How Long is Too Long?

First published 05 August 2005.

NOTE: This is really an extended update to my original post, “Good Landing?

Well the data keeps rolling in, and according to this CTV News report, the aircraft “landed long” on Rwy 24L.

So far, investigators have determined that the airliner landed farther down the runway from where a similar jet would normally land.

Levasseur said the plane landed “longer than normally, or longer than usual for this type of aircraft.”

Data and witnesses have suggested the Airbus A340 airliner was nearly halfway down the runway before it touched down Tuesday during a storm, although Levasseur said investigators are still trying to determine precisely how long or how far it landed.

— “Jet’s black box data good quality: investigator”,
CTV News, Fri. Aug. 5 2005 3:26 PM ET

Landed long is aviation-speak for “floated down the runway and missed the touchdown zone”.  Normally, on the approach, you aim to hit the runway “by the numbers”, as they say — and that’s where that phrase originates from, too.  Not actually on the runway numbers, because they are a little too close to the runway threshold, but on the touchdown zone markings placed after the runway numbers.  This usually means crossing the runway threshold at about 50ft AGL and then settling into a nice easy flare for a gentle touchdown.  Click here to see the sorts of markings (like threshold, aiming point, and touchdown zone) that runways usually have.

Now we don’t know at what altitude or speed AF358 crossed the threshold, but something tells me it was faster than her expected VAT (velocity at runway threshold, full flaps and gear) of 136-145kts.  And we know for certain that she did not land anywhere within the normal touchdown zone.  Does missing the touchdown zone mean you have to go around and try again?  Not necessarily — it all depends on how much runway is left in front of you, and what your aircraft’s landing performance is like.  Assuming AF358 did come in at the expected VAT, what kind of rollout distance could she expect?  Well, that’s largely a function of her weight and the runway conditions, as indicated by this chart, based on data lifted from the A340 operations manual:

Figure 1. A340 Landing Distance Chart


(click to view larger image)

We know that Rwy 24L is 9,000 feet (2,743 metres) long, all of it usable as far as landing distance calculations go.  What we don’t know is the landing weight of the aircraft, and exactly how much water was accumulating on the runway surface at the time.  But we do know that the thrust reversers were operational and deployed at the time of the incident.

Levasseur said there’s no evidence that a lightning strike hit the plane before the crash, saying the plane’s wings and wingtips are in good shape.

He also put down earlier reports that one of the engine thrusters used to slow the aircraft may have failed before the crash.

“We have confirmed that all thrust reversers on all four engines were in operation and were working fine,” said Levasseur at a Toronto news conference.

But the plane appeared unable to stop and was still travelling at about 150 kilometres per hour when it left the end of the runway and came to a crashing halt into a ravine.

— “Jet’s black box data good quality: investigator”,
CTV News, Fri. Aug. 5 2005 3:26 PM ET

150 km/h is not all that fast for a landing A340 — it works out to about 80kts, which is a little over half her landing speed (136-145kts).  So AF358 was decelerating somewhat successfully.  Normally, the thrust reversers would be disengaged at 60kts.  Not because they couldn’t help brake further, but because below 60kts there’s a danger that they might blow debris from the runway in front of the engines, and the engines might suck that debris in and damage themselves.  This is known as the dreaded FOD (foreign object damage), and it’s a big worry for aircrews as they are taxiing around the busy airport environment.  Below 60kts, with her thrust reversers disengaged, AF358 would slow down to 30kts or so using her main gear brakes, exiting the runway onto a high-speed taxiway as she did so.  Then she’d start to taxi to the terminal at about 20-25kts.

Since the TSB’s lead investigator has said that all thrust reversers were operational, there are corrections to be made to the landing distance calculation: factors like air density (it’s thinner the higher you go, of course), the presence of headwinds or tailwinds, and of course the reversers themselves.  These factors have known effects on landing distance and can also be calculated fairly accurately.

Figure 2. Landing Distance Corrections Chart

A340_Landing Distance_Corrections

(click to view larger image)

Now according to CTV’s article the aircraft was nearly halfway down the runway when it touched down.  Let’s be generous to the AF358 aircrew and assume that they touched down with 60% of Runway 24L remaining in front of them.  That would give them 5400 feet (1646 metres) to stop, which may be possible depending on their landing weight.  And weight will all boil down to a combination of passengers + cargo + fuel remaining.

