is offline  
Old 23rd May 2011, 23:05
  #2214 (permalink)  
 
Join Date: Jun 2009
Location: Oxford, England
Posts: 297
Received 0 Likes on 0 Posts
Please tell me I'm not putting my life in the metaphorical hands of a
potential priority inversion or interrupts left off for too long. And I
hope the CPUs and hardware on the flying planes are used at about 10% of
their ultimate capacity if they are using multi-tasking.
The PI bug was almost unbelievable, but one suspects that internal politics,
project cost sensitivity issues etc were a factor in that one. Not something
that all software engineers would have experience of or be trained to recognise
at the time either. It's easy to be smug and there's plenty of info on the web on PI
now, but have yet to see a book that deals with it in depth and how to prevent it.

As for rtos, my (avionics) experience in that area is not current, but iirc,
there are at least two rtos's that are fully qualified to DO178 level, so it's
quite likely that they are being used. I would need to check what level
they are qualified to, but no company invests that amount of development
time and resource to such a project without the objective of making a lot of
money in return. You could check the Wind River site, or whatever they
are called now, for starters.

I don't see any problem with rtos, (treading carefully) and there may be
no alternative in that the business pressures, regulatory environment
and general competitive pressure mean that ever more complex solutions
are and will be needed to satisfy requirements. There's only so far that
you can go in simplifying a complex system and still have it work at all.

Of course, development processes and tools need to keep up with it all and
this is also cost driven to a degree (Again :-)...
syseng68k is offline  
Old 23rd May 2011, 23:06
  #2215 (permalink)  
wozzo
Guest
 
Posts: n/a
WSJ: Air France's Black Boxes Point to Pilot Error

The pilots of an Air France jet that crashed into the Atlantic Ocean two years ago apparently became distracted with faulty airspeed indicators and failed to properly deal with other vital systems, including adjusting engine thrust, according to people familiar with preliminary findings from the plane's recorders.
More:
Air France's Recorders Point to Pilot Error - WSJ.com
 
Old 23rd May 2011, 23:15
  #2216 (permalink)  
 
Join Date: Mar 2008
Location: At home, retired 2012
Age: 75
Posts: 14
Likes: 0
Received 0 Likes on 0 Posts
"Old fart with no airline experience
Familiar with many planes
Extremely familiar with early operational FBW systems
Extremely concerned with flying in a safe, well-engineered airliner
A.A.G.G."

...and what a rich life, that of yours.
jcarlosgon is offline  
Old 23rd May 2011, 23:39
  #2217 (permalink)  
 
Join Date: May 2011
Location: BOQ
Age: 79
Posts: 545
Likes: 0
Received 1 Like on 1 Post
First off I want to apologize to CogSim for my wording. I had no intention of wanting to make the discussion a Yoke vs SS or an A vs B discussion. Only a commentary on hand strength and the predilection to prefer one or the other for precision…without detracting from the AF447 discussion.

Smilin Ed's points are valid and I agree it’s all a matter of being trained, experienced and ultimately comfortable.

My experience includes flying various transport category aircraft with a yoke assembly (control wheel and column), from both the left side and right side; a transport category aircraft simulator with sidesticks, from both the left and right side; and fighter aircraft with both conventional center stick and FBW sidestick. I also flew the rams horn in the Hawker 800 aircraft from both the left and right seat.

I also flew different configurations, both aircraft and simulator, fighter and transport, on the same day from different chairs.

It indeed is not a strength or a handedness issue only a training and experience one. The A330 sidesticks in either seat are excellent pilot interface control devices. I personally use the armrest for both the control wheel and sidestick configurations in either seat. Some people don’t.

As to whether you need two hands on a yoke control wheel is a matter of technique and experience also. In the old days some aircraft didn’t have auto-throttles or auto-thrust and actual manual movement of the thrust levers (gasp) was often required. The control wheel was designed to accommodate two-handed flying, but did not necessitate it.

As an instructor in the B727 aircraft (prior to getting a simulator), I’ve flown manual reversion (flight control switches off) to the allowed aircraft training minimum of 300’ AGL on approaches from both the left and right seat. I can assure you that strength-wise it was equally difficult from either side.
OK465 is offline  
Old 23rd May 2011, 23:58
  #2218 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
DozyWannabe,

"The problem is that such a scenario contradicts all the physical evidence we have so far."

I made errors in expressing a model i was discussing here because i was in a hurry. And edited the post # 2199

Looking to the parts recovered first (floating) it seems matching a "flat spin" tail port (CW) Why not?

I had a "deep spin" trying to post some details we imagined here

But recovered and the model (of a flat spin until impact) is "stronger" now.

Last edited by RR_NDB; 24th May 2011 at 00:16.
RR_NDB is offline  
Old 24th May 2011, 00:11
  #2219 (permalink)  
 
Join Date: Mar 2001
Location: us
Posts: 694
Likes: 0
Received 0 Likes on 0 Posts
By the Wall Street Journal's account:

The crew methodically tried to respond to the warnings, according to people familiar with the probe, but apparently had difficulty sorting out the warning messages, chimes and other cues while also keeping close track of essential displays showing engine power and aircraft trajectory.
Sensory overload?

Question: How long would it take the captain at rest in the module -- and perhaps sensing that something is wrong around 2:10 or 2:11 -- to get to the cockpit, have a crew member open the secured door, and step inside?

At that point, at 2:12?, 2:13?, if the FOs are simultaneously trying to explain to him what is happening and fly the plane, is his appearance more hindrance than help?
SaturnV is offline  
Old 24th May 2011, 00:20
  #2220 (permalink)  
 
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes on 0 Posts
CRM upside down

SaturnV asked

is his appearance more hindrance than help?
CRM also "stalling",

2:13:16 ~ 2:13:41 (Possible "Loss of Signal" with satellite) Unusual attitude

Capt. location? CVR will tell us

Last edited by Jetdriver; 24th May 2011 at 06:05.
RR_NDB is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service

Copyright © 2024 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.