Winterymix - convection intensity over the open ocean doesn't vary much day/night since sea surface temperatures have very little diurnal variation. However convection over the land not far from the LKP would decrease overnight leaving more 'airspace' over the adjacent sea for CBs to develop and peak towards dawn (without going into the broader scale MET dynamics). Also, the developing phase of CBs is the most active in terms of turbulence/hail/lightning/vertical motion so it's the brand new (towering CU/CB) cell which has just become visible to radar which is often the most dangerous.

Thank's, Triskel
 

Nice clarification about convective threats in the environment of interest. So...what did the other several flights in the area at the time of interest know or did they simply get lucky?

Wave action
 

"Boomtown

Quote:
With regards to the twisted and crushed metal we are seeing - early reports cited gale force winds and waves up to 5 meters high. Is anyone able to comment what those sort of storm conditions would do to a (temporarily) floating airframe?

Are those loads capable of tearing apart what is an otherwise largely intact airframe?

The short answer is no. Most damage is done when waves push an object against rocks or grind it against a rocky seabed. A loose collection of objects floating offshore can also be ground against one another by wave action causing damage primarily around the edges, but not tearing and compression damage of the sort seen in the pictures currently circulating. All the indications are that the damage was done by a high velocity impact with the sea surface."

I would concur with the no. Having been aboard a 50' boat in such conditions in water equally shallow, for stupid reasons I'm not going to get into, the impacts were severe, but not enough to break bones, bend seats, break glass, of anyone and anything secured. If you take a look at the fuselage of US 1549, it skidded across the Hudson and was bumped and hit by numerous rescue ships with metal hills and banged against the concrete wall at Battery Park, yet does not display the massive accordion crumple that we have seen in the instant incident.

If you want to see descriptions of flow management - known in the FAA as Traffic Management Initiatives then you could look at:

http://www.fly.faa.gov/Products/Trai...oklet_ca10.pdf

Things have not changed significantly since that was written.

This paper: Analysis of overshooting top detections by Meteosat Second Generation: a 5-year dataset - Proud - 2014 - Quarterly Journal of the Royal Meteorological Society - Wiley Online Library shows that, at least in Tropical waters off Africa, convection peaks at around 08 or 09 local time. In the ocean far from land it's around midnight and over land it's in the evening. The same is, at least on the day of the accident, true for the region around Indonesia/Singapore. Because there's so much land over there the sea is never truly isolated so we get the 'coastal' convection system with peak intensity around daybreak.
I think a better question (assuming that weather was a factor) would be: What made this particular flight unlucky?

I will pick up a few points from what you say:

Yes FAA, NAV Canada, and the Member States of ECAC - EUROCONTROL, Air Services Australia have fully fledged Traffic Flow Management (TFM) Systems. Under ICAO coordinated flow management is being developed with GREPECAS, the states of Caribbean and South America. The Asian Pacific Region Air Navigation Service Providers (ANSP) are aware that they do not have a networked TFM system and work is in hand using the auspices of ICAO to put one in place (see http://www.icao.int/APAC/Meetings/20...attachment.pdf )

Miles/Minutes in trail are going to be different as the aircraft moves from surveillance based control to non-radar procedural time based control. In consequence the flow management at boundaries between these sectors becomes complex. As surveillance based control normal separation minima are considerably smaller than procedural.

The statement "The pilot(s) can see in real time what the actual weather is , ATC cannot." is not necessarily correct - and even if true it is too simplistic to just avoid the next heavy rain radar return. Both the pilot and the controller may be able to see weather, their systems are different so they will see different weather. The controller can also see a lot further ahead and can see sucker traps where some flight paths would lead into dead ends, the controller also may have had lots of PIREPS on turbulence. The controller's weather comes from different radars so it will present weather information that is hidden from the aircraft radar by attenuation. It makes real sense for both the controllers and the pilots to work together there is no competition, both want the same outcome and the pilot should accept any help that can be offered.

If the weather is getting a bit too exciting then the pilot should say so early don't wait till things are really bad or until ATC gives you an instruction you can't take. If you can tell the controller that the weather is looking really bad ahead of you or to one side and you may need to take some avoiding action, then the controller will start ensuring that other traffic is kept clear of both you and the weather you are reporting.

The time for 'Communicate' is before things get so bad that all you can do is 'Aviate' then perhaps things won't get to that stage.

