If you just go by what the NTSB’s probable cause says, you’ll likely be at least partially misled about what happened in the crash of a Raytheon B200 King Air into a building in the FlightSafety International complex at Wichita Mid-Continent Airport (KICT), Wichita, Kansas. The accident occurred on October 30, 2014, and the NTSB adopted its probable cause this past March. The pilot, who was the only person onboard, was killed. Three people in FlightSafety’s simulator building also were killed, while two others received serious injuries and four sustained minor injuries. The weather was good VFR, with no ceiling, visibility 10 miles, and the wind 350 degrees at 16 knots.
If you focus only on the NTSB’s probable cause statement, you might think the pilot was at least slightly inept when it came to handling an engine-out situation and, therefore, should never have been trusted with a King Air. However, details buried in various investigative reports put together by NTSB staff indicate something else to me: that the pilot was well-qualified in twin-engine airplanes, including the King Air, and likely was making a valiant effort to plant the troubled airplane safely on the ground within the confines of the airport.
The NTSB’s probable cause statement says the accident was due to “the pilot’s failure to maintain lateral control of the airplane after a reduction in left engine power and his application of inappropriate rudder input. Contributing to the accident was the pilot’s failure to follow the emergency procedures for an engine failure during takeoff. Also contributing to the accident was the left engine power reduction for reasons that could not be determined because a postaccident examination did not reveal any anomalies that would have precluded normal operation and thermal [the post-crash fire] damage precluded a complete examination.”
“At 9:47:52, the pilot said, ‘We have 80 knots, feathers armed...,’ his confirmation that the autofeather system was armed.”
The accident airplane was a twin-engine Beechcraft B200 King Air turboprop twin, which had been built in 1999 and put into service in 2000. It was powered by a pair of Pratt & Whitney PT6A-42 engines. The Safety Board’s report says the airplane was equipped with six seats. However, registration information for a previous owner showed that it had 11 seats. Such discrepancies with cabin-class airplanes with reconfigurable cabins are common. The four-blade Hartzell propellers had full feathering and reverse capabilities. King Air B200 airplanes like this one, using Hartzell propellers, are required to have an autofeather system operable for all flights. The system has to be armed for takeoff, climb, descent and landing. If an engine loses power, the system automatically feathers the prop on the bad engine. At the same time, autofeather turns itself off for the good engine, so that the propeller on the good engine isn’t accidentally feathered. This can relieve the pilot of a tremendous burden: figuring out which engine just went dead, figuring out which foot you need to use on which rudder pedal to keep the airplane straight, and making sure you’re not about to feather the wrong prop. You can concentrate on other things and take comfort that automation is working hard to save your butt. On the other hand, if the problematic engine doesn’t lose enough power to trigger the autofeather system, the constant-speed propeller will keep turning and, in theory, help keep you in the air.
The airplane also had a rudder boost system to make it easier for the pilot to maintain directional control if there’s a significant difference in power output between the two engines. The rudder boost system was supposed to be turned on and tested before flight.
The B200 emergency procedures dealing with engine failure during takeoff say the pilot needs to do the following: be sure power is at the maximum allowable; maintain airspeed at or above takeoff speed; select landing gear up; don’t retard the failed engine’s power lever until the autofeather system has completely stopped prop rotation; and verify the prop is feathered or feather manually if autofeather isn’t installed.
The King Air had been purchased by a corporation based in Georgetown, Texas, two days before the accident. The accident flight was to reposition it to the Mena Intermountain Municipal Airport (KMEZ), Mena, Arkansas. It had undergone maintenance, and was returned to service nine days before the accident. The maintenance included hot-section inspections of the engines and overhaul of the right prop. During a test flight after maintenance, it was found that the pressurization system leaked too much, and engine temperature readings showed that the right engine was operating more efficiently than the left. The left throttle lever had to be pushed farther forward than the right lever to produce the same power. Further maintenance was performed, and during a test flight four days before the accident, it was noted that the throttle positions during flight now matched.
The pilot was 53 years old. He was ATP-rated with type ratings for Raytheon (Beech) King Air BE-300, Raytheon BE-400 (Beechjet), Cessna 525 (CitationJet), Falcon 10, Learjet 45, 60 and JET, and Mitsubishi MU-300 twin-turboprop airplanes. Unlike the larger King Air 300 in which the pilot was type-rated, the slightly lighter B200 didn’t require a type rating. In a case of tragic irony, the pilot had undergone training at FlightSafety International (FSI) in Wichita the month before the accident. Investigators reviewed his training records from FSI, which showed he had 3,139 total hours with 2,843 hours in multi-engine airplanes. His latest training at FSI was in the King Air 350 Proline 21 model. The training included emergency procedures.
The pilot had a history of various medical issues, and had been receiving his second class medical certificate via the special issuance route. The issues included thyroid disease, hernias, kidney stones, vertigo, anxiety and depression. In 2013, he had a recurrence of anxiety, and underwent lithotripsy to break up kidney stones, but didn’t report these to the FAA. Toxicology testing after the accident found use of Buspirone, an anti-anxiety medication, and Citalopram, which is used to treat depression. Both are labeled with warnings that they may impair mental and physical abilities needed for performing hazardous tasks. The pilot also had reported to his personal physician that he was having trouble sleeping. The pilot didn’t mention the drugs or sleep problems to the FAA. The NTSB determined that none of the medical issues was a factor in the accident. The sleep issue didn’t provoke the NTSB to examine fatigue as a possible contributing factor in this accident.
