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

AF 447 Search to resume

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

AF 447 Search to resume

Old 27th Jun 2010, 10:38
  #1621 (permalink)  
 
Join Date: Aug 2005
Location: Belgium
Posts: 35
Likes: 0
Received 0 Likes on 0 Posts
"A heavy enough bank left wing down would take the satellite out of the ACARS antenna beam perhaps past the antenna aiming algorithm's attempts to correct it."

The a/c is within the beams of two satellites, it has AOR-E on its right and AOR-W on its left.
I am pretty confident Inmarsat knows through which of them the signals transited.
iakobos is offline  
Old 27th Jun 2010, 10:55
  #1622 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
RE: What If?

Originally Posted by GreatBear
By way of revisiting some of the discussion on the earlier thread, is it then possible that the aircraft was upset by really bad (severe turbulence being an inadequate description) up- down- shear forces at 02:10, stalled (easy in the coffin corner), and entered a developed and unrecoverable flat spin. In such a spin, would there not be greater than 30 kt disagreement between port and starbord pitots as the a/c rotated about its CG? Sufficient difference to trigger the PROBE-PITOT ACARS message?
GreatBear;

Pitots are relatively insensitive to variations in the local airflow angle. Significant differences due to angle of sideslip are likely to occur earlier in static pressure and AoA.

EDIT:: Re: "easy in the coffin corner": I've added a few datapoints to the 'coffin corner' graph posted earlier. According to information kindly provided by PJ2, it takes about 4 minutes with T/L at IDLE to decelerate from M.8 to M.6, which is V_alphaMax (Vs1g). It seems that V_alphaMax at FL350 is defined by buffet onset, which probably implies that the airplane can decelerate even lower into increasingly severe buffet before it stalls.

EDIT2::
The graph showed the Mach corresponding to the stall warning AoA of 4.2 degrees at 1g. Since this is the SW threshold at M=.8, it is appropriate to show it as loadfactor at M.8. The graph has been changed accordingly.

regards,
HN39

Last edited by HazelNuts39; 27th Jun 2010 at 20:18. Reason: new graph and link
HazelNuts39 is offline  
Old 27th Jun 2010, 12:54
  #1623 (permalink)  
 
Join Date: Jul 2009
Location: UK
Posts: 134
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by HazelNuts39
GreatBear;

Pitots are relatively insensitive to variations in the local airflow angle. Significant differences due to angle of sideslip are likely to occur earlier in static pressure and AoA.
Also difficult to see how the 2 pitots on same side would disagree.

@HN39

How would your plot change with alt = FL370 and/or T = ISA+30

Last edited by sensor_validation; 27th Jun 2010 at 13:09.
sensor_validation is offline  
Old 27th Jun 2010, 13:07
  #1624 (permalink)  
 
Join Date: Jun 2009
Location: 6m below sea-level
Posts: 8
Likes: 0
Received 0 Likes on 0 Posts
update seabed worker

mm43

UNKNOWN
1.. Why the aircraft got into a LOC situation
2.. How long the aircraft continued flying.
3.. Where it impacted with the ocean.

but will this be all affecting its actual position on the seafloor?
shape of wreck, weight, currents at higher depts, I wonder if a simulation has been worked with different scenarios depending on how fast the wreck sank, while going down affected by different current speeds (and possibly directions), with different types of damage to hull and wings. For instance sinking while positioned horizontally with effects from remained wingshape. Or would it possible that the wreck was still drifting occasionally over the ocean floor, until it topped over into the abyss of >4000m, are there any dynamics not taken account for, presumed static
Peter-1959 is offline  
Old 27th Jun 2010, 14:38
  #1625 (permalink)  
 
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by sensor_validation
How would your plot change with alt = FL370 and/or T = ISA+30
The data on the graph are not affected by temperature. At FL370 the LF's for buffet onset and SW are reduced by a factor of 1,10; and the severe turbulence speed is M.8. The FCOM gives Vs1g up to M.425 and I have one datapoint at FL350 only. However, I guess that if you read V_alphaMax on the curve reduced by 1,10 at an unchanged LF you wouldn't be far off.

