ERA25FA230

The pilot and the passenger were attempting to take off when surveillance video captured, and a witness heard, the sound of the airplane’s engine backfiring followed by a sudden decrease and increase in engine power. Dark smoke was also seen trailing under the belly of the airplane. Data downloaded from the airplane’s onboard engine monitor recorded an rpm drop from 2,600 to 1,500 rpm then back to 2,500 rpm shortly after the airplane became airborne. Fuel flow, manifold pressure, and horsepower readings uniformly tracked with the decrease/increase of rpm, consistent with the pilot quickly reducing the throttle to idle then back to full power. During the takeoff, the pilot advised the air traffic controller that the airplane was having an engine problem and that he planned to “head up and turn back.” The airplane subsequently made a right turn toward the intersecting runway shortly after the engine backfired and it’s possible the pilot reduced power in an attempt to land but changed his mind and decided to circle back to another runway. The airplane then entered a steep left turn and descended, in what a witness described as a stall-like state. The last radar return indicated the airplane was about 110 ft above the ground at a groundspeed near its published stall speed. According to data downloaded from the airplane’s engine monitor, the engine was operating near full power as the airplane descended to ground contact. Postaccident examination of the engine revealed the No. 2 cylinder bottom spark plug was loose, and stains/residue from combustion exhaust gas leakage was visible around the spark plug port consistent that it had been loose prior to the accident. Review of the airplane’s engine maintenance logbook revealed that the spark plugs were, “cleaned, gapped, tested, and rotated” at the last annual inspection, about 37 hours before the accident. There were no other entries in the engine maintenance logbook regarding removal of the spark plugs after the annual inspection. According to the engine manufacturer’s service instruction on how to install spark plugs, the spark plugs should have been torqued to a specific torque value to prevent them from coming loose. Examination of the fuel system and its components revealed evidence of water/corrosion and debris; however, testing found no indication that would have precluded fuel from reaching the engine. While the data showed that the cylinder head temperature was low on the No. 1 cylinder, there was no evidence of a stuck valve, and the ignition system tested normally. An erratic exhaust gas temperature (EGT) reading for the No. 1 cylinder was also recorded for the entire flight. The EGT data had been erratic since the engine’s last annual inspection. The No.1 cylinder’s EGT connector was found undamaged and the yellow (positive) wire for the EGT probe was not connected (the red wire was connected). As such, the probe had most likely not been working properly since the last annual inspection.Based on the available information, it is most likely that the backfiring heard coming from the engine was due to the loose No. 2 cylinder spark plug, which was not properly torqued at the last annual inspection. Despite the fact that the engine was still producing close to full power following the backfire, the pilot entered a steep turn, likely in an attempt to return to the airport, after which the airplane most likely stalled from an altitude that was too low to recover.

The airplane’s last annual inspection of the engine was completed on October 11, 2024, about 37 hours prior to the accident. At the time of the accident, the engine had accrued a total of 2,268.9 hours and 553.7 hours since overhaul. A laminated Emergency Checklist was found in the wreckage. According to the checklist, the first item listed after losing engine power immediately after takeoff was to maintain an airspeed of 100 miles per hour (mph) or 87 knots indicated airspeed. According to the airplane’s owner’s manual, the stall speed of the airplane with 15 degree of flaps extended (takeoff flap setting) in wings level flight (0-degree bank) and landing gear retracted was 64 mph or approximately 55 knots. The stall speed rose to roughly 93 mph (approx. 80 knots) in a 60° bank. On June 19, 2025, at 0847 eastern daylight time, a Mooney M20F, N9573M, was involved in an accident near Beverly, Massachusetts. The commercial pilot/owner and the passenger were fatally injured. The flight was conducted as a Title 14 Code of Federal Regulations Part 91 personal flight. A review of air traffic control tower communications and radar data provided by the Federal Aviation Administration (FAA) revealed the airplane was cleared for takeoff at 0845 from runway 16 at Beverly Municipal Airport (BVY), Beverly, Massachusetts. According to the tower controller on duty, when the airplane was about 500 ft above the runway, he observed dark smoke trailing under the belly of the airplane. The controller asked the pilot if everything was okay and cleared him to land on any runway. The pilot responded he was having engine trouble, and they were going to “head up and turn back.” The controller cleared the pilot to land on runway 09. The control said the airplane initially turned right toward runway 09, then turned left. The airplane continued banking to the left into a steeper turn and descended behind trees. A witness, who was a flight instructor at BVY, watched the airplane depart. He said there was a “significant reduction in power” when the airplane