Qatar. Take-off FUBARed. Again.

[HighAndFlighty]

Qatar Airways, whose chief executive Akbar Al Baker has claimed runway overruns that damage aircraft "happen quite often", was forced to abort a take-off of an A350 filled with media guests at high speed in New York on Thursday evening.
'Please switch off the camera': Qatar Airways' dramatic auto-brake aborted takeoff at JFK (http://www.smh.com.au/business/aviation/please-switch-off-the-camera-qatar-airways-dramatic-autobrake-aborted-takeoff-at-jfk-20151211-gllm1d.html#ixzz3tze6A8Fx)

https://youtu.be/MKtPlj932YY

[wanabee777]

MKtPlj932YY

[TopBunk]

Very interesting.

Does the A350 really allow the decision making control to be totally in the hands of the computers for an abort when it computes the runway is too short / acceleration rate is insufficient to achieve the required speed in the remaining length?

Is this a computer decision up to v1, or just to, say, 80 knots, which I could possibly understand, or is it just a synthetic warning that the pilot has to action at all speeds?

I don't like some of the Airbus philosophies that take control away from the pilots, warnings I don't mind.

In this case, it seems to be quite an early abort decision, in that they apparently weren't going that fast by my eye.

It will be interesting to see whether human factors were involved here, as at Miami, in the performance calculations, interpretations thereof, etc.

[Airbus A320-family pilot for 5 years]

[msbbarratt]

No doubt the reason why will emerge before too long. There's all sorts of possible reasons. The A350 is still quite young, could be a gremlin. JFK is quite a busy and highly controlled airport, so presumably it's unlikely that they started off from halfway down the runway.

Nice to see that Airbus's engineers have thought of this kind of thing in the automation, and (SLF) also slightly worrying that they thought it necessary.

It's also slightly vindictive. Suppose a crew have made a configuration mistake. It's better to have that advertised in such an abrupt and public manner by their own aircraft stopping all by itself. But the crew then have to explain it to their passengers and management... The alternative outcome of the mistake is to have their crash or collision damage plastered all over the news headlines.

[hoss183]

Very early. Runway incursion? Technical fault?

[ShotOne]

"Stopping all by itself..." No.

[jack11111]

Throttle up to abort, 18.5 seconds. Maybe past 80 knots.
I like the sound of those engines, though.

[wiggy]

Reading the original article is it just possible that someone has misconstrued the meaning of "RTO auto brake" (or whatever Airbus call their equivalent system)......:)

I've obviously no idea what happened here but FWIW on some Boeings it's certainly possibly to get some distracting low speed RAAS nuisance call outs at JFK in certain circumstances due to the crossing runways.....(e.g When departing 22 Right if you cross 31 Right before you've hit the inhibiting groundspeed ).

Edit to add...having now had a look at the video wannabe has posted it appears to me as if the aircraft stops just short of the very displaced threshold...after the stop you can see piano keys, the numbers ( but I can't read them :8 ), the start of the TDZ markings and the PAPIs in the distance. There also appears to be a taxi way or lights thereof heading off into the 11 o'clock which might be a clue as to the location..............................................:oh:

[fox niner]

Well, at least the approach lights and localizer are still in place. You have got to give them that....

[peekay4]

Portion from ATC audio (extracted from LiveATC.net) after RTO RWY 22R.

http://bit.ly/1jStMo5

(Very unofficial transcript, times approximate UTC):

00:48:29 [JFK Tower] Qatari QR7452 Heavy, what is the reason for the aborted takeoff?
00:48:34 [QR7452] Uhh…
00:48:41 [QR7452] Uh, warning for a take off, that's the cause of (the aborted) takeoff, uh, we just need a couple of minutes to decide if we can, uh, give it another try from RWY 22R for now
00:48:53 [JFK Tower] Qatari QR7452 Heavy, thank you, hold short of Yankee-Alpha, you can take as long as you'd like over there, let me know if you require any assistance from us
00:49:02 [QR7452] That's a, copy thank you and a, we'll let you know, Qatari QR7452 Heavy


QR7452 then spent around one hour at the Yankee-Alpha intersection before eventually departing from RWY 31L.

[Capt Fathom]

The rubbish that gets regurgitated on here just makes me laugh!

[Jet Jockey A4]

Well I don't know whether I should laugh or cry...

One thing is for sure it could not have happened with a better crowd aboard this aircraft.

[Lonewolf_50]

As a neutral observer I see this: an event transpires wherein an aircraft is on takeoff roll, for some reason the take off roll aborts, they contact tower, clear the runway, and about an hour later determine (doubtless with some comms back to company base) that it's now OK to continue with the trip.

