ERA14FA112

The pilot's failure to maintain airspeed with one engine inoperative, which resulted in a loss of control while on approach. Contributing to the accident were airframe ice accumulation due to conditions conducive to icing and the loss of engine power on one engine for reasons that could not be determined due to the extent of damage to the airplane.

The instrument-rated private pilot was conducting a personal cross-country flight in the multiengine airplane under instrument flight rules (IFR). As the flight neared its destination, the controller issued clearance for a GPS approach, and, shortly thereafter, the pilot informed the controller that he needed to review the approach procedure before continuing the approach. The controller acknowledged, and, after the pilot reported that he was ready to proceed with the approach, the controller again issued clearance for the GPS approach. Radar data showed that, during the approach, the airplane tracked a course that was offset about 0.5 miles right of the final approach course until it was about 1 mile from the runway threshold. The airplane then turned left towards the threshold and descended to an altitude of about 145 ft above ground level over the runway threshold before the pilot performed a missed approach. It is likely that the pilot performed the missed approach because he was unable to align the airplane with the runway before it crossed the threshold. The controller provided radar vectors for the airplane to return to the approach course and cleared the airplane a third time for the GPS approach to the runway. Radar data showed that the airplane was established on the final approach course as it passed the initial approach fix; however, before it reached the final approach fix, its airspeed slowed to about 111 knots, and it began a left turn with a 25 degree bank angle. About 18 seconds later, while still in the turn, the airplane slowed to 108 knots and began descending rapidly. The airplane's rate of descent exceeded 10,000 feet per minute, and it impacted the ground about 9 miles from the destination airport. Examination of the accident site showed that the airplane was severely fragmented and fire damaged with debris scattered for about 450 feet. Postaccident examination of the wreckage did not reveal evidence of any preimpact failures; however, damage to the left engine indicated that it was not producing power at the time of the accident. The severity of impact and fire damage to the airplane and engine precluded determination of the reason for the loss of left engine power. Weather conditions present at the time of the accident were conducive to super cooled liquid water droplets, and the airplane likely encountered moderate or greater icing conditions. Several pilot reports (PIREPs) for moderate, light, trace, and negative icing were reported to air traffic control but were not distributed publicly into the national airspace system, and there was no airmen's meteorological information (AIRMET) issued for icing. However, the pilot received standard and abbreviated weather briefings for the flight, and his most recent weather briefing included three PIREPs for icing conditions in the area of the accident site. Given the weather information provided, the pilot should have known icing conditions were possible. Even so, the public distribution of additional PIREPs would have likely increased the weather situational awareness by the pilot, weather forecasters, and air traffic controllers. The airplane was equipped with deicing and anti-icing systems that included wing and empennage deice boots and engine inlet heaters. Due to impact damage to the cockpit, the positions of the switches for the ice protection systems at the time of the accident could not be determined. Although the airplane's airspeed of 108 knots when the steep descent began was above its published stall speed of 77 knots, both bank angle and ice accretion would have increased the stall speed. In addition, the published minimum control airspeed was 93 knots. It is likely that, after the airplane passed the initial approach fix, the left engine lost power, the airplane's airspeed began to decay, and the asymmetric thrust resulted in a left turn. As the airspeed continued to decay, it decreased below either stall speed or minimum control airspeed, and the airplane entered an uncontrolled descent.

