Jazz DH8A at Mont Joli on Feb 3rd 2016, control problems on approach

Last Update: August 8, 2017 / 16:26:18 GMT/Zulu time

Bookmark this article

Photo of Jazz Aviation C-GJMO, De Havilland Dash 8 (100) — Jazz Aviation C-GJMO, De Havilland Dash 8 (100) (Photo credit: Lord of the Wings© / Flickr / License: CC by-sa)

Incident Facts

Date of incident
Feb 3, 2016

Classification
Incident

Airline
Jazz Aviation

Flight number
QK-8964

Departure
Montreal, Canada

Destination
Mont Joli, Canada

Aircraft Registration
C-GJMO

Aircraft Type
De Havilland Dash 8 (100)

ICAO Type Designator
DH8A

Airport
Mont Joli Airport, Mont-Joli

Airport ICAO Code
CYYY

A Jazz de Havilland Dash 8-100, registration C-GJMO performing flight QK-8964 from Montreal,QC to Mont Joli,QC (Canada) with 24 passengers and 3 crew, was descending towards Mont Joli in icing conditions, the crew detected ice build ups, the anti-ice systems worked normally however. While intercepting the final approach course the aircraft encountered turbulence and the crew noticed reduced elevator efficiency. The crew increased airspeed, was able to regain control and continued for a safe landing at Mont Joli.

The Canadian TSB reported that a post flight examination found the anti-ice systems on the left hand wing as well as the left hand stabilizer had not worked.

On Aug 8th 2017 the Canadian TSB released their final report concluding the probable causes of the incident were:

Findings as to causes and contributing factors

- The landing gear extended overspeed occurred because the aircraft encountered significant increased performance shear while flying out of a low-level jet with the autopilot engaged in vertical speed mode.

- The combination of turbulence and shear contributed to the temporary difficulty with aircraft control effectiveness on approach.

- It is likely that the expectation of significant icing, the high workload, moderate turbulence, and attention narrowing contributed to the pilot flying’s perception that a reduction of elevator control effectiveness had occurred.

Findings as to risk

- If reportable occurrences are not reported to the TSB in a timely manner, perishable information may be lost, which could preclude the identification and communication of safety deficiencies to advance transportation safety.

- If cockpit recordings are not available to an investigation, this may preclude the identification and communication of safety deficiencies to advance transportation safety.

- If crews do not use the weather radar near frontal activity with embedded altocumulus castellanus, there is a risk that convective activity will not be detected and aircraft may be flown into conditions that could affect the safety of the flight.

- If long significant meteorological information messages are transmitted on the frequency while flight crews are in a high-workload phase of flight, it may disrupt the performance of high-priority tasks.

Other findings

- The aircraft did not encounter significant ice accumulation while on approach.

- Since there is no disagreement between the elevator position and the aircraft response to control inputs, it can be concluded that no elevator control issues occurred during this event.

- Analysis of the weather conditions, aircraft configuration, and flight data recorder data confirm that a tailplane stall did not occur.

The TSB reported that the flight crew reviewed weather forecasts before departure and voiced concern regarding a fast moving weather system, that was about to move a warm front towards the Mont Joli aerodrome associated with severe clear icing and turbulence near Mont Joli. The crew consulted with dispatch also raising concern that the return flight might not be able to depart due to freezing rain. However, the weather still met all legal and company requirements, hence dispatch and crew agreed the flight should depart on schedule.

While enroute Montreal Area Control Center inquired with the crew whether they were aware of the significant meteorological information (SIGMET) J8 indicating freezing rain and severe icing below 6000 feet about 30nm west of Mont Joli, which the crew affirmed.

While descending towards Mont Joli the crew levelled off at 14,000 feet about 55nm west of Mont Joli for about 3 minutes to remain clear of cloud and turbulence. The aircraft subsequently continued the descent at about 1100 fpm rate of descent. About 12nm from the initial approach fix EPMAL the crew was handed off to Mont Joli Radio and confirmed they were planning an area navigation approach to runway 06. Mont Joli radio queried whether they were aware of SIGMET J8, the crew affirmed. Two minutes later Mont Joli Radio transmitted the whole SIGMET on their frequency, the crew queried whether this was normal procedure, Mont Joli affirmed and the crew commented that this was a distraction not needed in the already high workload.

A few seconds later the aircraft passed intermediate waypoint APNAM descending through 4300 feet at 170 KCAS. The crew reduced power to idle, lowered the gear and stabilized the speed at 150 KCAS, the autopilot was maintaining a vertical speed. The speed subsequently increased to 171 KCAS as the aircraft descended from 3340 to 2480 feet.

