AINsight: Fatal In-flight Upset Poses Important Lessons

 - May 25, 2023, 6:54 AM
Bombardier Challenger 300 registered as N300ER on airport ramp prior to fatal accident
Bombardier Challenger 300—N300ER—was involved in an inflight upset accident on March 3, 2023, near Windsor Locks, Connecticut, fatally injuring a passenger. (Photo: Twitter)

On March 3, 2023, a Bombardier Challenger 300 was involved in an accident en route from Keene, New Hampshire, to Leesburg, Virginia, killing one passenger. Initial media reports suggested that an encounter with severe turbulence was the cause, but this was later ruled out. 

A recently published NTSB preliminary investigative report on the accident identified an in-flight upset caused by a pitch trim malfunction. The report raises many questions about the pitch trim system on this super-midsize business jet and yet more questions about the actions of the flight crew leading up to the in-flight upset event.

The Challenger 300/350 has been in service for nearly 20 years, with more than 800 examples delivered worldwide. 

In recent years, both the FAA and Transport Canada have issued airworthiness directives (AD) related to the aircraft’s horizontal stabilizer trim system. Each AD identifies a scenario following a “STAB TRIM FAULT” advisory and autopilot disconnect, where flight crew commands for a nose-up trim resulted instead in the nose-down trim movement of the horizontal stabilizer.

On June 6, 2022, the FAA published an AD for all Bombardier Challenger 300/350s after “multiple incidents” involving malfunctioning horizontal stabilizers. This AD ordered pilots to perform additional preflight safety checks of the pitch trim system and revised trim malfunction procedures. The AD also includes revised “AP STAB TRIM FAIL” caution and “STAB TRIM FAULT” advisory procedures. 

Earlier this month, on May 19, Transport Canada issued a similar AD for the Challenger 300/350 related to the horizontal stabilizer trim switch and unintended horizontal stabilizer motion. In this document the agency provided additional context to the problem: “In some events, the horizontal stabilizer reached full aeroplane nose-down trim position before the crew recognized the nature of the problem, which resulted in increased crew workload and reduced safety margins.” 

This is problematic since the crew may not be able to regain control of the horizontal stabilizer, resulting in reduced controllability and high control forces. 

Now let’s look at the accident flight focusing on the flight crew and the events leading up to the in-flight upset. 

According to the FAA airman records, both the pilot-in-command (PIC) and second-in-command (SIC) of the accident airplane held an airline transport pilot certificate. The PIC had more than 5,000 hours of flight time and the SIC had just over 8,000 hours.

Both pilots were new to the Challenger 300. Each completed initial ground and simulator training and earned their PIC type-rating in the Challenger 300 only six months before the accident flight. At the time of the accident, the PIC had 88 hours in type, while the SIC had 78 hours in the make and model. The pilots were fresh out of type training and each had limited operating experience in the twinjet.

The NTSB preliminary report did not specify where the pilots were trained. However, with a history of horizontal stabilizer trim issues—including recent ADs—it is a safe bet (let’s hope) that this was a topic during initial ground and simulator training.

Accident Flight

 According to the NTSB preliminary report, on the day of the accident the pilots reported a routine preflight inspection, engine start, and taxi for departure. During the takeoff roll, the SIC reported that the airplane accelerated normally, but he observed that the right primary flight display (PFD) airspeed indicator did not agree with the left side airspeed indicator. This prompted with the PIC to reject the takeoff. 

Following the rejected takeoff, the PIC exited the runway onto a taxiway. At this point, the left engine was shut down and the SIC opened the main cabin door and walked to the front of the airplane where he found a “red pitot probe cover” installed on the right-side pitot probe. He then removed the cover, did not see any damage, and returned to the cockpit. 

This error—overlooking the pitot cover during preflight—had several consequences that not only included the initial rejected takeoff, but subsequent aircraft faults and advisory messages, as well as other untrapped errors by the flight crew. 

