A crash that killed two Hageland Aviation pilots during a training flight near Kwethluk in 2014 may have begun as a training maneuver, according to a new report.
Wednesday’s National Transportation Safety Board factual report and docket of documents on the April 8, 2014, crash that killed Derrick Cedars, 42, and Greggory McGee, 46, said they were the Cessna 208 Caravan’s instructor and training pilot respectively. Cedars had more than 14,000 hours of flight time, including nearly 6,000 in Caravans, while McGee had 593 hours of flight time with none in Caravans.
The report was the board’s second on the same day involving a fatal Hageland Aviation flight, following the release of a factual report on the Nov. 29, 2013, crash that killed four people near St. Mary’s.
In the 2014 crash, investigators said, Cedars and McGee took off from Bethel at about 3:35 p.m. About 21 minutes later, data being transmitted by the plane’s ADS-B telemetry system tracked it crashing roughly 22 miles southwest of Kwethluk. Alaska State Troopers found both pilots dead at the site that evening.
No communications were received from the plane during the crash.
The Bethel airport, 27 miles northeast of the crash, reported winds from the north at 8 knots and clear skies three minutes before the Cessna's crash.
Johnathan Kapsner, a training pilot checked by Cedars who later became an instructor pilot at Hageland himself, told the NTSB that he and Cedars would typically start a training pilot’s stall maneuvers at an altitude of about 3,500 feet, with turns left and right before the plane entered a descent.
“The accident flight followed a typical routine and flight pattern from previous training flights,” investigators wrote. “The accident sequence initiated during the time when a simulated emergency and descent was typically initiated.”
Another Hageland training pilot, Jaime Burns, told investigators that after an emergency descent to 500 feet, Cedars had him descend further to an altitude of 200 feet and follow a waterway -- a maneuver he called “running the river.”
“He said that although he hadn’t done a low-flying maneuver like running the river before, he at no time felt unsafe,” investigators wrote.
A preliminary 2014 report on the Kwethluk crash said that the plane had experienced “altitude deviations” leading to a “rapid, steep descent” from an altitude of about 3,400 feet -- a pattern replicated by an NTSB computer simulation based on telemetry data from the crash and a physics model of the plane, which indicated instabilities beginning at 3:56 p.m.
“At this point, the airplane began a steep descent, which continued until impact,” investigators wrote. “The simulation descent rate from (3:56 p.m.) to the time of impact steadily increased to a maximum of about 16,000 feet per minute. The elapsed time from the initial upset to the point of impact was about 22 seconds.”
The NTSB gave heightened scrutiny in the report to the plane’s trim system, which was controlled automatically in normal operations by the plane’s autopilot. Investigators noted a warning in the manual for the Cessna’s trim system that “up to 45 pounds of force on the control wheel may be necessary to hold the aircraft level” in the event of an autopilot failure.
The board’s simulations of two possible scenarios for the crash indicated that moving the plane’s elevators would've required applying up to 130 pounds of pushing force by the time of impact to its central column in a “normal” scenario -- or 800 pounds of pulling force in a “runaway pitch trim” scenario.
When investigators visited the crash site, on land at a bend in the Kwethluk River, they found a 200-foot debris path from the plane as well as initial impact sites which suggested the plane hit the ground at a roughly 30-degree angle.
“Numerous sections of the severely fragmented airplane were located throughout the wreckage path,” investigators wrote.
None of the plane’s instruments were recoverable, in part due to a fire after the crash that consumed much of the cockpit and fuselage.
“The accident was not survivable,” investigators wrote.
Original article can be found here: https://www.adn.com
NTSB Identification: ANC14FA022
14 CFR Part 91: General Aviation
Accident occurred Tuesday, April 08, 2014 in Kwethluk, AK
Aircraft: CESSNA 208B, registration: N126AR
Injuries: 2 Fatal.
NTSB investigators either traveled in support of this investigation or conducted a significant amount of investigative work without any travel, and used data obtained from various sources to prepare this aircraft accident report.
