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

AF447

Old 12th Jun 2009, 02:14
  #1221 (permalink)  
 
Join Date: Jan 2008
Location: Vancouver, BC, Canada
Posts: 33
WhyIsThereAir, you said:

One can determine the VS was in place during the time the messages were sent by a far simpler test: the messages were received. Talking to a satellite requires a moderately stable platform. There are slew limits on how fast the antenna can track, even if it is a patch antenna or electrically steered antenna. The antenna has both pitch and roll limits and pitch and roll rate limits. It may also have limits on altitude change rate.
I am in the RF world...and for some reason I thought the Satcom antenna was a phased array antenna....therefore requiring no steering or any 'lock on time'...it simply needs to stay within a certain degree of vertical. That being said, if it is a phased array antenna and the aircraft become inverted or went in to an especially deep dive then no messages would transmit during such period.

I could be wrong about the antenna type...but after this thread and the other...reading several thousand posts...the antenna could be made of paper mache in my head by now.
DorianB is offline  
Old 12th Jun 2009, 02:27
  #1222 (permalink)  
 
Join Date: Mar 2009
Location: on the ragged edge
Posts: 80
Thank you "Safety Concerns" for that important post. I have to question a comment my colleague made which said:

The reason for such caution is, among other reasons, the fact that the ACARS is a maintenance reporting and messaging system, not a flight data analysis tool. The granularity of the data is very high, (stating the obvious).
While a good point, I'm not sure what he means by "granularity." I was under the impression that the most, if not all of the data path is digital, not analog on it's way up to the satellite. This means it is of a highly accurate and known behavior: we know the program, the version number and the data error bit checking subroutines that will not report the message if it is fragmentary (unlike micorsloth products which are not allowed in flight.) Error reporting of LMU items is very reliable stuff AFAIK: FAULT: The component experienced a subroutine shortcomming or a lack of signal that it needs to do it's thing. A fault occurred (not necessarily a complete failure or even a power loss). Some faults cause other faults in units that were depending on data from the upstream unit. This is my understanding of the process.

Mr Faser's attempt to delve into the aerodynamist's world of high speed, high altitude aerodynamic cause and effect sequences is ill-advised imho. These things do not have probable outcomes, even for a supercomputer. This is why most of the designs are scrapped at Edwards and Dryden after flight testing concludes, that for unknown reasons, the airfoil section when introduced on this particular airframe in this way behaves differently.

On the Rudder at altitude: whether or not a Yaw damper function was involved, or the pilot was involved is impossible to say. Airbus posters here suspect that yaw damping was available since in that alternate law mode it is still supposed to provide protection from dutch roll. My earlier comments were in the context of complete loss of yaw damping. I don't feel that's the problem now that Alternate Law has been explained to me.

All we should say for sure is that it appears the forward bolt hole of the VS yielded and the next two attach points did not, as evidenced by the attached lug carry though hardware. I wouldn't expect the VS/Rudder assembly to survive this way, attached and in good shape, if it was still attached to the airplane at sea level. But strong circumstantial SITA (french product for ACARS) data indicates it was attached at 0214z since the HF antenna coupler did not fault. (The SATCOM however, likely had it's own integral GPS for acquiring lock, I'm told, so this is no longer plausible evidence for an IRU platform available: hence ATT info.)

Seeing the same components faulted on other a330 mishaps is strong evidence. AB emphisizing unrelable airspeed preparedness is very suggestive that they feel this is the problem. Reportedly, the CEO said he doesn't think it's the pitot tubes. I don't either. Even though I think they all iced up, I suspect it's really the A330's interpretation of a complete loss of all air data that caused this accident. I had a number of accidental thrust latches on the FMA on the A310, where the AFS took the wrong action and required us to disconnect it.

Unfortunately, it appears I have sired a number of Captain experts with a total of one post. While they could have been banned for taking unpopular positions in the past, their posts suggest they have no experience flying in weather at altitude. I would suggest that from now on, that anybody with the title of Captain in his screen name is automatically suspect.

