AF447

[ArthurBorges]

Interpol's Helping Hand

INTERPOL to help co-ordinate international efforts to identify Air France crash victims

[Graybeard]

TCAS Again

Thanks for the TCAS block diagram on post 2650, PJ2. It confirms that airspeed is not an input to TCAS, hence not a reason for the NAV TCAS Fail on the ACARS list.
------

From Graybeard:

Quote:
"In one instance, (Delta/NW) it was noted the altitude indicated a drop of 300 feet, meaning the ADR was still outputting valid altitude calculation, but perhaps without airspeed trim. "
--------

DJ77 replied:

"Right, an uncorrected altitude has an error magnitude of a few hundred feet. That may be considered unacceptable for TCAS since RAs occur for an anticipated vertical separation of only 700 ft (above 20,000 ft)."
-----------


Altitude on the PFD was still valid during the Delta/NW event, so the altitude to the TCAS would still have been valid. The ADR will not send different altitudes to these two destinations, except below the QNH transition level.

The TCAS Fail message from AF447 remains unexplained, and not in concert with the other reported failures that can all be explained by erroneous airspeed.

GB

[PJ2]

GB;

The TCAS Fail message from AF447 remains unexplained, and not in concert with the other reported failures that can all be explained by erroneous airspeed.

Hm, it may be "unexplained" but I don't think it is inexplicable. I think the TCAS indicated "failed" because of the absence of good data from ADIRS 1. I don't think it's more complex than that.

[Daniel_11000]

Lomapaseo

LOMAPASEO

Agreed. The subject of my post was not on ‘pitot probes’ or something similar.
I was merely explaining that it is not sufficient to divide (catastrophic events) by (fleet flight hours) in order to verify the probability of an event.
What is important in the analysis is to be sure that the probability of an event to develop , for whatever reason, in a catastrophic event is below a certain level.
In our case , is not sufficient to divide one(AF 447) catastrophic event by the fleet FH, but in my opinion we should sum all similar events (the pitot problem that we know that could/can lead to catastrophic events) already happened and sum them in the formula. By this I am not saying that the pitot problem caused that AF447 event. I am merely saying that concurrent pitot problems can lead to catastrophic events , either alone or in conjunction with other more probable failures.

[takata]

Greybeard:
Thanks for the TCAS block diagram on post 2650, PJ2. It confirms that airspeed is not an input to TCAS, hence not a reason for the NAV TCAS Fail on the ACARS list.
[...]
Altitude on the PFD was still valid during the Delta/NW event, so the altitude to the TCAS would still have been valid. The ADR will not send different altitudes to these two destinations, except below the QNH transition level. The TCAS Fail message from AF447 remains unexplained, and not in concert with the other reported failures that can all be explained by erroneous airspeed.

One case of unreliable airspeed is always different from another. Look at the two cases investigated by the NTSB:

NTSB INVESTIGATING TWO RECENT INCIDENTS INVOLVING POSSIBLE A-330 SPEED AND ALTITUDE INDICATION ANOMALIES
The National Transportation Safety Board is investigating two recent incidents in which airspeed and altitude indications in the cockpits of Airbus A-330 aircraft may have malfunctioned.
The first incident occurred May 21, 2009, when TAM Airlines flight 8091 (Brazilian registration PT-MVB) flying from Miami, Florida to Sao Paulo, Brazil, experienced a loss of primary speed and altitude information while in cruise flight. Initial reports indicate that the flight crew noted an abrupt drop in indicated outside air temperature, followed by the loss of the Air Data Reference System and disconnections of the autopilot and autothrust, along with the loss of speed and altitude information. The flight crew used backup instruments and primary data was restored in about 5 minutes. The flight landed at Sao Paulo with no further incident and there were no injuries and damage.

Then, NW incident is not exactly the same as TAM, but the root is common. As well, the same variations (speed or altitude, speed and altitude) are shown in other cases from Air France, Air Tahiti and Air Caraďbes. They are not triggering exactly the same faults depending of the conditions (exposure time is an important factor here) affecting the lines of the pitot/static system.

But the general context is almost identical (as well as the crews reaction) and it is described also fairly well in the Airbus A330/A340 Flight Crew Training Manual:

FCTM A330/A340 Non-normal Operations 8.110.3 Miscellaneous REV 1 (6 JUN 05)

UNRELIABLE AIRSPEED INDICATIONS

Unreliable airspeed indications can result from blocked or frozen lines in the
pitot/static system.

