“I’m through flying old PoS aircraft!” and “They can’t pay me enough to do this again”
Seriously though. I doubt the crew did something to cause this accident. Probably something technical beyond their control. It’s not like they got bored and wanted to see what happened if they shut both engines of, right? Right…?
“I’m through flying old PoS aircraft!” and “They can’t pay me enough to do this again”
Oh no, I don’t mean to imply that. I’m just saying that often the post-accident crew interviews with the NTSB and the CVR/FDR data don’t always match; generally the crew has a story that puts them in as positive a light as possible, and omits any small errors. When I read an accident report that includes excerpts or a summary of the crew interview, it strikes me how funny it is the way our human memory works. Not that I blame them!
No, the only mistake I think the crew might have made (and we won’t know until the report comes out) is not turning back to the airport sooner after the engine failure. It seems based on the info I’ve seen, that they kept going for a bit, despite the controller offering them a turn back to the field a handful of times. I understand that they need to run checklists pertaining to the in-flight shutdown/securing of the engine, and SE approach/landing, but in a OEI situation with an unknown cause of failure, I suspect it might be best to run those checklists while headed toward the airport, or even holding above it!
Shamelessly copied from wikipedia, but matches my understanding of the flightaware data:
Publicly available flight data show the aircraft had only climbed to around 2,100 feet (640 meters).
The tower controller offered an immediate return for landing, but the crew instead requested “delay vectors” to “run a checklist”. They continued on a southwest heading, away from the airport. At around 1:46 a.m., the crew reported that the second engine had overheated, and they could not maintain altitude.
Not to play MMQB, they were dealt a crappy hand and did a great job surviving a ditching into the pacific at night, that’s incredible. Just something that struck me and that I’ve kept in mind ever since. It got mentioned in our last recurrent training as well; if you have the opportunity, get the airplane on the ground.
I’ll admit that I’ve actually been in the sim, after an engine failure/fire during an approach followed by a single engine missed approach and holding, asking approach for one more turn in the hold so that we could complete our checklists.
I wouldn’t even call that a mistake. Having an engine fail is rare. Having two fail is reeeally rare. At no point in my 22 years in this business have I ever been told to rush the handling of an engine failure. In fact, the “sit on your hands” and analyze has always been the way to go. Identify the problem and act on it. This is so you don’t end up in a dual engine fail situation by shutting down the live engine, for instance.
But sure, there are situations where you need to get down fast, or gravity will do it for you. Problem is knowing when you are in such a situation. But that’s Black Swan theory
So, in short, I doubt the NTSB will have a problem with them taking their time.
But my point is that I don’t think the crew had to paint a prettier picture than what actually happened, to get off the hook. Which is why I believe that the recorders will back them up.
If they hadn’t extended out to sea while running the checklists and instead turned back after the intial engine failure, what are the odds they would have crashed into a populated area and died after the second engine failed?
Where they took off from and would have turned back to, is the Lagoon Runway. Pretty clear of population. Not entirely, but sparse as its mostly over water.
We can speculate about what would’ve been possible if they had _____.
Problem is that they didn’t know the other engine would fail.
- The odds of one engine failing.
- The odds of two engines failing.
- The odds of dead sticking a 737 onto a runway, even if they were overhead.
- The odds of doing a fatal mistake due to stress.
Now, had they lost both engines at once, or known or feared that the other would fail, they would’ve acted differently. Then you are in the options gone, nothing to lose category.
One engine failing still leaves you with a number of options, that you may risk throwing out the window by doing something stupid under stress.
Their reluctance to rush the approach just means they didn’t want to risk creating another problem.
In retrospect there are always decisions that could’ve been better.
Right there and then, without the advantage of hindsight, they reacted to the problems at hand, while dealing with the associated stress as they progressed into a worsening scenario.
That’s the line of thinking I was using, just not as clear or eloquent. Perhaps their decision to extend out to sea played a part in their survival. I would think that dead sticking on a runway has much less margin of error compared to wide-open sea or grass, but my only experience is with frequent dead-sticking of warbirds in DCS
What I’m saying is that in hindsight the best option would’ve been to ditch the procedures and get back onto the runway.
But that’s not how you think when just one engine quits
All good points, and true. I’m not advocating ditching the checklist and rushing through things, that’s how you end up with Midland Flight 92. I’m not saying an engine failure should be reacted to by rushing through the procedures and risking a mistake; I’m a fan of slow hands in the cockpit too.
All that I’m saying is that I suspect the training will evolve as a result of this, as it always does.
I remember after the Colgan crash, the 142 training centers all changed the way they taught stall recoveries. One year we were holding pitch attitude, maximum power until recovery. The next year (and still today) it was reduce pitch until the stick shaker stops (which means a nose down attitude), and then maximum power, with no consideration to altitude loss, which I found troubling. Now we’re back to pitch to the horizon line and max power, but not simultaneously which is still not ideal, IMO. A few other things have changed over the last decade and a half, such as high speed aborts, but that’s one good example.
