WPR21FA286

The first officer’s (FO’s) improper decision to attempt to salvage an unstabilized approach by executing a steep left turn to realign the airplane with the runway centerline, and the captain’s failure to intervene after recognizing the FO’s erroneous action, while both ignored stall protection system warnings, which resulted in a left-wing stall and an impact with terrain. Contributing to the accident was the FO's improper deployment of the flight spoilers, which decreased the airplane's stall margin; the captain’s improper setup of the circling approach; and the flight crew’s self-induced pressure to perform and poor crew resource management, which degraded their decision-making.

The captain and first officer (FO) departed on a non-revenue flight operating under instrument flight rules with four passengers bound for Truckee, California. Most of the flight was uneventful. During the descent, air traffic control (ATC) told the flight crew to expect the area navigation (RNAV [GPS]) approach for runway 20. The captain (pilot flying [PF]) stated and the FO (pilot monitoring [PM]) calculated and confirmed that runway 20 was too short for the landing distance required by the airplane at its expected landing weight. Instead of making a request to ATC for the straight-in approach to runway 11 (the longer runway), the captain told the FO they could take the runway 20 approach and circle to land on runway 11, and the FO relayed this information to ATC. ATC approved, and the flight crew accepted the circletoland approach. Although the descent checklist required that the flight crew brief the new circletoland approach, and the flight crew’s acceptance of the new approach invalidated the previous straight-in approach brief, they failed to brief the new approach. ATC instructed the flight crew to hold, but the captain was slow in complying with this instruction, so the FO started the turn to enter the holding pattern and then informed ATC once they were established in the hold. About 20 seconds later, ATC cleared them for the approach. Before the FO confirmed the clearance, he asked the captain if he was ready for the approach, and the captain stated that he was. The FO subsequently commented that they had too much airspeed at the beginning of the approach and then suggested a 360° turn to the captain, but the captain never acknowledged the excessive airspeed and refused the 360° turn. After the FO visually identified the airport, he told the captain to make a 90° right turn to put the airplane on an approximate heading of 290°, which was parallel to runway 11 and consistent with the manufacturer’s operating manual procedures for the downwind leg of the circling approach. However, the FO instructed the captain to roll out of the turn prematurely, and the captain stopped the turn on a heading of about 233° magnetic, which placed the airplane at an angle 57° left of the downwind course parallel with runway 11. As a result of the early roll-out, the flight crew established a course that required an unnecessarily tight turning radius. When they started the turn to final, the airplane was still about 1.3 nautical miles (nm) from the maximum circling radius that was established for the airplane’s approach category. The FO also deployed flaps 45° after confirming with the captain (the manufacturer’s operating manual procedures for the downwind leg called for a flaps setting of 30°, but the manufacturer stated that a flight crew is not prohibited from a flaps 45° configuration if the approach remains within the limitations of the airplane’s flight manual). The airplane’s airspeed was 44 kts above the landing reference speed (Vref) of 118 kts that the flight crew had calculated earlier in the flight; the FO told the captain, “I’m gonna get your speed under control for you.” The FO likely reduced the throttles after he made this statement, as the engine fan speeds (N1) began to decrease from about 88% to about 28%, and the airplane began to slow from 162 kts. After the FO repeatedly attempted to point out the airport to the captain, the captain identified the runway; the captain's difficulty in finding the runway might have been the result of reduced visibility in the area due to smoke. The FO continuously reassured and instructed the captain throughout the circletoland portion of the approach. On the base leg to the runway and about 25 seconds before impact with the ground, the FO started to repeatedly ask for control of the airplane, but neither flight crewmember verbalized a positive transfer of control as required by the operator’s general operating manual (GOM); we could not determine who had control of the airplane following these requests. As the airplane crossed the runway extended centerline while maneuvering toward the runway, the FO noted that the airplane was too high. One of the pilots (recorded flight data did not indicate which) fully deployed the flight spoilers, likely to increase the airplane's sink rate. (The flight spoilers are deployed using a single control lever accessible to both pilots.) The airspeed at the time was 135 kts, 17 kts above the Vref based on the erroneous basic operating weight (BOW) programmed into the airplane’s flight management system (FMS). About 7 seconds later, the left bank became steeper, and the stall protection system (SPS) stick shaker and stick pusher engaged. The captain asked the FO, “What are you doing,” and the FO again asked the captain multiple times to “let [him] have