Loganair SF34 near Aberdeen on Oct 3rd 2014, approach to stall in icing conditions

Last Update: September 10, 2015 / 17:08:06 GMT/Zulu time

Bookmark this article

Photo of Loganair G-LGNM, SAAB 340 — Loganair G-LGNM, SAAB 340 (Photo credit: Bob the courier / Flickr / License: CC by-nd)

Incident Facts

Date of incident
Oct 3, 2014

Classification
Report

Airline
Loganair

Flight number
BE-6772

Departure
Aberdeen, United Kingdom

Destination
Shetland Islands Sumburgh, United Kingdom

Aircraft Registration
G-LGNM

Aircraft Type
SAAB 340

ICAO Type Designator
SF34

A Loganair Saab 340B on behalf of FlyBe, registration G-LGNM performing flight BE-6772 from Aberdeen,SC to Shetland Islands Sumburgh,SC (UK) with 25 passengers and 3 crew, was climbing out of Aberdeen cleared to climb to FL130. Prior to departure weather forecast had predicted a strong southsouthwesterly airstream with severe mountain waves between FL040 and FL280. Climbing through FL065 the aircraft entered clouds, the crew activated the aircraft's engine anti ice systems as well as the wing and stabiliser de-icing boots. Climbing through FL110 the vertical rate slowed, a small amount of ice was seen on the wind screen wipers, the captain assessed that the reduced climb rate was due to the mountain wave activity. Ice was seen forming on the propeller spinners, the propeller de-icing was activated. The captain, pilot flying, engaged the autopilot in vertical speed mode and reduced the airspeed from 160 (minimum climb speed in icing conditions) to 145 KIAS (icing exit speed) and continued to climb the aircraft believing the aircraft would soon climb out of the tops of clouds and thus out of icing conditions. The aircraft levelled off at FL130, about three quarters of an inch of ice were seen on the wind screen wiper at that point, the boots seemed to clear the ice from the wings however. The aircraft accelerated to 164 KIAS rather than more than 180 KIAS as expected. Having seen such a maneouver by other collegues the captain decided to climb the aircraft another 100 feet and return to FL130 hoping this would permit the aircraft to accelerate to normal speeds during the descend, however, during the climb the airspeed rapidly dropped to 150 KIAS and remained there during the descent. The crew concluded they were in icing conditions and needed to descend to FL110. ATC approved the descent and the crew initiated a descent at 1000 fpm. The captain commented that icing was worse than moderate, the airframe was picking up a lot of ice. Despite the 1000 fpm descent the pitch attitude remained unusually high at 5 degrees nose up, the rate of descent was increased to 2500 fpm. Vibrations occurred, the first officer described them like a car driving over a cattle grid, the captain commented: "feel that, that’s a stall ... i think ... icing stall". About 10 seconds after the onset of vibrations a stall warning sounded for a second, the stick shaker operated and the autopilot disconnected. The captain assumed manual control of the aircraft, pitched the aircraft down to 2 degrees nose down, more force than normal was needed to pitch the nose down, the airspeed increased to above 190 KIAS, the vibrations ceased and the aircraft appeared to be in trim without trim adjustments by the captain. The autopilot was re-engaged 9 seconds after it disconnected. 35 seconds after the stall warning the crew received a "ICE PROT" caution, the TIMER indications for both stabiliser and wing de-ice had activated. The related checklist was executed which cleared the indications, the flight continued to Shetland Islands without further incident and the aircraft landed safely at the destination.

The British AAIB released their final bulletin concluding the probable causes of the serious incident were:

G-LGNM probably encountered both severe icing conditions and mountain wave effect while climbing. The crew reduced the airspeed to VCLEAN+15, for optimum climb performance, but the propeller rpm and power were not increased to MCP. The co-pilot perceived that the command gradient between himself and the commander was too steep for him to feel comfortable advocating such a procedure, even though he believed it appropriate.

