N4082SBeech F33A · 1975

On August 8, 2019, about 0615 eastern daylight time, a Beech F33A, N4082S, was destroyed when it was involved in an accident in Hatboro, Pennsylvania. The pilot and two passengers were fatally injured. The airplane was operated as a Title 14 Code of Federal Regulations Part 91 personal flight.

According to the pilot's brother-in-law, the pilot and passengers were headed to St. Louis, Missouri, and then intended to fly to Sacramento, California. The pilot’s brother stated that the pilot intended to stop after the first day of flying in St. Louis, Missouri.

According to a fuel technician at the pilot's departure airport of Northeast Philadelphia Airport (PNE), Philadelphia, Pennsylvania, on the day before the accident, the pilot had returned in the morning from a flight in the accident airplane. The fuel technician had serviced the airplane numerous times in the past and was accustomed to interacting with the pilot, who would routinely help the fuel technician with his activities; however, the technician reported that, on this day, the pilot stated, "I'd be willing to help, but I've been having some chest pains." The fuel technician indicated that the pilot made no further comments regarding the chest pains, and, after fueling, he watched the pilot push the airplane back into his hangar with a motorized pusher.

According to Leidos Flight Service, the pilot called and received two outlook weather briefings the day before the accident, at 1153 and 1958. The morning of the accident, at 0351, the pilot called and received a standard weather briefing; he also filed two instrument flight rules (IFR) cross-country flight plans from PNE to Ohio State University Airport (OSU), Columbus, Ohio, for a fuel stop, with the final destination of Creve Coeur Airport (1H0), St. Louis, Missouri. The pilot reported to the weather briefer that he planned to depart PNE about 0500 due to the forecasted thunderstorms that were expected later in the day.

Review of air traffic control communications provided by the Federal Aviation Administration (FAA) revealed that about 0601, the pilot received a full route IFR clearance, which included instructions to fly runway heading and maintain 2,000 ft mean sea level (msl) on departure. About 10 minutes later, he was cleared for takeoff from runway 33. Shortly after takeoff, the pilot was instructed to contact Philadelphia (PHL) departure control; he subsequently attempted to radio PHL departure but was still transmitting on the PNE tower frequency. The PNE tower provided the pilot with the PHL departure frequency again, the pilot read back the frequency, and there were no further transmissions or distress calls received from the pilot to the PNE tower or PHL departure frequencies.

The National Transportation Safety Board (NTSB) conducted a performance study using automatic dependent surveillance–broadcast (ADS–B) data from the accident flight. The flight was about 6 nautical miles (nm) long and lasted about 2 1/2 minutes. The airplane departed PNE about 0612 and climbed while gaining airspeed. By 0613:20, the airplane had reached an airspeed of 140 kts and an altitude of about 1,000 ft msl (an NTSB weather study found that an overcast cloud layer was likely present from about 900 ft msl and extended to about 1,800 ft msl). About 0613:30, the airplane banked right to about 15° right wing down and turned from a heading of 325° to a heading of 350°. At 0613:55, the airplane banked left to about 20° left wing down, and 2 seconds later, it reached the maximum altitude of the recorded data of 1,186 ft msl. Figure 1 depicts the entire flight track of the accident flight and the location of the accident site.

Figure 1: The airplane's flightpath, denoted by the white line with the time and corresponding airplane altitude (ft msl).

Just after 0614:00, the airplane began a steep descent at a rate of 2,000 ft/min while rapidly gaining airspeed. In the descent, the airplane banked right to about 30° right wing down at 0614:15, before coming back to wings level just before the end of the data. The final calculated airspeed was 180 kts and the last recorded altitude, at 0614:20, was 586 ft msl, which was about 0.15 nm from the accident site. Terrain elevation in the area of the impact was about 230 ft msl. Figures 2 and 3 display the accident flight altitude, calculated groundspeed and airspeed, pitch, roll, and cloud layer information.

Figure 2: Accident flight altitude, calculated groundspeed and airspeed, and cloud layer data (Source: NTSB Performance and Weather Studies).

Figure 3: Accident flight altitude, pitch, roll, and cloud layer data (Source: NTSB Performance and Weather Studies).

Several witnesses near the accident site reported seeing and/or hearing the airplane. One witness described the airplane flying "extremely low" and "straight, not banking in either direction." An additional witness who was closer to the accident site described seeing and hearing a “loud low flying airplane” with the right wing lower than the left wing; about 2 or 3 seconds later, he heard an impact. Another witness described the engine as "loud" and indicated that it "wasn’t sputtering"; an additional witness described hearing the engine “screaming” just before sounds consistent with an impact.

According to FAA airman records, the pilot held a private pilot certificate with ratings for airplane single-engine land and instrument airplane. The pilot was issued an FAA second-class medical certificate on September 11, 2017. The pilot's logbook contained a record of flights from February 21, 2010, to July 28, 2019. The pilot had logged a total flight time of 985 hours, of which 396 hours were in the accident airplane and 35 hours were in actual instrument meteorological conditions. The pilot logged 7.3 hours in the 30 days before the accident and 17 instrument approaches in the 6 months before the accident. His most recent flight review and instrument proficiency check were completed on September 22, 2018.

