G-VIIO Las Vegas
 

The NTSB report into the engine fire at Las Vegas in 2015 has now been released. The reasons for the fire are pretty routine and remedies have already been implemented.

However the report does go on to criticize the actions of the Captain during the evacuation, which doesn't make for good reading either for the Captain or BA's training system.

Full report available at www.ntsb.gov

Fly26 Evacuation is a mandatory check-ride item, reviewed to a pass standard by every ATPL captain at least once a year, more likely 4 times. Calling for EVAC with engine(s) running would be a fail on a licence re-validation day.

Understandably, in a real-life event, there are stressors not available in the SIM for training - such as fear for life - and the performance will be degraded. That is a physiological fact: https://en.wikipedia.org/wiki/Yerkes%E2%80%93Dodson_law

At the same time, the training AND checking are in place for the sole reason of preparing us for that real-life event. It's hard to comment beyond what's displayed in the first post already.

It would be good to have a camera view of the engines since my strong impression from different cases of "an engine was kept running when the evacuation commenced" is about the mindset beside actions related to the checklist.

Even if a bang is heard and the aircraft yaws slightly and the fire warning goes on - it doesn't mean "catastrophic fire" automatically so it is fully possible to stop with the idea of taxiing clear of the runway after shutting the engine down and use the fire suppression system. The lack of visibility from the flight deck and therefore inability to visually assess the engine condition could lead to the "taxi mindset". A camera showing the engines would tell a different story and the need for immediate evacuation and therefore immediate shutdown of both engines would be evident already during deceleration after the take-off was aborted.

It is also worth to mention that ATC didn't notify the pilots of fire on the left side of the aircraft. An appropriate conversation could have been:

"Speedbird 2276, stopping."
"Roger, Speedbird 2276, there are smoke and flames on your left side, smoke and flames on your left side, fire services are being dispatched."

Even after stopping with more serious fire, no radio call about this fact were transmitted to the crew, which could have assisted in their assessment of the situation.

I don't think the case is that easy that "blaming the captain" is the right course of action - there were several factors involved including a lack of information about the extent of the fire.

It was a bit different in Manchester 1985 since the tower immediately told the captain about the fire when he radioed "28 Mike, we are abandoning take-off... it looks as though we've got a fire on the number one" - ATC: "Right, there's a lot of fire, they are under way now" - this provided immediate cues about the situation and got them to alert the cabin crew during the roll about "evacuate on the starboard side, please" before the aircraft stopped.

The whole BA2276 story seems to revolve around a mindset that they had an engine failure, a serious one but nothing that would prevent them from taxi clear of the runway after assessing the situation. All videos of the aircraft when it comes to a stop before the doors open is a pretty good indication that the lack of information also meant that the sense of urgency weren't there.

If I hear an engine "go boom" I would like to be able to view it through a camera so I know what it is all about. Especially knowing that immediate evacuation and shutdown will be necessary after stopping (i.e. no "taxi clear of the runway" mindset).

If a situation had not reached the point of evacuating yet(which it seemd to be during the initial stages of the event), wouldn't one perform the memory items for an engine fire first which would hopefully get the fuel shut off fairly quickly and then call for the appropriate checklist(ie engine fire or evacuation checklist depending on which route one has decided to take). It can take a certain time period for one to get to the proper electronic checklist page or a paper one for that matter. Even when going directly to the Evacuation Checklist(back page of the QRH book and therefore very quickly accessed) takes a certain number of long seconds prior to getting the the point where one shuts down the engines as one is playing with outflow valves first.

Being able to see the engine would make a couple of important differences, especially when it comes to the assessment of the severity of the situation and the following execution of the engine fire checklist with the associated closure of the spar valve and activation of the suppression system would be more prompt. It would also be evident that evacuation were necessary immediately after stopping.

It seems like the whole "slowness" were caused by a lack of cues about the gravity of the situation, which slowed everything down.

BA engineer

I must have missed that. All my quick scan caught was possible causes, not specific remedies

From the NTSB report:

Years ago evacuation was a many item memory drill checklist. Over time it morphed into a do-list with few or no memory items.

Some outfits took memory items to an extreme I'm told. A friend who upgraded to captain at Air Canada a couple of decades ago claims that he had to memorize all of the emergency checklists for the DC-9 even if the items weren't boldface or boxed for emphasis.

I've flown with folks who tried to impress me with their mental abilities by doing normal checklists from memory (and then sometimes messing them up :ugh:).