EDIT::
The graph has changed to show SW as a LF at M=0.8. Therefore reference to SW Mach has been removed from above text.

regards,
HN39

Last edited by HazelNuts39; 27th Jun 2010 at 20:31. Reason: Graph has changed
HazelNuts39 is offline  
Old 27th Jun 2010, 19:17
  #1626 (permalink)  
 
Join Date: Jun 2009
Location: NNW of Antipodes
Age: 81
Posts: 1,330
Received 0 Likes on 0 Posts
Originally posted by Peter-1959 ...

.... but will this be all affecting its actual position on the seafloor?
shape of wreck, weight, currents at higher depts, I wonder if a simulation has been worked with different scenarios depending on how fast the wreck sank, while going down affected by different current speeds (and possibly directions), with different types of damage to hull and wings.
In post #322 I made mention of some contributing factors that will have affected the progress of wreckage from the sea surface to the bottom. The process is complicated by not knowing the exact state of the a/c hull following its impact with the water, but I suspect that it is in three separate pieces, along with the cores of each engine. That makes for five separate paths to the bottom, of which the engines can be considered as straight-down items. Read the above post and then discount sub-surface current affects on the way to the bottom - they are minor. In other words, you are looking for a bottom debris field of about 400m radius, and probably smaller.

mm43
mm43 is offline  
Old 27th Jun 2010, 20:10
  #1627 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
Hazelnuts39 asks how far off satellite time might be from aircraft time or ground time.

The satellite would be very close to ground time. The satellite is constantly in touch with several ground stations which can provide it with corrected time certainly within milliseconds accuracy and probably much better. So I think we can stipulate that on terms of seconds the satellite is right on the mark with the ground stations.

Suppose the ACARS protocol does not require precise timing on the aircraft and as such can wander considerably. (I doubt this is the case.) A cheap crystal oscillator without any fancy temperature compensation is cut to about 50 ppm accuracy and wanders 25ppm to 50ppm over various military temperature ranges. Let's suppose it sits 100ppm, 0.01%, in error for an entire 12 hour (43200 seconds) flight. It might be as much as 4 and a third seconds off at the end of the flight.

This, however, is a very artificial scenario. The signal must be precisely on frequency or the satellite will have trouble extracting the modulation. The Inmarsat-M sets I worked on had temperature compensated oscillators, TCOs, good to the 1 ppm sort of level. ACARS is a simpler protocol. It MIGHT only require 5 ppm. I do not have the ACARS specification (or Inmarsat M) at hand anymore. Inmarsat marks them as proprietary so I properly discarded my copies when I left the program.

5 ppm is 20 times better than the scenario above or 0.216 seconds per 12 hour flight. It would be within a second over 10 days.

There is one more consideration, is there any procedure for setting the ACARS time estimate before each flight or is it "automagic?" I would "presume" there is a precision 1 second time tick and other means of sending that tick's actual GPS time value around the aircraft. I'd be pretty sure that ACARS snarfs up that estimate and uses it for continuous internal time updates. So having ACARS a millisecond off timing is not likely. The only real ambiguity is when the time stamp is applied to the signal. It seems to be part of the ACARS message itself. So it must be applied as the message enters or exists the transmission queue. (And, with time a part of each ACARS message it's self-time-correcting to fractions of a second.)

IMAO the time tick should be when it entered the queue. This accident shows that this might have helped unravel some of what happened. (And the NYSE recently discovered that makes a BIG difference when you have computer traders that will game a situation such that the queue becomes longer than 1 trade. There is a credible theory that's what brought the market down in 2008. That shows two things, this is an issue that is often poorly considered in system design and that this may also be the case in the ACARS system used on AF447.)

I hope that swats down ideas that the clock times could be materially off to say 10s of seconds or more.
JD-EE is offline  
Old 27th Jun 2010, 20:13
  #1628 (permalink)  
 
Join Date: Jun 2009
Location: I am where I am and that's all where I am.
Posts: 660
Likes: 0
Received 0 Likes on 0 Posts
iakobos - the question there is, of course, how fast can it resynchronize to a different satellite? The particular version of Inmarsat protocols with which I am most familiar is not particularly fast. It doesn't happen often enough to worry about for things that float or drive around countries other than the US. The time is measured in 10s of seconds as I recall.
JD-EE is offline  
Old 28th Jun 2010, 00:54
  #1629 (permalink)  
 
Join Date: May 2010
Location: Le Shed on the Tropic of Capricorn
Age: 62
Posts: 22
Likes: 0
Received 0 Likes on 0 Posts
Pitots may have worked properly, ADU error checking may have precipitated the event.