On April 8, 2014, about 1557 Alaska daylight time, a Cessna 208B Caravan airplane, N126AR, was destroyed after impacting terrain about 22 miles southeast of Kwethluk, Alaska. The airplane was being operated by Hageland Aviation Services, Inc., dba Ravn Connect, Anchorage, Alaska, as a visual flight rules training flight under the provisions of 14 Code of Federal Regulations (CFR) Part 91. The two crewmembers on board were fatally injured. Day, visual meteorological conditions prevailed at the time of the accident, and company flight-following procedures were in effect. The local training flight departed from Bethel Airport, Bethel, Alaska at 1535.
The flight was intended to be the first training flight of the newly hired second-in-command (SIC) pilot. The SIC was operating from the airplane's right seat during the training.
About 1745, personnel from Hageland Aviation in Bethel notified the Hageland Operational Control Center (OCC) in Palmer, Alaska, that the airplane was overdue. At 1754, the OCC called the Kenai Flight Service Station to initiate search and rescue operations. A company airplane was dispatched from Bethel to assist in the search, and, at 1839, the pilot of that airplane visually confirmed that the accident airplane had crashed. The Alaska State Troopers in Bethel, assisted by the Alaska Army National Guard, arrived at the accident scene at 2105 and confirmed that both pilots had died.
The National Transportation Safety Board (NTSB) investigator-in-charge, an inspector from the Federal Aviation Administration (FAA) Anchorage Flight Standards District Office, an investigator from Textron Aviation, and a representative of the operator traveled to the accident scene on the morning of April 10, 2014.
The check airman, age 42, held an airline transport pilot certificate with a multiengine land rating and commercial privileges for airplane single-engine land. His most recent FAA first-class airman medical certificate was issued on February 25, 2014, without limitations. His most recent FAA 14 CFR 135.293 and 135.297 proficiency checks were dated March 15, 2014, with approvals for single pilot, instrument flight rules, and lower-than-standard takeoff minimums. His last reported flight time to the company indicated that he had a total flight time of 14,417 hours, with 5,895 hours in the accident airplane type. The check airman's personal logbooks were not located.
The SIC, age 46, held a commercial pilot certificate, with single-engine land, multiengine land, and instrument ratings. He also held a single-engine airplane flight instructor certificate and an advanced ground instructor certificate. His most recent FAA second-class medical certificate was dated May 30, 2013, with a limitation that the pilot must wear corrective lenses for near and distant vision. He reported on his pilot qualification form that he had a total flight time of 593 hours, with no flight time in the accident airplane type. The SIC's personal logbooks were not located.
The accident airplane was a Cessna 208B Caravan, registration number N126AR, serial number 208B1004, manufactured in 2002. At the time of the accident, the airplane had accumulated 11,206 total flight hours and was maintained under an approved aircraft inspection program. The most recent inspection of the airframe and engine was completed on March 13, 2014.
The airplane was equipped with a Pratt & Whitney PT6A-114A turbine engine that was rated at 675-shaft horsepower. The engine was overhauled 4,286 hours before the accident.
Electric Trim System
The airplane was equipped with a Bendix/King KFC-150 Autopilot System that incorporated a three-axis autopilot and an electric pitch trim system, which provided autotrim during autopilot operation and electric trim to the pilot. According to the Bendix/King supplement to the Cessna 208 Operating Handbook, the trim system is designed to withstand any single in-flight malfunction. Trim faults are visually and aurally annunciated.
In Section 3 of the flight manual supplement, the emergency procedures for an electric trim malfunction (either manual electric or autotrim) were as follows:
1. A/P DISC/TRIM INTER Switch – PRESS and HOLD throughout recovery. 2. ELEV TRIM Circuit Breaker – PULL OFF. 3. Aircraft – RETRIM manually. WARNING – When disconnecting the autopilot after a trim malfunction, hold the control wheel firmly; up to 45 pounds of force on the control wheel may be necessary to hold the aircraft level.