After levelling-off, airspeed initially increased before reducing back towards VCLEAN+15, a speed that was only intended for use when climbing out of icing conditions, and in combination with MCP. It was apparent that the aircraft’s performance was being impaired by ice and it would have been appropriate to set MCP, as well as disengage the autopilot.

Pre-stall buffet was experienced and the recovery was delayed until after the stall warner had activated. Not all the stall recovery vital actions were implemented, although control was regained before a wing drop developed, as had happened in previous Saab 340 stall events.

The manufacturer is reviewing the guidance in the AFM and AOM, relating to flying the Saab 340B in icing conditions, and the operator has updated its advice to crews on the same subject.

The AAIB analysed: "The conditions at FL130 were capable of supporting SLD greater than 0.05 mm (50 microns) in diameter which can lead to the formation of severe icing. Analysis by the manufacturer concluded that the aircraft was affected by a large increase in aerodynamic drag. This could have been due to ice or downdraughts or a combination of the two. Even if the downdraughts had been twice as great as the Met Office calculated, the increase in aerodynamic drag indicated that the aircraft had probably encountered severe icing conditions."

The AAIB analysed: "The OM, as extant at the time of the incident, also stated that a speed reduction to VCLEAN+15 may be used for flaps 0 climb to exit icing conditions, when above MSA. However, it did not link this to an increase in power to MCP and the setting of maximum propeller rpm. Nevertheless, the pilots were familiar with NOTAC 66/13, which recommended that, if the climb were continued, the initial actions should be to move the condition levers to the max position, set MCP according to the appropriate chart, select half bank and climb straight ahead at a speed not below VCLEAN+15. The commander did not set MCP when reducing the IAS below VCM because he thought that downdraughts were primarily responsible for the reduction in the rate of climb, not icing. Furthermore, he believed that propeller rpm only needed to be increased to aid the shedding of propeller ice if severe icing conditions had been identified. The co-pilot did not advocate the setting of MCP because he considered that the commander would not value such a suggestion. The manufacturer’s analysis determined that, if MCP had been set, the aircraft could have been climbed to FL170 and vacated icing conditions."

The AAIB analysed stall recovery:

The AOM stated that in (certified) ice conditions, buffeting might be experienced at an IAS up to 25% above the clean stall speed. This equated to 131 KIAS at G-LGNM’s weight. However, the aircraft apparently encountered severe icing, which placed it outside certified conditions, and a strong vibration (the pre-stall buffet) was noticed at 137 KIAS. The co‑pilot was not sure of the significance of the vibration, while the commander was attempting to use the autopilot to lower the nose and increase airspeed. The Emergency Checklist states that if there is any doubt about the source of an unusual vibration or buffeting, a stall recovery must be performed.

As the IAS decreased, a roll to the left was corrected by the autopilot. This reflected a previous investigation, concerning N306AE, which noted a similar rolling motion prior to the stall of that aircraft. The OM recommends a reduction in AOA in the event of an uncommanded roll movement. This appears in a section relating to ‘SLD Detection’.

Of note, the stall warning would have been set to a lower AOA if the ‘Ice Speed Function’ had been operational. If so, the warning would have operated at an IAS of about 140 KIAS, before the pre-stall buffet and the uncommanded roll.

When the stall warning activated and the autopilot dis-engaged (at 132 KIAS, with an AOA of 13º), the commander effected a recovery by pitching the nose down to accelerate the aircraft. However, he did not call ‘stall - max power’ and use all available power, as specified in the AOM.

The pilots’ previous stall training had commenced from an approach configuration, with a low power setting and with the propeller rpm already at max. The pilots had practised carrying out a stall recovery by advancing the power levers but not the condition levers.

This anomaly was highlighted during the investigation and has been addressed by the operator in a new training schedule (see Safety Actions later in this report).

The AAIB analysed: "No definitive explanation could be given for the illumination of the ice prot caution and the timer light as the airspeed increased. It was likely to have been caused by a discrepancy in one of the sensors in the de-ice boot system but, as it cleared when the abnormal checklist actions were taken, there is no evidence that it was a contributory factor in the incident."