I have been silent for 12 months but would now like to raise the following for discussion.

1) The BEA#2 report indicates a sudden increase in speed inconsistency reports (7 times the frequency) from a point about 2 months before the peak in oil prices. Could this coincide with a renewed emphasis on flight plan optimisation with respect to fuel??

2) Speed indication of Mach by pitots can be over-estimated in a Cb and TAT will over-estimate SAT when using the over-estimated M.

Standard atmospheric air (mostly diatomic) has a Cp/Cv ratio referred to as G (gamma). Normally the low humidity, low pressure and low temp air at cruise level means G is relatively constant. Add abnormal % of water (triatomic) and G is no longer constant but decreases (BEA#2 'strong condensation' pg69).
The air density value is required for the pitot to measure air speed but M can be calculated from pitot results if G is assumed. As G decreases, M is overestimated by the pitots. This M value is used to correct static pressure giving an altitude calculation variation (error) even if the altitude has not altered and there has only been a change in G. This M is used to calculate SAT from TAT. SAT is also overestimated.

On a sudden change in Ps (altitude) calculation due to incorrect M, the TCAS (BEA#2 pg 35) ACARS would appear if Ps fails the 'credibility test' @2:10.
The IR may also reject the ADU sudden changes in calculated data (due to the variation in G), hence the ADIRU fault (2:11).

It seems possible that a sudden change in the air property (G) due to the Cb could start the cascading faults shown in ACARS due to internal error checking. This then leads to the A/P and A/THRUST disconnecting although the A/C may not have altered its velocity or altitude.

I don't recall this being raised previously. My apologies if it was. My concern is that AF447 may be somewhat repeatable if 'heat upgraded' pitots are installed as this does not alter the above, albeit rare, scenario.

thanks. Ian
FluidFlow is offline  
Old 28th Jun 2010, 11:16
  #1630 (permalink)  
 
Join Date: Jan 2008
Location: Herts, UK
Posts: 748
Likes: 0
Received 0 Likes on 0 Posts
Ian,

So that is a genuine instrumentation behaviour in exceptionally high humidity air masses...? Then it throws another potentially broken egg into the basket !
HarryMann is offline  
Old 28th Jun 2010, 12:21
  #1631 (permalink)  
 
Join Date: Aug 2009
Location: Texas
Age: 64
Posts: 7,196
Received 388 Likes on 240 Posts
Ian, are you suggesting precip/rain/moisture as the air anomaly that sets up your M error, or a vertical column of warm/moister air (which Tim Vasquez had suggested in his analysis), arising from the inherent instability within a Cb?

If I understand you correctly, this M error, due to the combination of factors, ought to have been experienced more frequently. Is it your suggestion that it has been, but the community has been generally unaware since the community wasn't looking for it?

Last edited by Lonewolf_50; 28th Jun 2010 at 14:44.
Lonewolf_50 is offline  
Old 28th Jun 2010, 12:43
  #1632 (permalink)  
 
Join Date: Jun 2009
Location: Bedford, UK
Age: 70
Posts: 1,319
Received 24 Likes on 13 Posts
Gamma - do the numbers add up ?

to trip the assumed accident chain
eg
http://www.dtic.mil/cgi-bin/GetTRDoc...c=GetTRDoc.pdf

Last edited by Mr Optimistic; 28th Jun 2010 at 12:44. Reason: not enough 's'es
Mr Optimistic is offline  
Old 28th Jun 2010, 14:17
  #1633 (permalink)  
 
Join Date: Jan 2005
Location: W of 30W
Posts: 1,916
Likes: 0
Received 0 Likes on 0 Posts
Ian,
The 3 Pitot heads, the 3 static ports, the 2 TAT probes will go through the same air mass at the same time, therefore they won't have any reason to disagree.
If I follow your theory, the only risk would be for the A/THR to compensate the over-estimated Mach and get the airplane closer to stall.
CONF iture is offline  
Old 28th Jun 2010, 14:54
  #1634 (permalink)  
 
Join Date: Jun 2009
Location: Chesapeake Bay
Age: 79
Posts: 57
Likes: 0
Received 0 Likes on 0 Posts
FluidFlow (Ian): Welcome to the workroom!

