Incident Details

Pilot and passenger (his wife, also a paramotor instructor) took off at 19:00 in northeast direction into wind toward the Loire. Microlight reached approximately 100m height, then headed northwest following the Loire overhead trees. After losing sight for several dozen seconds, witness saw microlight return toward strip as if pilot wanted to land, then fell like a stone from a few dozen meters height. Wing caught in tree, trike inverted on ground approximately 100m before runway 15 threshold. Passenger declared dead at scene, pilot succumbed to injuries the next morning.

Investigation unable to determine definitive cause of loss of control. Multiple anomalies found: 1) Fatigue cracks and welding defects in trike structure at engine mount connection and cage retaining arm welds, which could have caused in-flight rupture and destabilization due to engine moving left pulled by torque; 2) Asymmetric trim adjustment found (LH closed, RH open 10cm); 3) Asymmetric A-assist lines (LH 10cm shorter than RH); 4) Possibility of incorrectly attached wing cannot be ruled out. Pilot may have identified malfunction after takeoff affecting controllability and attempted to return, losing control during approach.

Pilot and passenger both fatally injured. Passenger died at scene, pilot died next morning from life-threatening injuries.

Male, 66 years old, paramotor pilot license holder, former French paramotor champion, very experienced, flying since the 1990s. Passenger was his wife, also a paramotor instructor.

2.1 m/s

Light wind from 230° at 4 kt (nearest station), varying between 200-260°

Very good visibility, high ceiling, light crosswind. CAVOK conditions, temperature 24°C, QNH 1018 hPa. Weather suitable for flight.

One pilot who took off just before reported being violently carried to the left after takeoff at around 30m height, despite weather conditions seeming very favorable

