"At 1640 and about 8,500 ft, the crew noticed the airspeed going up quickly and the speed trend excessively high. The first officer reduced engine power and used touch control steering to temporarily disconnect the autopilot before manually raising the nose to control the speed. The aircraft felt ‘heavy’, requiring the first officer’s two hands on the controls to move from the then -4° pitch angle (aircraft nose-up/down). The first officer expected that the pitch correction would be sufficient to arrest the speed trend.
The captain was unsure if the first officer’s control inputs were sufficient to avoid an overspeed so put one of his hands on the controls and disconnected the autopilot to raise the nose further. The captain believed he indicated his intention to take over control and while the first officer could not recall it being verbalised he was aware of the captain’s actions. The first officer recalled that he took his hands off the controls, releasing touch control steering in the process. Shortly after, concerned about a high nose-up attitude, the first officer put his hands back on the controls. To both crew members, what happened next was unexpected and unclear.
Suddenly, the crew felt high positive g,[3] the controls felt different and spongy, and cockpit warnings activated. The crew then verified that the aircraft was under control at a stable attitude and speed. It was level or in a slight descent at an airspeed of about 230 kt.
One of the cockpit warnings was ‘pitch disconnect’, indicating the left and right elevator control systems had been decoupled. This allowed for independent movement of the elevators via the captain and first officer control columns."
An explanation of the how the damage occurred is inferred, but not clearly explained in the updated ATSB Report IMHO.
From the report, it would appear that one pilot (Capt) was pulling on the pitch control and the other pilot (FO) was pushing and that opposite load caused the disconnect of the two elevators as per the system design for elevator jams. After the elevators disconnect each pilot still had independent control of his respective side (L/R) elevator only (no elevator jam). The sum of the two pilot's pitch inputs at the time of disconnect and conflict over who was in control resulted in the + g overstress. The flight loads from the asymmetric highly deflected (assumption from high +g loads generated) elevator control surfaces (one elevator up and the other down) at near to Vne also caused excessive loads which resulted in structural damage to both elevators and including their attachments to the horizontal tailplane. In turn the asymmetric loads imparted on the horizontal tailplane gave rise to over design limit loading to the vertical stabiliser attachment and surrounding structure resulting in permanent deformation. Some damage was also done to the rudder. So far no reference has been made to whether or not the vertical stabiliser attachment points to the fuselage have also been damaged...but one would assume that the manufacturer's inspection has cleared that area.
The message to aviators from this event so far indicates that a strict handover/takeover of control must always be implemented in SOPs and strictly enforced.
ATSB Report:
Investigation: AO-2014-032 - Flight control event involving an ATR72, VH-FVR, 47 km WSW Sydney Airport, NSW on 20 February 2014