From this link Investigator: Co-pilot was flying AirAsia Flight 8501:
Investigator: Co-pilot was flying AirAsia Flight 8501
we have a comment by one of the actual investigators.
Quote:
Ertata Lanang Galih, a senior pilot and investigator, said after
requesting permission to ascend, the plane was veering left and
wobbling, CNN reported. Siswosuwarno said it then ascended to 37,400
feet in about 30 seconds, according to the broadcaster.
The key word getting my attention is the use of the term "wobbling", and
the comment that this occurred before the sudden climb.
What would deserve use of the term "wobbling"? There is one thing that
comes to mind-roll oscillation causing a sinusoidal flight path.
What would cause roll oscillation? Basically a high gain pilot suddenly
dropped into Alternate law with roll direct. One who has never
experienced the change in roll sensitivity and decreased roll damping
that occurs at cruise altitude.
Now I can state with some confidence that 99.9% of the actual pilots
here have never had a roll PIO experience. So virtually none of you have
any comprehension of what it is like. Simply put, it demands your
complete attention. It is a draining experience. Less than five seconds
of roll PIO will get your adrenaline flowing. Thirty seconds is probably
enough to break down the average pilot's scan. It is as if your aircraft
has suddenly grown fangs and a mind of its own. PIOs do not require a
FBW system to happen. It is just that with their many reversion modes,
FBW aircraft are more likely to put a pilot into a part of the flight
regime he has never encountered before. At its core, PIO avoidance is a
training issue, assuming a properly designed flight control system.
The final piece of what could have happened is the basic additive (or
integrating) nature of Alternate Law. When PF's arm gets exhausted from
stick flailing he inadvertently starts moving the stick in an arc and
starts bumping nose up in the corners of roll travel. With the scan
broken and focused solely on roll, the aircraft starts a climb that
rapidly increases in attitude.
With the Captain out of his seat (perhaps performing a troubleshooting
procedure that Maintenance has briefed), we have all the Swiss cheese we
need to have an accident.
When the FDR data is finally published, look for a roll oscillation at
the beginning of the event. If it exists, you have an explanation here.
I use the term PIO because that is what I am comfortable with and one
interpretation of it is Pilot In-the-loop Oscillation. Also know by the
more modern term APC-Aircraft Pilot Coupling. I have had one short
encounter with roll PIO and that was an eye opener