Has anyone already considered that the initial upset could've been resultant of malfunctioning PCU's due deicing fluid?
I'm particularly referring to this in light of the recently published Norwegian PRELIMINARY REPORT ON SERIOUS AIRCRAFT INCIDENT report into LN-DYM dated 19-11-2013
Prior to the flight in question the aircraft had been parked outdoors for three days at low temperatures. A snow layer of approximately 25 cm had accumulated on the fuselage, and in order to remove the snow, the aircraft was de-iced using type I 30-60 % glycol de-icing fluid. A total of 1,807 liters of warm water and 1,136 liters of type I glycol was used during de-icing. The company that carried out the de-icing has informed that their personnel was experienced and adhered to standard procedures for de-icing. The temperature at the airport during de-icing was -17 °C.
LN-DYM had a normal take-off, climb, en-route flight and initial descent. NAX5630 was established on a straight-in ILS approach to runway 34 in Kittilä with configuration: Flaps 5, gear up, autopilot channel A in use, Auto Thrust engaged and established on localizer. When the aircraft was about to enter the glide path at an altitude of approximately 3,250 ft, the elevator/stabilizer trim started actuating for approximately 12 seconds. The trim moved towards Nose Up. As a consequence of the elevation of the nose of the aircraft and the ensuing loss of airspeed, the aircraft Auto Thrust System initiated full engine power. The high engine thrust contributed to a further elevation of the aircraft’s nose followed by rapidly decreasing airspeed. When the aircraft nose position rose above +20° both pilots started pushing with full force on the control column. The Flight Data Recorder (FDR) shows that the pilots used a combined force of 207 lbs in their attempt to take the aircraft out of the ascending path. The aircraft was at this time in clouds. The nose position eventually reached +38.5° and the airspeed dropped to 118 kt (TAS) before the nose position slowly started to decrease again. While control was being regained, the aircraft’s Stick Shaker and Stall Warning actuated for four seconds. Calculations afterwards show that the stall at 1G for the given configuration is 121 kt. The reason why the aircraft still didn’t stall was that the load on the wings was somewhat less than 1G. During the first phase, while the aircraft’s nose went up unintentionally, no attempt was made to disengage the aircraft’s autopilot, Auto Thrust System or to actuate the stabilizer trim towards Nose Down manually. One or more of these measures would have improved the situation.
- As a part of the examination of whether de-icing fluid can ingress into the Tail Cone Compartment in the direction of the total of four input arms of the PCUs on LN-DYM, the AIBN performed extensive simulated de-icing from different angles on the aircraft’s horizontal stabilizer and elevator. The examination showed that at times even quite considerable amount of fluid ingress into the compartment. Under certain circumstances it is possible that the input arms may be exposed to fluid which in turn freezes solid and blocks the PCU input arms. Aircraft manufacturer Boeing was not aware that significant amounts of fluid could ingress into the compartment in question before the AIBN’s examination revealed this.
- In view of the above, the AIBN has further conducted similar tests on another B737-800 Next Generation (NG) and a B737-300 Classic. The examinations showed that there had been ingress in the Tail Cone Compartments of all the aircraft and that this therefore is an issue concerning not only LN-DYM, but any B737.
- In a cold chamber test rig, aircraft manufacturer Boeing applied de-icing fluid onto a Power Control Unit and was able to simulate a comparable blocking of a PCU through de-icing fluid freezing solid on the input arms, thereby preventing them from having the freedom of movement that is necessary.
- Furthermore, the AIBN has conducted additional tests to determine the significance of different horizontal stabilizer trim positions for fluid ingress. The current procedure (published by Boeing and adopted by the airline) prescribes that the trim should be set in full forward position during de-icing. The AIBN’s examination showed that by changing the trim position to the middle position (the one used during take-off), fluid ingress was reduced.
- In view of the AIBN’s examinations, Boeing in October 2013 altered procedures in the Boeing 737 Aircraft Maintenance Manual (AMM) so that application of de-icing fluid should be carried out at an angle from the front and not from the side. Boeing also introduced new procedures into the Flight Crew Operations Manual (FCOM) prescribing that all B737 operators during de-icing should set the stabilizer trim to take-off position.