Radar Performance Study A performance study of the airplane's flight path was created by an NTSB airplane performance specialist. The study revealed that the airplane's airspeed was nominally 150 knots during the approach prior to the accident and showed good acceleration after the pilot declared a missed approach. During the accident approach, the airplane's airspeed had slowed to 111 knots, when it began a left 25 degree turn away from the final approach course, and 18 seconds later, while still in the turn, the airplane slowed to 108 knots and descended at over 10,000 feet per minute until impacting the ground. The study further noted that while the airplane's published stall speed for straight and level flight with the flaps and gear retracted was 77 knots, the stall speed would increase to approximately 108 knots in a 60 degree level turn. In addition, all speeds assumed a clean wing with no ice accretion. The airplane pilot operating handbook listed the minimum control airspeed as 93 knots. An autopsy was performed on the pilot by the Office of the Medical Examiner, Center for Forensic Medicine, Nashville, Tennessee. The autopsy report listed the cause of death as "multiple blunt force injuries." Toxicological testing was performed on the pilot by the FAA Bioaeronautical Science Research Laboratory, Oklahoma City, Oklahoma with no anomalies noted. The airplane's impact with the ground created an 11-foot-long, 11-foot-wide, 6-foot-deep impact crater. Broken tree branches that contained 45-degree angled cuts were observed at a height about 50 feet. The airplane impacted the ground at an approximate 70-degree angle, consistent with being in an inverted position. It was severely fragmented with debris scattered on a course about 320 degrees, for about 450 feet. In addition, a postcrash fire consumed a majority of the airframe. Portions of both outboard wings, the nose section, empennage and all flight control surfaces were located at the accident site; however, fragmentation of the wreckage precluded the ability to confirm control continuity to the respective flight control surfaces. A portion of the outboard left wing approximately 15 feet in length from the wing tip was located, as well as fragments of the left and right wing that exhibited crushing completely to the rear spar assemblies. The left main landing gear was in the retracted position, the right and nose landing gear were separated from their surrounding structure. Due to impact damage to the cockpit, the positions of ice protection system switches at the time of the accident could not be determined. Portions of the deice boot system were observed on wing debris and portions of the horizontal stabilizer. The boots were destroyed by impact forces and fire damage. Both propellers remained attached to their respective gearboxes, which separated from their respective engines. All three left propeller blades separated from the hub. Two right propeller blades remained attached to the hub, and one blade had separated. Both propeller assemblies were severely impact damaged and displayed evidence of rotational scoring; however, it was noted that the right propeller blades displayed significantly greater degree of rotational scoring, tears, and missing blade tips, then the left propeller blades. Both engines were impact and fire damaged. Their respective fuel pumps and fuel control units were separated. They did not display any evidence of catastrophic failure and were forwarded to the engine manufacturer for further examination under the supervision of an NTSB investigator. A subsequent teardown examination of both engines did not reveal any preimpact conditions that would have prevented normal operation. The type and degree of damage to the left engine was indicative of an engine that was not operating, with rotation consistent with a wind milling propeller at the time of impact. Static impact marks were observed on the first and second stage centrifugal compressor shrouds and no evidence of rotational scoring was noted on the turbine section. The type and degree of damage to the right engine was indicative of an engine that was operating under power at the time of impact. Rotational scoring was observed on the first and second stage centrifugal compressor shrouds and the first, second, and third stage turbine shroud. Almost all of the vanes on the first stage centrifugal compressor impeller were deformed and or separated. In addition, the forward curvic coupling of the second stage centrifugal compressor impeller was heavily smeared. The left engine fuel control unit was examined at Woodward, Rockford, Illinois, under the supervision of an FAA inspector. The fuel control unit was extensively damaged and could not be functionally tested. Subsequent disassembly did not reveal evidence any preimpact malfunctions. Computed Tomography scans of the left and right engine fuel shutoff valves revealed that they were both in the closed position; however, their respective preimpact position could not be confirmed. A subsequent teardown of the valves under the supervision of an NTSB investigator did not reveal evidence of any preimpact failures. The left engine fuel shutoff valve solenoids could be opened and closed by applying electrical power directly to the solenoids. The right engine fuel shutoff valve solenoids did not function when electrical power was applied. Examination of a vertical and directional gyro recovered from the wreckage was performed by the NTSB Materials Laboratory. The examination revealed that although both gyros were impact and fire damaged, internal damage was consistent with rotation at the time of the accident. A weather observation taken at JWN, which was located 9 miles north-northeast of the accident site, at an elevation of 495 feet, at 1655, reported wind from 360 degrees at 5 knots; visibility 5 statute miles; overcast ceiling at 800 feet; temperature 41 degrees F; dew point 25 degrees F; altimeter 30.29 inches of mercury. AIRMET Sierra issued at 1445, was valid at the time of the accident, and forecasted IFR conditions around the accident site with ceilings below 1,000 feet and visibilities below 3 miles. There were no AIRMETS for icing conditions valid at the time of the accident. The pilot received standard and abbreviated weather briefings from Lockheed Martin Flight Service. The last weather briefing requested by the pilot was at 1538, and included three pilot reports (PIREPs) for icing conditions in the Nashville area that were applicable to the pilot's flight. The National Weather Service (NWS) surface analysis chart around the time of the accident depicted a frontal system located across the southeastern United States with a surface trough in the vicinity of the accident site. Station models around the accident site depicted air temperatures in the mid-20's to mid-30's F, with temperature-dew point spreads of 3 degrees F or less, a north to northeast wind less than 10 knots, cloudy skies, and fog. Geostationary Operational Environmental Satellite number 13 (GOES-13) data indicated abundant cloud cover over the accident site with approximate cloud-top heights of 19,500 feet around the time of the accident. A review of pilot weather reports (PIREPs) that were publicly available in the National Airspace System (NAS) for the vicinity of the accident site revealed that, from about 3 hours before the accident to about the time of the accident, there were seven PIREPs that contained icing information that ranged from trace rime to a light to moderate mixed icing, with the reported icing conditions only occurring between 2,000 and 3,500 feet. Review of Current Icing Potential (CIP) images produced by the NWS Aviation Weather Center depicted light to moderate icing was likely at 2,000 to 3,000 feet around the time of the accident. It was noted that CIP data was intended to be supplemental to other icing advisories (e.g. AIRMET