So they do.

For the passengers, that's about a one hour delay in the trip.

To put this in perspective:

I have been stuck in an aircraft on a hot summer day in Texas for over four hours, as the crew await the new slot time up in DFW since Wx rolled in up there and all times are being redone. When we get there, Captain waves off (he explained it as interval too close to the passengers) and so we have to get sequenced in for another approach. All told, the delay in actual arrival time as two minutes short of five hours. Needless to say, connections to Boston missed.

With this A350: is this really a problem? Better safe than sorry. Nobody got hurt.

The provided video only raised this point to me: one of the pax is during takeoff roll messing about with what looks like a cell phone. I thought that wasn't correct.

[Noxegon]

one of the pax is during takeoff roll messing about with what looks like a cell phone. I thought that wasn't correct.

Perfectly legal provided it's in flight mode.

[Dani]

Does the A350 really allow the decision making control to be totally in the hands of the computers for an abort when it computes the runway is too short / acceleration rate is insufficient to achieve the required speed in the remaining length?

An Airbus never is making decisions.

It could be a warning from the RAAS, which gives warning if unsufficient runway available or taking of from a wrong intersection.

RAAS is not an Airbus invention.

Airbusses cannot take off with the wrong configuration, because it's tested before take off and newer ECAMs even give warnings when setting wrong configuration while setting them.

Dani

[viking767]

Do we know the speed at which the abort happened?
No requirement to go back to the gate for tire/brakes inspection?

[Airbubba]

Do we know the speed at which the abort happened?
No requirement to go back to the gate for tire/brakes inspection?

It's been a while since I've flown widebody 'buses but I would think that the A350 would have brake temps, brake fans and brake energy charts to mitigate and gauge the effects of the reject if it wasn't at a very high speed/brake energy. Perhaps in a new aircraft design the brake energy and warning zones are calculated for you.

If it was at high speed, as you say, get it fully checked out for CYA. Uh, I mean in the interests of safety, of course.

Most of the delay may have been finding out if whatever fault that caused the reject was deferrable for an ETOPS crossing. I went through a forty minute drill holding on the ramp for a minor deferrable item after pushback a few months ago. All the gurus had to have a Kumbaya session on the radio and maintenance came up with a bunch of codes and employee numbers for us to enter into the log book.

In the old days if you 'knew' it wasn't important, you'd just takeoff and write it up inbound to the next station. Or, so I'm told, of course I've never done that myself. ;)

The QR crew was very wise for not being more specific on the radio in my opinion. There are inevitably feds giving checks listening and a recorder running in the tower and on liveatc.net if you say something that might be second guessed later by the FAA, the company or PPRuNe.

The term 'abort' is recently considered insensitive and 'reject' is now used in the manuals that I have. Now, about that hoarfrost paragraph...

[777newbie]

I don't think the issue here is the RTO. **** happens even with a load of media on board.

The problem is the pathetic response from the CEO.

[ACMS]

Brake fans are an option on the 350.

Our mob in their usual wisdom haven't got them...:D

[Airbubba]

Brake fans are an option on the 350.

Our mob in their usual wisdom haven't got them...

They are great to have in my experience, especially when operating out of the Middle East.

But, the beancounters don't see any benefit and like fuel dump and thrust reversers (e.g. the KC-46A), some operators don't have them on widebodies.

[Piltdown Man]

I really love the "Please switch off the cameras" instruction. The naivety of it. Their thought process has to be "If no nobody records this event then no one will ever find out about it and we will never look stupid and will always be able to deny everything." This is a policy dreamed up by an eleven year old (from the middle ages). You would have thought the cabin crew's efforts would have been better spent thinking about what they might have to do next rather than being wasted to trying to get rid of evidence. This smacks of a company that has a very poor set of values.

PM

[ULMFlyer]

on second thought, maybe after the "Attention Crew at Stations" message, the FA was concerned with having to evacuate (BA at LAS comes to mind) and wanted pax to stop recording to focus on the instructions to come. Plausible?

[CDRW]

Yea - why let a smidgen of plausibility get in the way of a jolly good rant with a well thought out conclusion. Got to like the connection between the attendant asking to stop filming and company set of values.

[masalama]

Rejected takeoff as per the Indian DGCA is a return to gate event to download CVR . Does anybody else have such a rule?

The moment we advance thrust levers towards takeoff thrust and then retard them even if ATC ordered the reject , it's a return to gate and download CVR 😢😢😢

[ACMS]

Nope not for us. Hong Kong CAD.