I think we are all hoping for successful search and then recovery of the FDR and CVR. The sequence of ACARS messages, however, allows development of plausible scenarios and cause theories. The pitot disagree message has already effected industry changes (Thales to Goodrich).

Dean's 1979 paper on "Atmospheric Effects on the Speed of Sound" linked by Mr Optimistic points to the influence of water vapor at various temperatures in an ideal gas (see his Table 4) and talks about the heat capacity ratio Cp/Cv you mention. Seems the deltas are pretty fine, though...

Two thoughts:

1. Your proposition could be mathematically modeled if we knew the "standard altitude parameter" and the "credibility" test boundaries used by the TCAS software and mentioned by BEA to see if G can decrease sufficiently to increase M sufficiently to trigger an altitude calculation error and the ACARS warning message. Fluid dynamics, I've heard, is the trickiest of the physics disciplines, and the math is way beyond my own pay grade, but with the TCAS source code in hand I should think your what-if question could be ruled in/ruled out.

2. According to mm43, the WRN messages are sent in real time. Reciept of the TCAS NAV message follows the AUTO FLT AP OFF by 44 seconds. In your proposition, the TCAS NAV message, if the trigger, likely would have preceeded the other WRN messages.

GB
GreatBear is offline  
Old 28th Jun 2010, 16:52
  #1635 (permalink)  
 
Join Date: Aug 2005
Location: Belgium
Posts: 35
Likes: 0
Received 0 Likes on 0 Posts
JD, from experience it takes anything between 15 and well over 60 seconds to (re)synchronize on an Inmarsat, most cases falling in the 20-40" braket.
iakobos is offline  
Old 28th Jun 2010, 18:18
  #1636 (permalink)  
 
Join Date: Jul 2009
Location: UK
Posts: 134
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by FluidFlow
2) Speed indication of Mach by pitots can be over-estimated in a Cb
I am intrigued by this - the actual density of humid/ saturated air will be lower than that of dry air (the effect that keeps clouds up), so the total pressure observed by the pitot will be lower, and if calculated dry gas density used to convert to speed, the actual calculated air speed under-estimated. The wings will also see the lower density and produce less lift.

I am not aware of pitot tubes being compensated for moisture content of the air, are they ever?
sensor_validation is offline  
Old 28th Jun 2010, 18:54
  #1637 (permalink)  
 
Join Date: Jun 2009
Location: 6m below sea-level
Posts: 8
Likes: 0
Received 0 Likes on 0 Posts
deepsea search

mm43

"That makes for five separate paths to the bottom, of which the engines can be considered as straight-down items."

I cant find good information on how they are searching, what sort of sensory equipment.There are laserlight camera's these days, which can be operated till 6000m or more, mounted on a ROV. Certain laserlight wavelengths will give distinguishable reflections on metal surfaces, and meanwhile penetrate deep enough through low density organic sediment layers.
Peter-1959 is offline  
Old 28th Jun 2010, 21:19
  #1638 (permalink)  
 
Join Date: May 2010
Location: Le Shed on the Tropic of Capricorn
Age: 62
Posts: 22
Likes: 0
Received 0 Likes on 0 Posts
ABNORMAL air properties

I am currently on holidays with a pathetic internet connection.

