abelenky

rickair7777 is absolutely correct, and in software engineering and control systems, its referred to as a closed-loop system.

Basically once the plane "decides" to go nose-up to slow down, it also expects to see a decrease in speed. (A complex model, a lot like a flight-simulator, would model the amount of nose-up and effect on airspeed).

If the predicted and measured values are close, within some tolerance, then its fine. But if the scenario happens where the exact opposite happens (more nose-angle results in more airspeed), it is trivial for the computer to recognize that the model and measured data are diverging, and getting further and further apart.

It cannot correct for that, but it can certainly alert human-crew to conflicting data.

The closed loop is basically, "Predict what will happen. Do something, and observe the results. See if the prediction matches the results. Adjust parameters, and repeat the cycle for desired effect".

UAL T38 Phlyer

I've held off until now, but if they are reporting a huge rate of climb:

The Bus uses a "soft-cruise" mode to control airspeed in cruise flight, and has overspeed protection, and stall protection.

"Soft cruise" means for small airspeed disturbances, the airplane will climb or descend (I think the limit was about 100 feet, but this was in 2003) to fix the airspeed, rather than small throttle adjustments. The reasons were two-fold: passenger comfort (they are less likely to notice the small pitch changes than hearing the engine pitch change), and lower likelihood of engine failure, as a very high percentage of them happen during throttle changes. (This is what I was told during training).

There were ops notices/alerts when I flew the Bus that a couple of Buses had encountered the following during Mountain wave, when flown near their max altitude for performance limit, and the range between min speed (buffet limit) is not far from redline (mach limit):

A gust makes the airspeed increase, so in soft cruise, the airplane pitches up slightly. Airspeed stays high, and it pitches a little more, and starts to throttle-back.

Whether due to turbulence, or the g-load of the pull-up, the dynamic buffet-limit (which changes with g-load) is triggered. BOOM! Auto-TOGA thrust.

Uh oh!! Now it goes over redline, and throttles back again (can't remember if it would pitch-up again). Hmmm...going uphill, AND throttling-back...

I've seen it happen a couple of times on a minor scale in Mountain wave, with speed performance-improving and degrading, and obvious climbs or descents that don't register on the altimeter. I want to say the Captain hand-flew the throttles until the wave calmed down, but it was a long time ago.

In the notices we had, the aircraft had unintentionally been flown above the certified altitude due to the unexpected TOGA thrust.

Now, if this happened during a windshear-induced speed increase, while simultaneously hitting a 6000 fpm updraft?

You could end up between the two protections of stall and overspeed. Just a guess, but I'd think the protections could end up out of phase with the occurences, so that it makes it worse instead of better.

cardiomd

Many people say so, yes!

Perhaps you don't get as many compliments, dishing out eye-rolling sarcastic insults when people give you well-reasoned, thought out opinions?

When all of the yawns on the CVR mysteriously disappear, and the "I'm so tired" becomes "I'm ####" due to coziness between the airline and officials, perhaps you might change your mind. Pilot error will be the NTSB verdict.

Police hated dash cams when they first came out. They protect the public and increase safety. Now, most police realize that telling the truth can work in their favor. Sickos can always go to ******.com which has unimaginably worse stuff than a CVR. I don't see many efforts to get the older ones off the air, and nobody who is without an aviation mindset will really care about it after awhile.

What rickair and abelenky said. There is also cross-checking between the dual sensors, would trigger IAS unreliable when one becomes blocked and possibly drop to alternate law. No runaway pitch would result in a modern system. Perhaps confusing A/P disconnect without envelop protection would result, similar to AF447.

I'm surprised nobody has mentioned Pinnacle 3701. Even without the flameout after the climb to stall, if they lost envelope protection in the rapid attempted climb, could ride it back down fully stalled.

brianb

It seems to me that the whole problem would go away if you, gasp, turned off two ADRs and used the back-up speed tape on the PFD. Sure, you would be in alternate law but it's a hell of a lot easier then trying to chase down the normal airspeed indication. No numbers to deal with, just a green band surrounded by red. Keep it in the green and stay out of the hot salsa. Let the Captain fly, turn off ADR 2 and 3, wait till the pitot comes alive and continue on like nothing happened. It works.

Pogey Bait

Just my take. The bus got into some very warm air. The bottom side of the speed tape went from good to bad in a hurry.

ShyGuy

Not necessarily. AF 447 hit 7000fpm up and they were already at FL350, on their way to FL380.

This one was at FL320. We don't know specifics, but if it started with a high speed, TOGA thrust, and nose straight up (somehow also in alternate law) then maybe it could hit 11k fpm if only for a moment. But who knows? Not enough info here at the moment.

Hetman

http://www.atsb.gov.au/media/24550/a...503722_001.pdf

https://m.youtube.com/watch?v=1XNnEzFF5fg



Sequence of events derived from the FDR & CVR is on pages 27 and 28 of the report.

We have a bulletin in our FCOM regarding this event.

JoeyMeatballs

Why hasn't more Info been released? I can't imagine they don't know what happened?

ShyGuy

Indonesian authorities said the preliminary report will be out on the first month anniversary (couple more days from now) but that it won't be released publicly.

JoeyMeatballs

Thanks, hopefully it's leaked within hours.