QZ blackbox has been found on coordinate 03.37.21 S/109.42.42 E with depth of about 30 to 32 meters

Tim Penyelam TNI AL Berhasil Temukan Kotak Hitam AirAsia

QUOTE:
Winterymix - convection intensity over the open ocean doesn't vary much day/night since sea surface temperatures have very little diurnal variation. However convection over the land not far from the LKP would decrease overnight leaving more 'airspace' over the adjacent sea for CBs to develop and peak towards dawn (without going into the broader scale MET dynamics).
This paper: Analysis of overshooting top detections by Meteosat Second Generation: a 5-year dataset - Proud - 2014 - Quarterly Journal of the Royal Meteorological Society - Wiley Online Library shows that, at least in Tropical waters off Africa, convection peaks at around 08 or 09 local time. In the ocean far from land it's around midnight and over land it's in the evening. The same is, at least on the day of the accident, true for the region around Indonesia/Singapore. Because there's so much land over there the sea is never truly isolated so we get the 'coastal' convection system with peak intensity around daybreak.
Quote:
Nice clarification about convective threats in the environment of interest. So...what did the other several flights in the area at the time of interest know or did they simply get lucky?
I think a better question (assuming that weather was a factor) would be: What made this particular flight unlucky?

All the above omits mention of the other driver for rapid high level CB development. That is cooling from above, rather than heating from below. Cold advection [wind backing with height] can and does throw petrol on the fire of a hitherto modest CB. Believe me, it happens.

Hi,

With all the photos of aircraft parts available so far .. my feeling is a flat ditching of those parts .. little forward speed ... (my two cents)
Can't wait the analysis result of "black boxes" .........

@ Ian W Acoustics communications
 

Change the frequency - Remember this is not radio this is sonar a lower frequency in water has a lot more range.
Change the pulse recurrence frequency so that there is one pulse every minute or even every 3 minutes by charging a capacitor bank then discharge that for a greater power outputEncode the signal with the last position from the DFDR/GPS and the aircraft ID

You're right that VLF at 8Khz would give better range than 37Khz but not much more, certainly not in terms on NMs. Water presents just too much attenuation unless the source uses enormous power.

What of the idea that the ULB remains asleep until it is interrogated so extending battery life. , much like transponders that are used by Offshore oil and gas vessels which a give an xyz position Extra information could also be supplied easily depending on the firmware used. All this technology is already available but would imagine it would take years to get through CAA legislation.

If the crew failed to find the weather on radar that likely took them down, the first warning would likely be the sound of ice pellets hitting he aircraft.

Most of my tactical jet XC work (ancient history) was without available weather radar. The only operational restrictions applied from on high was, "do not fly into Weather Warning areas." As a result, we had the opportunity to fly into and through some otherwise interesting weather. Other than some spectacular St Elmo's fire, it was not that exciting.

For the meteorologists here, did conditions in the ITCZ on the airway that day equate to WW level storm intensity?

I'm betting that the weather is more likely to have interfered with the aircraft's sensors and thus precipitated this accident.

Flat spin background
 

Yes, I'm reading the wreckage as a flat impact with wings (+- 15 degrees) of level and low forward speed. This corresponds to a flat spin.

One ominous aspect of flat spins is the eyeballs out g that results from the rotation. This would rapidly disable the crew, particularly if they did not have their shoulder harness locked.

How to get into a flat spin? Not hard once you get real slow. Compressor stall one engine while at high power and high AOA.

Matching up debris
 

If this helps anyone ive matched up the tail to approximately where it fits to the rear section.


http://i60.tinypic.com/20l17q8.jpg

flat spin
 

"MACHINBIRD" IMHO if you'd ever done a flat spin, you wouldn't write :"One ominous aspect of flat spins is the eyeballs out g that results from the rotation. This would rapidly disable the crew, particularly if they did not have their shoulder harness locked."

I 've been teaching aerobatics and spin recovery ( positive, inverted, and positive flat and inverted) spins (in CAT "A" aerobatic airplanes...) for over 20 years and both my eyeballs are securely in place, or so say my FAA and JAR-FCL medicals.

You don't need a compressor stall - high AOA and a small input to the rudder can put you into a spin relatively quickly. It was the standard method to enter a spin way back. Do it a fraction before the stall and you can have max lift on one wing and the other stalled. I would think that one of the Normal Law protections would stop you getting there unless it was all tied up with an overspeed protection because of a sudden increase in OAT.