Descending through 2480 feet 6.1nm before the runway threshold the aircraft encountered moderate turbulence. A few seconds later the maximum gear extended speed of 172 KCAS was exceeded. The captain (ATPL, 14,200 hours total, 11,200 hours on type), pilot flying, immediately levelled off until the speed had decayed to 160 KCAS, then resumed the descent, however felt a sudden change in elevator force and perceived this as reduction of elevator effectiveness. The crew did not declare emergency however. A short time later the captain noticed the TAIL advisory light for the outer left de-ice boot did not illuminate. Due to the existing workload the related de-ice boot failure procedure was not executed.

The turbulence decreased as the aircraft descended through 1540 feet, the crew set the flaps at 5 degrees and further down to 15 degrees at 970 feet. About a minute later the aircraft landed safely, the crew did not submit a pilot report. Post flight examination showed all critical surfaces and airframe were free of contaminants and ice. The de-icing system was tested, the outer left de-ice boot showed indication of a possible malfunction.

Tests of the de-ice systems the following morning found no anomaly. An inspection for exceeding the maximum landing gear speed also found no damage.

The TSB analysed:

The flight crew was certified and qualified for the flight in accordance with existing regulations, and there were no indications that the flight crew’s performance was in any way degraded because of physiological factors, such as fatigue.

Throughout the approach and landing, forecasted meteorological conditions were conducive to moderate turbulence and severe icing below freezing temperatures, as the aircraft was descending through a warm front with embedded altocumulus castellanus ACC and a low-level jet stream.

When the aircraft was approximately 6 nm from the runway threshold, moderate turbulence was encountered during which the landing gear extended speed (VLE) was exceeded. The pilot flying (PF) disconnected the autopilot and immediately performed a level-off to reduce the airspeed. During this moderate turbulence encounter, the PF felt a sudden change in elevator force and perceived this as a reduction of control effectiveness, which the PF later interpreted as a tailplane stall.

...

According to subsection 8(1) of the Transportation Safety Board Regulations, “Every person having possession of or control over evidence relating to a transportation occurrence must keep and preserve the evidence unless the Board provides otherwise.”35 Shortly after landing, the pilot-in-command (PIC) reported the occurrence to the company. However, the occurrence was not reported to the TSB until 09 February 2016 by the PIC. If reportable occurrences are not reported to the TSB in a timely manner, perishable information may be lost, which could preclude the identification and communication of safety deficiencies to advance transportation safety.

The TSB analysed the approach:

At 1837, the crew initiated the descent into CYYY and later performed a level-off at 14 000 feet, just above the tops of the clouds, to minimize the exposure to icing conditions. The presence of ACC is indicative of the presence of convective activity and, therefore, the potential for significant icing and turbulence with associated local visibility reducing to ½ sm in snow showers and blowing snow.
The crew was unaware of the presence of convective activity on approach and did not use the weather radar because they did not think that convective activity was a factor during the occurrence flight (Figure 6). Therefore, if crews do not use the weather radar near frontal activity with embedded ACC, there is a risk that convective activity will not be detected and aircraft may be flown into conditions that could affect the safety of the flight.

With respect to icing the TSB analysed:

Throughout the descent below 14 000 feet, the flight experienced light to moderate mixed icing in cloud. The crew used the wing inspection and glare shield lights to monitor for icing. The system appeared to be functioning normally, even though, shortly after the reported difficulty with aircraft control, the PF noticed that the TAIL advisory light for the outer left de-ice boot did not illuminate.

The aircraft overflew the edge of the SIGMET J8 area, located approximately 20 nm from CYYY, as it was descending through 7800 feet, therefore avoiding the forecasted area of severe icing.

Analysis of the FDR data showed that the total air temperature (TAT) remained above freezing in the warm air aloft during descent from approximately 10 000 feet to 2720 feet. The reported cloud base at CYYY was 1972 feet (1800 feet AGL). Therefore, the aircraft flew in potential icing conditions from 2720 feet to 2350 feet, which is when the difficulty with aircraft control was noted. The aircraft flew out of potential icing conditions at 1972 feet. During the last part of descent in the cold air below the warm front, the temperature decreased from 0°C at 2720 feet to −7°C on the ground.

Considering that the aircraft flew in potential icing conditions for 12 seconds before the autopilot was disconnected and the temporary difficulty with aircraft control occurred, it is unlikely that airframe icing was present. Furthermore, the period of flight in potential icing conditions totalled 43 seconds, and ice accumulated during this period would have remained on the aircraft until after the aircraft landed in the colder air. Since no trace of ice was found on the aircraft when it was inspected after landing, it can be concluded that the aircraft did not encounter significant ice accumulation while on approach.