The supplement also provided the following information:
MAXIMUM ALTITUDE LOSS DUE TO AUTOPILOT MALFUNCATION
• Cruise, Climb, and Descent - 500 ft • Maneuvering - 100 ft • Approach - 100 ft METEOROLOGICAL INFORMATION
The nearest official reporting station was Bethel Airport, located about 27 miles northeast of the accident site. At 1553, about 3 minutes before the accident, a meteorological aerodrome report was reporting wind from 020 degrees (true) at 8 knots, visibility 10 statute miles, sky condition clear, temperature 19 degrees F, dew point 3 degrees F, and altimeter 28.87 inches of Mercury.
There were no communications with the accident airplane at the time of the accident.
The accident airplane was not equipped, nor was it required to be equipped, with a cockpit voice recorder or a flight data recorder.
Automatic Dependent Surveillance-Broadcast (ADS-B) Tracking and Recording
The airplane was equipped with ADS-B technology. In typical applications, an airplane equipped with ADS-B uses an ordinary GPS receiver to derive its precise position from the Global Navigation Satellite System constellation and then combines that position with any number of aircraft parameters, such as speed, heading, altitude, and flight number. This information is then simultaneously broadcast to other aircraft equipped with ADS-B and to ADS-B ground or satellite communications transceivers, which then relay the aircraft's position and additional information to air route traffic control centers (ARTCC) in real time.
A review of ADS-B data received by the Anchorage ARTCC showed that, after departure, the airplane flew northeast and began a series of maneuvers with track, altitude, and speed variations consistent with a training flight. About 21 minutes into the flight, when the airplane was about 3,400 ft mean sea level (msl), a slight fluctuation in altitude, followed by an initial upset, occurred. The airplane continued a rapid and steep descent until ground impact.
WRECKAGE AND IMPACT INFORMATION
The accident site was situated on a land thumb in a bend of the Kwethluk River at an elevation of about 75 ft msl. Ground scars extended from an area of frozen tundra and through an area of heavy willows, among which the airplane was found resting upright supported by a number of toppled willow trees. An extensive postcrash fire consumed the majority of the airplane's fuselage; the worst fire damage was situated near the forward fuselage and cockpit area.
The initial impact crater and wreckage path were oriented on a heading of about 128 degrees magnetic. Fragments of the belly pod structure and belly pod contents were sprayed forward from the initial impact point and scattered along the wreckage path. One of the main landing gear tires was the farthest piece of wreckage, located about 330 ft ahead of the main wreckage. Initial ground impact scars were about 200 ft west of the main wreckage. The fuselage was oriented on a heading of about 250 degrees magnetic.
An area of topped willow trees just behind and adjacent to the initial ground impact was determined to be the initial point of impact with an object. Initial strikes suggest an approximate 33-degree nose-down attitude at impact. A mark in the tundra that was about the same distance from the airplane centerline as the left main landing gear was also noted at the main impact area. A slight discoloration that resembled the shape and size of the left wing was also present at the initial impact area.
The propeller was separated from the airplane and located between the airplane and initial impact crater. Two of the three blades were loose in the hub, and one blade was fractured inside the hub and separated. Two blades exhibited torsional and aft bending. The separated blade was missing about 12 inches of its outboard section and had several large gouges in its leading edge.
Numerous sections of the severely fragmented airplane were located throughout the wreckage path. (See the wreckage plot and GPS coordinates for documented major portions of the wreckage in the public docket for this accident).
The main wreckage area consisted of the empennage, main fuselage, and cabin. The empennage was severely damaged during the impact but was relatively free of fire damage. The main fuselage, cabin, and cockpit area were mostly consumed by the postcrash fire. No identifiable instruments, gauges, or other equipment were recovered in the main fire-damaged area. Several pieces of instruments and the instrument panel were located throughout the wreckage path, but their condition at the time of impact could not be determined.
Both wings were separated from the fuselage and were located just west of the main wreckage area. Both wings had severe impact and thermal damage. The landing gear and wheel assemblies were separated from the fuselage and were fragmented.