reached 150 ft agl and was about halfway down the 5,001-ft-long runway. He said, “A few seconds later it seemed as though the engine got power back partially. The pilot initiated a left turn to the east.” While turning the engine sounded as though there was another significant reduction in power. The airplane then began to descend with its wings level eventually in what appeared to be a “stalled” state behind the tree line to the accident site. Airport security video captured the airplane shortly after it became airborne, and dark smoke could be seen trailing under the airplane. The video, which had no sound, showed the airplane initially turning to the right before it turned left. The airplane then entered a steep turn and descended behind trees. Another surveilence camera located at the airport captured the sound of the airplane as it was departing and the last few moments of the flight before impact. About 0846:28, thesound of the airplane taking off was heard followed by the sound of the engine backfiring at 1846:32. The sound of engine power then quickly decreased before a suddenly increase. The airplane appeared to continue to climb with the engine running before it entered a very steep left turn to the northeast. The airplane descended and disappeared behind trees. The last radar return received on the airplane was at 0846:55. At that time, the airplane was 217 ft mean sea level (110 ft above ground level) at a ground speed of 65 knots (see figure 1). Figure 1 - The airplane’s final two radar-derived positions with the wreckage location/accident site annotated. An autopsy was conducted on the pilot by the Commonwealth of Massachusetts, Office of the Chief Medical Examiner on July 20, 2025. The cause of death was determined to be multiple blunt force injuries as the manner of death to be an accident. Toxicology testing performed at the FAA Forensic Sciences Laboratory identified Minoxidil in the pilot’s urine. Minoxidil is a medication commonly used to stimulate hair growth. The pilot held a commercial pilot certificate with ratings for airplane single-engine land and instrument airplane. His most-recent FAA second-class medical certificate was issued on September 2, 2020, without limitations. A review of the pilot’s logbook revealed that his last logged flight was on April 4, 2025. At that time, he had logged 429 total flight hours of which, 77.2 hours were in the accident airplane. The airplane impacted a grass berm adjacent to a two-lane road with the left wing, then crossed the road and impacted a light pole where it came to rest upright and perpendicular to the road. There was no postimpact fire. The left wing separated from the airframe and was found about 15 to 20 ft forward and to the right of the main wreckage. The wing tip, aileron, and flap had separated, and the fuel tank was breached; however, when the fuel tank cap was removed, fuel was observed about 3-inches below the tab. The tail section separated from the airframe and came to rest to the left of the fuselage. The engine and propeller assembly remained partially attached to the firewall and was displaced about 70 degrees to the left. The right wing (including the aileron and flap) remained secure to its respective wing root. The right flap appeared to be extended about 15 degrees (takeoff setting), and the landing gear were retracted. The right wing fuel tank was not breached and about 32 gallons of 100LL were visually observed when the fuel tank cap was removed. The airplane’s flight control tubes were fractured in several locations due to impact; however, continuity was established for all major flight controls to the cockpit area. The airplane was equipped with two Electronics International engine monitors; however, only one monitor recorded the accident flight. Data downloaded from the monitor revealed that during the last 25 seconds of the flight, the engine rpm dropped from 2,600 to1,500 before it increased back to 2,600 and settled at 2,500 rpm. Fuel flow, manifold pressure and engine-horsepower readings all uniformly followed the decrease and increase of rpm. Just before the data ended, there was another small reduction/increase in fuel flow and rpm that coincided with an increase in the No. 2 thru 4-cylinder exhaust gas (EGT) temperatures. The data also indicated that the No. 1 cylinder EGT reading was erratic for the entire accident flight and had been erratic since the date of the last annual inspection in October 2024. Examination of photos taken of the No. 1 cylinder at the accident site revealed the OLC-2 connector for the K-type EGT probe was not damaged. However, the yellow (positive) wire was disconnected from the connector, and the red (negative) wire was connected. The yellow wire not being connected would result in erratic EGT readings. The engine was separated from the airframe for examination. The top spark plugs were removed from the engine, and the plugs appeared dark in color. The engine’s crankshaft rotated freely via manual rotation of the propeller, and compression/valve train continuity were established on each cylinder. The compression was low on the No. 2 cylinder compared to the other cylinders. The engine sustained impact damage and the No. 2 cylinder intake tube had been pushed into the No. 2 cylinder bottom spark plug. The spark plug exhibited a slight bend. Further examination revealed that the No. 2 bottom spark plug was loose and the area around the plug port exhibited dark combustion exhaust leakage as evidence by staining/residue on the