Most failure modes of the airspeed/altitude system are detected by the ADIRS
and lead to the loss of the corresponding cockpit indication(s) and the triggering
of associated ECAM procedures. The fault sensing logic relies on a voting
principle whereby if one source diverges from the average value, it is
automatically rejected and the system continues to operate normally with the
remaining two sources. This principle applies to flight controls and flight
guidance systems.

However, there may be some cases where the airspeed or altitude output is
erroneous without being recognised as such by the ADIRS. In these cases, the
cockpit indications appear normal but are actually false and pilots must rely on
their basic flying skills to identify the faulty source and take the required
corrective action. When only one source provides erroneous data, a simple
crosscheck of the parameters generated by the three ADRs allows the faulty
ADR to be identified. This identification becomes more difficult in extreme
situations when two or all three ADR sources provide erroneous information.

Normally, each PRIM receives speed information from each ADIRU and
compares the three values. Pressure altitude information is not used by the
PRIM. Each FE computer receives both speed and pressure information from
each ADIRU and compares the three values.
In a failure situation, various combinations of ADR faults may occur, each
interpreted differently by the PRIM and FE computers:

· One ADR output is erroneous and the two remaining ADRs are
correct. The PRIMs and the FEs reject the faulty ADR. On basic
A333/A343 aircraft, there is no ECAM alert, however one PFD will display
some incorrect parameters. On the A346 and enhanced A333 aircraft, if
one ADR output is erroneous, and if this ADR is used to display the speed
information on either PFD, a NAV IAS DISCREPANCY caution is
triggered. In all cases, CAT3 DUAL will be displayed as an INOP SYS on
the STATUS page.

· Two ADR outputs are erroneous but different and the remaining ADR
is correct, or if all three ADR outputs are erroneous but different. The
AP and A/THR will disconnect. If the disagreement lasts for more than 10
seconds, the PRIM triggers the NAV ADR DISAGREE ECAM caution.
Flight controls revert to ALTN 2 law. The SPD LIM flag is displayed on
both PFDs, however VLS and VSW are not displayed. This condition is
latched until a PRIM reset is performed on ground without any hydraulic
pressure. However, if the disagreement was transient, the AP and A/THR
can be re-engaged when the NAV ADR DISAGREE message has
disappeared.

· One ADR is correct but the other two ADRs provide the same
erroneous output, or if all three ADRs provide consistent and
erroneous data. The PRIMs and FEs will reject the “good” outlier ADR
and will continue to operate normally using the two consistent but faulty
ADRs.

Any erroneous speed/altitude indication will always be associated with one or
more of the following cues:
· Fluctuations in airspeed indications
· Abnormal correlation of basic flight parameters (IAS, pitch, attitude, thrust,
climb rate); e.g IAS increasing with large nose-up pitch attitude, IAS
decreasing with large nose down pitch attitude, IAS decreasing, with nose
down pitch attitude and aircraft descending
· Abnormal AP/FD/A/THR behavior
· Stall or overspeed warnings
· Reduction in aerodynamic noise, with increasing IAS
· Increase in aerodynamic noise, with decreasing IAS

The ADRs provide a number of outputs to many systems and a blockage of the
pitot and/or static systems may also lead to the following:
· SPD LIM flag on PFD
· Alpha floor activation (because AOA outputs from the sensors are
corrected by speed inputs)
· Wind shear warning (due to Mach input)
· Flap load-relief activation
· Flap auto-retraction from 1+F to 1
· Alpha lock on slats retraction (due to the speed logic part of the alpha lock
function)
· ALTI DISCREPANCY on ECAM
· RUD TRV LIM FAULT ON on ECAM

Always apply the ECAM procedure. If the failure is not annunciated on ECAM,
crosscheck all IAS/ALTITUDE sources (ADR 1, 2, & 3, and ISIS/STBY INST).
Early recognition of erroneous airspeed indications requires some familiarity with
the relationship between attitude, thrust setting and airspeed. If it is positively
confirmed that the outlier ADR is at fault and that the other two ADRs are
correct, select the faulty ADR OFF. This action will generate an ECAM
procedure, which should be applied in order to reconfigure the PFD to display
correct information.