Here's the pertinent BA Evac checklist from the NTSB report:

You tell us, you're the 777 Captain around here.

Well the remedy for the disk failure was increased inspection criteria for the web area of the disk. I would agree it doesn't 'fix' the basic cracking issue but it allows any defects to be caught before catastrophic failure of the disk.

Unless they can attached this to a lifing issue (wear out mode) simply saying look at it some more, may not address causes associated with maintenance or abnormal operation which are not covered in repetitive inspections..

Not my impression. Which accidents/incidents you took this conclusion from ?

One can't help but suspect that if this had been an Emirati or Chinese crew some of the posters on here would be adopting a slightly more critical stance........

I noticed on the Checklist it says shutdown the engines with the fuel cutoff switches & only then the firewall cutoff switches. Do not shut down the engine with the firewall shutoff valves. Why is that.

Having observed the laborious and time consuming delays by some crews in the simulator finding the right page in the QRH for Evacuation. Then reading going through each item step by step before finally reaching the part where the captain says "EVACUATE", I really get concerned that time is running out and the passengers are still strapped into their seats. With no shortage of real life examples of fire breeching the cabin and associated toxic heat and smoke overcoming those trying to get out, I firmly believe the former Recall or Memory Evacuation checklist was safer overall. We know that deadly mistakes were made in some accidents where the Memory items were inadvertently omitted and that was the rationale for Boeing and other manufacturers to conduct a Read and Do checklist policy. BUT as a captive audience member jammed in my window seat and the smoke rolling towards me while on the flight deck someone flipping the pages of a QRH that tell the crew what to do next, I would prefer they take the risk of a missed Memory item but just get me out of there fast.
I am aware that research has shown the Read and Do Evacuation checklist is the safest option, but I still prefer to have my crew whipping through a Memory checklist and giving passengers a better chance of survival. Illogical maybe, but a gut feeling nevertheless.

Some information would help the crew:

1. A report from ATC that flames and smoke are visible when the "stopping" call were made or independent from it (this would help a crew to understand the gravity of the situation where they are unable to see it for themselves).

2. The ability to view the engines from the flight deck (camera surveillance) in order to understand the situation immediately after the bang and engine failure.

The decision to evacuate must be based on information about the conditions outside (and inside) the aircraft and I don't think a fire warning is enough to reach such a conclusion (what happens if said warning goes out because the fire suppression system did its thing and the fire was extinguished?).

Making an automatic connection between a fire warning AND evacuation, i.e. that the latter is the default action simply because of the former doesn't feel like an optimum solution.

I understand the captain and his surprise that ATC didn't tell him about the fire (which is common during other fire events such as American 383) when they reported "stopping". It would be pretty logical to assume some attention from that direction when an aircraft report "stopping" shortly after commencing take-off.

The mindset of this crew seem to be "engine failure with fire warning" rather than "catastrophic engine fire" with the associated need for prompt evacuation after coming to a complete stop. Keeping the other engine running indicate a mindset about vacating the runway and that the issue were more of a technical "glitch" rather than a potential disaster.

It looks like this event were linked to a lack of appropriate information, which slowed the whole decision process followed by a rush to get things done (thus the start of evacuation with the no 2 still running).

A simple "Speedbird 2276, you have fire on your left side, repeat fire on your left side, fire crews are being dispatched" from ATC would have assisted the decision making here. It is actually pretty surprised that no call to that effect were being made (compare it with the American 383 and British Airtours 28M).

JammedStab: I think you are referring to the BOAC B707 which dropped an engine into the Staines reservoir and made a successful landing back at LHR on the then runway 05. Showing my age here! The 707 in those days had separate LP and HP fuel valves, the former on the engineer's panel - and yes, that item was missed but resulted in valve position lights being retrofitted to the aircraft to obviate a recurrence. More modern types combine the two valves functions.

50 years ago ...

the middle part of the SULLY movie spent all its time with the NTSB and others wondering why he didn't just land at any of the nearby airfields...

Maybe they should change sim sessions so you're in there for 2000+ hours then suddenly pull a scenario. Then see how perfect the drill is..

G

I think all videos of this event including the in-cabin ones where the "All passengers, please remain seated" shows a crew that try to assess the situation. They knew they had an engine failure and yes, a fire warning came on after a few seconds but they had no idea due to a lack of information that there were a fuel leak and severe fire between the fuselage and engine 1.