Any analysis that results in lessons learned from this is by virtue of the luxury of hindsight, as are all lessons learned in aviation safety. You’re right, they didn’t know the other one would go too. But no one ever knows that ahead of time, especially when we don’t know the cause of the first engine’s failure.
You’re also right that as of right now, we teach that an engine failure in a twin jet is almost a non-event, and the emphasis is always on taking your time to assess and react to it properly. Perhaps that pendulum has swung too far in the direction of counting on the luxury of time, and it might be time for the training centers to allow it to swing back into the balance point of “take enough time to do things right, but also prioritize the ultimate goal of getting the aircraft back on the ground safely in a timely manner”. Or something more eloquently put than that.
The accidents and incidents that scare me the most are ones where fuel contamination causes an engine failure; such as the medevac Citations that got DEF instead of Prist and both had dual engine failures. Both engines have been to the same places, got the same fuel, and if one rolls back, there’s a pretty good chance the other one will too. What great airmanship!
Regarding stall recovery, even in training I am pitching to the break the stall and the power is applied simultaneously. I figure it takes awhile for the engines to spool up, so I’m not waiting! I have never had an instructor try to correct me on that… I have recurrent the week after next, so we will see what happens this time.
Same here! It was an emphasis item for a couple years at CAE, to break the stall then add power. Seemed stupid to me, losing a thousand feet of altitude for no reason when if you do both at once (and don’t drop the nose that much!) you would lose virtually zero.
There is no way I would question the steps the crew took based on what little I know. When an engine fails we pilots see large bodies of water as wonderful expanses of refuge because they harbor no big rocks or buildings. I’d be a millionaire if I had a nickel for every time I said or heard something like, “If we lose one we’ll make a slight left to stay over the sea…”
Oh sure! As an industry we’ve been extremely successful in learning from our accidents. To the point where accidents are so few, that we can’t draw meaningful conclusions from them. This is where I think it gets interesting as the Safety 2 or Resilience Engineering concept comes into play and we try to learn from all those instances that had a happy outcome and those everyday flights where all went well.
There are rarely just one way to overcome a problem, so why are we teaching pilots a single way of dealing with them? Your stall recovery is a great example. We’re taught not to lose altitude because we assume that every stall is close to the ground. But aircraft fly just as good at 10’ as in 1000’, so why not use whatever altitude we‘ve got?
Ok, maybe trading energy and losing altitude is ok after all? Well, both are correct as it depends on the situation. Just as an engine failure.
Taking off in high surrounding terrain and losing an engine is one thing. Taking off over the sea, is another. Ref. @smokinhole’s post, I too do the more relaxed over water brief. It makes sense. But if there are low clouds filled with rocks, you need to climb. Two very different scenarios. When I started flying we only trained for engines to fail at or just above V1. At least before you’ve cleaned up. That’s the most critical phase, so it makes sense preparing for that. But that’s not when all engines fail. Got to train for other events too…
Anyway, it is always interesting to discuss accidents and incidents. And we must look at every one of them in hindsight, to learn what we can. The NTSB sure does too. But crew mistakes that are only identifiable with the use of hindsight and retrospective knowledge, are not mistakes that we can use to judge the crew by. Still, as you say, the NTSB may very well recommend considerations to the handling of engine failures, as a result of the investigation. But that’s a different matter.
On approach to HNL 4R yesterday we could see the recovery vessel working over the wreck.
That’s not really my point, and is a very good observation. I don’t care about judging the crew; that’s playing MMQB with the benefit of hindsight they didn’t have, and isn’t fair. If I came across as wishing to condemn the crew, that’s not what I intended at all.
That’s much more where my interest lies. I think all pilots study accidents in an effort to learn and be proactive, to prevent the same thing from happening to us. Our industry as a whole is reactive, especially the regulators and the training organizations. I think we’ll see a shift as a result of this, time will tell.
That doesn’t mean the crew did anything differently than they were trained, or ‘wrong’ necessarily, just that the training can evolve. My mindset about an engine failure due to unknown cause has certainly evolved because of this, and I think that’s a good thing. I won’t ask for another turn in the hold or a delay vector in the sim ever again.
No, I got that as the discussion progressed. No worries!
Well, I will if I feel I need it…
This will always be a decision one must weigh against the options.
I totally get what you’re saying about unknown engine failures. An engine just flaming out, without an indication as to why, one should bear in mind that this can happen to the other engine as well. But if you had to shut one down because of oilpressure, high temp or something else that indicates that something is affecting that engine only, then that’s a different scenario.
And then there’s a difference between stressing and working fast. There will be a certain level of stress, in any such situation, and one should be mindful of that and take the time needed to make sure all the steps are covered. But at the same time use the time efficiently.
Yup, well said.
Easier said than done, though.
That’s why training and practice is necessary.