It’s possible to calculate AF358’s likely route and from that, extrapolate fuel burn, but the fuel burn is also affected by factors like taxiing time, departure delays, headwinds enroute and things like that.  It will take me quite a while to figure all of that out, so I’m not too eager to get involved in it right away — it is Friday night after all!

What I will say is that, based on the charts, and factoring in wet runway performance, a slight headwind, and operational thrust reversers, continuing a landing at a weight of 230,000kg or over would have been foolish with only 60% of the runway ahead of you.  That’s a pretty high landing weight though, and I doubt that AF358 was actually that heavy.  More than likely other factors came into play, like an unexpected tailwind (from a microburst or windshear perhaps).

Later on tonight, or tomorrow, I’ll work out a likely route and factor in appropriate weather and fuel planning, and have a specific landing weight estimate.  From there it will become apparent whether the pilots were acting with due regard for the aircraft’s limitations and weather conditions.  Although I wouldn’t make the same call under the same circumstances, I think the numbers might just go their way — but I’m prepared to be wrong, too.

UPDATE 080330Z AUGUST 2005: Upon reviewing the operations manual I think it’s safe to say she was a lot lighter than 230,000kg, as she could not even have made it up to her transoceanic cruise altitude (FL350 or FL390) at that weight.  Her landing weight must have been well under 210,000kg, as that is the maximum weight for mach 0.84 cruise at FL350.

According to CTV, TSB lead investigator Real Lavasseur says there was no way AF358 could have stopped in time:

The Air France A340 Airbus that skidded off the runway at Toronto’s Pearson Airport last week had no chance of stopping on time, says the lead investigator for the Transportation Safety Board.

Real Levasseur told reporters Sunday that because of how far down the runway the plane landed, combined with the heavy rain falling at the time, the jet had little chance of stopping.

“Under those conditions, I am pretty convinced that there was no way the aircraft was going to be able to stop before the end,” he told a news conference.

Levasseur says he believes the plane landed about 4,000 feet down the 9,000-foot runway. Under normal runway conditions, 5,000 feet would have been enough for a plane to stop in time. But the rain changed all of that.

“With the runway conditions that we had, the water on the runway, and the braking action, which was poor, my preliminary estimate is that at that point, there was no way that this airplane could have stopped before the end,” Levasseur said.

Levasseur clarified that the brakes were likely working properly, but may have had trouble getting a grip because of the rain. “The brakes themselves are much less efficient on a wet runway than on dry runway,” he noted.

The plane was also travelling slightly faster than normal for a landing, moving at a speed of about 148 knots with a tailwind (275 km/h), instead of the usual 140 knots (260 km/h).

— “Air France jet couldn’t have stopped: TSB”,
CTV News, Sunday, 7 August 2005, 4:45pm ET

As you can see by the landing distance correction chart, a moderate 10kt tailwind can increase the landing distance on a wet runway by as much as 24%.  But since we now know that both her brakes and thrust reversers were fully operational, they would have reduced the landing distance by about 7% (leaving an overall increased landing distance of 17%.

The big question now is what happened in the landing phase — why did they drift so far down the runway, instead of plunking down in the touchdown zone?  Unfortunately I’m not familiar with Air France’s requirements and whether they insist the crew use autoland in severe weather, like some North American carriers.  Using autoland does increase the landing distance slightly, and the operations manual usually insists on a minimum 115% of landing distance when using autoland under CAT II/III conditions (Rwy 24L, though, is not CAT II/III capable).

It would be a little unusual for the autopilot to allow an aircraft to drift that far down the runway without touching down, so I’m inclined to think that AF358 was under manual control during the landing segment.  If that’s the case, Air France is going to be shelling out big bucks to settle some lawsuits.

Category: Aeronautics  Tags:
You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.
2 Responses
  1. Front Office Politics

    First published 15 September 2005. In the aviation world, the cockpit is often referred to as the front office. And just like any ordinary ground-based office, if it is to function well, the people in it need to communicate effectively,

  2. Air France 358 crash was avoidable

    First published 03 October 2005. But lazy, half-assed reporting was not. Sunday’s Toronto Star had a nice three-page story about the September 2nd incident involving Air France 358 (web version here). As citizens of this city are no doubt aware,