The lack of ATC information also made it "less urgent". My own mindset would be like:

"OK, we have an engine failure, fire warning came on but we did the suppression already and NO INFORMATION from ATC when we stopped either so it's a failure and we will taxi clear of the runway in a few minutes so engine 2 can be kept running."

The lack of an ATC call would certainly give the impression that NO FIRE could be seen. It is reasonable for a crew to expect a call about visible flames if there are any, either from ATC as a reply to their "stopping" message or by other aircraft in the vicinity or a combination. This crew acted pretty much the way it could be expected from a lack of appropriate information, i.e. "keep the engine running for taxi, assess the situation and taxi clear of the runway when things are checked".

It could even have been that way that the crew did decided to taxi clear of the runway before stopping due to their mindset under the circumstances (lack of radio calls about fire could very well produced a mindset of getting clear, especially if a taxiway were in a convenient location as they slowed down).

The focus should be on providing pilots with tools for situational assessment, i.e. the ability to know the environment outside the aircraft and make appropriate decisions based on such knowledge. It is pretty natural to consider "taxi clear" or even "taxi to gate" after an engine failure, both due to the risks of evacuation but also for passenger comfort.

Evacuation can't be the "supposed default action" just because of an engine failure, i.e. that the absolute first thing that should be done is the evacuation checklist immediately after stopping on the runway without any other information than "engine failure, fire warning came on a few seconds later".

A camera located in front of the nose gear (showing the landing gear and engines) and another one on the tail (showing the engines, wings and top of the fuselage) would provide all information needed, right away. The most appropriate course of action would be:

1. Engine failure or fire - throttle to idle, fuel cutoff, ignition off, fire handle pull
2. Assess situation (using the camera system and other sources such as ATC)
3. If serious fire: Evacuation checklist, shut down remaining engines
4. If minor failure: Keep other engine running, taxi clear of the runway when appropriate

If step 1 and 2 happens at the same time (i.e. the crew knows about the fire already while they are braking) - evacuation checklist immediately when the aircraft comes to a complete stop.

I can't blame the BA2276 crew for their attempts to assess the situation rather than just ordering evacuation when they had no idea about the conditions (it is pretty evident that their first clue were the visible smoke rising above the fuselage and it took a while before it happened since the plume was dynamic during braking and the major development started after they had stopped, this is also visible on the in-cabin video). It shows the aircraft accelerating normally, suddenly it shudders heavily but nothing particular is seen first, they stop and after a couple of seconds black smoke rises in front of the wing and the callout "Ladies and gentlemen, please remain seated" can be heard.

The crew couldn't see anything particularly amiss until the smoke become visible and the lack of ATC information certainly did slow them down.

Even the American 383 did show a bit of the mindset issue when the number 1 was running when the doors opened (seems like the pilots also thought about taxiing clear of the runway). It is very audible that the engine is idling when the first passengers went out of the L1 door before it spools down and stops.

It is appropriate to arrange ways for a crew to get the information needed to do a prompt evacuation decision when needed.

There is another video of BA2276 commencing take-off where the engine failure can be seen and it doesn't look dramatic at first; it is possible to see smoke being emitted but the aircraft slows down and stops without something that would indicate the need for immediate evacuation. The fierce fire became visible after stopping when the plume became static and the fuel started to flow onto the runway.

Edit: It is pretty interesting to notice that even when they called "Speedbird 2276, mayday, mayday, request fire services" - there was still no information from the ATC regarding fire despite the smoke now being way more serious. When they made their mayday call a decent exchange of information would have been: "Speedbird 2276, fire services are on their way, there's a lot of smoke on the left side of your aircraft".

If I would see an aircraft on fire like that I would certainly call the crew and report it, especially when it is pretty evident from the lack of evacuation AND one engine still running that they don't really understand the situation.

It is almost painful to hear the lack of fire communication from ATC in this case, it is like crew were "expected" to figure everything out on their own. It is pretty clear that ATC saw what's happened since they started to issue go-around commands and closed the runway but not a single word about the fire being serious.

Since the fire was located between the engine and fuselage, such information is important since looking out of the flight deck windows would produce a "normal" view of the wing, without anything amiss.

Give me prior warning to duck before the handbag goes flying over my head! :E

I would have thought that it is both possible and useful for a single "EVAC" button to perform the required cutoffs and fire extinguishers and therefore eliminate the need for (most of) the Evacuation Checklist (or a significant part thereof). We live in an age of technology. Non?