exterior and cooling fins, consistent with the plug being loose at the time of impact (figure 2). The spark plug’s position was marked and when it was manually moved to a hand tight position (or zero-torque position), it took about 1/8th to 1/4 - inch turn to do so. When it was attempted to tighten the spark plug to facilitate better compression, the plug broke at the base of the threaded section that housed the electrodes as soon as torque was applied. The broken spark plug was examined by the NTSB Materials Laboratory, and the fracture surfaces were consistent with overstress. No preexisting anomalies were identified. The No. 2 cylinder top sparkplug was installed in the bottom port, and a another compression test was performed. The compression was consistent with the other cylinders. Figure 2 - Combustion exhaust leakage stains and residue around the No. 2 cylinder bottom spark plug port. Review of the airplane’s engine maintenance logbook revealed that the spark plugs were, “cleaned, gapped, tested, and rotated” at the last annual inspection in October 2024. There were no other entries in the engine maintenance logbook regarding the removal of the sparkplugs after the annual inspection. According to Lycoming Service Instruction 1042AH, which was active at the time of the accident, Spark Plug Installation Guidelines, the installation torque values were:All spark plugs - 420 in.-lbs. (or 35 ft.-lbs.) (47.5 Nm) 5/8-24 (16 mm-24) lead nuts - 80 to 90 in.-lbs. (9 to 10 Nm) 3/4-20 (19 mm-20) lead nuts - 110 to 120 in.-lbs. (12 to 13.6 Nm) The magnetos remained secured in the accessory housing. A timing check was performed; the left magneto timing was late, near top dead center. The left magneto flange was fractured and appeared to have shifted during the impact sequence. The magneto was removed, the drive was secure, and it sparked from all four posts when the impulse coupling was snapped by hand. The right magneto was secure to the engine and was correctly timed. The magneto was removed, the drive was secure, and spark was observed from all four posts when the drive was rotated with a drill. The fuel selector was found on the right-wing fuel tank. When the selector was moved by hand, it moved freely to the “left” and “off” detents. The firewall fuel strainer was removed, and a small amount of debris and water were in the bowl. A few strings of unidentified debris were on the screen. Per the M20F Owner’s Manual, the electric boost pump was to be turned on for takeoff; however, the toggle switch on the instrument panel was found in the off position. During the postaccident examination, when power was applied to the pump, and the pump motor operated. The fuel line from the boost pump to the main fuel supply line to the engine was breached from impact. When compressed air was blown through the main fuel line at each wing root, fuel sprayed out of the lines at the fuel selector strainer indicating that fuel was capable of reaching the fuel selector from both tanks. The breached fuel lines and the main fuel supply line to the engine were also clear of debris. The fuel injector servo was removed from the engine, and clear 100LL fuel was drained from the servo body. The fuel screen was removed and absent of debris. The fuel manifold was removed along with the fuel injectors. When the No. 3 fuel injector was removed, the insert separated. The injector and insert were clear of debris. All of the other injectors/inserts were clear of debris. The manifold was disassembled, and fuel was in the manifold body. The fuel was absent of debris/water and the diaphragm was intact. The fuel servo and manifold were removed and sent out to the manufacturer to be bench tested. A small amount of water, corrosion and sand-like debris were found in the servo; however, the servo and manifold/injectors tested to manufacturer specifications. The fuel pump remained secured to the engines accessory housing. The fuel pump was removed, and fuel was present in the fuel pump. It created pressure, and suction when the lever was actuated. The pump was disassembled and a white sealant-like material was observed in the fuel outlet chamber and pump chamber. Evidence of corrosion was also observed in both chambers. The oil filter adapter broke from engine due to impact, and the back of engine was covered with oil. The filter was absent anymetallic debris. The oil sump screen was removed and absent of any debris. The air filter was clean. The engine monitor also indicated that the cylinder head temperature (CHT) for the No. 1 cylinder had a consistent low temperature reading for the entire flight of about 275 degrees F. A lighted borescope was used to examine the interior of the No. 1 cylinder, and no evidence of a stuck valve was observed. The cylinder’s valve guide was measured using a Lycoming valve gage guide (R-ST-310). The test indicated the valve’s play was tight and below minimums per Lycoming Service Bulletin 388. The No. 1 cylinder was then removed from the engine along with the remaining cylinders for comparison. The interior walls of the No. 1, 2, and 3 cylinders exhibited some lead bromide deposits. The interior walls of the No. 4 cylinder appeared new (replaced at the last annual) and did not exhibit any lead bromide deposits. The exhaust valve was removed from the No. 1 cylinder. The valve was easily removed from the cylinder, and no deposits were observed on the valve stem. No mechanical deficiencies or malfunctions were observed that would have contributed to the low CHT reading.