@HarryMann. This is genuine once we agree we are referring to an abnormal situation. I raise it for discussion to see if it is applicable.
@Mr Optimistic. (#1631). Thanks for the link. Most of this appears to be at 1 atm and within the normal humidity levels of air. A link that shows G variation over a larger range of variables would be http://spaceagecontrol.com/AD-InFlig...easurement.pdf fig 62(a) This graph too is for normal humidity but you can see the trend of decreasing G with added moisture ('saturated air' in this case). I cant post the actual formula at the moment but G decreases if air has more triatomic gases in it ie also more CO2 or O3. ie the N2, CO or O2 mix doesnt matter much. This is also one reason why the term 'dry air' is used as the reference numbers quoted (ie R) to the 10th decimal place would all be 'wrong' if the air was not dry.
@Lonewolf. Yes, I am referring to high levels of moisture (not just a saturation of air with water vapour though). ABNORMALLY high amounts of moisture that are brought into the cruise level by a Cb. The M variation falls into the 'who cares' category but it is the effect of this potentially sudden change in M on the calculation of other parameters which are then scrutinised by error checking. If there was no error checking then I suggest there would not have been a problem as the error in flight data is not sufficintly significant within itself. Hence it is the handling of the variations in air properties that I believe may be an issue and may have caused HAL to give up.
@Conf. Yes, the pitots and TAT's will agree with their partners ignoring turbulance. However for a change in G the M error from a TAT/SAT calcualation is about 4 times that from a pitot M calculation. So a program checking for errors may notice an anomoly here as it checks the validity of other calculated parameters even though it is 'turbulance trained'. It all depends on the range allowed before a parameter is deemed 'not valid'.
@GB. Thanks for your welcome and comments. My point could also be phrased in terms that we did not have an 'ideal gas' in this Cb. Yes, it all hinges around the 'credibility' tests within the software which one would expect to have been determined by the range of air properties expected to have been encountered. Enter a Cb with properties outside of this range and HAL may get confused.

rgds Ian

Last edited by FluidFlow; 28th Jun 2010 at 23:32. Reason: smiley added
FluidFlow is offline  
Old 28th Jun 2010, 22:40
  #1639 (permalink)  
 
Join Date: Jul 2009
Location: UK
Posts: 134
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by FluidFlow
Hence it is the handling of the variations in air properties that I believe may be an issue and may have caused HAL to give up.
No, the Auto thrust and Auto Pilot disengaged because the speed measurements disagreed, which is most likely due to uneven blockage of pitot tubes and their drains due to ice.

Air properties and computations will be a common-mode problem, but are of considerable interest in understanding the state of the A/C when handed back to the pilots for manual control.
sensor_validation is offline  
Old 28th Jun 2010, 22:53
  #1640 (permalink)  
 
Join Date: May 2010
Location: MA, USA
Posts: 126
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by Peter-1959
I can't find good information on how they are searching, what sort of sensory equipment.
I presume you have looked at the information provided by BEA on the search equipment: Sea Search Operations

If you follow the links there, you will find the specs for the sonar on the Orion towed vehicle. If you dig deep, the specs for the REMUS AUVs from Waitt Institute are here: AUV Specifications | Search for Amelia though I don't see mention of the sonar frequency and range. Specs for the sonar on the Geomar vehicle are here: IFM-GEOMAR: The AUV Abyss where it mentions the frequency but not the range.

The search is being conducted with side-scan sonar. The sonars operate with frequencies in the span of 50-250kHz, depending on the required trade-off of range and resolution. The typical ranges used are about 300-500m either side of track, and the achievable resolution allows detection of an object about the size of a 55 gallon oil drum.

There are laserlight camera's these days, which can be operated till 6000m or more, mounted on a ROV. Certain laserlight wavelengths will give distinguishable reflections on metal surfaces, and meanwhile penetrate deep enough through low density organic sediment layers.
Laser scanning cameras have the advantage of reducing the common-volume scattering (water volume that is common the both the path from the light source to the target, and from the target to the camera), and thus the fog effect of underwater photography. That can greatly increase the optical range above the 10-20m typical of conventional photography in the deep ocean. A quick scan of the literature shows that ranges of 4-6 attenuation lengths are achievable, limited by forward scattering, which is not eliminated. See for example: http://jaffeweb.ucsd.edu/files/bathy...m%20L-bath.pdf

Attenuation length (distance in which intensity drops by 1/e) in the deep ocean does not exceed 50m: http://www.nestor.noa.gr/2nd/files/247_252_bradner.pdf Typical AL in the deep ocean is probably less than half that. That puts the range of a laser scanner at a maximum of 6*50=300m, but more typically 4*25 = 100m. That does not compete well with sonar for a wide area search for large objects that are likely sitting proud on the seafloor.

I'm not sure about your point regarding wavelength. The range of wavelengths is quite narrow, in the blue-green range, for which attenuation in seawater is reasonably low. Red, infrared and UV are strongly absorbed.

Do you know of some equipment with specifications that exceed the above numbers?

Last edited by auv-ee; 29th Jun 2010 at 01:12.
auv-ee 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.