A flight instructor who had flown with the accident pilot in the past, most recently in July 2019, described the pilot as "competent, thorough, and meticulous" with regard to his flying and care of his airplane. He reported that he and the pilot discussed the long cross-country trip and that, in July 2019, the pilot completed chandelles, steep spirals, and short-field approaches and landings, as the pilot expressed to him that he may need to land at shorter-than-normal runways and wanted to practice the maneuvers and landings. He reported that in his experience with the accident pilot, his "habit" was to not turn the autopilot on until cruise flight.

In May 2019, the airplane was equipped with a new Garmin G5 electronic flight display instrument with electronic stability protection (ESP) and a Garmin GFC 500 autopilot. The ESP system had the capability to use autopilot servos to correct the flight controls if the airplane exceeded attitude or airspeed preset parameters. The last known system configuration on the accident airplane was set to activate if roll exceeded 45°, pitch exceeded 20° nose up or -15° nose down, or airspeed was less than 70 kts or greater than 198 kts. A review of the pilot's logbook found that he had accumulated a total of 20.8 hours of flight time since the installation of the G5 and GFC 500, of which 5.3 hours were dual instructional hours.

According to the avionics mechanic/private pilot who installed the G5 and GFC 500 on the accident airplane, he and the accident pilot completed two flights together in May 2019 to familiarize the accident pilot with the equipment and to ensure it was functional. He reported that the G5, its ESP system, and GFC 500 functioned as expected. However, he reported that the accident pilot "wasn't getting it" with regard to how the autopilot and G5 ESP worked during the first flight. He reported that during the second flight a few days later, the accident pilot was "behind the aircraft." He described that that pilot would engage the autopilot but would not be prepared for what it was commanded to do; for example, when he programmed in a descent with the autopilot, he would not manage the power and would question why the airplane accelerated.

The mechanic observed another example with a climb with the autopilot, when initiated, the airplane would begin to slow because the pilot would not adjust the power, and he would ask "why is it decelerating?" The mechanic stated that he observed a "pattern" of the pilot mismanaging the power settings while the autopilot was engaged. The mechanic recalled that the accident pilot deactivated the G5 ESP system during portions of their flights because he "did not like it."

According to an NTSB weather study, an overcast layer of clouds was likely present around the accident site and flightpath with bases about 900 ft msl and the top of the clouds about 1,800 ft msl. The automated surface observing system (ASOS) at PNE reported visibility of 5 statute miles and haze.

The Wings Field Airport (LOM), Philadelphia, Pennsylvania, which was located about 9 miles west-southwest of the accident site, reported via an automated weather observing system visibility of 2 1/2 statute miles and mist. The Doylestown Airport (DYL), Doylestown, Pennsylvania, which was located about 10 miles north of the accident site, reported via an ASOS visibility of 10 statute miles or greater. There was no significant precipitation identified around the accident site.

Numerous witnesses near the accident site described the weather conditions about the time of the accident. Most of the witnesses stated that it was overcast, several mentioned that it was “not foggy,” several noted that it was “light out” or “bright,” and one noted that it was hazy.

According to the Astronomical Applications Department at the United States Naval Observatory, the beginning of civil twilight was at 0535, and sunrise was at 0605.

In May 2019, the airplane was equipped with a new Garmin G5 electronic flight display instrument with electronic stability protection (ESP) and a Garmin GFC 500 autopilot. The ESP system had the capability to use autopilot servos to correct the flight controls if the airplane exceeded attitude or airspeed preset parameters. The last known system configuration on the accident airplane was set to activate if roll exceeded 45°, pitch exceeded 20° nose up or -15° nose down, or airspeed was less than 70 kts or greater than 198 kts. A review of the pilot's logbook found that he had accumulated a total of 20.8 hours of flight time since the installation of the G5 and GFC 500, of which 5.3 hours were dual instructional hours.

According to the avionics mechanic/private pilot who installed the G5 and GFC 500 on the accident airplane, he and the accident pilot completed two flights together in May 2019 to familiarize the accident pilot with the equipment and to ensure it was functional. He reported that the G5, its ESP system, and GFC 500 functioned as expected. However, he reported that the accident pilot "wasn't getting it" with regard to how the autopilot and G5 ESP worked during the first flight. He reported that during the second flight a few days later, the accident pilot was "behind the aircraft." He described that that pilot would engage the autopilot but would not be prepared for what it was commanded to do; for example, when he programmed in a descent with the autopilot, he would not manage the power and would question why the airplane accelerated.

The mechanic observed another example with a climb with the autopilot, when initiated, the airplane would begin to slow because the pilot would not adjust the power, and he would ask "why is it decelerating?" The mechanic stated that he observed a "pattern" of the pilot mismanaging the power settings while the autopilot was engaged. The mechanic recalled that the accident pilot deactivated the G5 ESP system during portions of their flights because he "did not like it."

