Cirrus Design CORP SR22 accident near Bloomfield, New Mexico, October 28, 2016

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

The private pilot was conducting a cross-country personal flight. He reported that, during cruise flight, the engine starting "missing" and that, within 2 minutes, lost all power. His attempts to restore engine power were not successful. He was unable to locate a suitable forced landing site within the airplane's glide range, so he chose to activate the ballistic parachute system. The aft fuselage separated from the airframe during the accident sequence. The airplane came to rest in an area of low brush and small trees. Postaccident engine examination revealed that the fuel port cap was not installed on the throttle body metering unit. The cap was subsequently found in the cylinder baffling immediately below the metering unit. Both the fitting and the cap appeared to be undamaged. The cap was subsequently reinstalled, and an engine test run was conducted with no anomalies noted. Data downloaded from the airplane's onboard avionics indicated that the engine speed and fuel flow were stable for the initial portion of the flight. However, about 45 minutes into the flight, the engine speed and fuel flow decayed abruptly, and the airplane entered a gradual descent. The engine speed did not recover during the rest of the flight. Before the loss of engine power, the fuel flow initially decreased from about 18.0 gallons per hour (gph) to 9.9 gph, before increasing to about 30.0 gph after the loss of power. Examination of the fuel system revealed that the fuel flow transducer was installed between the fuel pump/mixture control and the throttle body metering unit, which was upstream in the fuel flow relative to the fuel port fitting. The loss of the fuel port cap would have allowed unrestricted fuel flow into the engine compartment. The transducer would have continued to measure fuel flow passing through the unit even though that fuel was ultimately not reaching the engine because it was installed upstream to the fuel port fitting. Further, the increased fuel flow measured after the loss of engine speed was consistent with an availability of fuel, a functioning fuel pump, and the mixture control being at or near the full-rich position. Maintenance records revealed that, about 3 weeks before the accident, the engine's fuel nozzles were replaced, followed by an operational flight check and a ground run, to verify that the fuel system setting was in accordance with the Engine Maintenance and Overhaul Manual. The manual specified that the fuel port cap needed to be removed to connect external fuel pressure gauges and then reinstalled once testing and necessary fuel system adjustments were completed. The airplane had been operated 9.2 hours since the fuel nozzles were installed. Based on the available information, it is likely that maintenance personnel did not fully torque the fuel port cap before the airplane was returned to service, which allowed the cap to loosen due to engine vibration during normal operation and then back off. At that point, the fuel supply to the engine would have been interrupted, and the engine would have lost all power due to fuel starvation. Restoring fuel flow to the engine in-flight in this situation would not have been possible.

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.