Final Report: Causes of loss-of-control accidents undetermined
CESSNA CARAVAN 208B, ROCKFORD, ILL., DEC. 17, 2002–Caravan N277PM crashed on a night ILS approach to Runway 7 at the Greater Rockford Airport (RFD) following a loss of control. The pilot was killed and the airplane was destroyed. The NTSB determined that the probable cause of the accident was the pilot’s failure to maintain control of the airplane during the approach. Factors were the low ceilings, high winds, crosswind and wind-shear conditions.
The airplane hit trees and the ground approximately 2.1 miles from the approach end of the runway at 10:51 p.m. A witness said the airplane was at “mid-throttle” as it flew over and he then heard the power increase before the sounds of impact. He reported no precipitation but “severe winds, mostly from the south, shifting [and] volatile directly from the east,” and said visibility was “extremely poor.”
Three of the five pilots who landed transport-category airplanes around the time of the accident experienced a crosswind that varied between 15 and 50 knots during approach. Four reported airspeed fluctuations that varied between +8 knots and +10 knots. Three broke out of the clouds between 200 and 300 feet agl. None of the flight crews encountered icing conditions.
Radar data indicates that the Caravan was approximately 500 feet too high at the outer marker, and the turboprop single then entered a rapid descent from an altitude of about 2,300 feet at approximately three nautical miles from the approach end of the runway. Examination of the airframe, engine and propeller governors failed to reveal any failures or malfunctions that would have resulted in loss of control.
The Part 135 nonscheduled cargo flight, operated by Planemasters as Flight 1627 (PMS1627), was carrying cargo for United Parcel Service (UPS). The aircraft was operating in IMC on an IFR flight plan and had departed Decatur, Ill., at 9:54 p.m.
At 10:43:59, the RFD approach controller asked the pilot of PMS1627 if he could fly at 170 knots. The pilot responded, “Ah, be pushing it sir.” The controller then asked if he could maintain 150 knots. The pilot responded in the affirmative. After a caution about wake turbulence at 10:49:30, the aircraft was cleared for the approach. At 10:50:58 the pilot contacted the tower. The local controller reported the wind as 140 degrees at 18 knots and cleared the aircraft to land. The pilot’s acknowledgement of this transmission was the last radio contact with PMS1627.
Local weather at 10:54 p.m. was wind 110 degrees at 17 knots, gusting to 20 knots; visibility 1.25 miles; light rain and mist; clouds 300 feet overcast; temperature 2 degrees C; dew point -1 degree C; and barometer 29.73 inches. A warm front extended across the area, with a band of southerly low-level wind at 40 knots. Nearby weather stations reported wind from the south at 30 to 40 knots, an approximate 90-degree shift from surface winds.
The pilot had been employed at Planemasters since March 2002 and had 1,841 hours TT–1,686 as pilot-in-command, 1,497 hours in make and model, 257 hours of actual instrument time and 1,065 hours of night flight time. He received recurrent training on Aug. 21, 2002.
GATES LEARJET 24D, SIERRA BLANCA, TEXAS, DEC. 10, 2001–Learjet N997TD, operating as Turbodog 36, was destroyed when it experienced loss of control during descent to its destination and crashed at 6:21 p.m. The NTSB determined the probable cause of the accident was loss of control during descent for undetermined reasons. Both the ATP captain and commercial copilot were killed. The flight crew had established a 4,000-fpm descent from FL390.
As the airplane descended through FL219, ATC asked the pilot to contact approach control. The pilot read back an incorrect frequency and spoke an unintelligible word. The controller attempted to correct the pilot, but no additional communication was received. Within a pause in the pilot’s last transmission, a 1680-Hz frequency noise could be heard for 0.1 seconds. Only two systems in the airplane had aural warning systems within that frequency range–the cabin altitude warning and the overspeed warning (both were destroyed in the accident).
Soon after the pilot’s last transmission, radar data depicted the airplane climbing to FL231 before entering a steep and rapid descent (approximately 42,000 fpm). Just before the loss of control, the airplane exceeded its maximum operating airspeed of 300 kcas. (According to the manufacturer, the airplane has been successfully flown at airspeeds up to 400 knots without loss of control.)
The right wing and sections of the right horizontal stabilizer/elevator separated from the airplane just before the crash and were located approximately 200 to 250 feet from the main impact crater. No anomalies with the airframe or engine were found that would have led to the loss of control. A cockpit voice recorder was installed in the accident airplane; however, it did not record the accident flight.
The airplane was registered to and operated by Air Cargo Express of Fort Wayne, Ind. Night VMC prevailed, and an IFR flight plan was filed for the Part 91 positioning flight. The flight originated from Harlingen Valley International Airport (HRL), Texas, at 5 p.m. and was destined for El Paso International Airport (ELP). Weather was not determined to be a factor.
The accident site consisted of an impact crater measuring 17 by 30 feet. The main crater was excavated to a depth of six feet when a rock layer was met. The southeast quadrant of the crater was excavated four feet until undisturbed dirt was met, and the other quadrants were excavated six feet until undisturbed dirt was met. Aircraft debris was located in a fan distribution emanating outward from the crater to the north-northwest and ending approximately a quarter mile away. Both engines were found within the crater, and no significant portions of the cockpit were recovered. What remained of the flight control cables were examined, with no anomalies noted.
The NTSB Vehicle Recorders Lab examined the pilot’s radio transmissions for differences in the final transmission from earlier transmissions that might indicate changes in the pilot’s psychological state. Mistakes in communication can be a sign of impairment or distraction. The pilot’s only mistaken readback was in his final transmission, stating “nineteen five” when repeating his new radio clearance of “one one niner point one five,” followed by a two-syllable unintelligible word. Overall observations indicated there were no dramatic speech changes when compared with earlier transmissions. However, the pilot was a little slower at responding to the controller, and spoke somewhat slowly. The reason for the differences could not be determined from the available data.
As an airplane approaches the speed of sound, shockwaves will form and transonic effects begin to develop, such as drag rise, buffet and longitudinal stability changes, also known as Mach tuck, and control-surface buzz. Mach tuck is a term used to describe a nose-down pitch tendency in the transonic speed range (Mach 0.75 to 1.20), when a decrease in downwash on the horizontal tail will create a diving moment and the aircraft will “tuck under.” Some aircraft, including the Learjet 24D, use a stick-puller system, which will cause the aircraft to climb in the event of an overspeed.