I wonder if we are getting to a point where test pilots are going to be required to test departures from the flight envelope to see how the aircraft behaves and how to get it back? Otherwise a LOC for whatever reason that puts the aircraft immediately out of the envelope requires the crew involved to become test pilot capable in seconds.

Wouldn't the g experience in the cockpit depend very much on the center of rotation of the fuselage? In a Cat A aerobatic aircraft the cockpit is not a lot forward and indeed may be on the center of rotation - probably with the design intent to reduce g when the aircraft is rotating. Now in a larger non-aerobatic aircraft that is not a design decision.

Spin?
 

There is no where near enough wreckage for me to start concluding what the impact angles or energy may have been. However, there is one piece of evidence that has come in and must be taken into account. With the destruction of the aft pressure bulkhead I don't think it is possible that the impact of the tail was low energy. One POSSibility may be that the tail came down as a separate piece and the open end of the tube hit first with the hydraulic effect opening the pressure bulkhead and the bottom of the fuselage.

VS damage: sideways on impact, starboard side down :confused:

Edit: with linear witness mark from subsequent impact with port HS?

I have been wondering about this strange tail section torn off and went and took a look at a pretty recent water event - this one with a very successful outcome for everyone onboard. Also an A320.

Sorry, the pictures are small but I have links to highres pics at the end of the post.

Notice the similarity in the tail section. This is fron the transport of the recovered plane so no stabilizers...

http://telstarlogistics.typepad.com/...910934e28a4-pi


The actual plane on the museum. Stabilizers are put back to place. Surprisingly similar damage as seen on AirAsia.

http://s3-media1.fl.yelpcdn.com/bpho...e2ZSvng/ls.jpg

Some links to highres photos:

http://journeysbyjill.files.wordpres...2/img_0168.jpg

http://smithsonianscience.org/wordpr...light-1549.jpg

http://tinyurl.com/m978xrt

EDIT; Seems like the last picture won't show properly sometimes. I have saved it on my computer if someone is really really interested...

EDIT 2: Followed advice and used tinyurl.

I can no longer find the NTSB report for 1549. Angle of the plane on impact would be interesting for those of us unfamiliar with it, if anyone can recall. Believe speed was approx. 153kt.

Modern airborne radar is better than the "crap" we had 30 years ago? Surely you jest.

I could see the control tower on an airport from 5 miles out with the old radars, I could see individual airplanes parked on the ramp. I could see the weather clearly and make my own decisions as to the amount of water carried by parts of the cloud and thus make my own decision as to the parts to avoid. I flew years and years in the tropics and experienced hundreds of severe encounters at some of the worst levels (around 13,000 to 15,000 feet).

I give you the new radars are easy to use and they do the deciphering for you, but often they are wrong and always exaggerate. They are useless for fine work, cannot do even a small part of what was done by the older radars. They are cheaper to buy and to maintain, and are much lighter, so I see the reason for them, but don't kid yourself that they are better for the purpose they were built.

susier
 

Here is, I believe, the full report:

http://www.ntsb.gov/investigations/A...ts/AAR1003.pdf

I just found another angle showing how/why an aircraft can be ripped apart from below...

http://blog.flightstory.net/wp-conte...age-damage.jpg

flat spin
 

Good point, IAN G, about center of rotation. But, as you know, its real tough to get into a flat spin... which requires deliberately held full back stick (as AF# 447) AND out-spin aileron, i.e. a crew trying to stop the spin with ailerons (toward the high wing) instead of rudder. Anyone who has ANY unusual attitude/spin training KNOWS this is wrong. Dead wrong. From the French BEA report, AF # 447 wasn't in a spin, but a long (4 minute...) falling leaf configuration... in and out of stall, with wings alternately rising and falling as the nose remained close to, or above, critical AOA.

boofhead, thanks for that - have often wondered about these "modern" radars too

RE Numerous posts
 

Safety vs cost and who decides: Black boxes have nothing to do with safety other than supplying new knowledge to prevent a similar future event. Ingenuity as per IanW, not bales of money, should be aimed at crash location improvements. But focusing money, attention, and ingenuity on wreckage location is attention not aimed directly at accident cause. Crash location improvements directed by hindsight have to wait for identical circumstances to recur, but even then black boxes cannot fully inform about the nature or extent of pilot confusion.

