ERA13FA083

The noninstrument-rated pilot’s improper decision to depart in dark, night marginal visual flight rules conditions, which resulted in his spatial disorientation and subsequent loss of airplane control.

Before departing at night for his destination airport, the noninstrument-rated pilot received a weather briefing, which advised of marginal visual flight rules (MVFR) conditions. The briefing also included an airmen's meteorological information advisory for developing instrument flight rules conditions due to low ceilings and mist. Shortly after takeoff, the pilot contacted a radar approach controller for visual flight rules flight-following services, and he was advised to squawk a beacon code, but, before the approach controller was able to identify the airplane on the radar, the pilot radioed, "I'm in trouble." Shortly after, both radar and radio contact were lost. Review of radar data indicated that the airplane's climb rate was steady until the airplane reached an altitude of about 2,100 ft msl. The airplane then began descending rapidly while turning right until it impacted terrain. Examination of the wreckage did not reveal any evidence of preimpact failures or malfunctions of the engine or primary flight controls. However, examination of the elevator trim system revealed that the elevator trim cable was disconnected from the trim control wheel in the cockpit and that it had pulled out of a swaged rod end (bolt), which displayed a longitudinal crack on the outer surface of the swage. Examination of the fracture surface revealed that the fracture occurred due to overstress. The examinations were not able to determine if the cable pulled out of the fitting during the accident sequence or if it was a pre-existing condition. Regardless, review of the elevator pitch control system revealed that, even if the elevator trim cable had disconnected in flight, it should not have led to an uncontrollable situation due to its redundant design. At the time of the accident, both the sun and the moon were more than 15 degrees below the horizon. Further, warm, moist southerly wind ahead of an approaching cold front was producing variable clouds, and a band of low stratiform clouds with their tops near 4,000 ft existed over the area. Operating in MVFR conditions increases a pilot's workload and stress level because navigation becomes more difficult and reduces the margin of safety. As a result of the increased workload and stress level and the pilot's minimal simulated instrument time (about 7 hours); his minimal night experience (about 3 hours); the dark, night MVFR conditions; restricted visibility, including a lack of ambient light; and the sustained right turn and descent, it is likely the pilot experienced spatial disorientation and subsequently lost control of the airplane.

