Subscribe by E-mail

Archives

Recent Assets

  • Hendell PC-12.jpg
  • Cropped heli.jpg
  • FIN-HELITRAK-Xa.jpg
  • Goldenrules 1.jpg
  • Golden rules.jpg
  • Flying close to terrain.jpg
  • ICAO ANC assembly.jpg
  • ICAO ANC screen.jpg
  • ICAO.jpg
  • Jamestown.jpg

June 2011 Archives

AF447 and the locked cockpit door

| | Comments (3) | TrackBacks (0)

At the Paris Air Show today I was talking to a person who has reason to be interested in the loss of AF447, and she raised an interesting point: did the fact that the cockpit door is a post-9/11 security door make a material difference to the time it took for the captain to re-enter the flightdeck after the accident sequence had begun?

The BEA (the French accident investigation agency) may be able to determine the answer, and if they can, they will reveal it in the accident report.

Maybe even if the door did NOT delay the captain's re-entry the accident would have happened anyway, but there are a few hints - among the currently incomplete summary of what the pilot flying was doing with the controls - that, shortly after the captain's arrival, the pilot flying momentarily changed his attempted recovery tactics in a way that might have been successful if he had maintained the change.

Whatever we ultimately find out about this from the BEA's investigation, it is a valid time to review whether the cockpit security door is a net safety benefit or disbenefit. In less than three months it will be the tenth anniversary of 9/11. Maybe the aviation industry should use the occasion to review all the changes, on the ground and in the air, that were adopted in the immediate aftermath of that truly shocking event.

Sukhoi Superbounce: how to arrive at Le Bourget

| | Comments (7) | TrackBacks (0)

Careful, watch that sink rate...

Flare. Flaair...!

sukhoi_ heavyland_bigPic.jpg

Oops!...overcooked it...here we go...

sukhoi superjet 170611_1694.jpg

...whoaa!...

sukhoi superjet 170611_1696.jpg

...don't drop the nose - the spoilers are out...! Here we go again...

sukhoi superjet 170611_1697.jpg

...that's better...

sukhoi superjet 170611_1701.jpg

...we've arrived!

Okay that'll do for today. 

Just demonstrating what robust landing gear the Superjet has.

Tomorrow's flying lesson

| | Comments (0) | TrackBacks (0)

Rockwell Collins is going to provide pilots with a button to push if they've lost track of which way up they are. No need to train for upset recovery, then.

The same Pro Line Fusion suite will take care of emergency descent even if you've passed out from hypoxia, and if some Canada Geese or volcanic ash take out your engines it'll take you to the nearest aerodrome.

-----------------------------

Right, laddie, so you want to learn to fly? Okay, strap in. That's it.

Now, you see that button? Push it.

Very goood!!

Right, now for the debrief...

 

 

Lauber on AF447 and autopilot disconnect

| | Comments (3) | TrackBacks (0)

Do you think it's a bad idea that the autopilot/authrottle on AF447 tripped out just because two pitot sensors provided different airspeed values for a short time?

Dr John Lauber, aviation psychologist and engineer, formerly of the US Navy, NASA, and the NTSB explains why it's not.

I advise those who are algebra-averse to persist. Clarity emerges through his language. But, dear reader, you might struggle a bit without some knowledge of basic aerodynamic principles.

(I've copied this in from the Mifnet, of which he and I are both members)

Quote Dr Lauber: "L=cl*A*.5*r*V^2  i.e., Lift=coefficient of lift x (density x velocity^2)/2 x A.  A is the area of the wing, and is fixed, certainly in cruise flight.  r (rho) is air density, and is for all practical purposes fixed when in cruise flight.     The term (density x velocity^2)/2 is called dynamic pressure,"Q" for short.

"Airspeed sensing systems measure Q, and the display is calibrated in knots.  Authothrust and autoflight systems must have an accurate and reliable measure of Q to maintain L at the desired value. 

"If there is a discrepancy between two independent sources of Q, which one is correct?  Because of the inherent stability of modern aircraft, disconnecting auto-systems has no direct, immediate consequences, and the book procedure for handling airspeed discrepancy is to basically maintain thrust and pitch attitude while the issue is resolved. 

"The QRH [quick reference handbook] has tabled values for thrust and pitch attitude settings as a function of weight/altitude--the crew has to look up these values, and then set thrust and pitch accordingly.  In the short term, pilots/systems can manipulate Cl through changes in alpha; but the long-term solution requires Q, i.e., airspeed."

QED

It is a relief to have an independent expert who knows his stuff explain things about this accident which, although not essentially a complex one, seems to have destabilised the thinking of a lot of people.

Dr Lauber independent? If you knew him as I do, you would know he is. But conspiracy theorists should know that he also worked for Airbus North America for some time after he left the NTSB. 

How much ash is too much?

| | Comments (1) | TrackBacks (0)

 

Maybe this much...?

 

emb 190 BRC.jpg

Latin America is more used to volcanic activity than Europe is. Something's always erupting somewhere between Mexico and Tierra del Fuego, usually Popecatepetl in Mexico which seems always to be up to something.

This time it's Puyehue-Cordon Caulle in Chile, and the cloud is drifting east messing up aviation in much of Argentina. But at least you can see where the plume is. Mathematical models not needed right now!

The Daily Mail online has assembled a series of amazing images showing just how dramatic this has been, on the ground and in the air.

