Following this afternoon's press conference there has been a flurry of developments in the last hour. The first, which is a surprise, is that Airbus has just announced that it is going to study the possibility of realtime datalinking of flight data recorder information to the ground in future.

The press conference is still going on. It's being given head of investigation Alain Bouillard. Not a huge amount to say, but here it is so far.

He says: "As of today we are far from having any real idea of the causes of this accident."

• No pre-existing faults reported by crew or ACARS
• Aircraft did experience inconsistent speed indications
  
• Aircraft "was not destroyed while it was in flight and the aircraft seems to have hit the water on bottom of fuselage with a sharp vertical acceleration" - derived from damage to recovered parts, especially galleys
• The aircraft reached the sea whole - it had not been damaged at that point
  
• Aircraft did not blow up in flight
  
• More than 640 pieces of debris found from all over the aircraft. Notably internal furnishings.
• No distress call of any type
• No inflated lifejackets been recovered
• Black box pingers last for 30 days nominal with up to 10 days possible after that
• Search with submarines and robots to continue to 10 July
• Second phase using different techniques to begin 14 July
• Other aircraft encountered storms and diverted by 10-80nm and all had trouble communicating with Dakar.

More to come

The incident at Johannesburg last month in which a British Airways crew very nicely avoided a potentially catastrophic situation on take-off has resulted in an airworthiness directive on the Boeing 747-400.

The AD finally confirms precisely what happened. Prior to V1 a thrust-reverser amber warning was generated and a few seconds later, after V1, a second warning was received. (It's not specified, but I think the warnings must have been either both outboards or both inboards because of how the logic works.)

The flap control unit (FCU) is designed to to retract the leading edge flaps when the reversers are deployed in order to avoid "impingement of efflux air" from the reversers (ie FOD damage). It worked as advertised, but of course the aircraft was taking off not landing and the warnings were spurious. Result: aircraft rotates (at hot 'n' high Jo'burg remember) with no leading-edge flaps. The AD confirms that the crew experienced pre-stall buffet and stick-shaker activation.

Once the aircraft lifted off, the air/ground logic system re-commanded extension of the L/E flaps after five seconds, and the deployment took 10-15 seconds during which there was more buffet and stick-shaker before it climbed away, and the crew dumped fuel before returning to land for tea and medals.

What the FAA's AD does is to mandate a Boeing service bulletin which has already been issued which "describes procedures for modifying certain thrust-reverser control system wiring to the FCU in the P414 and P415 panels. The modification includes re-routing and re-terminating one wire for each engine and replacing the wire if necessary."

It affects only aircraft with Rolls-Royce engines. Ho hum....

The FAA adds without comment: "In addition, one operator reported 12 single-engine REV indications during take-off over the past three years, leading to seven rejected take-offs. The incident described above was the first known simultaneous two-engine event."

Obviously nobody has any idea how the Yemenia Airbus A310 crashed at Moroni in the Comoros, but pilots familiar with the airport are cheerfuly admitting that it's a challenging place to go. In particular the approach to runway 20 at night, in poor weather, is not great - although obviously it doesn't in itself explain the accident.

Here are a couple of charts, the first showing the visual procedures just to help orient yourself, and the second showing the procedure for the 20. Given the strong south-westerly wind, gusting to 37kt, at the time there's not much doubt that it was in use. You can see the uncomfortable manoeuvre over the dark sea and potential use of land-based lights for visual guidance with all the known potential for disorientation that it brings. May turn out not to have anything to do with it of course.

France's transport minister Dominique Bussereau has said that there is talk of the flight having made one approach followed by a go-around and then the accident on a second approach. But he stressed that the information wasn't confirmed. Local reports talk of wreckage within eight miles of the coast.

Visual chart.

Runway 20 procedure.

Let me say first that I have no doubt that the floor in Lufthansa Technik's new Malta hangar is a truly fabulous example of its kind. And also that I do sympathise with poor Jane Shepherd whose job it is to create press releases on, well, floors. (Inasmuch as I ever sympathise with the PR industry, which is rarely.)

