A Slippery Runway Causes A Phenom Excursion, Part 2

Even though the investigation of a runway excursion at Elk River Airport (NC06) was classified as a limited probe, it drew interest from multiple parties. There were representatives from the Transportation Safety Board of Canada (TSB) as well as from Embraer and Brazil’s accident investigation agency.

Like many investigations during the COVID shutdown era, NTSB investigators did not travel to the accident scene. They relied on an FAA inspector from Charlotte, North Carolina, who documented the site and participated in the investigation.

The leading edges of both wings of the Phenom 300 were damaged, and the left wing attach point link fitting had fractured. When the co-pilot was interviewed, he said the right main landing gear went off the runway. When he exited the aircraft, he noticed a “small pine tree limb” in the right winglet.

The Phenom was equipped with an L-3Harris FA5000 series cockpit voice and data recorder (CVDR). It was retrieved and its data downloaded at the NTSB laboratory. In total, 103 parameters of flight data were recorded, enabling a performance specialist to analyze the accident sequence in detail. A CVR specialist auditioned the voice recording and provided a summary report of the voices and sounds made during the flight.

The jet’s brake control unit (BCU) was removed and shipped to its manufacturer—Meggitt PLC—in Coventry, England. With all five members of the investigative group observing, the unit passed its production acceptance test with no faults related to the accident.

There was no official weather reporting facility at NC06. An airport 12 mi. away was reporting 10 mi. of visibility, winds 190 deg. at 7 kt., temperature 28C, dew point 18C and altimeter 30.30. There was lightning north and southwest. 

The NOAA National Severe Storms Laboratory estimated the rainfall rate near NC06 to be about one-half to two-thirds in. per hour. Airport cameras recorded spray being thrown up behind the airplane during its landing roll.

Data Versus Assumptions

The performance study used a variety of sources to determine the jet’s trajectory, including ADS-B, FDR and CVR data, ground scars, runway texture and manufacturer’s thrust and aerodynamic data. A specialist determined that the Phenom could have stopped—with only 290 ft. to spare—if the crew had been able to conform exactly to the manufacturer’s assumptions. Those assumptions were:

• A steady 3-deg. angle approach at Vref in landing configuration   
• Vref airspeed maintained at runway threshold   *Idle thrust established at runway threshold   
• Attitude maintained until main landing gear (MLG) touchdown   
• Maximum brake applied immediately after MLG touchdown   
• Antiskid system operative

Data showed the flight differed from manufacturer’s assumptions:

• Indicated airspeed was 8 kt. faster than Vref over the threshold   
• The throttles were not at idle until 740 ft. past the threshold   
• The spoilers were not deployed until 1,770 ft. past the threshold, and the brakes were not fully applied until 2,200 ft. past the threshold.
• The jet was 36 ft. above the assumed 50 ft. threshold crossing height   
• The jet was in a 10-deg. left bank over the threshold   
• The brake pedals were not held to maximum

The challenging characteristics of the visual approach to Runway 12 made it highly unlikely the crew could have met the manufacturer’s assumptions. In addition, when the pilot began steering the aircraft back toward the centerline, the resulting cornering force reduced the braking force.

A year after the incident, manufacturer Embraer published wet runway landing distances for the Phenom 300 in a supplement to the Pilot Operations Handbook and the Quick Reference Handbook. Using that data, a specialist found that for the accident conditions, the no-wind, wet-runway landing distance would have been 4,437 ft., just 168 ft. less than the length of Runway 12 at NC06. This was very close to the number the NTSB came up with.

The specialist found the wheel braking friction coefficient achieved was less than certification standards assumed. This has become a common finding in wet runway accidents. He found that, for preflight planning, if rainfall may be possible at the time of landing, operators should consider using a Runway Condition Assessment Matrix value of four rather than the standard five normally used for wet runways.

FAA Safety Alert For Operators (SAFO) 19001, “Landing Performance Assessments at Time of Arrival,” pertains to the Elk River accident pilots. As they approached NC06, the captain said, “we can land in the rain as long as it’s VFR.” The SAFO argues otherwise. It says, “121.195, 135.385 and 91.1037 wet or slippery landing data may not provide adequate runway length for landing on a wet or contaminated surface.” It reminds Part 135 companies to take appropriate action to ensure crews have the needed information.

The SAFO provides multipliers to use with dry unfactored landing distances to correct for precipitation. For turbojets with no thrust reversers, landing on a “good” but ungrooved runway, the multiplier is 2.6. Using that guidance, the crew would not have attempted to land.

The two pilots were employed by Nicholas Air Services, a Part 135 on-demand operator. The 36-year-old captain was an Airline Transport Pilot (ATP) with type ratings on the EMB-145, EMB-505 and Learjet 45. His current first-class medical required that he use corrective lenses. He had been employed at the company since March 2020. He had 3,946 total flight hr., of which 873 were in the EMB-505 (Phenom 300). He had been busy, flying 95 hr. in the last 30 days.

The co-pilot was 39 years old. He held a commercial certificate with a type rating on the EMB-505 (second-in-command privileges only). His last proficiency check was on June 20, 2021, but his type rating was issued only 15 days before the accident. His first-class medical certificate said he “must wear lenses for near and distant vision.” His last job before joining Nicholas Air was flying aerial surveying flights. He had logged 2,089 total flight hr., with 247 hr. flown on the Phenom.

We offer the conclusions of the investigation and our comments in Part 3 of this article.

To read “A Slippery Runway Causes A Phenom Excursion, Part 1,” click here.