[Capt Fathom]

Please switch off the cameras
If I thought I was about to be thrust into a emergency situation, I would not want people filming me!
The world we live in has become very litigious!

[zz9]

on second thought, maybe after the "Attention Crew at Stations" message, the FA was concerned with having to evacuate (BA at LAS comes to mind) and wanted pax to stop recording to focus on the instructions to come. Plausible?

I'd be inclined to believe that and give them the benefit of the doubt. When you abort a takeoff that way the flight attendants will clearly be thinking "Is this an evacuation situation?" and be ready to take charge and issue orders. This could easily have been a copy of the BA incident at Vegas a couple of months ago for all she knew that that moment. PR is the last thing on her mind.

As for this incident and whether "computer said no" the last plane I flew the fuel gauge consisted of a clear tube on the side of the tank so this sort of tech is way beyond me. I'll be interested in hearing exactly what happened.

[armchairpilot94116]

Her asking him to stop filming is not uncommon. Even on a normal takeoff much less a possible emergency. She spoke up only when he was filming her.

Many airlines frown on video taking during takeoff and landings. And many cabin crew male or female do not want you filming them . They consider it an invasion of their personal privacy.

Rightfully so.

[G-CPTN]

An aircraft is not a 'public place', therefore the operators are within their (UK Law) right to prohibit photography.

[crewmeal]

Those poor 'terrified' pax.......

Qatar Airways aborts takeoff on first Airbus A350 flight from US | Daily Mail Online (http://www.dailymail.co.uk/travel/travel_news/article-3358102/Qatar-Airways-aborts-takeoff-Airbus-A350-flight-US.html)

[Astra driver]

This looks to me like the crew loaded Runway 22L in the Take-off init page and then departed from 22R.
22L at 8,400ft available (vs 12,070ft on 22R) was still likely long enough so the perf calculations came out OK.

As they began the take off roll down 22R the RAAS system would have began squawking repetitively
"On Runway 22R! On Runway 22R!", Warning the crew they were departing from the "wrong" runway.
The crew likely thought "WTF? Why's it doing that?" And sensibly decided to abort the take off and figure out WTH was going on.

[wiggy]

An aircraft is not a 'public place', therefore the operators are within their (UK Law) right to prohibit photography

That may be the case but I know of at least one operator who has decided that these days trying to enforce any such "ban" is going to cause more trouble than it's worth. It has even told it's crews they must assume there will be cameras running if something " interesting " happens.

Astra

Any idea if the 350 "RAAS" is the same as installed in some Boeings? If (big if) it uses the same or similar logic there should have been a couple of aural prompts before spooling them up.

[Plastic787]

Reading through the comments on that article on the mail website really makes you despair at the level of the general public's understanding of what we do. Utterly clueless comments like the aircraft can take off by itself and others expressing in equal parts horror and comfort that it wasn't the pilot's decision to reject the takeoff! :ugh:

Having said that, with that level of intelligence and understanding surely they could all be in the frame to be the next CEO of Qatar Airways!

[Astra driver]

Quote, "Any idea if the 350 "RAAS" is the same as installed in some Boeings? If (big if) it uses the same or similar logic there should have been a couple of aural prompts before spooling them up."

Wiggy,

I'm not an Airbus bloke, but I believe RAAS is a 3rd party system and therefore should behave similar in all aircraft types.

What I'm guessing happened here is that they inadvertently loaded 22L into the perf/take-off init page of the FMC's. Upon holding short of 22R, RAAS would have stated "holding short runway 22R" as expected. When lining up on the runway the RAAS would again, as expected, have stated "on runway 22R".

If they had departed from the same runway as loaded in the FMC that would have been the last call out they would have received from the RAAS under normal conditions, (The exception being in case of an abort when the system would announce runway distance remaining in 1,000's of ft).

The problem would occur when you start to move down a runway above a certain speed that is not the runway you have loaded into the FMC take-off init page, it then logically assumes you are attempting take-off from the wrong, potentially shorter, runway and announces repetitively, "on runway XX, on runwayXX" to warn the crew of the error.

In this case it would have been a case of "no harm, no foul", since the runway 22R is in fact longer than 22L, but the crew had no way of knowing for sure what the error was and wisely chose to abort.

This is of course just speculation on my part, but I think it's a highly likely scenario.

[peekay4]

An aircraft is not a 'public place', therefore the operators are within their (UK Law) right to prohibit photography
This was a special flight -- being the first ever A350 departure from a US airport -- filled exclusively with invited journalists, influencers and other VIPs. There were no regular passengers on board.