=== https://bea.aero/fileadmin/user_upload/BEA2023-0413_VA.pdf === SAFETY INVESTIGATION REPORT www.bea.aero @BEA_Aero November 2024 BEA2023-0413 Accident to the APCO F3 Bi / Power 2Fly Fenix without an identification mark on Thursday 14 September 2023 on Beaulieu-sur-Loire microlight strip Time Around 19:001 Operator Private Type of flight Local Persons on board Pilot and one passenger Consequences and damage Pilot and passenger fatally injured This is a courtesy translation by the BEA of the Final Report on the Safety Investigation. As accurate as the translation may be, the original text in French is the work of reference. Loss of control, collision with vegetation and then ground 1 HISTORY OF THE FLIGHT Note: the following information is principally based on statements. The only witness to the accident observed the pilot, accompanied by his wife, take off at around 19:00 from Beaulieu-sur-Loire microlight strip. The pilot took off in a north-easterly direction, into the wind, towards the Loire. The witness saw the microlight reach a height of around a hundred metres. The microlight then headed north-west, following the Loire, overhead the trees on the other side of the bank. After losing sight of it for several dozen seconds, the witness saw the microlight return to the microlight strip as if the pilot wanted to land, and then fall like a stone from a height of a few dozen metres. He rushed to the accident site situated a few hundred metres away (see Figure 1) and discovered the trike on the ground, inverted, with the wing caught in a tree and still attached to the microlight. He immediately alerted the emergency services who declared that the passenger was fatally injured and that the pilot had life-threatening injuries. The pilot succumbed to his injuries the morning of the next day. 1 Except where otherwise indicated, the times in this report are given in local time. - 2 - Figure 1: satellite view of the accident site (map source: IGN, annotations BEA) 2 ADDITIONAL INFORMATION 2.1 Meteorological information There was no weather station on Beaulieu-sur-Loire microlight strip. The various people present on the strip at the time of the accident indicated that the weather situation was suitable for flight, with very good visibility, a high ceiling and a light RH crosswind. The 19:00 METAR for Orléans - Saint-Denis-de-l’Hôtel aerodrome situated around 30 NM north- west of Beaulieu-sur-Loire, indicated wind from 230° of 4 kt, varying between 200° and 260°, conditions CAVOK, temperature 24°C, dew point temperature 12°C and QNH 1018 hPa. The 19:00 METAR for Avord airbase 702 situated around 30 NM south west of Beaulieu-sur-Loire, indicated wind from 210° of 4 kt, varying between 170° and 250°, conditions CAVOK, temperature 24°C, dew point temperature 11°C and QNH 1018 hPa. 2.2 Pilot and passenger information The 66-year-old pilot was one of the paramotor pioneers in France. Over the course of his career, he won the title of French paramotor champion in the single-seater foot-launch category at the first paramotor championships in 1992 and the title of European paramotor champion in 2008 in the two-seater foot-launch category. He had been a paramotor pilot since 1992, flex-wing pilot since 1988 and fixed-wing pilot since 2020. He was a paramotor and flex-wing microlight instructor since 1998. Wreckage Beaulieu-sur-Loire microlight strip LF4554 Threshold of runway 15 Position of witness - 3 - It was not possible to determine his total experience and his recent experience. Based on the information collected during the investigation, the pilot had significant total and recent experience which included the two-seater trike paramotor. The pilot was also the manager of the Beaulieu-sur-Loire microlight strip which he ran with his wife, the passenger of the accident flight. The latter was also a paramotor pilot since 2008 and had participated in competitions. She was an instructor since 2019 and flex-wing microlight pilot since 2012. The results of the autopsies carried out on the pilot and passenger did not bring to light any element likely to explain the accident. 2.3 Microlight information The microlight was a two-seater trike paramotor. It was composed of a 36 m2 APCO F3 Bi wing, a Power2Fly Fenix trike and a Bombardier-Rotax 447 two-stroke piston engine. It was not equipped with an airframe parachute. Based on the information provided by the manufacturer, this wing is positioned as the highest performance Reflex tandem wing in their range and is not recommended for beginner pilots. Witnesses indicated that the pilot had already flown with this wing numerous times. It can carry a maximum weight of 430 kg which was consistent with the take-off weight on the accident flight. The wing had been manufactured in April 2023 and sold to the pilot in May 2023. No steps had been taken to obtain an identification card. As a consequence, the microlight did not have an identification mark. 2.4 Statements The accident occurred the day before an airshow organised by the wing manufacturer, APCO, on the microlight strip and several participants had already arrived on site. The event was cancelled following the accident. The eyewitness to the accident, who had no aeronautical knowledge, told the BEA that the people present on the microlight strip when the paramotor took off, were preparing their wings and had their backs to it. None of them therefore saw the accident. The person who took off just before the accident pilot indicated that she was violently carried to the left after take-off, at a height of around 30 m. She added that the weather conditions nevertheless seemed very favourable for the flight. Another person mentioned that the pilot was in the habit of carrying out a strip circuit at low height before setting off on his flight. 