The PIC then restarted the left engine and resumed the taxi for another takeoff attempt. Shortly after the engine started, the crew reported that a “RUDDER LIMIT FAULT” advisory message was annunciated on the engine indicating and crew alerting system (EICAS). The PIC attempted to “clear” the message using a ground avionics “stall test,” which failed to clear the message. At this point, the crew decided to continue the flight since it was an advisory message and not a caution or warning message. 

During the second takeoff, the SIC noticed that the V-speeds were not set, but the acceleration was normal. From memory, the SIC called “V1” and “rotate” at 116 knots. The takeoff and initial climb were normal.

The PIC reported that the autopilot was engaged as the aircraft passed through 6,000 feet MSL. It is important to note that a review of flight data by the NTSB indicated three separate unsuccessful attempts to engage the autopilot before reaching 6,000 feet. 

During the subsequent climb to FL240, the flight crew observed multiple EICAS caution messages, including “AP STAB TRIM FAIL,” “MACH TRIM FAIL,” and “AP HOLDING NOSE DOWN.” Neither crew could recall the exact order that the EICAS messages were presented.

The flight crew then located and agreed to execute the “PRI STAB TRIM FAIL” checklist—a message that was not displayed. 

Locating the correct checklist or “running” the wrong checklist in the quick reference handbook (QRH) have been identified as issues in the checklist design. Checklist titles should be the same as EICAS messages. Providing more context to a flight crew using “conditions” can help lead the pilots to the correct checklist. Note: there was an “AP STAB TRIM FAIL” checklist available.

According to the NTSB preliminary report, the first step in that checklist was to move the primary trim switch from PRI (primary) to OFF. At this point, the aircraft was climbing through 22,780 feet. 

As the switch was turned to OFF, removing power from the primary stab trim and disconnecting the autopilot, the aircraft abruptly pitched up to +11 degrees aircraft nose up at +3.8 gs and then the pitch decreased, and the vertical acceleration changed to -2.3 gs. The aircraft then pitched up to 20 degrees (+4.2 gs). 

During these pitch oscillations, the PIC instructed the SIC to move the stabilizer trim switch back to the primary position. The report noted that normal aircraft handling was available after the primary trim switch was moved back to normal.

In reviewing recent accident reports—including this one—where passengers are injured or killed, they often have recorded cyclic gs (positive and negative) that are two to three times the certified structural limits of the aircraft.

Remember, transport category aircraft are typically certified to -1.0 to +2.5 gs. The maximum recorded values during this accident flight ranged from -2.3 to +4.2 gs. 

In real life, these events are very violent episodes. Unrestrained occupants would forcefully be tossed around the cabin, causing injury or death.  

Following the upset, the flight crew were alerted by a passenger that another passenger had been injured. The SIC exited the cockpit to check on the passenger and provide medical attention. Soon afterwards, the SIC advised the PIC that there was a medical emergency and they needed to land. 

The flight crew informed ATC of the medical event and began a diversion to Bradley International Airport (KBDL) in Windsor Locks, Connecticut. After landing, the airplane taxied to the ramp, where an ambulance was waiting. The passenger would later die from her injuries at the hospital. 

This was a tragic event. In the coming months, the NTSB will publish its final accident report. From this document, there will be many more lessons, a singular probable cause, and interesting points for further conversation. 

From the preliminary NTSB report, we know that there was an in-flight upset caused initially by a pitch trim anomaly. Unfortunately, a passenger was killed on this flight.

For Bombardier Challenger 300/350 pilots, a thorough QRH review of the horizontal stabilizer trim failures is warranted. 

For the rest of the business aviation community, a renewed interest in upset prevention and recovery training (UPRT) would be appropriate. Airline pilots have been required to complete a UPRT program since 2019. To date, there is no requirement for Part 91 or 135 pilots to complete any UPRT program.

The opinions expressed in this column are those of the author and are not necessarily endorsed by AIN Media Group.