We don’t know from the NTSB’s report what the pilot was doing in the hours before the accident, but we do know that it was about 9:38 a.m., when the pilot radioed KICT clearance delivery with the ATIS information and requested his IFR clearance to KMEZ. The controller advised that he was cleared as filed, to maintain 5,000 after departure and to expect flight level 270 after having been airborne for 10 minutes. The pilot then contacted the ground controller and was cleared to taxi to runway 1R. They exchanged pleasantries over the radio, since the pilot had spent 20 years working as a controller at KICT and retired in 2013.
The airplane was equipped with a cockpit voice recorder (CVR), which picked up comments made by the pilot, as well as mechanical sounds. There are enough comments to surmise that if he wasn’t going through the preflight checklists, he at least was checking the major items from his own memory. The CVR wasn’t locked to local clock time, so I’ve calculated approximate times. The transcript shows that, at 9:46:13 a.m., while conducting a run-up, the pilot whispered out loud, “Prop test.” That was followed by the sounds of engine runs lasting 35 to 45 seconds. Then, the pilot whispered, “Trim.” Eight seconds later, he uttered an expletive ending with the word “it.” Could he have seen something out of the ordinary with the left engine, but decided to ignore it? There’s no way to know for sure. Nevertheless, the pilot radioed the local controller that he was ready for takeoff. The tower replied, “...Runway One Right, fly runway heading, cleared for takeoff.” The pilot acknowledged. The CVR picked up the sounds of engine power increasing. At 9:47:52, the pilot said what sounded like, “We have 80 knots, feathers armed.” That last comment presumably referred to his confirmation that the autofeather system was armed.
At 9:48:01, the CVR picked up the kind of beating sounds made when propellers aren’t operating in synchronization. The change in sounds should have caused the pilot to sit up and take notice. At 9:48:05, the pilot recognized that a major problem was unfolding because the CVR picked him up saying, “the (expletive).” About 11 seconds later, he radioed the tower, “and tower just declaring an emergency, ah, we just lost loss the left engine.” Almost immediately, the CVR picked up the stall warning horn for about a second. It then sounded again, just as the controller radioed to acknowledge the pilot’s emergency call. The stall warning horn sounded briefly two more times. At 9:48:25, the pilot made a comment to the effect that the plane was “going in.” The CVR continued picking up sounds for another nine seconds.
If you go by the NTSB’s probable cause statement, even before radioing that he had an emergency, the pilot should have been executing emergency procedures: pulling up the gear; feathering the left prop because the autofeather system hadn’t kicked in; adding right rudder. However, we can speculate that the pilot probably realized from the engine instruments, engine and prop sounds, and seeing both props still rotating, that he was dealing
NTSB investigators calculated that the right engine continued to produce normal takeoff power, while the left engine continued to produce low to moderate power. Asymmetrical thrust would have had the airplane flying in a relatively gentle left bank, which the pilot may have thought would help put it into position to land in an open area on the airport or even downwind on runway 1L/19R. That runway is 10,301 feet long by 150 feet wide. The track the airplane was following likely would have put it about one-quarter of the way along runway 19R at touchdown.
A B200 at gross should be able to achieve several hundred feet per minute single-engine climb. Keep in mind that this airplane was lightly loaded, so climbing straight ahead should have been possible in the proper configuration, especially with the left engine still producing some power.
Controllers in the tower who witnessed the accident confirmed that after the pilot radioed about the engine problem, the airplane entered a shallow left turn. It then continued turning left before descending into the FlightSafety building. Witnesses who were on the east side of the runway said the airplane didn’t get higher than about 150 feet. It struck a top corner of the FlightSafety building, then broke up and burst into flames as it slid across the roof.
An ATP-rated pilot, who had about 7,500 hours total time and more than 2,000 hours in King Air airplanes, witnessed the accident from the parking lot at the FlightSafety complex. He gave a description of an airplane that was under control, in a gradual descending left turn of about 5 to 8 degrees left bank, 7 degrees nose down, with the landing gear extended and both propellers turning. Then, the bank suddenly increased to about 20 degrees and the nose went down to about 20 degrees. At that point, the left wing struck the building.
Video from surveillance cameras at the airport allowed investigators to calculate that the airplane’s ground speed was at about 85 knots, consistent with the airplane’s minimum controllable airspeed of 86 knots, increasing to 92 knots just before impact. The descent rate had been about 0 feet per minute, but increased to 1,600 feet per minute just before impact.
An aerodynamic study by NTSB investigators concluded that just before impact, the airplane went into a 29-degree nose-left sideslip, which only could have been achieved with a substantial amount of left rudder input by the pilot. The report didn’t speculate why the pilot would deliberately push on the left rudder pedal and enter a destabilizing maneuver.
Severe damage from the impact and fire made it difficult for investigators to determine whether all aircraft systems were operating properly. Although the propellers were broken into fragments, marks created at impact confirmed that neither propeller was feathered. Examination of the engine cores confirmed that both engines were producing power at impact. The fuel metering systems, including the fuel control units and compressor discharge pressure lines, were too damaged by fire to be fully tested. The left engine fuel pump was destroyed. The reason for the power reduction couldn’t be determined.
The NTSB concludes that the pilot did the wrong things by not following procedure. But, would it have been right for the pilot to feather the left prop and give up whatever power the left engine still was putting out? Would it have been right for the pilot to pull up the gear if he had a belief he could land on open space or even on runway 19R? Would it have been right for the pilot to attempt climbing away from the airport environment with a developing engine problem? The pilot of this King Air can’t review his decisions in the hindsight enjoyed by investigators, but we can. In the context of the 26 seconds of flight before impact, I’d say they don’t seem to be as unreasonable as the NTSB suggests.
Peter Katz is editor and publisher of NTSB Reporter, an independent monthly update on aircraft accident investigations and other news concerning the National Transportation Safety Board. To subscribe, visit www.ntsbreporter.us or write to: NTSB Reporter, Subscription Dept., P.O. Box 831, White Plains, NY 10602-0831.