Yes including me. Feel free to challenge me on anything.

Cheers,

CC

All the above, is only my opinion only, and I could be wrong.
Captain-Crunch is offline  
Old 12th Jun 2009, 03:52
  #1223 (permalink)  
 
Join Date: Jan 2006
Location: US
Posts: 2,160
Picture of an A310 vertical stabilizer, after impact with water at approx. 215 KIAS, can be found on page 98.

Looks indentical to the AF 447 vertical stabilizer.

http://www.bea-fr.org/docspa/2000/5y-n000130a/pdf/5y-n000130a.pdf
misd-agin is offline  
Old 12th Jun 2009, 03:59
  #1224 (permalink)  
 
Join Date: Jun 2009
Location: Chelan, WA
Age: 43
Posts: 43
Originally Posted by barit1
Yes - NWA 727 on ferry out of NYC had exactly this problem (1 Dec 74) and the outcome was as PTH and NimSim described. If anyone has access to full NTSB reports, the # is NYC75AN070.
Just to be helpful, the report can be found at ASN Aircraft accident Boeing 727-251 N274US Stony Point, NY
einhverfr is offline  
Old 12th Jun 2009, 04:22
  #1225 (permalink)  
 
Join Date: Jun 2009
Location: Austin, TX
Posts: 1
To Captain Crunch

(lapsed student pilot here, but lots of computerized embedded systems expertise)

Regarding the "The granularity of the data is very high" statement, I interpreted this as saying "We don't know if or how the ACARS system batches transmissions".

In communications systems there is an overhead associated with obtaining the channel to communicate data, and to maximize utilization of the channel its appropriate to make the overhead (from a overall channel perspective) as minimal as possible. So the ACARS system might not transmit a single notification as soon as it receives it from the onboard systems... Hypothetical: Upon getting a notification it could set a countdown timer for 2 minutes, and when that 2 minute timer expires it transmits the original notification and anything else it receives in that two minute period. This is perfectly acceptable given what ACARS is intended for, and makes the overall communications system more efficient, but it also means that: a) The timestamps might not reflect the *exact* moment something was detected, rather it's a close approximation to the time. and b) There might have been unsent notifications in the ACARS system that never got transmitted (even if the transmitter and antenna stayed intact all the way til the end)
-RB
Roastbeef99 is offline  
Old 12th Jun 2009, 04:28
  #1226 (permalink)  
 
Join Date: Sep 1998
Location: wherever
Age: 51
Posts: 1,612
Picture of an A310 vertical stabilizer, after impact with water at approx. 215 KIAS, can be found on page 98.

Looks indentical to the AF 447 vertical stabilizer.

http://www.bea-fr.org/docspa/2000/5y...y-n000130a.pdf
misd again:

It's on page 122 and it looks nothing like the AF447.

Last edited by FE Hoppy; 12th Jun 2009 at 05:29.
FE Hoppy is offline  
Old 12th Jun 2009, 04:58
  #1227 (permalink)  
 
Join Date: Nov 2006
Location: SoCalif
Posts: 898
CC: "I suspect it's really the A330's interpretation of a complete loss of all air data that caused this accident."

The ACARS reported ADR fail, which could mean any one or more of the outputs of the ADR: airspeed, altitude, CAS, TAT, etc. . The pitot tubes on the A330 feed only airspeed computation, and airspeed is all that had to fail to lead to this accident. Altitude Fail in addition to airspeed fail is really, really tough to overcome. There was probably not a loss of altitude data, and I see no indication there was, except for the TCAS, which could be explained differently.

Do you see any other evidence of Altitude Fail?

GB
Graybeard is offline  
Old 12th Jun 2009, 05:09
  #1228 (permalink)  
 
Join Date: Jul 2004
Location: Found in Toronto
Posts: 609
Originally Posted by FE Hoppy View Post
misd again:

It's on page 122 and it look nothing like the AF447.
From page 122: http://www.bea-fr.org/docspa/2000/5y...y-n000130a.pdf

Lost in Saigon is offline  
Old 12th Jun 2009, 05:12
  #1229 (permalink)  
RWA
 
Join Date: Sep 2007
Location: Melbourne
Posts: 180
Checked the thread as best I could, apologies if this has already been covered.