However, in very extreme circumstances, two or all three ADRs may provide
identical but erroneous data. If there is any doubt, then do not instinctively reject
the outlier ADR, although the temptation may exist if the other two ADR outputs
are consistent. In most cases, this decision would be correct, but not in the case
where two speed/altitude indications are consistent but wrong. Apply the initial
actions of the UNRELIABLE SPEED INDICATION QRH procedure from memory
as they quickly provide a safe flight condition in all phases of flight and aircraft
configuration. Rely on the primary flight parameters of pitch attitude and thrust
setting.

Because the displayed information may be erroneous, the flying accuracy cannot
be assumed. Incorrect transponder altitude reporting could cause confusion.
Therefore, declare a MAYDAY to advise ATC and other aircraft of the situation.
Reference to the QRH should only be made when a safe flight path has been
established. The QRH provides pitch attitude and thrust settings for each flight
phase and for different weights.

After applying the QRH procedure, and when the aircraft flight path is stabilised,
attempt to identify the faulty ADR(s). Once the faulty ADR(s) has/have been
positively identified, it/they should be switched OFF. This will trigger the
corresponding ECAM procedure, which should be applied.

Depending of the cause of the failure, the altitude indication may also be
unreliable. However, there are a number of correct indications available to the
crew:

Unreliable Parameter: Altitude

Disregard:
.. Altimeter
.. IAS/TAS
.. Wind
.. V/S
.. FPA

Use:
... GPS altitude (on GPS monitor page)
... RA (low level)

Unreliable Parameter: Speed

Disregard:
.. IAS/TAS
.. Wind

Use:
... GPS GS (on GPS monitor page)

When flying the aircraft with unreliable speed and/or altitude indications, it is
recommended to change only one flying parameter at a time; e.g. speed, altitude
or configuration. Consequently, plan to be at VAPP by the final approach fix.
______________
[addition for people not familiar with airliners procedures: this kind of Non-Normal Operation is to be found in every aircraft manual for such an issue; this is not only related to A330/A340]

[Lemurian]

Hello,
This is a joint effort at describing what the messages tell us.
It's about how these ACARS messages could have been translated into the flight-deck environment and the piloting issues that we could see. I tried to keep as factual as possible except in two cases where there is some uncertainty as to the meaning of a message.
I also refrained from speculating as to the check-lists they could have performed, except the "unreliable airspeed" QRH procedure.

THE COCKPIT ENVIRONMENT

AUTO FLIGHT AP OFF : Red Master Warning light / Cavalry Charge

REAC W/S DET FAULT : Red Flag on PFDs

F/CTL ALTN LAW : Amber Master Caution / Chime

Flags on PFDs : Red “FD” Flags on PFDs

AUTO FLIGHT ATHR OFF : Amber Master Caution / Chime *

NAV TCAS FAULT : Red Flags on PFDs / NDs

Flags on PFDs : Red “SPD LIM” Flags on PFDs

F/CTL RUD TRV LIM FAULT : Amber Master Caution / Chime

Flags on PFDs : Red “FPV” Flags on PFDs

NAV ADR DISAGREE : Amber Master Caution / Chime

ISIS SPD/MACH Flags : Red Flags on ISIS SPD /Mach

IR2 FAULT : Amber Master Caution / Chime
(With extreme caution: It was only a “Maintenance Status at 0211Z,
If an ECAM fault was announced later, it would mean the loss of HDG
And ATT for the F/Os PFD and ND.)

F/CTL PRIM1 FAULT : Amber Master Caution / Chime

F/CTL SEC1 FAULT : Amber Master Caution / Chime


As one can see from this list, there is only one red warning : The Auto-pilot disconnect.
All the rest of the ECAM or PFD/ND messages are at a lower level and present no real urgency.
On the other hand, the annunciations, both aural and visual coming at a very high rate couldn’t have helped the crew to keep or regain any SA they had.

(*) See the A/THR issue below

FLIGHT MANAGEMENT AND PILOTING SITUATION

Task attributions in an abnormal situation are generally well defined in a modern flight-deck: One flies and communicates, the other manages the systems and the check-lists..

