PPRuNe Forums - View Single Post - AF356 tailstrike in yyz
View Single Post
Old 23rd Jan 2024, 21:25
  #42 (permalink)  
fdr
 
Join Date: Jun 2001
Location: 3rd Rock, #29B
Posts: 2,956
Received 861 Likes on 257 Posts
Originally Posted by CVividasku
I explained everything but you don't believe me.
They're telling in on the radio that the cause was a long landing. They touched down slightly after the 600m mark. Some airlines won't allow pilots to touch "in the TDZ" which can sometimes extend for more than one kilometer.

Why they pulled so hard is probably explained here :
AF356 tailstrike in yyz

Although the AAIB report doesn't explicitly identify the elevator delay and PIO, as a contributing factor, they are clearly visible on the curves. If we get curves for this incident, I expect to see a similar thing. I don't think they've updated the flight control laws between the two incidents.
?

You have the video, you know which airport it is, there is the Demi god, google. The video show fairly well when they touched down, and when they added thrust, and when they rotated, when they hit their tail and when they finally stopped de-dagging the butt of the bus.

"The radio"... ATC? I've added some points of interest to a screen capture of the location. They were well within the normal TDZ, but that is their choice, up until they have selected reverse thrust at least. There is no magic about doing a GA, the aircraft follows the SSC input intros case, AS IT DID AT LHR, and neither of those events has indication of a PIO, or APIO or anything similar, BEFORE the tail strike. Have another look at the SSC trace for the LHR event that you have linked to, the tail strike occurs before the pilot reverses the SSC pitch command, and that is a high amplitude input as a consequence of the attitude achieving ANU sufficient to achieve the tail strike. During these events, by memory and assumption of Airbus architecture, the pitch is in a direct law, the pilot is not commanding a g load, he/she/it is commanding a deflection of the elevator.

The thrust line is providing a considerable ANU moment, and what is interesting is the sequence of the pilot input; When increasing the pitch up moment from the engines, adding full aft SSC at the same time is going to always result in a wild ride until the gain of the input and the direction of the input is resolved by the pitch moment attained by the engines. For a normal takeoff, the thrust moment is nominally stable (nominal, as the ATR iis commanding an EPR or RPM, which is gross thrust, but net thrust reduces as TAS increases, (the -mdot.V0 component)). In the GA case pulling back while thrust is increasing will hive a total ANU pitch rate that is undesired. stabilising the engine at GA and then putting in the SSC command would avoid the excessive pitch rate that occurs, and there is nothing novel in that, it is what we have done for 75 years with underwing pylon mounted engines.

Having an excessive resultant pitch rate is not a "PIO", it is the wrong applied control deflection given the total moments that are being generated. LHR was not caused by a PIO, nor is YYZ.

There is no time criticality in a GA from the TDZ unless it is related to an obstruction such as could have been the case in HND 34R or LAX 24L. In a GA for a long landing, you have far more runway available than a takeoff case from the same runway, I would think that getting the thrust applied first while limiting ANU pitch up inputs is rational to the point that is what we used to teach; hold the attitude if before TD, and when stable, pitch to a normal TO attitude. No magic read, just not adding limit control inputs needlessly.





G-XWBC [AAIB ] [1]

The co-pilot initiated the go-around, selected Take Off Go-Around (TOGA) on the thrust levers and applied a pitch up demand on his control column, briefly reaching full aft control movement. Engine response from idle to go-around thrust takes several seconds and with the low energy state the aircraft briefly touched down. As it did so the pitch attitude was increasing in response to the co-pilot’s control inputs and reached a maximum of 15° nose up.

This bit is not a PIO, it is a poor technique to apply which arises from either lack of understanding or knowledge of the dynamics of the aircraft. Just because the aircraft has smart systems don't mean that it is smart at all times. Having a big pull on the prong and adding thrust moments thereafter is going to be sporty, when the driver is controlling the elevator deflection proportionally to the SSC, as this and all Bus' do on/near ground. At the time of the tail strike, the driver d'busses has grabbed a handful of back stick, and plonked the thrust levers (switches? toggles? variable resistors) to the loud setting, and the blenders take some time to crank up, and when they do they give an additive ANU pitch moment, and stuff happens. Holding an attitude, and applying thrust, getting thrust set and then pitching thereafter by input to achieve some semblance of a rational target works pretty well, but is regretfully far less exciting. We do not teach GA in the flare, or touchdown sufficiently, these drivers are not the problem, the problem is the assumption that without proficiency in this fairly simple control requirement, when it occurs, the crew react and potentially overreact in a state of anxiety.

From a great bit of reading [2]:

4 . 8 . 3 Pilot-Induced Oscillations
The pilot-induced oscillation (PIO) can be defined as sustained oscillations or instabilities resulting from the pilot being in the control loop. These oscillations would not occur if the pilot had not closed the loop, since with few exceptions the airplane alone is dynamically stable. It follows that control system dynamics as well as airframe and pilot dynamics enter into this phenomenon. In other words, it is the total system that must be considered when evaluating PIO.


[1] AAIB 27939
[2] USNTPS FTM103 FIXED WING STABILITY AND CONTROL
[3] Flight Investigation of Longitudinal Short Period Frequency Requirements and PIO Tendencies, by Dante A. Difranco, Cornell Aeronautical Laboratory, Inc., Buffalo, New York, AFFDL-TR-66-163, June 1967.
[4] Frequency Response Method of Determining Aircraft Longitudinal Short- Period Stability and Control System Characteristics in Flight, by Henry A. Klung, Jr., Captain, USAF, Aerospace Engineer, AFFTC-TR-66-24, August 1966.

Last edited by fdr; 24th Jan 2024 at 00:36.
fdr is offline