Learjet 23 accident at Windsor Locks, Connecticut, 4 Jun 1984

Accident narrative

On June 4, 1984, Night Air 4, an unmodified Gates Learjet 23, crashed while on final approach to Bradley International Airport in Windsor Locks, Connecticut. When the airplane was about 200 feet over the approach lights, witnesses saw it level off, roll about 90 degrees to the right, and descend into the ground. The airplane was destroyed by impact forces and a postcrash fire. Both pilots and the single passenger on board were killed.

### The flight

Air Continental Inc. operated the airplane on a regularly scheduled cargo flight transporting cancelled bank checks under 14 CFR 135. The flight departed Cleveland Hopkins International Airport, Ohio, at 2200. To stay within maximum gross weight limitations, the airplane departed with full wing and tip tanks and minimum to zero fuel in the fuselage tank. After an uneventful flight, Night Air 4 arrived at Syracuse Hancock International Airport, New York, at 2245. Following routine ground cargo handling without refueling, the flight departed Syracuse at 2311.

The flightcrew was properly certificated and qualified. The captain had 11,039 total flying hours, with approximately 1,130 hours in Learjets. The first officer had 5,263 total flying hours, including 189 hours in Learjets. The passenger was a former employer of the captain and a guest on the flight.

The airplane's landing weight at Bradley was computed at 9,266 pounds, which required a landing approach speed of about 120 knots. Radar data indicated the approach speed actually flown was about 128 knots.

### Sequence of events

The en route portion of the flight was uneventful. At 2332, Boston Center handed Night Air 4 off to Bradley Approach Control at 16,000 feet. Approach control cleared the flight for a visual approach to runway 33 and provided a turn to position the airplane on final approach 10 miles from the airport. At 2338:22, the crew reported on final approach, and the control tower cleared the flight to land. Recorded radio transmissions revealed normal handling, and the co-pilot's voice was calm. Air Continental policy required the non-flying crewmember to handle the radios.

Weather observations at 2350 reported no ceiling, scattered clouds at 4,000 feet, 20 miles of visibility, and winds from 280 degrees at 8 knots. The approach and runway light systems were on low intensity settings, and the sequence flashing lights were off.

Fifteen witnesses described basically the same accident sequence. The airplane was on a normal approach to runway 33 with no apparent abnormalities. When it was about 200 feet over the approach lights, witnesses heard an increase in engine thrust and saw the airplane halt its rate of descent in what two pilot witnesses thought was an apparent attempt to go around. Immediately afterward, the airplane began a level turn to the right. The roll continued until the wings were nearly vertical to the ground. The nose then dropped below the horizon, and the airplane descended into the ground in a nose-low attitude. Witnesses reported an explosion followed by intense ground fire. None reported seeing the landing lights or the landing gear extended, and there was no inflight fire, smoke, or airframe separation.

### What the investigation found

The accident was not survivable because impact forces exceeded human tolerances. The crash site was adjacent to the airport fire station, and crash/fire/rescue response was immediate. The airplane was not equipped with a cockpit voice recorder or a flight data recorder, and neither was required. Ground navigation facilities were checked after the accident and found to be functioning satisfactorily.

The initial impact point was approximately 1,200 feet to the right of the runway 33 threshold. The airplane disintegrated and scattered over an area about 400 feet long and 200 feet wide. The initial impact damaged a chain link fence, and pieces of the right tip tank structure were found in a ground gouge just beyond it. The tail section was heavily damaged by fire and crushed against a telephone pole. The left wing was recovered inverted, while the right wing was broken apart and heavily damaged by fire. Postmortem examinations showed both crewmembers were seated in their assigned seats. Toxicological tests were negative.

Examination of the wreckage revealed no evidence of preaccident failure or malfunction of the aircraft structure or powerplants. Both engines were estimated to have been operating at 90 to 92 percent rpm at the time of the accident. The landing gear actuators were in the extended position, and the nose gear down lock balls were in the down and lock position.