Weather Briefing Review of the outlook weather briefing requested by the pilot from the Princeton Contracted Flight Service Station (FCFSS) revealed that prior to departure; he had been advised of AIRMET Sierra. Additionally, he had also been advised of the weather conditions that had been reported in the Valdosta area which included scattered clouds at 1,600 feet, broken clouds at 2,200 feet, and broken clouds at 4,300 feet. Furthermore, He was also given the temperature and dew point which at the time were within 4 degrees of each other, and was advised that it was marginal VFR, which is defined by the NWS as a ceiling between 1,000 and 3,000 feet or visibility in the three- to five-mile range. Radar Data Review of correlated radar data indicated that at 19:46:13.11, the airplane's climb rate was steady until reaching an altitude of approximately 2,100 feet msl. Approximately 7 seconds later the airplane's altitude dropped to 1,900 feet msl, and the airplane had begun to turn right. At 19:46:22.927, the airplane was still turning to the right and had descending through 1,700 feet msl. Approximately 5 seconds later, it was still continuing to turn right, and was at 1,400 feet msl. Further examination of the radar data, indicated that the last radar contact occurred at 19:46:32.507, when the airplane was still at 1,400 feet msl. The average rate of descent up to that point was approximately 3,420 feet per minute. Comparison of the last radar contact to the location of the accident site, indicated that the last radar contact had occurred when the airplane was approximately 359 yards from its initial impact point with the trees. Elevator Trim System The DA 40 elevator trim system included a mechanically operated trim tab. This allowed the pilot to trim the airplane for different speeds and center-of-gravity positions. The elevator trim system had three main parts: - The handwheel assembly with trim indicator. - The Bowden Cable (Elevator Trim Cable) which connected the handwheel to the trim tab. - The trim tab actuator assembly. The handwheel assembly on the center console controlled the elevator trim system. The assembly had a metal mounting frame. The frame attached to the rear of the engine control assembly and the top of the control bulkhead. A long bolt through the mounting frame carried the handwheel. The bolt also held friction disks, plain washers, and spring washers, which were mounted against the handwheel. Two jam-nuts could be used by the pilot to adjust the friction by applying compression to the friction disks and washers. A small gear wheel attached to the handwheel. The small gear wheel engaged with a large gear segment with internal teeth. The gear segment had a pivot bolt at the bottom of the mounting frame. A ball-stud attached the eye-end of the Bowden cable to the gear segment. An extension to the mounting frame to the rear made the anchor point for the outer sheath of the cable. The gear segment was also the trim indicator. The top face of the segment had a white line across it midway between the front, and back. The top face could be seen through a slot in the cover plate. The sides of the cover plate had markings to show the trim position. The Bowden cable connected the trim handwheel assembly to the trim tab. The cable went through holes in the front and rear main bulkheads, the baggage frame and each of the ring frames. It then went up the front face of the front web of the vertical stabilizer and through a slot near the top, and then through a large hole at the top of the rear web of the vertical stabilizer to the trim tab actuator assembly. The cable had an inner core with threaded end fittings. Spherical end fittings at each end connected to the gear segment and trim tab actuator assembly. Clamp blocks held the outer core to the mounting frame at the front and a bracket from the horizontal stabilizer at the back. The trim tab was a one-piece FRP molding. The tab had two integral levers. Two cranked actuating levers were attached to the integral levers. The left cranked actuating lever connected to the Bowden cable, and the right actuating lever connected to a friction damper. The friction damper had a clamp-block with a hole for a rod. The rod connected to the right actuating lever on the trim tab. The friction on the rod in the clamp block was adjustable. When the top of the trim handwheel moved forward: - The small gear wheel moved the top of the gear segment forward. - The gear segment pulled the inner core of the flexible cable forward. - The inner core of the flexible cable pulled the left cranked actuating lever forward. - The left cranked actuating lever pulled the trim tab lever forward to move the tab up. - The up movement of the trim tab would push the elevator down in flight giving nose-down trim. When the top of the handwheel was moved aft, the gear segment moved aft, the cable moved aft and the trim tab would move down. This would push the elevator up and give nose-up trim. In each case, the pilot could see the trim position from the white mark on the gear segment. Laboratory Examination of the Elevator Trim Cable and Swaged Rod End Examination of the elevator trim cable (Bowden cable) and swaged rod end from the accident airplane by the NTSB Materials Laboratory revealed that the core cable consisted of the cable strand with a plastic coating that was removed in the area where the bolt was swaged onto it. Examination of the cable strand revealed twelve outer wires that exhibited bands of deformed material along their outermost surfaces. The strand contained nineteen wires in the configuration of 7 inner wires and 12 outer wires (e.g. 7 x 12, 19-wire strand). The average strand outside diameter was about 0.142 inch (3.6 mm). Measured by digital microscopy, the average wire diameter was about 0.027 inch (0.7 mm). The outer wires were wound in a right lay. Manufacturer's drawings indicated that the strand had a diameter of 3.5 mm (0.138 inch) and consisted of nineteen zinc-coated wires, each with a diameter of 0.7 mm (0.027 inch). The drawings however illustrated and specified a left lay. The ball end and the nut were removed from the bolt and measurements revealed that the bolt satisfied the dimensional requirements in the manufacturer's drawings. The drawings specified the bolt material as DIN 1.4305 which is compositionally equivalent to UNS S30300, and specified the bolt material as stainless steel. Examination revealed that the swaging tool during manufacture had pressed five flat regions equally spaced around the circumference of the swage. Ribs of extruded material were formed between the flat swaging jaws. Further examination revealed that the fracture had occurred along one of the ribs of material that was extruded between the swaging jaws during manufacture. Brown-colored deposits were present in the fracture. Examination of the inner surface of the bolt swage area adjacent to the fracture revealed circumferential lines consistent with a drilling operation and helical impression marks consistent with contact with the cable wires. The deposits in the crack were evaluated by scanning electron microscopy (SEM) and standardless semi-quantitative energy dispersive spectroscopy (EDS). The EDS spectra, revealed the presence of carbon, oxygen, iron, zinc, magnesium, aluminum, silicon, phosphorous, sulfur, chlorine, potassium, calcium, chromium, manganese, and nickel. The fracture surface of the swage area was also examined by SEM and a typical scanning electron fractograph which revealed that the fracture micro-mode was microvoid coalescence (MVC) due to overstress. The longitudinal orientation of stringers within the fracture surface, were consistent with a resulfurized steel such as UNS S30300. The inside surfaces of the bolt swage area exhibited helical impression marks (e.g. grooves) consistent with permanent deform