M-Squared SPRINT 1000 accident near Madeira Beach, Florida, September 5, 2009

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

The float-equipped, pusher-configured experimental light sport airplane overflew an intercoastal waterway, landed and water-taxied for several minutes, then took off again. At an altitude estimated to be between 200 to 300 feet, witnesses heard a "pop," saw parts come off the airplane, and then saw it nosedive into the water. An installed ballistic recovery parachute partially deployed just before water impact. No witnesses reported any bird activity in the area at the time of the accident. The airplane was recovered from the water, except for the propeller and propeller gearbox, which were never located. After a Federal Aviation Administration (FAA) examination, and unknown to the pilot's wife, who thought it was in storage, the airplane was disposed of. The pilot's wife subsequently recovered the engine with the three-pronged engine-side gearbox coupling flange still attached, but after moving out of state disposed of that as well. Photographs provided by the FAA and sheriff's department indicated that once the gearbox and propeller separated from the engine, they severed the airplane's tail support structure, which resulted in a loss of pitch control. Additional photographs indicated that the propeller gearbox adapter that joined the gearbox to the engine was fractured and mostly missing and that three bolts, one of which would have normally been in each of the three prongs that connected the engine-side gearbox coupling flange to a rubber damper inside the adapter, were missing. Threadlocking material, which would have helped maintain torque, was not observed in the photographs. The pilot had purchased the airplane about 2 1/2 weeks before the accident as non-flyable parts. The former owner had previously removed the engine, had work performed on it by the engine distributor, and reinstalled it on the airplane prior to selling it to the accident pilot. No maintenance entries were recorded in the airplane logbook. According to the engine distributor, after repair the engine would have been shipped with the gearbox attached loosely, in an "up" position, to fit into the shipping box. Upon installation, the gearbox would have needed to be rotated 180 degrees to the "hanging" position to maintain the thrust line below the engine. After the sale of the airplane, the accident pilot was seen performing maintenance on the airplane, which was legal since it was experimental. However, the airplane had no annual condition inspection, which was required by its operating limitations, and could have only been performed by an FAA-certified mechanic or repairman. Sometime after the pilot had begun flying the airplane, the previous owner observed that the gearbox retaining bolts were loose, and advised the pilot that they had to be tightened. The previous owner subsequently saw the pilot tightening the bolts without using a torque wrench, and advised him that he needed to use one. The pilot then acquired a torque wrench, but the torque values he used could not be determined. The pilot was the last known person to work on the airplane, and several days before the accident had advised the kit manufacturer that he had cleaned and retorqued all the bolts. Whether he meant only the gearbox-to-adapter bolts or whether he included the flange bolts among those torqued is not known; there were no maintenance entries in the airplane's logbook. Because of the missing gearbox and propeller, and the disposal of the engine, it was not possible to definitively determine the sequence of events leading to the separation of the gearbox. However, if the gearbox adapter had failed first, there would likely still have been some remnants of the flange bolts remaining in the flange, and possibly bent flange prongs along with torn threads in the bolt holes. Because the majority of the bolt hole faces were pristine, it is likely that there was a loss of torque on the flange bolts and that they backed out of their holes. It is unknown if the loss of torque resulted from inadequate torque during bolt installation or whether a different part of the gearbox became loose first, which then overcame the proper torque. Because of a lack of record-keeping, the person who last torqued the bolts 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.