The flaps were partially extended. The flap actuator was recovered in a position corresponding to about 34 degrees of extension out of a maximum of 40 degrees. The horizontal stabilizer actuator was trimmed nearly to the full airplane nose-up position. The Board noted this is not an abnormal position for landing, as it relieves back pressure on the control wheel, but it is also indicative of inflight spoiler deployment, since extending the spoilers causes a nose-down pitching moment. The rudder trim tab was in the faired position. The aileron trim actuator was recovered in a position corresponding to 9 degrees of trim tab deflection for left wing down, which the Board noted indicated the pilot attempted to counteract an uncommanded right roll using aileron trim.

The left wing spoiler actuator rod was found in the retracted position. The right wing spoiler actuator rod was extended 1.97 inches, corresponding to 39 degrees of spoiler extension. A laboratory inspection of the right spoiler actuator revealed a heavy band of discolored metal on the interior wall corresponding to the extended position, indicating the spoiler was extended while exposed to the postaccident ground fire. Based on the physical evidence and the airplane's reactions, the Safety Board postulated that the spoilers were asymmetrical at impact.

Examinations of the spoiler, flap, aileron trim, and automatic flight control systems did not indicate any potential source of inflight malfunctions. A small droplet of solder was found in the flow restrictor of the right spoiler actuator hydraulic line, but tests showed it did not affect flow rates. A review of Service Difficulty Reports since 1979 for Lear 20 series airplanes disclosed three reports concerning the spoiler system, but none indicated if operation had been affected in flight. The Board noted a leaking or broken hydraulic line would be the most likely cause of an asymmetric position, but postcrash fire damage prevented evaluating the accident airplane's lines.

Examination of the warning light bulbs indicated the left and right fuel pressure lights were illuminated at impact, which the Board attributed to the rolling maneuver uncovering the fuel pumps in the tanks. The spoiler warning light bulbs were off at impact. The Board noted that the impact sequence might have caused an electrical interruption, or the sensor wiring might have been defective.

### Analysis of the flight and controls

The Safety Board reviewed previous Learjet takeoff and landing accidents involving steep banking and loss of control. Common maneuvers in these accidents included steep banking with high roll rates immediately before loss of control, an inability of the flightcrews to recover, and the addition of engine thrust appearing to aggravate the severity of the bank attitude. In those investigations, the Board had cited factors such as ice accumulation, wake vortex turbulence, mistrimmed controls, and asymmetrical thrust. However, the Board found that none of these factors appeared to be present in this accident.

Flight tests conducted by the Safety Board in an unmodified Learjet 23 involved four profiles flown at Bradley International Airport. The first two profiles evaluated descent rates from FL200 to 10,000 feet at idle power. With spoilers retracted, the descent rate was 4,000 to 4,200 feet per minute. With spoilers extended, the vertical speed indicator pegged at 6,000 feet per minute, with a calculated rate of about 6,500 feet per minute. The tests indicated that the descent rate achieved by the accident airplane as it descended through 7,000 feet was achievable only with the spoilers extended. The Board believed that the pilot extended the spoilers during the turn to final approach to increase the rate of descent.

The Learjet flight training manual cautions pilots that spoilers and flaps should not be used simultaneously because of probable fatigue damage to the flap surface. The flight tests showed that maintaining level flight with the gear and flaps down and spoilers extended required about 90 percent engine power, which matched the engine rpm found at impact. The Board noted that there is little difference in cockpit background noise with spoilers extended or retracted when the gear and flaps are down, and that the spoilers-extended configuration requires only 3 percent more power than the retracted configuration. The Board postulated that a pilot could overlook this difference, particularly if the spoiler warning lights were not lit.

The Safety Board postulated that as the airplane passed over the approach lights, the pilot realized the spoilers were extended and retracted them. The left spoiler retracted, but the right spoiler did not, causing the airplane to roll to the right. While the pilot added power, he apparently did not use enough aileron and rudder control input to arrest the roll. Flight tests indicated that an asymmetrical spoiler condition is controllable if prompt and correct rudder and aileron inputs are made, but the Board noted that at night on final approach, delayed recovery inputs by an unwary pilot might not be sufficient to stop the roll. The Board could not determine the reason for the right spoiler malfunction.

### Probable cause

The National Transportation Safety Board determined that "the cause of the accident was an uncommanded roll to the right which caused the airplane to roll about 90° and descend into the ground. The cause of the uncommanded roll was an asymmetric retraction of the flight spoilers wherein the left spoiler retracted and the right spoiler did not. The Safety Board could not determine the reason for the right spoiler malfunction."