The QR flight attendant asking one of the journalists to "switch off the camera" did not go very well with the group. No doubt the PR damage control is continuing.

[wiggy]

Astra

Looking at our FCOM our RAAS is a lot dumber than the one you describe.

There are lots of warnings in our "book" about it's limitations, one being that at low ground speed it will tell you that you are on or approaching any runway in it's database - hence my comment in an earlier post about occasional nuisance warnings at JFK in certain circumstances. Certainly when departing from 22R, full'ish length, heavy, you can possibly end up approaching the 31R intersection before the RAAS groundspeed inhibit kicks in...so at around 60-70 knots and just as you're hopefully about to call "power set" you can get a "approaching runway 31 Right" RAAS call....... I've had that gem personally on two or three occasions in a 777-300.

There are also dire warnings in our FCOM about the fact that our RAAS certainly cannot do any dynamic performance analysis, though thankfully it will of course call out if you're on a .."short runway" (company defined fixed distance)..and also calls out distance remaining to runway end in the roll out.... it's got various other warnings but compared with what you describe we must have version 3.1...:bored:,

I'd hope the likes of the 350 must have version 10, it would certainly be interesting to know ...............

[Mark in CA]

An aircraft is not a 'public place', therefore the operators are within their (UK Law) right to prohibit photography.

OK, but this was in the US, not UK and this was a plane loaded with journalists! :ugh:

[Astra driver]

Wiggy,

"Certainly when departing from 22R, full'ish length, heavy, you can possibly end up approaching the 31R intersection before the RAAS groundspeed inhibit kicks in...so at around 60-70 knots and just as you're hopefully about to call "power set" you can get a "approaching runway 31 Right" RAAS call....... I've had that gem personally on two or three occasions in a 777-300. "

That sounds like another highly plausible scenario.

I personally really like having RAAS, others not so much because of the nuisance alerts.

I remember one occasion approaching Runway 25 at Twin Falls, Idaho and the RAAS authoritively announced "Approaching runway 26!"
A brief moment of "Uh-oh" in the cockpit followed by re-checking the localizer frequencies and confirming the correct airport on the map display confirmed we were lined up with the correct runway and airport. Found out later that the runway had been recently re-designated 25 due to magnetic correction.

[wiggy]

I remember one occasion approaching Runway 25 at Twin Falls, Idaho and the RAAS authoritively announced "Approaching runway 26!....Found out later that the runway had been recently re-designated 25 due to magnetic correction..

Ouch...good job you didn't have RAAS linked to some form of fancy "Auto-Go Around"...I'm sure somebody somewhere thinks it's a good idea....:ok:

[Hispeedflutr]

In October some FSF guys talked about new tech invented by Airbus: Runway Overrun Protection System. Looked good on video. (See link below):D:D

AFAIK With Honeywell RAAS/Smartrunway it will be only some audio shouting like "4000 ft remaining" and so on..
But with this ROPS the system can and WILL max auto brake by itself without pilots intervention.

If it's true then on that flight the pilots were expecting to takeoff normally and then suddenly ROPS kicked in. Their first thought must be WTF.....

I guess Airbus didn't see it coming:ugh:

https://rudypont.wordpress.com/tag/rops/

[aa5bpilot]

Some of the headlines suggest the aircraft came to a stop from a speed in excess of 100mph. I also saw from the video that the aircraft came to a stop basically at the displaced threshold. This got me wondering: What is a typical accelerate-stop distance for a widebody passenger jet? How does an A350 go for 30 seconds and end up at the *beginning* of a runway??

As it turns out, JFK's 22R has a 3425' displaced threshold. From the video, I counted about 32 seconds from first motion to stop, split about equally between acceleration and deceleration. Assuming textbook physics using those numbers gives an acceleration of 13.3 ft/s^2 (0.4g, which looks reasonable for a jet) and max speed of 125kts. If the actual max speed achieved was only 88kts (ref: https://twitter.com/AirlineFlyer/status/675154919339499521), that allows for some slop and lower acceleration numbers given the apparent runway used.

FWIW, my accelerate-stop distance is 2000'. And my home airport runway is *shorter* than the displaced threshold at JFK's 22R!

[DaveReidUK]

In October some FSF guys talked about new tech invented by Airbus: Runway Overrun Protection System. Looked good on video. (See link below)

No, this wasn't ROPS in action.

