N236BCBeech B36TC · 1996

On March 15, 2021, about 1458 eastern daylight time, a Beech B36TC, N236BC, was destroyed when it collided with a vehicle and the ground after takeoff near Pembroke Pines, Florida. The private pilot and pilot-rated passenger in the airplane and a passenger in the vehicle were fatally injured; the driver of the vehicle sustained serious injuries. The airplane was operated as a Title 14 Code of Federal Regulations Part 91 personal flight.

A witness at the departure airport reported that he heard the accident airplane during its before-takeoff engine run-up and stated that the engine was sputtering, “like [a] rough idle.” He heard cycling of the propeller, “a few times” and the engine backfired when power was increased during each sequence. He further reported that the engine sounded “…high, very high. From low to full rpm repeatedly which was more than a normal run-up. He was doing it fast.” The witness did not observe the airplane’s takeoff.

According to Federal Aviation Administration (FAA) air traffic control information, the airplane was cleared for takeoff from runway 10L at 1457:05 and the pilot was instructed to enter the left downwind leg of the traffic pattern, which he acknowledged. There were no distress calls received from the airplane during the flight.

A pilot-rated witness located about 775 ft northwest of the departure end of runway 10L saw the airplane in a very low climb at a “very slow rate.” He attributed the takeoff to be a soft-field takeoff. The witness’s attention was diverted, but he reported hearing a sudden total loss of engine power, adding that it “failed completely.” At that time, the airplane was about 100 to 200 ft past the departure end of the runway and at an altitude “definitely lower than 300 ft above ground level.” The airplane remained at the same attitude for 1 to 2 seconds, then started a “gentle” right bank while maintaining same pitch attitude. The airplane then “stalled,” spun, and pitched nose down. He heard the sound of impact and noted an explosion.

After manufacture in 1996, the airplane was exported to South Africa, where it remained until de-registered for export to the United States in January 2021.

According to an individual who inspected the airplane as part of its last annual inspection, the engine was operated twice and, “All engine operation was normal.” To his knowledge, the airplane had been flown twice since the inspection was completed. After one short flight on March 2, 2021, the accident pilot and another individual were seen working on the engine for an unknown reason. After the second flight on March 9, 2021, the pilot (right seat occupant of the accident flight) informed him that the airplane flew, “perfectly well,” but the knob for the aileron trim was slipping. Follow-on inspection of the aileron trim knob revealed it was satisfactory.

According to the airplane type certificate data sheet, the maximum engine rpm and manifold pressure were 2,700 and 36.0 inches of mercury (inHg), respectively.

As part of the Before Takeoff checklist contained in the airplane flight manual (AFM), the three-position auxiliary fuel pump was to be in the OFF position. A note in the “Takeoff” checklist indicated that takeoff rpm would be between 2,600 and 2,650 and would increase to 2,700 during the takeoff roll, while a warning in that same checklist indicated, “DO NOT TAKE OFF WITH THE AUXILIARY FUEL PUMP ON HI. EXCESSIVELY HIGH FUEL FLOWS CAN CAUSE ENGINE COMBUSTION TO CEASE DURING THE TAKE-OFF ROLL. If fuel flow exceeds the red line (34.2 gph), manually lean to 33.2 GPH prior to takeoff.”

The AFM also indicated that the auxiliary fuel pump was a single-speed, dual-pressure, electrically-driven, vane-type pump, which was located below the pilot's seat and controlled by a single three-position switch. The switch was equipped with OFF (bottom), LO (middle), and HI (upper) positions, was located on the pilot's subpanel to the left of the landing gear handle, and was used to:

1. Purge fuel vapors and prime the engine prior to start.

2. Provide a low boost to the fuel flow for all flight conditions.

3. Provide an alternate source of fuel pressure to the engine in the event the engine-driven fuel pump fails.

The airplane was fueled from a truck on March 2, 2021, with a total of 79.3 gallons of 100 low lead (100LL) aviation fuel added. The person who fueled the airplane could not recall if he filled the interconnected inboard and outboard fuel tanks on each wing. The general manager of the fuel facility reported that, since the time of the fueling, there were 551 aircraft fueled from the truck, totaling 11,735 gallons, with no issues regarding fuel quality reported. Postaccident inspection of the fuel truck by Federal Aviation Administration personnel revealed no contaminants; the fuel was blue in color, clear, and consistent with 100LL. Fuel testing was not performed.

Information from a commercial flight tracking site indicated that the airplane was flown from HWO on a local flight from 1555 to 1615 on March 9, 2021. There were no other flights between the fueling date and the accident flight.

