Amateur-built Buttercup accident near Lyndonville, New York, May 28, 2023

The NTSB's narrative final · quoted from the NTSB record

The pilot, who was the owner and builder of the experimental, amateur-built airplane, departed with a pilot-rated passenger on a 15 to 20-minute local flight. Witnesses reported that the airplane broke up in flight and descended to the ground. The debris path extended for about 1,435 ft, beginning with debris from the wings, consistent with failure related to the wing structure. Postaccident examination of the engine and propeller revealed no mechanical anomalies that would have precluded normal operation. No preimpact flight control deficiencies were found. There was no evidence of flutter. The wing separations were consistent with upward and aft bending and/or torsional loading based on deformation signatures of the wing attachment brackets, strut brackets, and front strut attachment fittings (outboard end). Fracture surfaces of the wing attachment brackets and front strut fittings were consistent with overstress separation. No flight track information was available for the accident flight. The pilot built the airplane based on dimensions from another airplane for which no plans had been produced. The accident airplane was issued a special airworthiness certificate about 21 years before the accident. At the time of the accident, the airplane had been modified from its original configuration with the installation of a heavier, higher-horsepower engine and the installation of a different propeller. Because no maintenance records were located for the airplane, there was no information available regarding this modification, the airplane’s operational time, or its inspection/maintenance history. The airplane’s design/build plans were not based upon a structural engineering analysis to determine a structural design envelope, structural loads, or structural performance. Accordingly, the basis for any design safety margins and limitations are unknown. There were no regulatory design requirements for the airplane due to its experimental classification. Metallurgical examination revealed no conclusive evidence of pre-existing or progressive damage signatures that would have initiated the accident, and fractured structural components were consistent with overstress separation. One small area of possible fatigue fracture was observed emanating from the weld face on the lower-forward left strut end fitting at the outboard end; however, based on the amount of deformation observed in this area, it was not a likely fracture initiation point. The pilot’s toxicological testing detected several volatile solvents; however, the volatile solvent levels measured in the pilot’s tissues cannot be used to reliably predict the route of his exposure or specific effects. Thus, whether the pilot had inhaled volatile solvents recreationally or was experiencing any impairing effects of volatile solvent exposure, could not be determined. Toxicological testing also detected ethanol in liver tissue; n-butanol (a potential indicator of postmortem microbial activity) was also detected in this specimen; however, it is likely that the detected ethanol was from sources other than alcohol consumption and did not likely contribute to the accident. The pilot’s toxicology results also demonstrated the presence of central nervous system depressant medications, including cyclobenzaprine and gabapentin. Cyclobenzaprine use may be associated with significant performance impairment. Use of gabapentin can also increase the user’s somnolence, dizziness and fatigue, potentially impairing pilot performance. Both cyclobenzaprine and gabapentin have potential postmortem redistribution; therefore, reliably associating levels with impairing effects is not possible. Overall, whether the pilot’s use of cyclobenzaprine and gabapentin contributed to the accident cannot be determined. The pilot-rated passenger’s mild-to-moderate coronary artery disease conveyed some increased risk of a sudden impairing or incapacitating cardiac event, including angina, arrhythmia, or heart attack. There is no autopsy evidence that such an event occurred. However, such an event does not leave reliable autopsy evidence if it occurs immediately before death. The aircraft was configured such that it could be controlled from either the left or right front seat. Thus, it is unlikely that the pilot-rated passenger’s coronary artery disease contributed to the accident. In addition, the pilot-rated passenger’s toxicological testing indicated use of citalopram and diazepam. Citalopram and diazepam, and conditions that may be treated with these medications, can adversely affect pilot performance and judgment. However, it is unclear if the pilot-rated passenger’s performance or judgment were a factor in the outcome; whether effects of the pilot-rated passenger’s use of citalopram and diazepam or of any associated underlying conditions contributed to the accident cannot be determined. Although the available information is consistent with an in-flight failure of the wing structure due to overload, given the lack of flight track information or recorded data from onboard the airplane, the circumstances of the in-flight breakup could not be determined.

The complete narrative as the NTSB published it. The NTSB's docket holds the report as a PDF and any photographs, statements and other documents from the investigation.