This was addressed two posts up by a retired commercial airline pilot
So there is no battery power? I find that very hard to believe. It is certainly not enough to power all flight systems but I’m pretty sure the CVR/FDR will survive for a while.
There IS battery power. But at that point in extremis the important thing is saving the airplane & people, not recording what happened for posterity.
The engineering standards say to cut power to the FDR & CVR, along with most of the computers that provide data to them. To buy additional survival time for the people.
Or rather more accurately, the engineering standards don’t mandate the battery also supply power to all the rest of that stuff that is doing nothing to enhance survival at the time. With the result that the battery lasts longer.
Bottom line:
By the time a modern jet airplane is down to the battery, it’s a one in 100 million bad day event. And is designed for accordingly. Prevention is a much larger part of the deal than is mitigation.
AIUI, some planes have a ram-air turbine that pops out to provide useful amounts of electrical power in the event that both engines are dead. Is the 737-8AS not so equipped?
Air accident investigations are taken very seriously as you know so it is much more than a record for posterity.
In this case the flight was low and the crew shot for the runway but in most situation it takes 2-3 minutes to deploy the APU and restore power to the airplane so it seems illogical to cut off the FDR/CVR to save a tiny amount of power and leave a gap in the flight record. Maybe some sensors can be unavailable but a complete failure after engine loss? Very odd.
Also as mentioned by @MachineElf the ram-air turbine should deploy automatically in modern passenger jets and provide usable power for critical systems, including the flight recorders.
I will try to for definitive sources on this.
my guess is 99.99% of all airliner crashes are prefaced by something “very odd” … so there’s that - seems to fit the theme of them crashing.
Not really. Most accidents are not very odd in my experience, there is a plausible scenario of cascading failures or pilot errors that fit the available data.
That was the point though. Accidents are extremely unusual, and so the things that lead to them must be unusual. More precisely, the combination of things - it’s never one thing - lining up to cause a fatal accident is ALWAYS odd, That’s why it hardly ever happens.
I got the point, it’s just not true. The speculation following an accident from ADSB data, control tower records, flight path, etc, paints a picture for aviation experts that turns out to be roughly accurate the vast majority of the time. Here we don’t have that and have an unexplained (as stated in the news piece) flight recorder failure in flight. This is highly unusual.
I found this:
Short version: controls are hydraulically assisted, but there’s a mechanical connection to the yoke, so the plane can be flown without hydraulics (though it reportedly takes some muscle). So even if both engines crap out, a ram-air turbine isn’t necessary; the expectation is that you can get by with batteries powering the avionics and muscles powering the flight surfaces until you can get the APU started.
It seems likely that birdstrikes took out both engines, and the pilots were lucky enough and skilled enough to deadstick their plane, Sully-style, into a fully survivable belly landing on a smooth, flat runway - and then had the sad and terrible misfortune to have that runway terminate at a plane-shredding wall of concrete.
It still appears there were critical mistakes, or at least things not done because of time pressure. The landing gear could have still have been deployed manually, and the loss of hydraulics didn’t directly force the crew to land the aircraft too far down the runway.
So apparently the B737-800 is a crap design and the black boxes have no alternate power source in case of dual engine failure.
And by the way the FAA has come out with stricter requirements that require alternate power sources to these systems but these requirements postdate this design.
If it’s a Boeing, you’re certainly going (into terrain)
Preliminary report has been released. The link is direct to a PDF which auto downloaded when I clicked it, so don’t do that if you don’t want that to happen.
It doesn’t have much more info than we already knew. The CVR and FDR stopped functioning one minute after the tower warned the crew of bird activity and a Mayday was declared 6 seconds after the recorders failed. Bird remains were found in both engines.
We now have a reasonable scenario for this crash. They initiated the go-around BEFORE the bird strike (this is crucial), retracted the landing gear and started climbing, then hit the birds. At this point they lost both engines and did a decent job flying the plane to land in the other direction, but they came in very fast and were very unfortunate to hit a poorly designed concrete embankement at the end of the runway.
The recorders stopped working because (shockingly) there is no alternate power source for this model in a critical situation.
This is the only thing that makes sense, otherwise they would have continued to land the first approach with the plane properly configured.
Official guidance is kinda ambiguous on continuing an approach after a problem has occurred and simply landing in the next 90 seconds versus taking the 10-20 minutes to follow all the book procedures in detail, then landing. That’s where real pilot judgement and scenario-specific pre-planning versus pettifogging rules-obedience comes in.
Another possibility is they took some generic birdstrikes not endangering the engines & commenced a go-around to do all the post-birdstrike processes. Which is itself not a bad decision. But about then the engine(s) started suffering from yet more bird ingestion and suddenly they were engine-free. Oops.
I’ve addressed this already several times in this thread.
Under total loss of AC power very few things are operating. The absolute bare minimum to give one pilot the info necessary to keep the airplane under control and navigate after a fashion. It does no good to have powered recorders when all the systems feeding data to them are dead. A nice neat recording of flatline “data missing” accomplishes nothing except to shorten the airplane’s battery life at a time when you may really need it to last as long as possible to reach safety.
In this case near an airport battery life doesn’t much matter. But that’s sure not the scenario the systems are designed around. However screwed you are in a battery-only 737, you’re a lot more screwed in a no-battery 737.
And the other point is they didn’t touch down until half way down the runway.
Even with both engines out after the bird strike, could they have flown differently so they would land safely?
Figuring out how to maneuver from wherever you are so you run out of speed and altitude just at the approach end of the runway all properly aligned and configured is a tall order. Landing even 500 feet short is highly likely to be catastrophic at most airports while landing 1000 feet long most likely puts you near, but not beyond the far end, when you stop. As such, aiming to be long, not short is a natural tendency.
In their case they did not realize what Sully did: that the only high probability successful way to execute an unplanned forced landing at a nearby airport is to not do that. The only way to win is not to play.
Had those guys aimed for the large amount of watery bay conveniently located alongside or beyond that airport they’d be alive and probably heroes. Instead they aimed for a teeny spot of pavement and almost did it right and were killed for their efforts. Along with 180 other folks.
For damned sure the airport design with the non-frangible berm did not help. But they hit that thing at such a speed that if it hadn’t been there to hit, they’d instead have hit the next big obstacle. And if it had not been there to hit, they’d have hit the next. etc. Coming off the end of that runway was never going to go well due to what immediately surrounds that airport.
And in that, the Muan airport is hardly unique. Airports should never leak airplanes because damned near everything outside the chain link perimeter fence is just not airplane-compatible.
I know you addressed it, and I addressed you addressing it by pushing back on the idea that this is a non-critical system that should be turned off in the event of transient power loss to save a tiny amount of power. The procedure is to turn on the APU, the FDR/CVR will not endanger the flight, and as far as I can tell every one in aviation agrees with this.
And once the APU is running and the electricity switched on, the CVR/FDR will begin running normally again. And once all the rest of the computers finish booting up the various data traces will become meaningful.
I don’t have a good timeline for how long from AC power loss to impact, but at low altitude it’s ~90 seconds for the APU to go from “switch turned on” to “providing electricity”.
The challenge here is that they apparently never started the APU, or if they did for whatever reason it did not automatically connect electricity to the busses as per normal, nor did they manually make that connection. Or they did start it at some point in their reaction to the crisis but the ground arrived while it was still starting.
In the formal dual engine failure checklist it’s a bit down the list of things to do. In the informal muscle memory response as taught at at least some US carriers, it’s one of those almost immediate memory action items.