In recent years, the best data from such Yaw x Roll = DIVE
discrepant-rudder INTIAL upset comes from JTSB's investigation of
Air Nippon, September 6, 2011, B737-700, JA16AN;
Yaw x Roll upset (LoC-I),
CRZ 41,000 Ft, approximately 69 NM east of Kushimoto,
Wakayama Prefecture, Japan; ≈ 22:49JST.
Data and images are included in the JTSB "Serious Incident Report"
The apparent step-changes were recorded, documented, by the Data Recorder.
Experienced investigators had seen this before: an artifact of the friction, of the heading stylus inscribing on the metal foil.
An earlier lesson from the AA Flt One investigation, regarding FDR's HDG- trace during initial-upset.
The investigation of sudden aberrations, almost step-changes, in the FDR's recorded Mag-HDG were found on the foil for the AA Flight One initial-upset, 1Mar62 (see CAB's AAR for AA Flt One):
"...absolutely precise time correlation of the four traces was impossible and errors of one to two seconds between them exist. An additional error is introduced by the friction and play in the recorder, resulting in short period (up to two seconds) aberrations of ... heading traces from the smoothly varying changes made by the airplane. Characteristic of these are the numerous steps . . . Although ... headings are usually expressed in terms of exact values throughout the flight recorder discussion, it must be understood that they are approximate but generally accurate to within plus or minus one second...."
[CAB's Report on AA Flt One, pg 10.]
"... Starting at time 1008:07 the heading trace began to record changes the significance of which cannot be precisely defined. From this point in time, therefore, the recorded heading changes will be described but the possible meanings to be ascribed to them, except where the reason for a change is clear, will be discussed under analysis...."
[CAB's Report on AA Flt One, pg 14.]
The human-analyst chose these points, a human's choice. The times of the chosen pairs were not coordinated between the different traces (the points selected on the G-trace differ-in-time from the points selected on the Hdg-trace).
The FDR- raw-data, shows a "buzz" (approximately oscillating +/- .04 in the G-trace, that starts about five seconds before the first activity in the Mag-HDG (the INITIAL upset-yaw was preceded by the "buzz" in the G-trace).
The FDR- raw-data, shows a "buzz" (approximately oscillating +/- .04 differences in the G-trace), that starts at recorder time 23:23, about five seconds before the first activity in the Mag' HDG-trace. The changes in the HDG-samples at 23:30 appear as almost "step" changes (suggesting the FDR HDG-stylus' scribe on the foil is unrealistic or aberrant), appearing as a perhaps stiction-delayed STYLUS jerk-response to HDG-inputs that had occurred over a longer time-interval.
After reviewing several presentations (various graphic-plots of time-history) of the data from the mishap- FDR,
the most useful data are those raw X-Y pairs sampled, and presented in the tables of raw data.
. . .
Failure - Interactions
Discrepant-Rudder, of a mere 2-degrees, excited a roll-rate of 6 degrees/second
AirAsia QZ8501 / 28Dec2014, A320, PK-AXC
Another recent military mishap report documented Discrepant Rudder inputs from a faulty Yaw Damper:
The dynamic-stability of the KC-135R was upset,
after an unrecognized failure-mode,
which excited this "hunting rudder" response.
"Shell-77" / 3May2013, KC-135R, 63-8877,
discrepant-rudder: Dutch Roll, LoC-I, Breakup
Board's Rpt, 64-pages:
Hand-on Rudder Trim Knob
pilot-input of LEFT RUDDER Trim