PD2691 - Incident: Porter DH8D at Sault Ste Marie on Apr 16th 2023

A Porter Airlines de Havilland Dash 8-400, registration C-GLQB performing flight PD-2691 from Toronto City,ON to Sault Ste Marie,ON (Canada) with 52 passengers and 4 crew, landed on Sault Ste Marie's runway 12 at 22:21L (02:21Z Apr 17th) but overran the end of the runway and came to a stop on soft ground about 85 meters/280 feet past the end of the runway. There were no injuries, the aircraft sustained minor if any damage.

The runway was closed and is still closed about 15 hours after landing.

The airport reported the aircraft went off the runway and came to rest at a grassy area. There were no injuries and no damage. The passengers were taken to the terminal with airport limousines.

On May 1st 2023 the Canadian TSB reported: "Upon arrival, the flight crew performed the RNAV approach for runway 12; however, after touchdown the aircraft overran the end of the runway. The aircraft came to a stop about 250 beyond the end of the runway and was stuck in soft grass/mud."

On Nov 20th 2024 the TSB released their final report concluding the probable causes of the serious incident were:

Findings as to causes and contributing factors

These are conditions, acts or safety deficiencies that were found to have caused or contributed to this occurrence.

- Once the aircraft was over the runway, the flight crew’s focus briefly shifted to other tasks and, with the limited visual cues available during the night landing, they did not recognize that the aircraft was further down the runway than expected.

- The wing spoilers deployed 2 seconds after touchdown and the lowering of the nose wheel and application of full braking was delayed because the flight crew were unaware of the aircraft’s proximity to the end of the runway. As a result, significant deceleration did not commence until the captain recognized the runway end lights, when the aircraft was 850 feet from the end.

- When the captain applied full braking with 850 feet of runway remaining, the aircraft should have been theoretically able to stop; however, braking performance was degraded, likely due to the runway surface texture on the last 300 feet of the runway and possible air contamination in the hydraulic system. As a result, the aircraft overran the end of the runway.

Findings as to risk

These are conditions, unsafe acts or safety deficiencies that were found not to be a factor in this occurrence but could have adverse consequences in future occurrences.

- If operational procedures and pilot training do not emphasize the circumstances in which propeller reverse must be used and the aircraft’s behaviour when it is used, there is a risk that flight crews will not recognize and respond correctly to situations that require the use of this function in time to avoid a runway overrun.

The TSB analysed:

In this occurrence, while touching down on Runway 12 at Sault Ste. Marie Airport (CYAM), Ontario, the flight crew were not initially aware of the aircraft’s proximity to the end of the runway, and when they did become aware, the actions taken were not able to bring the aircraft to a stop before it reached the end of the runway, resulting in a runway overrun.

The analysis will therefore focus on runway situational awareness; flight crew landing technique, including the use of reverse thrust; as well as degraded braking performance stemming from technical system faults and runway surface conditions.

Runway situational awareness

The occurrence aircraft’s approach into CYAM was conducted during night lighting conditions and in light rain. The crew’s plan was to touch down between 1000 and 2000 feet from the beginning of the runway, with Taxiway J (located approximately 2000 feet from the beginning of the runway) marking the end limit of the expected landing zone. The primary visual indicators that were available to assist them in their landing were the runway edge lights, the taxiway lights and signs, the lights of the intersecting runway, and the runway end lights.

While the aircraft was passing Taxiway J during the final approach, the captain was checking the pitch indicator to monitor for a pitch angle greater than or equal to 5°, per company procedures, to avoid the risk of an aft fuselage strike. At that point, the first officer (FO) was focused primarily on the runway surface to ensure the aircraft’s proper contact and alignment given the visual challenges associated with conducting a landing at night and in light rain. It is therefore likely that the crew did not either perceive or process Taxiway J as they passed it during this final phase of flight.

Based on where the first weight-on-wheels signal was triggered and on the expectations of the flight crew given their targeted landing zone, they may have mistaken Runway 04/22, which intersects with Runway 12, for Taxiway J, creating an inaccurate mental picture that they touched closer to the threshold than they actually did. This mental picture of their location is supported by the fact that the captain focused on matters less pressing than decelerating the aircraft by instructing the FO to correct a minor lateral deviation while the nose wheels was touching down when the aircraft was approximately 2000 feet from the end of the runway.

When the nose gear touched down completely with 1700 feet of runway remaining, the FO initially applied only light braking in accordance with company After Landing procedures, which was not immediately corrected by the captain, indicating that the crew was still not aware of the aircraft’s proximity to the end of the runway.

One of the ways in which a flight crew’s situational awareness on a runway can potentially be improved is the use of signage dedicated to indicating remaining runway length. Some other airports throughout the world, and most Canadian military airports have adopted distance-remaining signage. These signs assist flight crews by showing the distance remaining on the runway in increments of 1000 feet.

Situational awareness can also be improved by having yellow runway edge lights in the last portion of the runway indicating to a flight crew that they have reached the last third or the last 600 m of the runway. Although newer runways are approved under these requirements, many airports like CYAM were certified under previous standards and still have white runway edge lights.

Landing technique

Before landing on Runway 12, the flight crew conducted a stable approach and were aware of the wind and runway surface conditions. According to the aircraft flight manual (AFM), the landing was planned with sufficient runway length for the given aircraft configuration and the wet runway condition. The aircraft’s approach speed was higher than the minimum reference speed (Vref); however, this approach speed was briefed by both flight crew members during the approach briefing and was within the requirements of Porter Airlines’ standard operating procedures (SOPs).