While holding the interest of many (including me), exactly how 8501 impacted the ocean is in many ways irrelevant to 'safety' because final impact most resulted from loss of control long before. Impact speed and attitude were not chosen by the 8501 pilots (unless Mr Snuggles is on to something). Falling leaf, spin, flat spin, or other descent profiles are mostly irrelevant because pilots don't train for them and pax jets aren't designed to complete or withstand recovery from them.

Accidents are ~0.001% (or pick a tiny %) of all flights. Some accidents such as airframe failure resulting from completely unforeseen forces or airframe inadequacies are unavoidable, but they get included in the 0.001%. 'Normal' rarely applies. What does usually apply is a confluence of events/circumstances peaking, like ocean rogue waves, in a very short amount of time, from a few seconds to hardly more than a minute. The last opportunity to avoid an accident is the time between the next to last and last decisions in a short chain. Accidents seemingly surrounded by normality, and accidents of an exceedingly rare confluence of factors, both share pilot confusion and inattention as high ranking primal causes.

One improvement would be better real time wx information (Langleybaston and ATC Watcher) to avoid the series of brand new surprises involved in 'picking your way through'. But there are numerous previous posts about the current limitations of both equipment and the operators of that equipment. Horizontal separation of five or fifteen miles from preceding flights no guarantee of identical weather.

The best solution would be to focus attention on how to elongate the time available to pilots to react to conditions to enable good decisions, better real time wx being one aspect. Time elongation is otherwise currently and systemically limited by both a very narrow range between overspeed and stall, and by momentary (where 60 seconds is a long time) failure of necessary instrumentation or agreement of automation components. More pilot time for thinking would prevent some accidents, time not currently available as events prove. My point is that accident prevention can't ignore the coffin corner of time, so while time needs to be addressed, the impediments are systemic. IanW's last paragraph in 1784 also applies.

Great set of Flt 1549 pics in MrSnuggles posts! Actual Utoob videos of the event show the angle of fuselage to water, which was notably glass smooth. The same impact angle where the point of impact chanced to be 20-30' fwd of that of 1549, and with the impact being not glass smooth but a 15' swell/wave instead could have holed the fuselage bottom of 8501 and directed a torrent of water into the fuselage, overpressuring the upper fuselage, parting the lower half of pressure bulkhead, and carrying away the aft floor, horizontal stab mounts (which escaped in 1549), FDR and APU. A lot of "could have's" remain.

Have to agree with boofhead about the old radars. The "C" band would punch through everything and give you a clear picture of what lay behind the storm immediately ahead. Definition was absolutely superb. It was a sad day when they were removed for "progress" (cheaper).

Modern Radar
 

I'm new to the blog, but very experienced with "old" radar. Many would argue that the old green screen radar scopes were more accurate. I agree in part, but I do not recall ever seeing a control tower or aircraft on the ground. I will admit I did not like the new digital color radar when it first emerged many years ago. In all fairness I always wanted to compare the old with the new in numerous difficult situations. Having 37 years with the airlines added greatly to my decision making process when interpreting course corrections for weather.

If you don't have this skill, avoid the situation to begin with or declare an EMERGENCY and turn around.

Many thanks Mr. S, for the Hudson report link :)


Looks as if the pitch was 9.5 degrees and speed was 125kts, so I was mistaken about that.


Airbus ditching parameters:


'The January 21, 1988, Airbus certification test report stated that the fuselage of an A320 would "undergo no destruction liable to create a water passage" if the airplane ditched with the following parameters:

1. landing gear retracted,
2. 11° pitch,
3. -0.5° glideslope, and
4. flaps in landing configuration for minimum speed.

According to Airbus, the ditching certification criteria also assumed that engine power was available, that the descent rate was 3.5 feet per second (fps), and that the airplane landed longitudinal to any water swells. These criteria are consistent with the test results published in the NACA reports.'



There are similarities in that FR66 and aft of that were most severely affected (that is the point just forward of the rear door) however the VS was not caused to detach in that incident.


This point is interesting:


'As discussed previously, because of the operational difficulty of ditching within the Airbus ditching parameters and the additional difficulties that water swells and/or high winds may cause, it is very likely that, in general, after ditching an A320 airplane without engine power, the "probable structural damage and leakage" will include significant aft fuselage breaching and subsequent water entry into the aft area of the airplane. Therefore, it should be assumed that, after a ditching, water entry will prevent the aft exits and slide/rafts from being available for use during an evacuation.'

