Hi all,

A question from a curious non-pilot; apologies in advance if i may be on the wrong formum! With regard to uncommanded thrust roll back on one engine in the cruise, presumably there is an immediate effect on the yaw/bank angle off the aircraft? Is this compensated for by anything automatically, for example rudder trim or is that not enough? Presumably the pilots would have to do something else to stabilise the situation?

Just curious about the aerodynamic effects during an event of this nature. Thanks in advance. As I said, if i'm in the wrong place, apologies and please redirect me if so.

Thanks, Ian

The aircraft would yaw in the direction of the non-working engine side due to asymmetric thrust, with possible autopilot disconnect. Rudder, and then trim would be required to hold the desired track, and the autothrottle would be disconnected by the pilot, with power applied to the good engine. I believe in a 777, the application of yaw is also automatic (but not in older aircraft, like a 737/757, where it is all manual). If the loss of thrust was for any length of time, and at high altitude, the aircraft would need to descend.

Just to add another element to the hardware/software debate, all structural design and modelling of airframes, bridges, skyscrapers, etc, is now done in software, so there is no getting away from software, no matter which side of the arguement you are coming from. Quite clearly, there is a big difference between realtime software controlling safety critical systems and software used in CAD, etc, but the principles are the same, the outcome is dependent upon the integrity of the design and development process.

I'll refer you to some older posts I made on earlier incidents, because I don't want to go round the houses with it again, but the gist of it is that the software engineering discipline for real-time safety-critical systems cannot be compared to any other type of software engineering in terms of reliability and testing.

http://www.pprune.org/rumours-news/3...ml#post4321101

http://www.pprune.org/rumours-news/3...ml#post4046089

Flight data recorders
 

Hello (please ignore if redundant or off topic),

Would there have been more information available if the pilot had landed as soon as possible versus continuing to Atlanta?

Thanks.

Unlikely.
Present-day FDRs and QARs run for something like 25 hours before they start over-writing previous data. Older CVRs looped in 30 minutes, but more recent ones loop and over-write only after two hours... And in this case the pilots were right there to tell what happened anyway.

Even if they had landed ASAP, the rollback gremlin would probably have quit the aircraft well before....
On BA038 we've been looking at ice... which would have melted by the time the investigators arrived.
This time, we may be looking at ice again, or a transient, or whatever... it came and went.
So let's hope this time the rollback gremlin left a bit more of a DNA footprint.... I'm sure everybody would like to catch him... and wring his neck.

CJ

Christiaan J:

Right you are. Ice, or not, one trusts the eventual science shall present an opportunity for refining design requirements rather than perpetual procedural refinements. The aviation industry always welcomes science to further its capabilities. From the 1920's when the science was that there was "not enough lift in the air" to the capabilities of todays aircraft, science will bring us through. We enjoy the performance of todays professional pilots, aircraft and supporting systems. Under it all though is the requirement to continuously learn and improve. Happy New Year to all.