2.5 Examination of site and wreckage The wreckage was situated around 100 m before the threshold of Beaulieu-sur-Loire microlight runway 15. In their accident report, the rescue services specified that on their arrival, the trike/engine assembly was on its back in a ditch east of the road visible in Figure 2. - 4 - The wing was lying flat in a tree. There were no particular marks in the surrounding vegetation. The wing was detached from the trike by the first responders; it was not possible to know how it had been attached to the trike before the accident. Figure 2: aerial view of accident site (Source: GTA, annotations: BEA) No anomaly was found on the powerplant. The damage observed, in particular on the propeller blades, suggested that the latter had been rotating at the time of impact but it cannot be concluded that there had been engine torque. The wing was examined by a workshop specialised in the maintenance and repair of paraglider and paramotor wings. The general condition of the wing was close to new. The trims on the risers were found with an asymmetric adjustment. The LH trim was in the closed position while the RH trim was found open by around 10 cm. It seems unlikely that the trim adjustment was modified when the wing was detached from the trike. The line locking system was found in good condition and was still effective in preventing untimely sliding of the lines. The microlight was equipped with two A-assist lines2 which were found attached to the trike. The LH line was around 10 cm shorter than the RH line. The lines may have been handled by the persons who detached the wing from the trike and the adjustment of these lines may have been modified at this point or on impact with the ground which means that it is not possible to conclude with certainty about when the asymmetry in the line lengths appeared. 2 According to the wing flight manual, the A-assist system improves the take-off characteristics. The system functions by reducing the angle of attack when the wing is behind the trike and progressively disengages when the wing rises above the trike. The A-assist lines are adjusted on the ground. Wreckage Beaulieu-sur-Loire runway 15-33 LF4554 - 5 - Figure 3: diagram of A-assist system (Source: Apco F3 Operating Manual) The damage to the trike’s structure seemed to be consistent with a nose-down attitude with a slight LH bank on impact with the ground. The rear section of the trike had ruptured at two welds of the trike structure. One at the RH arm holding the protective cage and the other at the RH arm connecting the engine mount with shock absorbers to the structure of the trike. Figure 4: ruptures at trike welds (Source: BEA) After analysis in the BEA laboratory, fatigue cracks and welding defects were observed on the trike structure at the longitudinal weld connecting the engine mount to the trike and at the RH and LH circular welds of the cage retaining arm. The defects observed, cracks in fusion zone, collapse of welding bead and penetration/fusion defects, were the result of the welding process. The welding defects probably contributed to the appearance of fatigue cracks under the bead. Engine mount-to-trike connecting weld Cage retaining arm weld - 6 - Given the damage on the trike, the fatigue cracks and welding defects observed pre-dated the accident. Furthermore, the substantial number of load cycles required for the initiation and spread of the fatigue cracks observed could not be produced in just the accident sequence. It should be noted that it is very difficult to visually detect this type of defect. When contacted by the BEA, Power2Fly, the Portuguese trike manufacturer indicated that it was not aware of ruptures of this type on other trikes. Neither was the BEA aware of other similar occurrences. The manufacturer also indicated that it did not have specific welding procedures when building the trikes. Conclusion of microlight examinations Fatigue cracks and welding defects were observed on the trike structure at the longitudinal weld connecting the engine mount to the trike and at the RH and LH circular welds of the cage retaining arm. This fatigue damage could have caused the in-flight rupture of one of these welds. The in-flight rupture of the longitudinal weld connecting the engine mount to the trike could have caused a destabilisation of the paramotor due to the engine moving to the LH side pulled by the engine torque, and thus modifying the centre of thrust and centre of gravity position. In this case, the pilot may have wanted to return and land as quickly as possible. The trims and A-assist lines were found with an asymmetric adjustment during the examinations carried out by the BEA. Flying with an asymmetric adjustment of the trims and A-assist lines is not normal and affects the controllability of the microlight. These adjustments could have been the cause of a controllability problem, or potentially indicative of a problem impairing the controllability of the paramotor that the pilot may have tried to correct by changing the trim settings on his wing. It cannot be completely ruled out that these asymmetries were the consequence of the impact with the ground or the lines being handled by the first responders after the accident. The BEA was unable to obtain any information on how the wing was attached to the trike at the time of the accident. In this context, it cannot be ruled out that the pilot took off with an incorrectly attached wing, for example attached in a single-seater configuration to the rear attachment points whereas the microlight was being used in a two-seater configuration, or attached asymmetrically with one riser on the front attachment point and another on the rear attachment point. An incorrectly attached wing would have modified the centre of gravity of the microlight and compromised the performance of the flight. - 7 - 3 CONCLUSIONS The conclusions are solely based on the information which came to the knowledge of the BEA during the investigation. Scenario Shortly after the microlight had taken off, a witness saw it return to the microlight strip and then suddenly drop. The investigation was not able to determine the reason for this loss of control. However, given the different findings and anomalies observed on the paramotor (see paragraph 2.5), it is possible that the pilot had identified a malfunction on the microlight on taking off, potentially compromising the controllability and that he had wanted to return and land on the microlight strip. It can be envisaged that he then lost control of the microlight during the approach. The BEA investigations are conducted with the sole objective of improving aviation safety and are not intended to apportion blame or liabilities.