On the Aviation Herald site, referring to this accident, there's a passage about two Air France A340s that had similar problems (thankfully without such a tragic ending) which says:-

"The captain released control of the airplane to the first officer and tried to switch his display from ADIRU1 to ADIRU3. 2 minutes later autopilot and autothrust disconnected and the fly by wire changed into alternate law. The crew noticed icing conditions (static air temperature [SAT] -29 degrees Centigrade) and switched anti ice including pitot heating systems from automatic to on. The speed indications became normal again and agreed again.........

'From automatic to on.' I'd heard that the A330/A340 have automatic pitot heating, don't know if all airliners do nowadays? But if there's an automated system controlling the pitot heaters, could it be possible that the whole cascade of 'computer malfunctions' was caused by.......a computer malfunction........?

Crash: Air France A332 over Atlantic on June 1st 2009, aircraft impacted ocean

Also - I looked through Airbus's 'advisories' as best I could, and don't recall seeing any mention of 'Check Pitot Heat ON.'
RWA is offline  
Old 12th Jun 2009, 05:23
  #1230 (permalink)  
 
Join Date: Mar 2009
Location: on the ragged edge
Posts: 80
Greybeard,

I have to defer to the A330 techs to tell me the likelihood of that. I'm just going on the experience that on the early bus where all three altimeters sprung into very different readings, and began to drift, so we were unsure of our altitude. We lost all air data for a time imho, and I suspected at the time that it got the static ports and tats as well.

Moral of the Story: Don't fly within 100 nm's of a typhoon near the equator or in the ITCZ with a bus. The ice won't show up on the radar and even benign looking stuff in the arm of cyclonic weather can take everything out for a short time.

And it's important to point out that this was a lot of years ago, and I'm retired now. A check pilot who did work for Boeing told me of the B-747 engine structure mistakes. Chalk it up to Test Pilot Urban Legend if you will:

Last edited by Captain-Crunch; 12th Jun 2009 at 07:10. Reason: deleted off-topic 747 proof
Captain-Crunch is offline  
Old 12th Jun 2009, 06:06
  #1231 (permalink)  
 
Join Date: Apr 2006
Location: Land of the Raj
Age: 65
Posts: 350
If the VS failed in a downward compression direction as has been intimated in preceding writings, I would think that it is quite possible that this caused deformation and damage of the aft pressure bulkhead and the speculated subsequent pressurization failure.

Comments?

KW
kwachon is offline  
Old 12th Jun 2009, 06:20
  #1232 (permalink)  
 
Join Date: Jun 2008
Location: Dubai
Age: 60
Posts: 29
SATCOM AE operation

From the RC OPtion Data (SDU-906)

The SDU receives airplane heading and attitude data from the inertial reference unit (IRU). The SDU system processor determines the elevation and azimuth angle necessary to point the high gain antenna beam at the applicable satellite. The SDU sends this data to the beam steering unit (BSU).
The Anetnna is a phased arraray type It needs to know which Active elements to activate to 'Steer the beam; ( done electronically, no moving components)

If anyone knows AF's fit Manufacturer &/or High Gain/Low Gain Antenna(s) please let me know
avspook is offline  
Old 12th Jun 2009, 06:41
  #1233 (permalink)  
 
Join Date: Jun 2009
Location: dublin
Posts: 7
re rudder:

We tried this in the sim:

FL 350 M .81 normal law,

Quickly apply full rudder, the a/c rapidly banks to 50-60 degrees bank, with only 4 degrees of rudder travel available.

Conversely, an engine failure in cruise is counteracted with the lightest touch of rudder, maybe 1 inch and can barely be seen on the flt controls page.