As the video in your link states: "ROPS assists the flight crew during the final approach and roll-out". It's not operational during a takeoff, though the A350 FMS features a broadly similar Takeoff Surveillance function, which does what the name says in the tin.

[ExV238]

For the avoidance of doubt and misinformation...

The A350 has no automatic rejected take-off mode, and nor does any other aircraft that I know of. In common with other comparable types, the auto brake is normally armed for take-off and then operates IF the pilot rejects the take-off.

The ROPS function is operational only during landing. If a potential runway overrun is detected (one of the top few aviation hazards in recent years) then max auto brake will be applied if auto brake is already selected by the pilot. If auto brake has not been selected, then the system will issue warnings only.

There's some ill-informed 'automation hysteria' on this thread...

[Dufo]

If I am not mistaken, you get 'xx remaining' if speed decreases more than 7 knots during takeoff rols as the system assumes you are aborting.

[ExV238]

Dufo,

I'm not certain exactly what you're describing, but there's no numerical display of runway remaining during an RTO in the A350 (if that's what you mean?)

[Dufo]

I am not familiar with A350 specific system but this is from RAAS manual:


4.3.7 Distance Remaining – Rejected Take-Off Advisory




RAAS provides distance remaining advisories to provide the flight crew with position awareness during a Rejected Take-Off (RTO), refer to Figure 4-12.




4.3.7.1 Annunciation Criteria




The advisory is generated if:


Aircraft is on a runway, and


Ground speed is greater than 40 knots, and


Aircraft ground speed during the take-off roll decreases by 7 knots from its maximum, and


Aircraft is on the last half of the runway (default) or a specified distance from the runway end.


RAAS will provide distance remaining advisories as detailed in section 4.2.4. When the ground speed decreases below 40 knots, these advisories will terminate.

[Piltdown Man]

Excluding the technical and operational aspects of this event, asking/ordering journalists to stop using cameras, whether allowed or not is a pointless thing to do. The first job of cabin crew is to ensure the safety of the passengers. Their time might have been better spent thinking about how to execute an evacuation. And let's remember, the culture of a company is expressed in the actions of its employees. The world in which we now live is recorded in detail. Nothing we say or do will stop that and it is naive in the extreme to think otherwise. Therefore, to avoid harsh judgement by social media, make sure you have a practical and common sense approach to dealing with the inevitable problems we face doing our job. A good start might be to stop your senior executives saying that serious events like over-runs are a common event. So unless we can uninvent things, let's learn how to deal with the world as it is.

PM

[ExV238]

Dufo,

Ah, OK.

No, not in A350. Airport Nav Function and ROW/ROPS are broadly similar equivalents, but no voice call-outs of runway remaining during RTO.

[Astra driver]

Quote,

"The TORA did not take into account the extra distance of the displaced threshold."

If that was the case, I'm curious as to why this didn't show up during the take-off performance calculations since they would use the same database.

[FlightDetent]

Because it is not the same database. :suspect:

[underfire]

The root cause of the event was a CODING ERROR in the Jeppesen Airport Map Database.
This was a Jeppesen error.
The TORA did not take into account the extra distance of the displaced threshold.

In RNP procedure design, I am constantly dealing with the errors in the different navdatabases, on every cycle. Waypoint locations, runway endpoint locations/elevations, etc.

Now that there is an ac that tracks runways length, I am certain that many, many more errors in these databases will be uncovered. Similar to RNP, the navdatabase information wasn't really used, and errors went unnoticed. I have seen may runway lengths that are complete bull****.

I found it really irritating that the runway endpoint in the database, according to FAA standards, is not measured at the threshold, but at the physical end of the runway pavement, leading to many errors in navigation when looking at TCH and the waypoint. (just like the displaced TCH at PSP, which are significant. Look at the NOTAMS, still do not reflect the DT)

The ac now have the ability, with the new systems to track and automate many functions. This will cause many problems in the short term, just as with the data for RNP, as the data has never been used, checked, nor vetted.

It will be necessary for many agencies to get off whatever they have been riding and do their jobs...

Back to the issue...

I guess my question would be, why does the error show up after roll, not prior to?

[wiggy]

why does the error show up after roll, not prior to?

How does the system know you're going to actually "roll" on the runaway you are aligned with..until you actually roll? (Actually that's just reminded me of an old Isaac Asimov story)....

I'm guessing the designers don't want the system triggering a nuisance warning every time the aircraft is on a "short runway" that's being used deliberately, at low speed, perhaps as part of a taxi route.

FWIW some of the caution/warning features on the RAAS on our Boeing are inhibited until you reach 40kts ground speed for exactly that reason.