The initial impact point was located in a residential backyard with the wreckage path oriented on a 350° heading. The airplane sustained extensive impact damage and was heavily fragmented. The debris was scattered about 330 ft from the initial impact to the main wreckage, and an odor of fuel was present along the debris path. All major components of the airplane were accounted for at the accident site, and flight control continuity was established for all flight control surfaces. There was no evidence of fire. All three landing gear actuators, when measured, were consistent with a retracted position. The Garmin G5 electronic flight display instrument sustained extensive impact damage and separated from the cockpit. The SD memory card slot did not contain an SD card, and no data was retrieved from the unit.

The engine remained partially attached to the airframe and displayed significant impact damage. The crankshaft was rotated by hand; crankshaft continuity was established from the forward section of the engine to the accessory section. All six cylinders remained attached to their cylinder bays and displayed varying amounts of impact damage. Internal examination of the cylinders with a lighted borescope showed that they displayed normal operating and combustion signatures.

All cylinders, with the exception of Nos. 4 and 5, displayed thumb compression and suction when the crankshaft was rotated by hand. A significant amount of wood was embedded in the No. 4 exhaust lifter location. Thumb compression and suction was established on the No. 4 cylinder after the exhaust rocker arm with the debris was removed. The No. 5 cylinder sustained the most impact damage, and thumb compression and suction could not be established due to the damage.

The left and right magnetos displayed varying degrees of impact damage; however, when the magneto couplings were rotated by hand, spark was observed on all six posts for both magnetos. The top and bottom spark plugs for each cylinder, when removed and examined, displayed normal operating signatures, with the exception of the No. 5 spark plugs, which sustained impact damage. Fragments of the three-bladed propeller had separated from the propeller hub and were found in the debris path. The propeller blades each displayed varying degrees of twisting, s-bending, and chordwise scratching.

Spatial Disorientation

The FAA Civil Aeromedical Institute's publication titled "Introduction to Aviation Physiology" defines spatial disorientation as a “loss of proper bearings” or a “state of mental confusion as to position, location, or movement relative to the position of the earth.” Factors contributing to spatial disorientation include changes in acceleration, flight in instrument meteorological conditions (IMC), frequent transfer between visual meteorological conditions and IMC, and unperceived changes in aircraft attitude.

The FAA's Airplane Flying Handbook (FAA-H-8083-3B) describes some hazards associated with flying when the ground or horizon are obscured. The handbook states, in part, the following:

“The vestibular sense (motion sensing by the inner ear) can and will confuse the pilot. Because of inertia, the sensory areas of the inner ear cannot detect slight changes in airplane attitude, nor can they accurately sense attitude changes that occur at a uniform rate over a period of time. On the other hand, false sensations are often generated, leading the pilot to believe the attitude of the airplane has changed when in fact, it has not. These false sensations result in the pilot experiencing spatial disorientation.”

At his last FAA medical examination, the pilot reported taking atorvastatin for high cholesterol, hydrochlorothiazide for high blood pressure, and levothyroxine for hypothyroidism; none of these were considered impairing. The pilot had a remote history of premature atrial and ventricular contractions, which were first reported in his initial FAA medical certification examination in 2010. Cardiac work-up was unremarkable with normal exercise stress tests and cardiac monitoring, and there had been no recent recurrence. No other medical concerns or issues were identified.

Personal medical records from the pilot's primary care health center were obtained for the 3 years before to the accident. Records were consistent with the reported medications and medical conditions in the FAA medical certification files. His blood pressure was well controlled, and he denied chest pain or syncopal events (fainting) at multiple clinical visits. For his application to obtain hospital clinical privileges, the pilot, who was a medical doctor, underwent a cognitive screen on October 15, 2019, and it was reported that there was no evidence of cognitive dysfunction. A family member interviewed after the accident reported that he was not aware of any medical issues with the pilot.

According to the Montgomery County Coroner's Office, Norrington, Pennsylvania, autopsy report, the pilot’s cause of death was “multiple injuries reportedly sustained as the operator of an aircraft involved in an aircraft accident." Due to the pilot’s extensive injuries, the brain and heart were not available for examination.

Toxicology testing performed by the coroner's office was positive for caffeine in muscle tissue. FAA Forensic Sciences Laboratory toxicology testing was negative for ethanol in muscle and kidney tissue and negative for drugs. Carbon monoxide and cyanide testing was not performed.

The NTSB performance study used ADS-B data from the accident flight to determine if any of the G5 ESP parameters were exceeded. The study found that none of the G5 ESP airspeed, roll, and pitch parameters set at the time of installation were exceeded, even during the final descent. No data was retrieved from the unit, so it was not possible to know if the pilot had disengaged the system, as he had done during previous flights, or if he changed the attitude, bank, and airspeed engagement parameters.