Respect the rudder, it is a very powerful control.
cpdlcads is offline  
Old 12th Jun 2009, 06:57
  #1234 (permalink)  
 
Join Date: Jun 2009
Location: Czech
Age: 51
Posts: 3
pj, safety concerns

concerning the VS issue - you cannot tell if there were other "1001 msg" or not because you cannot be sure that the 02:14Z msg was the last one, or are you?

do some of the experts present here consider the backloops between adirus and computers?
curious digger is offline  
Old 12th Jun 2009, 06:59
  #1235 (permalink)  
RWA
 
Join Date: Sep 2007
Location: Melbourne
Posts: 180
As I understand it, cpdlcads, the Rudder Travel Limiter does indeed restrict rudder movement to 4 degrees or so at high speed. As opposed to 30 degrees or so at low speed.

BUT, it surely relies on the ASIs for the speed? If the pitot tubes were partially blocked, presumably the ASIs would understate that speed?

So could the RTL, in those circumstances, have permitted a much greater rudder deflection than was safe?
RWA is offline  
Old 12th Jun 2009, 08:16
  #1236 (permalink)  
 
Join Date: Jun 2009
Location: dublin
Posts: 7
RWA

Yes thats why with rud trvl limit fault ecam says rudder: handle with care
cpdlcads is offline  
Old 12th Jun 2009, 08:26
  #1237 (permalink)  
 
Join Date: Jun 2009
Location: Who knows...
Posts: 8
avspook

It is possible, that SAT antenna was smcelectronic CMA-2102 SATCOM High Gain Antenna System.

http://www.cmcelectronics.ca/pdf/satcom2102.pdf

No build in GPS, electronically steerable, phased-array antenna, 12-17 dB gain.
Ground Brick is offline  
Old 12th Jun 2009, 08:37
  #1238 (permalink)  
 
Join Date: Jan 2008
Location: Irvine, CA
Posts: 94
Should we care ?

Yes thats why with rud trvl limit fault ecam says rudder: handle with care
Which means exactly what on a flight control without feedback? Do you have some sort of "care" marks, or artificial resistance when you reach those "care" limits?
CARE ON .................. CHECKED

Who has experience with careful rudder inputs flying manually at altitude at M 0.80 in alternate law?...in severe turbulence...
Test pilot territory I guess...

Not saying this is relevant. We are fishing in the dark.

Last edited by Interflug; 12th Jun 2009 at 10:35. Reason: spelling
Interflug is offline  
Old 12th Jun 2009, 09:04
  #1239 (permalink)  
 
Join Date: Aug 2002
Location: Europe
Posts: 229
I have my own idea

I have read quite a bit about this incident.
It is my personal opinion, according to the facts available at this precise time, that any other professional pilots who would have been on duty in that cockpit at that precise time, left exactly in the same scenario as the AF447 flying through that weather, would have done all possible to find a solution and fly the damn thing but whatever action would have been taken this would have resulted in a guaranteed disaster:
-Coffin corner and loss of all speed data (including GPS due to bad signal) is a recipe for disaster whatever the pilots on board.
Combine the above with a sudden and multiple list of serious faults and the plane very quickly either goes too fast or too slow, whatever corrective action is taken.
In both cases the plane would destroy pretty quickly.
If they could have precisely controlled their speed nothing would have happened. They could not guess their speed.
I am sure excellent pilots were dealing with an impossible event.
Airbus vs Boeing does not apply, any aircraft, whatever the make and model, would have gone beyond the coffin corner speed limits if flown without speed data.
Airbus was fully aware of the pitot tube problem, AF did not react.
Corners can never be cut in aviation especially when dealing with essential flying data, speed is everything. GPS signal in weather is lost.
ILS27LEFT is offline  
Old 12th Jun 2009, 09:27
  #1240 (permalink)  
 
Join Date: Jan 2008
Location: southwest
Age: 74
Posts: 287
Ils27left

Let's begin at the beginning.
Are you sure that 'any other professional pilots' would have been flying through that weather, and not around it?
Dysag is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information

Copyright 2018 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.