Buenos Aires Volcanic Ash Advisory Centre (VAAC) has plenty to report, but I wish they'd provide graphics instead of Lat/Long coordinates you have to plot yourself. 

Understanding crew behaviour in AF447

| | Comments (15) | TrackBacks (0)

If you are puzzled by crew behaviour in Air France 447, perhaps a quick read of this will provide food for thought.

That article's not about Air France 447, it's about the airline training status quo, the kind of training the Air France crew would have received. 

Ultimately, it's about the regulators' failure to modernise training requirements as the aeroplanes themselves have been transformed.

There is a case for saying that aeroplanes are getting better all the time (I would argue that is indisputable), but pilot competency is in reverse.

It's not the pilots' fault, however. And it's not just me being an old grump and saying "things aint what they used to be".

Recurrent training is the same today as it was for Lockheed Constellations in the 1950s. As Dr Kathy Abbott says in my article, pilots are being given fantastic tools and then not taught to understand nor manage them.

Can technology kill the most common airline accident?

| | Comments (2) | TrackBacks (0)

In the last 25 years there have been 1,020 runway overruns by airliners or business jets.

Most overruns were non-fatal, but they were highly damaging, expensive, and shocking to passengers. Many, however, were fatal, and a total of 1,082 people died in them.

Airbus has invented and developed a system - the runway overrun prevention system (ROPS) - that could have the same effect on runway overrun accidents as Honeywell's invention of the EGPWS had on controlled flight into terrain disasters. EGPWS has virtually eliminated CFIT for aircraft fitted with it.

Already more than 60% of the A380 fleet are fitted with ROPS, and it can be retrofitted to the rest of the fleet - all its types - starting next year. It's a software upgrade.

 

Thumbnail image for A380 landing.jpgImagine what an A380 overrun would cost. Is it worth the risk?

This could be a serious product differentiator for Airbus, because it has patented the system, and although Honeywell has a software upgrade for its EGPWS which provides some very useful assistance to pilots in reducing the risk of runway accidents, it does not come close to offering the comprehensive protection ROPS provides.

But Airbus has decided not to keep ROPS to itself, because runway overruns cost the industry as a whole too much in damaged aircraft and damaged reputation. So the manufacturer has just announced that it's offering ROPS commercially to competitors.

Two years ago at Toulouse I took a trip in an A380 and saw ROPS operating live. I was seriously impressed. If you want to see for yourself how ROPS works, Here is the fully illustrated blog I wrote following the trip

The big stall recovery debate

| | Comments (22) | TrackBacks (0)

Air France 447 and the Colgan Air crash at Buffalo have injected new vigour into a debate that has been going on quietly in the industry for some years now.

It's about how to teach airline pilots to handle the aircraft when it's close to the edge of the flight envelope.

The debate about airline training for upset recovery (recovery from unusual attitudes) is a part of this, but it's stall recovery specifically that I want to concentrate on here.

For brevity, I am going to assume that readers know that stall is ultimately about angle of attack and that they understand the aerodynamics of a stall, so I can concentrate on existing perceptions about recovery technique, because technique is what the industry debate is about.

Some of stall recovery's finer points are in dispute. Let me set out what I think the case is, and then please feel free to tear my perceptions to shreds; the objective is to discover the truth, if there is a single truth.

During basic training, most pilots are taught to recover from a stall by dealing with the attitude first, accepting some height loss as inevitable, then applying power. 

However line pilots, who spend most of their time in controlled airspace, have been taught for decades that they should apply power first, then adjust attitude if necessary to minimise height loss.

I suspect that this technique, blessed by the FAA, was was taught because it was assumed pilots would be reacting to a stall warning, rather than waiting for the actual stall. And the controlled airspace factor means that any loss of altitude would mean loss of cleared vertical separation.

Buffalo has caused a lot of heart-searching at the FAA and National Transportation Safety Board, leading to a re-discovery that the "power-then-attitude" stall recovery technique was not the one that the manufacturers' test pilots were required to demonstrate to win certification for the aeroplane type in the first place: test pilots use the classic "attitude-then-power" recovery, accepting height loss as inevitable. So the FAA, seemingly without noticing, had authorised a line training technique different than the one they required for type certification.

Of course if the "power-then-attitude" technique is applied skilfully, with a due appreciation of basics like phase of flight, density altitude and height, it works. But if applied when the aircraft is thoroughly stalled it will delay stall recovery and risks making the situation worse. 

When and why the "power-then-attitude" technique change was accepted is not completely clear, but I have a personal theory. Please feel free to shoot it down:

In the 1950s when transport aeroplanes were powered by big piston engines and props, the power response was instant and gave you lift-producing propwash over the wings, which could assist stall recovery and restore airspeed. Also the thrust line and drag line were almost the same, so there was no pitch-up with power application as there is with today's underwing jet engines. And wings were not supercritical, so the stall was less dramatic.

My theory is that the FAA approved the power-then-attitude system for propliners and then forgot to review it for jets.

I have put this to many worthy senior airline pilots, but have not been able to confirm it. Maybe it's lost in the mists of time.

Finally, has the "power-then-attitude" technique got anything to do with the apparent fact, as demonstrated over the Atlantic and at Buffalo, that pilots both sides of the Atlantic - in big jets and turboprops (Colgan was a Q400) - are at risk of reacting wrongly to stalling?