But in all forms of writing there is such a thing as knowing when to stop. Like before you write this sentence for example.

"The new development was officially inaugurated by Maltese Prime Minister Lawrence Gonzi, who walked all over Flowcrete's floor to applaud Lufthansa Technik for its investment towards achieving excellence in high value added manufacturing."

You remember the Air New Zealand body-paint ad based on the "nothing to hide" tagline - as opposed to low-cost carriers who hide extra fees in their booking conditions - well they've, er, extended the concept to the flight-safety announcement. (Once you start writing about risque stuff, everything sounds dodgy.)

Great videos.
More videos in the series, including bloopers, below. Great stuff.

Happily the February loss of a Bond Helicopters EC225 Super Puma in the North Sea was non-fatal, and now it turns out that luckily it was in benign conditions for the survivors. That's because of three serious and unsuspected phenomena that led to the search operation being much more difficult than you'd imagine in the circumstances. The AAIB reports on them today.

The helicopter, you may recall, descended fairly gently into the sea in darkness, visually observed by individuals on the rig to which it was heading. It floated and the 18 occupants quickly got into the liferafts. Between the aircraft, the liferafts and the individuals they had plenty of locator beacons. Nevertheless, it took a 27 minute search to find the survivors.

Here's why.

What if they put out a press release and nobody believed it? Slightly surreally the NTSB issued a press release last night giving brief details of two AF447-like incidents of which they've become aware, but because of how they did it some people are very sensibly questioning if it's real. Quite right too.

The problem is that although the press release was e-mailed via the NTSB's automatic notification system, it wasn't, and still hasn't been, placed on their website. So if you weren't on the e-mail address list then you had to rely on news reports for the information. And the news reports, as is sadly the way these days, carry no additional information or quotes from the NTSB confirming the authenticity. So given the reputation of the media, people are doubting the press release's provenance.

A couple of people are even saying the grammar is so lousy on the press release that it can't be authentic. A lovely thought, but it's hard to think of a public document that would survive that test in 2009. (AAIB report maybe??)

Anyway I've come in to find this situation this morning. I've got the release direct from the NTSB overnight myself in the usual way that I do and I'm happy that it's for real. (Though I confess I don't know that for a fact, so I'm adding to the problem.)

Below is the text, followed by the text of an informal document circulating on the internet via email and websites before the NTSB document emerged, and is purportedly from a pilot on the Northwest flight mentioned by the NTSB. There's no proof of its authenticity of which I'm aware, but please leave a comment if you can help with that.

The TAM incident is interesting in that the crew reported a sudden drop in outside air temperature rather than the real or indicated increase that has been associated with other incidents (such as Air Caraibes). The NWA incident recorded in the supposed pilot report includes an account of a sudden increase too.

The French BEA investigators have just announced that their next press conference will be in Paris at 15:00 local (13:00GMT  09:00EST) for up to two hours next Thursday 2 July. There is an entirely unconfirmed story around that the preliminary report will be issued next Tuesday 30 June.

A company in Lafayette, Louisiana called C&C Technologies has added some important detail to the way the black-box pinger behaves at the end of its battery life, but more importantly it describes the unmanned submarine that it may use to help the search for the FDR/CVR and sunk debris. C&C say they've already been contacted by the investigators to that effect.

The first key fact, which addresses something I've been wondering about, is that once the 30-day specified duration of the pinger is reached, it's expected to die very quickly. It's a feature of the lithium batteries used. That's in a week's time.

The submarine (or autonomous underwater vehicle - AUV) it turns out is a pretty capable beast. It's untethered, has sonar and camera, and an acoustic link to transmit the sonar and video images to the surface in realtime. Crucially, the most advanced version is rated to 4,500m (15,000ft) depth which is what we're potentially talking about in the crash area. (Earlier ones to 3,000m)

Full details and pictures below.