With respect to the SIGMET transmission the TSB analysed:

At 1856, the FSS asked JZA8964 if they had received SIGMET J8 and the crew confirmed having the information. Approximately 2 minutes later, the FSS transmitted the contents of SIGMET J8 in both French and English. This transmission lasted 90 seconds. The PF communicated with the FSS about the distraction caused by this transmission while the aircraft was just about to turn final at 12 nm.

Although procedurally required by the FSS Flight Services Manual of Operations, such broadcasts may introduce a potential source of distraction during the approach while preventing any other communication on the frequency.

Because human attentional resources are limited, high-priority tasks may be delayed or dropped when a well-established sequence is interrupted by an unexpected external event. Therefore, if long SIGMETs are transmitted on the frequency while flight crews are in a high-workload phase of flight, they may disrupt the performance of high-priority tasks.

With respect to the turbulence and performance shear the TSB analysed:

The tephigram soundings confirm the presence of the low-level jet in the vicinity, with a 50- to 60-knot southwesterly wind at 4000 feet, while the surface wind was southeast at 10 knots.

As the aircraft descended from 3340 feet to 2350 feet on final, the autopilot was engaged in vertical speed (VS) mode and power was set at idle. During that time, the airspeed increased from 150 KCAS (knots calibrated airspeed) to 171 KCAS, and FDR data shows a gradual decrease in pitch attitude as the autopilot maintained the set VS. This increase in speed was the result of an increased performance shear caused by the diminishing tailwind component. This is consistent with the aircraft flying out of the low-level jet.

At 2480 feet, the aircraft encountered moderate turbulence. The FDR data confirmed a sudden increase in aircraft oscillations, vertical acceleration up to 1.5g, and an abrupt 15-knot increase in airspeed in 2 seconds, leading to the VLE overspeed to a maximum of 185 KCAS. No power change was observed during this period. Therefore, it can be concluded that the VLE overspeed occurred because the aircraft encountered significant increased performance shear while flying out of a low-level jet with the autopilot engaged in vertical speed mode.

With respect to aircraft response to control inputs the TSB analysed:

The FDR data were analyzed in an attempt to explain the PF’s reported difficulty controlling the aircraft. Specifically, the FDR data were analyzed to evaluate the aircraft’s response to control inputs to determine if the aircraft encountered an elevator jam, a reduction of elevator control effectiveness, or a tailplane stall.

The limited FDR parameters do not include control column position or force and, therefore, do not allow for ruling out an elevator jam with FDR data exclusively. However, the FDR data shows considerable elevator movement, the PM observed control column movement, and the PF did not perceive any blockage of the elevator. Therefore, it is unlikely that an elevator jam occurred.

During the approach, the PF noticed that the TAIL advisory light for the outer left de-ice boot did not illuminate. Therefore, it is possible that the de-icing boot on the outboard section of the left horizontal stabilizer did not fully inflate. However, the investigation concluded that the aircraft did not encounter significant ice accumulation while on approach (see Section 2.4.1).

Throughout the event, there is no indication of an unintended departure from the approach profile. The initial elevator-up control input to level the aircraft when the VLE overspeed was noticed is apparent on the FDR and is followed immediately by an increase in pitch attitude, with a simultaneous g increase to 1.7g, and a level-off at 2340 feet.

Metars:
CYYY 040200Z 11008KT 10SM -PL OVC020 M05/M08 A2978 RMK SC8 PRESFR SLP091
CYYY 040100Z 11009KT 3SM -PLSN OVC020 M06/M09 A2985 RMK SC8 SLP115
CYYY 040000Z 12011KT 3SM -PLSN OVC020 M07/M10 A2990 RMK SC8 /S07/ PRESFR SLP133
CYYY 032338Z 13010KT 3SM -PLSN OVC018 M07/M10 A2993 RMK SC8 SLP141
CYYY 032300Z 10008KT 1 1/4SM -SN OVC018 M08/M10 A2996 RMK SC8 PRESFR SLP150
CYYY 032200Z 12011G17KT 1 1/4SM -SN OVC018 M08/M10 A3001 RMK SC8 /S05/ PRESFR SLP167
CYYY 032100Z 12010KT 1 1/4SM -SN OVC018 M08/M11 A3005 RMK SC8 /S04/ SLP183
CYYY 032031Z 07005KT 1 1/4SM -SN OVC018 M09/M11 A3007 RMK SC8 PRESFR SLP188
CYYY 032009Z 08006KT 010V110 2SM -SN OVC020 M10/M12 A3008 RMK SC8 PRESFR SLP193
CYYY 032000Z 05006KT 010V110 1SM R06/6000FT/U -SN OVC015 M10/M12 A3010 RMK SN5SC3 /S02/ PRESRR SLP199