There is some uncertainty, though, as to who the “Pilot Flying” was. With the Captain in the LH seat, either one could have been it. During the Captain’s rest, only the most experienced co-pilot could have been PF and only from the RH seat, the second co-pilot being in charge of the systems, radio-com and navigation.(AF SOPs)
It was then up to him to keep an update of the weather at alternates, keep a sharp eye on the PETs between the alternates in an ETOPS environment, help with the radar… It’s also his job to compute at every reporting waypoint engine and speed parameters for turbulence penetration, altitude capability and possible step climb, the fuel situation…etc.,..

-What piloting instruments were available ?

It looks from the messages that they, quite early in the chain of events, lost SPEED and Mach information, both on the PFDs and the ISIS, and the FPV, flight path vector.
They, of course were without Flight Directors and auto thrust. The A/THR disconnect, apart from the amber caution light would have triggered a “Thrust lock” indication on the ECAM, indicating the need to use manual throttle.
Nothing else was lost…….until the possibility after 0211Z of an IR #2 failure, causing the loss of attitude and heading indications on the RH side. A “switching” to the IR #3 would have been needed to recover these indications on the F/O’s side.
The ISIS, also minus speed indication could have been used...but from the RHS ? Not easy and vertigo-inducing.
So, we’re left with the necessity of an “UNRELIABLE AIRSPEED INDICATION” check-list, in itself a non-event in most conditions. In this case, they had the speed, they had the thrust..just match the T/Ls with the present N1 and they would have been very close to the needed parameters for level flying at the same speed.

-Alternate Law handling issues.
With the “ADR DISAGREE” situation the flight control laws would have reverted to “ALTERNATE LAW 2”, with most of the envelope protections unavailable except the load factor in pitch.
-Roll Direct : there is a direct stick-to-surface relationship. All ailerons are available and spoilers 2, 3 and 6 are used for roll augmentation (until 0213Z that is, as the loss (?) of SEC1 would render spoiler #6 on both sides unavailable).But there is enough surfaces to guarantee a minimum of 20°/s roll rate if needed.
By personal experience, alternate law flying is hardly noticeable : the ailerons feel a bit heavier and slower to respond but that’s about it.

[Dysag]

JuggleDan

A draft of a final report will always be reviewed by manufacturer and operator before release. Is it so surprising that they also get sight of a first report?

Furthermore, the Air France CEO is more or less re-stating the BEA's terms of reference: reveal the established facts at this stage and not enter into speculation of cause.

[marchino61]

Furthermore, the Air France CEO is more or less re-stating the BEA's terms of reference: reveal the established facts at this stage and not enter into speculation of cause.

I agree - I think you will find this is standard practice for most (all?) countries' accident investigators, that the initial report is only a statement of the facts known so far.

[gonebutnotforgotten]

Handling in alternate 2?

Re Lemurian's instructive post about the likely cockpit scene. quote (emphasis added):

So, we’re left with the necessity of an “UNRELIABLE AIRSPEED INDICATION” check-list, in itself a non-event in most conditions. In this case, they had the speed, they had the thrust..just match the T/Ls with the present N1 and they would have been very close to the needed parameters for level flying at the same speed.

By personal experience, alternate law flying is hardly noticeable : the ailerons feel a bit heavier and slower to respond but that’s about it.

This is a point which I, as a non airbus ex-pilot, find not only interesting but possibly the key to the question often raised in this thread about why AF447 came to grief while others survived a similar emergency.

As I understand it, in all laws other than direct, the basic pitch law remains C*, which means that the stick demands vertical acceleration, and the aircraft exhibits no natural speed stability. So while alternate may well be a no-brainer in smooth conditions, just how easy is it in turbulence? In a conventional aircraft in severe turbulence one is taught not worry about altitude but to gently control attitude within bounds and leave the trim severely alone. But from the A320 onwards, the side stick idoesn't control attitude directly, and the Horizontal Stabiliser is free to do its own thing (unlike conventional controls where if you don't touch it, it stays put).

The thread on LH landing accident at Hamburg last year revealed widespread lack of understanding of the Airbus control laws in roll (specifically what to in a cross wind, with many contributors swearing that it was totally conventional, it isn't), is there a similar lack of knowledge being shown here, in an admittedly much less common situation?

Even if the conditions in which AF447 found itself were not as awful as some have suggested, I can well see how skills appropriate to conventional aircraft may have been inadequate on that night.