After manufacture in 1996, the airplane was exported to South Africa, where it remained until de-registered for export to the United States in January 2021.

According to an individual who inspected the airplane as part of its last annual inspection, the engine was operated twice and, “All engine operation was normal.” To his knowledge, the airplane had been flown twice since the inspection was completed. After one short flight on March 2, 2021, the accident pilot and another individual were seen working on the engine for an unknown reason. After the second flight on March 9, 2021, the pilot (right seat occupant of the accident flight) informed him that the airplane flew, “perfectly well,” but the knob for the aileron trim was slipping. Follow-on inspection of the aileron trim knob revealed it was satisfactory.

According to the airplane type certificate data sheet, the maximum engine rpm and manifold pressure were 2,700 and 36.0 inches of mercury (inHg), respectively.

As part of the Before Takeoff checklist contained in the airplane flight manual (AFM), the three-position auxiliary fuel pump was to be in the OFF position. A note in the “Takeoff” checklist indicated that takeoff rpm would be between 2,600 and 2,650 and would increase to 2,700 during the takeoff roll, while a warning in that same checklist indicated, “DO NOT TAKE OFF WITH THE AUXILIARY FUEL PUMP ON HI. EXCESSIVELY HIGH FUEL FLOWS CAN CAUSE ENGINE COMBUSTION TO CEASE DURING THE TAKE-OFF ROLL. If fuel flow exceeds the red line (34.2 gph), manually lean to 33.2 GPH prior to takeoff.”

The AFM also indicated that the auxiliary fuel pump was a single-speed, dual-pressure, electrically-driven, vane-type pump, which was located below the pilot's seat and controlled by a single three-position switch. The switch was equipped with OFF (bottom), LO (middle), and HI (upper) positions, was located on the pilot's subpanel to the left of the landing gear handle, and was used to:

1. Purge fuel vapors and prime the engine prior to start.

2. Provide a low boost to the fuel flow for all flight conditions.

3. Provide an alternate source of fuel pressure to the engine in the event the engine-driven fuel pump fails.

The airplane was fueled from a truck on March 2, 2021, with a total of 79.3 gallons of 100 low lead (100LL) aviation fuel added. The person who fueled the airplane could not recall if he filled the interconnected inboard and outboard fuel tanks on each wing. The general manager of the fuel facility reported that, since the time of the fueling, there were 551 aircraft fueled from the truck, totaling 11,735 gallons, with no issues regarding fuel quality reported. Postaccident inspection of the fuel truck by Federal Aviation Administration personnel revealed no contaminants; the fuel was blue in color, clear, and consistent with 100LL. Fuel testing was not performed.

Information from a commercial flight tracking site indicated that the airplane was flown from HWO on a local flight from 1555 to 1615 on March 9, 2021. There were no other flights between the fueling date and the accident flight.

The airplane crashed in a densely populated area east of the airport. The main wreckage, without the engine and separated propeller, came to rest adjacent to the airport perimeter fence about 1,831 ft and 138° from the departure end of runway 10L. The accident site was located adjacent to a north/south-oriented road bordering the east side of the airport. The airplane came to rest upright on a magnetic heading of 150°. The vehicle that was impacted came to rest heading 180° on the west edge of the east perimeter road near the resting position of the main wreckage. The separated propeller came to rest near the front of the car on the passenger side. Two portable electronic devices from the airplane wreckage were retained for examination, but the units sustained heat, moisture, and impact damage, which precluded the recovery of data.

Further examination of the accident site revealed two broken powerlines. The first impacted powerline, located north of the accident site, was an east/west oriented, 7,620-volt line that was installed about 35 ft above ground level (agl). The second impacted powerline, located immediately adjacent to the accident site, was a north/south oriented 120-volt insulated line that was installed about 23.5 ft agl.

Examination of the wreckage revealed that the right wing, cockpit, cabin, right horizontal stabilizer, and right elevator were nearly consumed by fire, while the left wing exhibited extensive fire damage. The left side of the stabilizer, elevator, elevator trim tab, and trim tab rod remained attached. Both spars of the vertical stabilizer were fractured and heat damaged. The fuel selector was positioned to the left fuel tank. Some liquid consistent with fuel admixed with a foamy material consistent with that used for fire suppression was noted in the 14-gallon outboard fuel tank, while no fuel was noted in the inboard 40-gallon fuel tank. The left fuel supply was free of obstructions from the sump assembly in the inboard tank to the engine-driven fuel pump inlet, with the exception of the wing root area, where the line was crimped for recovery. There were no obstructions of the left fuel system. The right fuel supply line was heat damaged in the middle of the cockpit. The three-position auxiliary fuel pump switch, which was between the LO and off positions, was examined by the NTSB Materials Laboratory, which revealed no impact marks on any portion of the switch assembly; therefore, the switch position at the time of impact could not be determined.