The engine was separated from the fuselage and was located about 67 ft east of the main wreckage. The engine case was crushed, and several portions of the case and accessories were fragmented or separated.
MEDICAL AND PATHOLOGICAL INFORMATION
A postmortem examination was conducted on the check airman under the authority of the Alaska State Medical Examiner, Anchorage, Alaska, on April 11, 2014. The cause of death for the pilot was attributed to multiple blunt force injuries.
The FAA Civil Aerospace Medical Institute (CAMI) performed toxicological testing of the check airman on May 12, 2014, which was negative for carbon monoxide, drugs, and ethanol.
A postmortem examination was conducted on the SIC under the authority of the Alaska State Medical Examiner on April 10, 2014. The cause of death for the pilot was attributed to multiple blunt force injuries.
The FAA CAMI performed toxicological testing of the SIC on April 20, 2014, which was negative for carbon monoxide and drugs. The testing detected 14 mg/dL, mg/hg ethanol in the SIC's muscle and 10 mg/dL, mg/hg ethanol in his liver. CAMI's review of the toxicology results was unable to determine if the ethanol was from ingestion or from postmortem sources.
The accident was not survivable.
TEST AND RESEARCH
Postaccident Airframe Examination
The NTSB Airworthiness Group convened to examine the airplane wreckage at the facilities of Alaska Claims Services on April 22 and 23, 2014. The wreckage was highly fragmented, and a substantial portion of the fuselage was consumed by the postcrash fire. The remaining major portions of the airplane were laid out in a hangar to facilitate examination.
The flap jackscrew was recovered in two pieces, and the screw was fractured at the ball nut. The jackscrew extension measured 6.5 inches, which indicated the flaps were in the "up" position at the time of impact.
The empennage was recovered in two major pieces that were significantly damaged. The largest piece consisted of the empennage structure from the canted bulkhead at fuselage station (FS) 436.68 aft to the end of the airplane at FS 509.50, the vertical stabilizer and rudder, the right horizontal stabilizer and elevator, and the left horizontal stabilizer to stabilizer station (SS) 80. The smaller piece consisted of the left horizontal stabilizer from SS 80 to SS 123 and the entire left elevator.
The right elevator and trim tab remained attached to the right horizontal stabilizer and had significant crushing damage. The right elevator torque tube separated from the center fitting along the rivet line. The right elevator trim tab actuator extension measured 2.2 inches, which equates to about 8 degrees tab up. The left elevator remained attached to the outboard section of left horizontal stabilizer. The left elevator torque tube separated from the center fitting along the rivet line. The left elevator trim tab actuator extension measured 2.4 inches, which equates to about 14.5 degrees tab up. The right and left elevator trim cables were cut about 4 ft forward of the canted bulkhead during recovery. The right elevator trim cable was continuous from the cut to the right pulley in the horizontal stabilizer and to the chain at the right trim tab actuator. The balance trim cable was attached to the chain at the right trim tab actuator and ran through the center section of the horizontal stabilizer to about left SS 80, where it was cut during recovery. The left trim tab actuator chain remained on the actuator. The left elevator trim tab cable was cut at the chain and forward of the canted bulkhead and was pulled free of the empennage.
The rudder cables remained attached to the bell crank at the lower end of the rudder, and the rudder autopilot cables remained fastened to the rudder cables aft of the canted bulkhead. The rudder and rudder autopilot cables were cut about 4 ft forward of the canted bulkhead during recovery, but all had individual broken strands and were kinked in several places. The elevator cables and one of the elevator autopilot cables remained attached to the control horn. The upper elevator autopilot cable was broken at the aft cable end, and the cable end remained attached to the control horn. The lower elevator autopilot cable was broken about 6 ft forward of the canted bulkhead and had a splayed appearance consistent with overload failure. The two elevator cables were cut about 5 ft forward of the canted bulkhead during recovery, but all had individual broken strands and were kinked in several places. The elevator push-pull tube was fractured at the forward end and forced forward into the control horn. The aft end of the push-pull tube remained attached to the bell crank at the center elevator torque tube fitting. There was no evidence of binding or damage on the elevator push-pull tube.