[Lemurian]

gonebutnotforgotten :

and the aircraft exhibits no natural speed stability

Careful here with the vocabulary.
"high - or low - speed stability" refers to a somewhat degraded protection that uses pitch control to prevent an over -or under -speed situation : too fast and a gentle pitch-up is introduced, too slow and a nose-down pitch comes in.
Aerodynamically, the aircraft is just like any other "un-protected" product, and they generally don't fall from the air in droves !

[Meikleour]

Lemurian

I beg to differ with your comment about handling in ALT2 being little different. Have you experienced this in flight rather than in the sim? I have flown an A330 in ALT2 due to a twin ADR disagree situation (one blocked pitot + a different failed ADR) and I was quite surprised at how rapidly the aircraft banked versus the normal roll rate. It also exibited a strong natural tendency to return to wings level with the stick neutral. It is very easy to overbank in ALT2 law.

[Lemurian]

Meikleour,
Your experience is better than mine, then. I bow to it.
I understand that you adapted and didn't overbank, or did you ?
The tendency towards wing level is what I remember, and I put it into the "heavier feel" that I experienced.
A contribution like yours is important as, past the factual treatment of the ACARS messages, the hypothesis on how this cockpit could have looked like (and the resultant conclusion that - on the basis of the ACARS messages - that situation was quite manageable), the airplane still crashed.
So, IMO, we have to string in more information.
I hope that we'll learn some more this afternoon.

[gonebutnotforgotten]

RE Handling in Alternate

Lemurian,

Aerodynamically, the aircraft is just like any other "un-protected" product, and they generally don't fall from the air in droves !

Agreed, I'm not arguing that the A320+ series aircraft are naturally speed unstable , I know they are not, but the question is how do they appear to the pilot who is interacting with them through the side stick via some subtle control law logic? If I have misunderstood how C* works, then perhaps someone could put me right, but as I see it, in normal and alternate laws, the stick commands vertical acceleration, while the speed is controlled by the AT. if AT is disconnected then the pilot has to control speed directly through via the thrust levers, but the aircraft will have no tendency to regain any particular speed once disturbed. If this is not the case, could someone explain how the pitch control laws change when the AT is disconnected? It was noticeable that the encyclopaedic description of control systems posted a few hundred posts ago dealt with the 'hip bone's connected to the thigh bone' aspect of the system, and not with how the confounded thing actually works. In partial answer to your reply to Meikleour, I think 'the airplane still crashed' because it wasn't easy to control it in those circumstances (night , turbulence, no speed indications, warnings coming at you every 3 secs, and no natural speed stability), and understanding of the subtle differences between these FBW aircraft and conventional ones is not good.

[UNCTUOUS]

Facts from the FCTM......

Takata’s citation (post 2686) from the FCTM (Flight Crew Training Manual) would seem to be endorsing the theory put forward by UNCTUOUS at posts 2219 and 2412

Most failure modes of the airspeed/altitude system are detected by the ADIRS
and lead to the loss of the corresponding cockpit indication(s) and the triggering
of associated ECAM procedures. The fault sensing logic relies on a voting
principle whereby if one source diverges from the average value, it is
automatically rejected and the system continues to operate normally with the
remaining two sources. This principle applies to flight controls and flight
guidance systems.

However, there may be some cases where the airspeed or altitude output is
erroneous without being recognised as such by the ADIRS. In these cases, the
cockpit indications appear normal but are actually false and pilots must rely on
their basic flying skills to identify the faulty source and take the required
corrective action. When only one source provides erroneous data, a simple
crosscheck of the parameters generated by the three ADRs allows the faulty
ADR to be identified. This identification becomes more difficult in extreme
situations when two or all three ADR sources provide erroneous information. etc etc ...

From a private email: AF447 FPV & Ps data
Excellent piece of lateral thinking on the Static ports. I have been struggling to interpret the Flight Path Vector flag on the PFD (see ACARS PFR). Whilst the FCOM & FCTM make no reference to static pressure data as an input to the FPV I have in the past come across reference to the FPV being unusable when static pressure data is unreliable.

With detected erroneous static pressure data one expects airspeed/Mach, altitude, vertical speed, FPV, AP/FD and ATC altitude reporting to be affected. Setting the ultimate component aside, and noting the ACARS shows flags for the FD, shows an AP involuntary disengagement, shows PFD FPV flags, along with speed/Mach function for the ISIS and a variety of detailed messages regarding pitot-probes, one is at a loss* to explain the lack of reports for altitude and vertical speed.

