bearfoil, your questions in black font, my thoughts in blue:
At "unreliable airspeed" and a/p disconnect, was the a/c experiencing attitudes/control inputs that had exceeded the autopilots programmed range? Or did the a/p decouple simply because of uA/S?
The autopilot will disconnect if the pitch attitude exceeds 25deg NU, 13degND or 45deg Roll. I believe the AF330 in question was equipped with EGPWS which in some installations will have a "Bank Angle" warning. I would expect at some point that an ECAM message to be generated but we dont' have enough information to make this assumption. For further, the BEA 2nd Interim Report,
"1.16.3 Study of losses of or temporary anomalies in indicated speeds occurring in cruise on Airbus A330 / A340", can be referenced. The autopilot and autothrust disconnected in ten of the thirteen recorded cases of this series of events.
Other a/c had encountered unreliable a/s, without turbulences forte, could 447 have been in still air? How quickly did the a/c revert to Alternate Law II, then Direct?
Assumptions made that the a/c was in turbulence would come from the many weather and satellite images available and the most-likely track information provided by many very skilled posters.
Such conclusions may even be reasonable but there is nothing to directly indicate or necessarily conclude that the a/c was in turbulent air or convective weather.
None of us (I assume) has actually examined the wreckage first-hand and with sufficient engineering knowledge come to conclusions which are substantially at odds with the BEA statements regarding the likely trajectory and attitude of the aircraft.
This is not stating that the BEA conclusions are necessarily what happened but a lot of people who kick tin and who can "read" metal have not put forward theories explaining the evidence from the wreckage which supports contrary theories. Many I'm sure would be very interested in serious theories based upon alternate interpretations of the traces in the metal and the cabin material which support for example, mid-air loss of the VS.
One possibility we might anticipate when the recorders are found and read is the loss of data after the loss of control. The recorders are powered off the AC1, AC2 and DC2 busses. I have long theorized that once LOC occurred and the descent established, due to extreme angles of airflow across the engine inlets upsetting the flow through the engines that the possibility exists that they may have flamed out at . Loss of the AC1 and AC2 busses would deploy the RAT for hydraulics and the AC and DC Essential busses, (again, no recorders powered) but again very high angles of airflow relative to the longitudinal axis/chord of the aircraft would preclude its use; if the RAT is the only source of power, load-shedding begins if the CAS is < 260kts. Whether the APU would start in such circumstances, (using APU-dedicated battery) and whether that occurred is also cannot be known.
Could the pilots have been so quickly left with an a/c needing manual control only that they overcontrolled into upset?
Highly speculative when involving human and environmental factors but in what may have been more benign circumstances we have thirteen other examples of loss of ADR data due loss of pitot information in which no loss of control occurred. We cant' conclude that no loss of control would occur as a result of a sudden requirement for manual flight. I can't see experiened pilots "over-controlling" the airplane into a loss-of-control merely from loss of airspeed information, (fly pitch, bring the thrust levers back out of the CLB detent and set the N1 that was being used before the problem, get the checklist out), but that said, we have also seen a number of industry events in which wild excursions occurred as a result of untoward pilot input so we know it can occur.
RTLU. Is travel limited by mechanical (dynamic) stops? Or is it restricted by pressure/sensing of hydraulics?
Rudder travel is limited in two ways both of which are mechanical: the PTLU and RTLU are similar in structure. According to a schematic, they are co-located in the VS, the RTLU about 1/2 the way up the VS, the PTLU at the base of the VS where the rudder and VS structure join. From the BEA 2nd Interim Report:
"1.12.3.5.5 Examination of the Rudder Travel Limiter Unit (RTLU)
The RTLU was found in its place in the fin and disassembled. An examination
was performed at the manufacturer’s and showed that it would allow travel
of the rudder measured as 7.9° +/- 0.1°. As an example, at FL350, this travel is
obtained for Mach 0.8 +/- 0.004, corresponding to a CAS of 272 +/- 2 kt."
At 4/8total sweep, (degrees) and mechanical stops, without hydraulics diminished, can the Rudder experience rapid reversals and subject the a/c to cycling divergent Yaw?
I'm not qualified to say it isn't possible from a design or engineering pov. I have done in the past a survey but not an exhaustive study of Advsories and Directives on other matters and there was not a reference to such an issue or event. To my awareness there is no known occurrence of uncommanded rudder reversal or rapid reversal movements recorded or discussed in any Airbus literature/incident/safety reporting system I have seen or have access to and have studied nor have I experienced it in any aircraft.
The BEA Report indicates that the RTLU permitted a rudder travel of 7.9deg.
ACARS cadence. Could the a/c have remained under control through all ACARS, and at altitude?
There are important understandings about the ACARS messages one of which I stated very early in the first AF447 thread which was eventually closed, and which was clearly stated in the 1rst Interim Report. That comment was, "It is therefore possible in a CFR to find an ECAM message preceding a fault message that is nevertheless timed one minute before it." So the assumption that the ACARS messages are in some meaningful order from which possible cause may be derived, is incorrect. Rather, the need for examining what caused within each affected system a ECAM Class 2 message to trigger as well as understanding what the protocols of that system to send the notice-of-failure to the CMC and for the CMC to prioritize the message and send it thence to the ACARS, must be appreciated.
I do not see direct evidence in any of the ACARS messages that the a/c was out of control. The "Advisory Cabin Vertical Speed" has been cited as one indication of loss of pressurization but the conditions are a rate < or > 1800fpm as compared
Cabin. Could the Cabin Pressure FAIL transmission have been an auto alert that pressurization was lost, at altitude?
As described in the previous response, although it is possible we cannot conclude this. There are a number of interpretations from the FCOM; the AMM does not do a good job of covering Advisory messages. Essentially, the FCOM actions required for this message are to switch CPC's...Cabin Pressure Controllers. Control is switched to MANual then back to AUTO which changes the CPC's. Although one would believe the 1800fpm rates to be those actually sensed, I can imagine at least one interpretation where that is the rate commanded by a CPC and a switch to the other CPC is required. Digging this deeply in the corners for any hint that the cabin itself was structurally compromised is at best a tenuous endeavour.
I hope this is somewhat helpful. I think re-reading the BEA first and second interim reports is well worth while.
PJ2