The aircraft initially made a soft touchdown on the runway. As a result, the wing spoilers did not activate until 2 seconds later, when the main landing gear wheels made solid contact, compressing the main landing gear wheel struts enough to activate the weight-onwheels signal to the aircraft systems. At this point, the aircraft was already 2850 feet down the runway.

Following this solid contact, propeller DISC mode was selected; however, the nose landing gear contacted the runway about 7 seconds after the weight-on-wheels signal, with approximately 1700 feet of runway remaining.

Following the nose touchdown, the FO initially applied only partial braking. It was not until 6 seconds after the nosewheel touchdown, approximately 850 feet from the runway end, that the captain recognized the runway end lights and the lack of deceleration, given the runway distance remaining, and took over to input full braking action.

At this time, the runway length remaining was close to the limit of the aircraft’s theoretical landing performance for the given conditions. As a result, there was no remaining margin for the crew to react to unforeseen circumstances, such as possible degraded braking performance.

Use of propeller reverse thrust

Owing to the fact that the Q400 Airplane Flight Manual and the DHC-8 Series 400 Aeroplane Operating Manual do not provide specific guidance on the circumstances in which propeller reverse should be used, the decision to use this function is based mostly on the pilots’ training and familiarity with the aircraft’s behaviour when reverse is activated.

Porter Airlines Inc.’s procedures state that the use of reverse should be kept to a minimum and should be used only if required for operational or safety reasons. Cautions for its use are also provided. In 2019, Porter Airlines Inc. acknowledged the limited familiarity among its pilots on the usage of the propeller reverse and issued an internal Safety Alert. The Safety Alert identified circumstances in which the use of reverse should be considered, such as an inoperative anti-skid or normal brake system, or any situation in which the stopping distance is critical.

Training provided to company pilots includes practising the use of reverse in the simulator and having discussions about the use of disc/reverse as part of their line indoctrination training. However, the function is rarely used in normal operations given the reportedly high efficiency of propeller discing during landings.

In this occurrence, the deteriorated braking performance and reduced runway length remaining were only recognized approximately 8 seconds before departing the end of the runway. Considering the additional 3 second system delay that follows a pilot’s command of reverse before the propellers actually start providing additional deceleration, the investigation could not establish whether the selection of reverse would have avoided the overrun at that time.

However, the use of reverse thrust early in a landing roll would significantly increase aircraft deceleration, reducing potential for a runway overrun. Reverse was not considered by the flight crew at any stage of the landing roll as a means to provide additional deceleration.

Air contamination in the hydraulic system

During the aircraft’s deceleration, control of the aircraft was transferred from the FO to the captain when they both noticed that the aircraft was not decelerating as expected. The brakes were fully applied with 850 feet of runway remaining and while the aircraft’s ground speed was still 78 knots. At this point, the propellers were both in DISC mode and all spoilers had deployed.

The anti-skid system activated immediately; however, the flight data recorder data show a pressure fluctuation between 2100 and 1000 psi for the left brakes and a direct increase to 3000 psi in the right brakes following pedal application until the aircraft departed the runway.

Even considering the operational factors (the delay in lowering the nose, applying the DISC setting, and braking) and the environmental conditions that were present at the time of the occurrence, when full braking was applied, there was theoretically still sufficient runway length available for the aircraft to stop before the clearway.

However, post-occurrence maintenance activities revealed that the brakes could not hold the aircraft in position during high-power engine runs because of the presence of air in the hydraulic system. Although the investigation could not determine how this air, which was subsequently bled from the brake assemblies, had been induced in the hydraulic system, it possibly affected the aircraft’s braking performance during this occurrence.

Runway surface texture

Five days after the occurrence, the TSB laboratory conducted a survey of the surface microtexture and macrotexture of Runway 12 at CYAM. A general degradation in the surface microtexture was observed. The macrotexture survey also showed significantly low measurements on the concrete pavement of the last 300 feet of Runway 12 and on the asphalt pavement of the 100 feet of overrun. The aircraft’s deceleration decreased from 0.29 to 0.16g when the aircraft moved from the asphalt portion of the runway to the concrete portion, indicating a degradation of the braking performance.

The theoretical landing performance considered different runway surface conditions (from a nominal, new surface to a moderately polished one) and varying “sharpness” factors of the asperities on the paved runway surface. The results indicate that regardless of the runway surface microtexture or macrotexture, the aircraft should have been theoretically able to stop within the runway distance available, even if it came very close to the limit. However, the degraded surface texture of the concrete portion of the runway and the asphalt overrun had an impact on braking performance.

Related NOTAMs:
J2035/23 NOTAMR J2033/23
Q) CZYZ/QMRLC/IV/NBO/A/000/999/4629N08431W005
A) CYAM B) 2304170510 C) 2304172100
E) RWY 12/30 CLSD DUE DISABLED ACFT.

J2033/23 NOTAMN
Q) CZYZ/QMRLC/IV/NBO/A/000/999/4629N08431W005
A) CYAM B) 2304170244 C) 2304171100
E) RWY 12/30 CLSD DUE DISABLED ACFT.

Related Flight: PD2691, Twitter: #PD2691, Porter Airlines News