Regarding convection and why this flight had problems whilst others didn't, I feel it could have something to do with the fact that it was at a lower level (FL320) than any other flight (FL340-380).

In equatorial oceanic cumulonimbus, updraft strength can be enhanced above FL200 due to heat of sublimation of ice. Therefore it is possible that somewhere above this height and up to near FL320 the updrafts were especially strong but also still containing some supercooled water droplets and hence icing (observations showed -29 C at FL320, but in updrafts it could have been several degrees warmer than that). Flights higher up (-35 C at FL340, -40 C at FL360) were more unlikely to encounter supercooled water, therefore they may have escaped.

So, for 8501, possible airframe/sensor icing, affecting performance as the plane climbed to higher levels, and possible some ice crystal engine shutdown then to finish the job?

I have done a detailed analysis here.

Ber Nooly,

The higher you are the less airspeed due mach limitations. Hence less control surface effectiveness.

4Greens

That's what I mean. Performance is compromised due to the combination of icing and altitude.

Interesting that on a pilots forum that this is a concern... shouldn't the concern be elsewhere?

With all due respect in this case you are wrong.
We are not talking about ejector seats for every passenger. Just agreement by commercial airlines to use what Inmarsat are offering for free - a basic location service. Nor are we talking about full flight data telemetry by the micro-second that some nerds want.

For those speculating intensely over the style of damage to the tail - you might want to see the following photos of the tail recovery from the Java Sea.
It wasn't lifted on board carefully, as it should have been - it was dragged aboard the recovery ship via a low-mounted winch with a horizontal winch cable.
This technique shows the Indonesians are inadequately prepared for wreckage recovery, with a shortage of adequate marine craneage, and they care little about preserving the wreckage in "as-found" condition, to be able to figure out impact forces, angles, speed, etc.
They are obviously relying on the FDR to provide all the info they need as regards the flight path and impact attitude.

http://resources2.news.com.au/images...59e7d08be1.jpg

http://resources0.news.com.au/images...59e7d08be1.jpg

http://resources3.news.com.au/images...59e7d08be1.jpg

Given the number of accidents in Indonesian waters (mainly ferries), you'd think they'd be experts. But presumably they rarely bother to retrieve the wreckage.

I'm surprised the tail wasn't at least contained in some kind of netting to keep the pieces together.

In that last picture the aft exit isn't doubled over as it was in subsequent photos, and it looks like the VS may have been fully attached prior to recovery. I can even see a chord-wise crack forming as it is bent over the ship's stern. I wonder now about the fractured bulkhead as well.

neville_nobody wrote: [11th Jan 2015, 06:41]

JSmithDTV responded:
JSmith, why do you find that concern so unusual or "interesting"? Suppose you viewed a forum in which American police officers were discussing their professional concerns.

I'd wager a primary topic right now is recent demands to have American police officers always wear cameras and microphones while on duty, so that everything the officers say and do is recorded.

Some officers would be expressing concern how this could adversely impact them by invading privacy (e.g, recording comments about fellow employees, issues at home, or even that pretty female dispatcher). It also could lead to management overhearing adverse comments about supervisors or political figures, which might result in some officers being disciplined or fired.

Few, if any, employees are thrilled about having management monitoring every word said, let alone potentially storing it in a permanent (and searchable) database. Airline pilots are no different.

Access to CVRs presently is limited to serious incidents. Pilots understandably prefer to keep it that way. The concern expressed by Neville_Nobody is that if real-time streaming begins, it will be impossible to maintain that level of protection, regardless of what promises are made initially.

That seems like a legitimate concern, and certainly well within the bounds of what is an "appropriate" post on a professional pilot forum. (Speaking as a passenger who has a personal interest in airline safety, yet also recognizes the perils of surrendering all of our rights and privacy in exchange for an illusion of safety.)

Umm, they actually have engineers from Airbus advising. I believe the general plan is to cut up the tail section into several pieces and then ship them to Jakarta.

It can't be beyond technical capabilities to stream both CVR and FDR, but keep the voice channel private, separate and only accessible with legal permissions.

For example, the voice data could be sent encrypted while the technical data is unencrypted, with the voice recording only able to be opened via a decryption code embedded in the FDR data. That way, it would only be accessed under the same circumstances under which CVRs are accessed today. After all, the main information that's needed is the technical data - I doubt many pilots' last words include co-ordinates?