Static isn’t just the hash detected by the ears. Obstructed/iced static lines may prove to be the glue that ties in all the AF447 esoteric failure indications. Read Takata’s entire FCTM citation and you get the impression that the safety factor afforded by triple redundancy is more precisely a fallible fallacy foisted upon a gullible coterie of aviating automatons.
.

[Grayengineer]

WX

It is a little disturbing to see experienced captains arguing strongly on opposite views on how to use weather radar. The view that 447 with two FOs flew into a monster is not hard to imagine now. I think that out of this investigation will be the need to make this equipment easier to use and part of sims.

[takata]

UNCTUOUS:
Static isn’t just the hash detected by the ears. Obstructed/iced static lines may prove to be the glue that ties in all the AF447 esoteric failure indications. Read Takata’s entire FCTM citation and you get the impression that the safety factor afforded by triple redundancy is more precisely a fallible fallacy foisted upon a gullible coterie of aviating automatons.

Hi,
I agree with the general point saying that the static pressure was at some point contamined as the TCAS and Cabin controller reacted also, altitude being also unreliable, but we don't know the error magnitude for sure. It could be a function of dynamic pressure correcting static pressure.

As for the pernicious contamination from the ground, I do not share your point because:
1. the system reacted and disconnected the flight envelope protections, showing that it was an abrupt change of air data stream at 0210.
2. the flightpath doesn't show any acceleration of F-GZCP during her whole flight.
3. GPS altitude could be used to control Indicated altitude at any time.
4. Your premice being "loss of control between 0210-0214" is based on nothing acertained until now; this is by far not the only possibility behind this catastrophe. We'll certainly learn more in a few minutes about it.

Then, your "fallible fallacy foisted upon a gullible coterie of aviating automatons" is falling short of your main point. A triple redundant air data system linked to automation or a triple redundant system linked to no automation would not change the problem of icing all the probes beyond specification or beyond design limit (here is pointed the Thales 'AA' probe as underperforming due to drainage system). Beside, triple analogic instrumentation showing three different unreliable value-sets won't help the pilot very much for the same reasons.

S~
Olivier

[blueloo]

Any updates on the FDR/CVR recovery - a couple of false(?) reports of the pingers being detected early on, and unfortunately no hint of the signals being heard since then....

[Graybeard]

Thanks for the replies, PJ2 and Takata.

First:
"Hm, it may be "unexplained" but I don't think it is inexplicable. I think the TCAS indicated "failed" because of the absence of good data from ADIRS 1. I don't think it's more complex than that."

-----------

Each LRU, Line Replaceable Unit, as in the ADIRU, is a conglomeration of separate, often independent, functions, i.e.:
Altitude - independent computation, save for low speed trim.
Airspeed - depends on altitude for computation.
Vertical Speed - depends on altitude/rate
Pitch
Roll
Heading
Ground Speed
Lat/Long
etc.

Each of these functions outputs one or more words on a common Arinc 429 data bus. The receiving device recognizes and sorts the words by their three digit identification labels. Each word has its own SSM, Sign Status Matrix to tell the receiving device its health. The receiving devices may also receive the same words from the other sensors, to use in case of Fail Warn from the primary, or for comparison.

A blocked pitot, for example, will not disable altitude computation, nor IRU functions. This was seen in the Delta/NW incident.

TAM and others loss of airspeed AND altitude indication to the pilots indicates a different situation entirely. It provides evidence of a static source problem, such as frozen static lines, rather than iced pitot, or in addition to iced pitot.

GB

[deSitter]

briefing

Upshot - BEA say it did not break up in flight but entered the water vertically at high speed.

So, explain the perfectly intact tail.

-drl

[takata]

deSitter:
Upshot - BEA say it did not break up in flight but entered the water vertically at high speed.

Absolutely incorrect:
"L'avion est entré au contact de l'eau en ligne de vol avec une forte acceleration verticale"

Meaning:
en ligne de vol = horizontaly (not verticaly)
+ acceleration verticale = stalled

No mention of the global impact force from the conference, I need to read the report. But no surprise from the press headlines, it was too technical to be understand rightly (certainly inacurately translated). I have watched the conference on TV.