All of the examined fractures had a dull, grainy appearance consistent with overload failure. No evidence of preexisting corrosion or cracking was found on any of the examined parts. No evidence of preimpact fire or bird impact was observed on any of the examined wreckage. (See the Airworthiness Group Factual report and appendixes in the public docket for additional information.)
Aircraft Performance ADS-B and Simulation Study
The NTSB's Office of Research and Engineering conducted an aircraft performance ADS-B and simulation study using data from the archived ADS-B data transmitted from the airplane, crash site information, and a simulator model of the Cessna 208B. The simulation provided a physics-based estimate of the position and orientation of the airplane throughout the accident flight. The performance observations noted below are based on the results of this simulation.
The simulation indicated normal flight conditions from the beginning of the transmitted ADS-B data until 1556:30. At this point, the airplane began a steep descent, which continued until impact. The simulation descent rate from 1556:30 to the time of impact steadily increased to a maximum of about 16,000 ft per minute. The elapsed time from the initial upset to the point of impact was about 22 seconds.
At the last recorded ADS-B plot, the simulation indicated a calibrated airspeed of about 256 knots and a bank angle of about 2 degrees right; just before impact, the pitch angle changed to -34 degrees from the maximum calculated pitch during the descent of -40 degrees. The simulation indicated that, during the initial upset, the left bank angle momentarily exceeded 60 degrees.
Simulated calculations of required engine power settings throughout the flight were consistent with normal engine operation. The calculations were also consistent with the airplane remaining upright throughout the upset sequence, with maximum simulation roll angles of about 45 degrees left and 47 degrees right.
Calculations of elevator position, elevator trim tab position, and control column force were also made using normal and runaway pitch trim scenarios. Elevator position calculations were consistent in both the normal and runaway pitch trim scenarios. Elevator trim tab position remained constant at near 0 degrees in the normal scenario and steadily increased from near 0 to about 19 degrees trailing-edge-up near the end of the simulation in the runaway pitch trim scenario. Control column force required to change elevator position under normal trim conditions steadily increased from 0 pounds of push force to about 130 pounds of push force at the time of impact. In the runaway pitch trim scenario, the control column force required to change the elevator position quickly increased from near 0 pounds of pull force to about 800 pounds of pull force at the time of impact. (See the NTSB ADS-B simulation study in the public docket for this accident for detailed graphs.)
ORGANIZATIONAL AND MANAGEMENT INFORMATION
Hageland Aviation Services is a Part 135 air carrier that holds on-demand and commuter operations specifications and is authorized to conduct business exclusively under the business names "Hageland Aviation Services, Inc." or "Ravn Connect." The company headquarters are located at the Ted Stevens Anchorage International Airport, Anchorage, Alaska. The president and chief pilot in place at the time of the accident resided in Anchorage. The director of operations and director of maintenance resided in Palmer, Alaska.
Check Airman's Training Routine
Numerous previous training flights performed by the check airman were reviewed using archived ADS-B data and interviews with other pilots. The review of these flights showed that the construct of the flights were all similar in sequence and nature. After departure, the check airman would fly a set of standard maneuvers, including steep turns, slow flight, and aerodynamic stalls. At some point after the maneuvers portion of the flight was completed, a simulated emergency would be initiated, followed by a descent to an altitude that averaged between 100 and 200 ft above ground level. This descent would normally occur over a river bed, after which the pilot-in-training would be instructed to "fly the river." Company personnel were aware that this maneuver was being performed and stated that the maneuver was used to "teach coordination." After this low-level maneuver was completed, the check airman would then direct the airplane to an airport where numerous takeoff and landings would be accomplished before returning to the home base.
The accident flight followed a typical routine and flight pattern from previous training flights. The accident sequence initiated during the time when a simulated emergency and descent was typically initiated.