the airplane.” The stick shaker and stick pusher then briefly disengaged before engaging again. The airplane then entered a rapid left roll, consistent with a left-wing stall, and impacted terrain. A postcrash fire consumed most of the wreckage. Analysis of data retrieved from the flight data recorder (FDR) indicated that the engines were functioning normally at the time of impact and there were no indications of a flight control or system malfunction. Most of the wreckage was consumed by postcrash fire, and the flight control linkages were destroyed either by high energy impact forces or the postcrash fire, which precluded a complete examination of the wreckage. Examination of the primary flight control surfaces did not reveal any preimpact mechanical anomalies. Engine data from the accident flight did not show any interruptions in power or suggest any mechanical anomalies with the power production capabilities of either engine. Flight Crew Performance The captain and FO were appropriately qualified to perform their respective duties as pilotincommand (PIC) and secondincommand of the accident flight, which was the first pairing of this crew for the operator. A review of operator documentation revealed that the flight complied with the requirements of Title 14 Code of Federal Regulations (CFR) Part 91, General Operating and Flight Rules, and was not conducted under the operator’s 14 CFR Part 135 certificate. Although toxicology testing detected ethanol in the FO’s tissue, given the different concentrations of ethanol, the presence of npropanol, and the state in which the body was found, it is likely that the identified ethanol was from sources other than ingestion. The flight crew elected to conduct a circling approach to runway 11 and never asked ATC for the straightin RNAV (GPS) approach to the desired runway. The crew also failed to brief the new circling approach after previously briefing the anticipated straightin approach. The flight crew’s failure to brief the circling approach prevented them from sharing a mental model for how the approach should have been conducted and points to poor crew resource management (CRM) because they failed to prepare for adverse situations and contingencies, such as a missed approach. Because of their lack of preparation, they made critical errors on the approach that reduced the safety margin, which included: o flying the circling approach at a higher airspeed than the upper limit specified for the airplane’s category C approach category; o failing to establish the airplane on the downwind leg of the circletoland approach; and o failing to visually identify the runway early in the approach, likely due to obscuration by smoke. The airplane’s higher airspeed reduced the flight crew’s time to configure the airplane, assess their position relative to the runway, and make corrections to their trajectory, which further reduced the safety margin. During the approach, the FO made several announcements to the captain that the airplane was fast. The captain rejected the FO’s suggestion to take a 360° turn early in

NTSB Safety Alert SA-084, Circling Approaches: Know the Risks, cautions that circling approaches can be riskier than other types of approaches. Specifically, circling approaches often require maneuvering at low altitude and low airspeed during the final segment of the approach, increasing the opportunity for loss of control or collision with terrain. These risks are heightened when conducting circling approaches in marginal or reduced visibility conditions. Captain The captain, who had signed an employment contract with the operator but was not yet an employee at the time of the accident, was operating the flight under contract to the operator until he could be onboarded. A review of training records from the captain’s training provider showed that the captain, who was the PF and PIC during the accident, completed his most recent proficiency training as PIC in a Challenger 605 simulator 10 days before the accident flight. This most recent ground training included CRM training, and the captain’s overall rating for ground training was proficient. Records showed that the captain passed his checkride, which included a nonprecision approach, stall prevention, and a goaround/rejected landing; however, the instructor comments for the practice simulator sessions noted that he rushed checklists, needed to slow down and read the checklist requirements, and needed to setup approach procedures without PM prompts. The captain also enrolled online for the accident flight operator’s basic indoctrination training, which included instruction on the flight operator’s GOM, 12 days before the accident flight. Federal regulations do not require any leadership and command training for Part 91 or Part 135 operations, and the captain had not taken any leadership training. However, according to both 14 CFR 91.3(a) and the flight operator’s GOM, the captain as PIC would have been “directly responsible for, and is the final authority as to, the operation of” the accident airplane. First Officer The FO, who was the PM and second-in-command during the accident flight, was not the accident operator’s employee at the time of the accident and had been hired as a contract pilot for the accident flight. The flight was the first pairing of this crew with the operator. A review of training records from the FO’s training provider showed that the FO completed a simulator session in a Challenger 604 in May 