Examination of the flight controls revealed no evidence of preimpact failure or malfunction. The left elevator trim tab actuator was extended 1.35 inches, which equated to between neutral and 5° tab trailing edge down (nose-up). The aileron trim actuator was extended 1.4 inches, which equated to 2.25° tab trailing edge up (wing down). The left flap actuator was extended 1.85 inches, which equated to flaps retracted while the right flap actuator was heat damaged which precluded determination of flap extension.

Examination of the separated engine revealed that a section of insulated electrical powerline about 3/16 inch in diameter was wrapped about four times around the crankshaft aft of the crankshaft flange. Impact damage was noted to the exhaust, air induction, cylinder, fuel supply, alternator, and ignition systems components. No fuel contamination was noted in the manifold valve, or of the small quantity of fuel expelled from the engine-driven fuel pump during hand rotation. All b-nuts of the fuel supply hoses were tightly installed. The throttle cable was connected at the controller, but was fractured consistent with impact. All spark plugs except the No. 2 cylinder bottom (which was not removed) and the No. 1 cylinder bottom were dark in color consistent with a rich fuel-to-air ratio. The engine was retained for operational testing at the engine manufacturer’s facility.

Operational testing of the engine at the manufacturer’s facility was performed following replacement of impact-damaged components and installation of a test propeller pitched to the lowest specified blade angle. The engine was started and operated using only the engine-driven fuel pump during multiple full throttle engine runs. During one of the runs, while at full throttle with an operative engine-driven fuel pump, black smoke emitted from the exhaust and the engine suddenly lost total power following activation of the test cell’s electric fuel pump. The engine was then restarted for the last engine run and operated for about 5 minutes with full throttle applied and only the engine-driven fuel pump, achieving a maximum of about 2,451 rpm and 32.9 inches of mercury (in Hg). At full throttle, the maximum recorded fuel flow was about 144 pounds-per-hour (PPH), which was about 35% less than the minimum required, and the unmetered pump pressure was 23.4 psi, which was about 41% less than the minimum required; however, those were the required values for maximum continuous power at 2,700 rpm and 36 inHg. Following the last engine run, the upper spark plugs from each cylinder exhibited a normal color and were not dark or sooty as they were when first viewed postaccident. The engine-driven fuel pump was operationally tested to production standards, which revealed out-of-tolerance pressures at 600 rpm (too low) and at 2,700 rpm and 36 inHg (too high); no adjustment was made to the pump.

Following the engine run, the turbocharger was examined at the manufacturer’s facility, which revealed no evidence of preimpact failure or malfunction. The pressure relief valve tested within limits for the “crack” pressure, while examination of the wastegate revealed that the cylinder was about 8° out of alignment with the valve housing assembly, consistent with impact damage. The wastegate tested within limits at the fully open and start to close tests, but was out of limits at the start to open and closed positions. Following testing, the turnbuckle of the valve housing assembly was separated from the shaft of the cylinder, and the tests were repeated with nominal results. The controller tested within limits at all test points.

All three propeller blades remained in the hub. One blade exhibited a slight aft bend beginning at the blade root, a gouge on the leading edge midspan, and scratches on the cambered side of the blade. The second blade was at a higher blade angle and exhibited a leading-edge gouge at the blade root. The third blade was loose in the hub, was bent aft about 110° near the tip, exhibited a gentle radius aft bend from the root outward, and heavy/coarse chordwise scratches on the cambered side of the blade.

Examination of the impacted car revealed that the entire right side, hood, and portion of the roof were extensively damaged. A child car seat was located on the right side of the car in the rear seat. A crease in the right front door was noted, consistent with propeller blade contact.

A video with sound from a camera located at the airport, and several video recordings from private residences near the accident site, one of which contained sound, were provided to NTSB. According to the NTSB specialist’s factual and study report, the engine rpm was determined to be about 2,600 rpm, or about 100 rpm less than the specified full-rated rpm at the start of the takeoff roll. No sound energy consistent with the airplane’s engine was noted on the audio recording from a camera near the accident site. That was likely due to a low signal-to-noise ratio, or because the engine was not operating prior to impact.

A postmortem examination of the pilot was not performed due to religious concerns.

Toxicology testing performed by the FAA’s Forensic Services Laboratory on the pilot’s heart blood and urine identified no evidence of impairing drugs.