2021 and completed Challenger 604 recurrent training as PIC in June 2021. The recurrent training program included 15 hours of ground training and 8 hours of simulator training, which were split evenly as the PF and PM with five nonprecision approaches and one circletoland approach. The simulator training sessions included CRM, in which the FO was rated proficient. The FO’s training was not specific to the operator’s policies and procedures and the FO did not receive training on the operator’s GOM. According to Federal Aviation Administration (FAA) Order 8900.1, Volume 5, Chapter 2, Section 19, the accident airplane, a Challenger 605, was under the same type rating designation as the Challenger 604. Toxicology testing performed by the FAA Forensic Sciences Laboratory detected ethanol in the FO’s muscle and kidney tissue concentrations approximately equivalent to 0.059 grams per deciliter and at 0.028 grams per deciliter, respectively. FAA toxicology testing also detected npropanol, another form of alcohol, in his muscle tissue. FAA toxicology testing of the captain’s muscle tissue was negative for ethanol and other testedfor drugs. Ethanol detected in postmortem specimens may result from ethanol production by microbes in a person’s body tissues after death. Flight Management System The airplane was equipped with a triple FMS, including three control display units in the cockpit, and three flight management computer (FMC) units in the underfloor avionics equipment bay. The flight crew used the control display units to input, modify, and execute flight plans; calculate airplane performance; and determine the airplane’s approach speeds (including Vref), maximum landing weight, and landing field length. The year before the accident, a maintenance facility serviced the airplane’s FMC units to comply with a scheduled battery replacement. Servicing the FMC units involved removing the FMCs from the aircraft, sending them to a third-party repair facility, and subsequently reinstalling them in the aircraft. Bombardier’s (the airplane manufacturer’s) maintenance manual instructed owners to “make sure that the default values and BOW value are appropriate for the aircraft” after reinstalling the FMC. According to the maintenance facility, it reinstalled the FMCs and the required databases, which included the approach speeds and performance databases, but did not input the BOW specific to the accident airplane. During subsequent testing, the manufacturer of the FMS confirmed that the BOW defaults to 24,000 lbs after battery replacement and reinstallation of the databases. The FMC manufacturer also confirmed that the installation of the databases did not prompt the user to enter a BOW. Flight Spoilers The airplane was equipped with flight spoilers that provided lift dumping and speed control while airborne. A flight spoiler control lever on the center pedestal allows the pilot to select variable amounts of flight spoiler deployment up to a maximum of 40°. An amber caution message is posted on the engine indicating and crew alerting system if the flight spoilers are deployed in flight and another condition is met, such as an altitude between 10 and 300 ft above ground level (agl); the left or right N1 is greater than 79%; or, if radio altitude is not available, any time the landing gear is extended. Stall Protection System According to a technical memorandum prepared by Bombardier, the natural stall characteristics of its Challenger 600 series airplanes (including the Challenger 605) include an abrupt load factor reduction and uncontrollable roll at the instant of stall with no pre-stall warning. The memorandum notes that these natural stall characteristics are not certifiable to 14 CFR Part 25 Transport Category Airworthiness standards. Consequently, Challenger airplanes incorporate an SPS that provides certifiable stall characteristics (most critically, a stall warning and pitch down at the point of stall) by mechanical means. To provide a stall warning, which is absent in the natural stall, Challenger control columns have stick shakers that engage when the airplane AOA increases above a predefined threshold (the shaker firing angle). To obtain certifiable stall characteristics if the AOA increases further, a stick pusher device abruptly commands full nose-down elevator once the AOA crosses a second higher threshold (the pusher firing angle). This nose-down elevator produces a nose- down pitching moment and pitch response similar to the pitch response associated with certifiable natural stall characteristics. The nose-down motion reduces the AOA and wing lift. The pusher firing angle defines the airplane stall AOA for certification purposes and is tailored to provide stick pusher engagement and AOA recovery without encountering the natural stall. The pusher firing angle also accounts for higher pitch rates and consequent AOA overshoots beyond that point. The shaker firing angle is offset to a lower AOA than the pusher firing angle to provide the stall warning margin required by certification standards. Bombardier programmed the shaker firing angle assuming that a crew experiencing the stick shaker will lower the nose and recover AOA promptly and thus avoid the stall AOA defined by the pusher firing angle. Two AOA vanes mounted on the left and right sides of the forward fuselage provide the AOA signals for the SPS. If either AOA signal exceeds the shaker firing angle, the stick shaker on the corresponding control column will enga