Cirrus Design Corp SR22 N112TR accident near Twin Lakes, Colorado

Probable cause

The failure of the left turbocharger at FL230, which resulted in a rapid loss of manifold pressure and subsequent complete loss of engine power due to an overly rich fuel-air mixture setting. Contributing was the smoke in the cabin in IMC, which caused the pilot to proactively shut down the engine and deploy the airframe parachute.

NTSB analysis

The pilot reported that he initially leveled off at 17,000 ft mean sea level (msl) but ultimately climbed to flight level (FL) 230 to stay above the cloud tops. During the climb to FL230, the pilot recalled a brief alternate air indication on the primary flight display (PFD). The engine was performing normally and producing full power at that time. Upon leveling off, he reduced engine power to approximately 65% to 75% and adjusted the fuel flow. The alternate air indication extinguished, and all engine indications were normal.

The pilot noted that, after leveling off at FL230, a “drastic reduction” in engine power occurred, and he observed an indication of 10% to 15% engine power. He immediately set the mixture to full rich and the fuel pump to high boost. He then adjusted the power lever in an attempt to restore engine power. None of the actions had any effect. During the descent the flight reentered instrument meteorological conditions (IMC). After declaring an emergency and establishing best glide airspeed, the pilot established a course toward a valley east of the current position. About that time, smoke started to enter the cabin and he recalled thinking that restarting the engine was not going to be possible. As the airplane descended through the minimum safe altitude for the mountainous terrain in the area, he noticed a small circle of yellow/mountain peak appear on the multi-function display (MFD) indicating terrain was near; he decided to set the engine mixture control to idle/cutoff and deploy the Cirrus Airframe Parachute System (CAPS). Once the parachute fully deployed, he turned off the electrical system and secured the cabin for touchdown. He recalled contacting the ground “hard” a few seconds later.

Review of the available engine data revealed initial reductions in engine power that were likely associated with the airplane leveling off at cruise altitudes. However, three of those power reductions were accompanied by further momentary reductions that appeared to be consistent with the pilot’s attempts to close the alternate air door. Each consisted of a single power reduction followed by a return to cruise engine speed, indicating that the alternate air door likely closed and the indication cleared.

A subsequent fluctuation in engine power occurred that lasted about 90 seconds. This fluctuation was accompanied by a substantial loss in manifold pressure, exhaust gas temperatures (EGT), and turbine inlet temperatures (TIT). Afterward, the engine speed stabilized about 2,600 rpm and then gradually increased until it abruptly decreased and went to zero consistent with the pilot shutting down the engine. (This is when the pilot decided to shut down the engine due to cabin smoke and approaching terrain.) The subsequent increase in engine speed and relatively stable fuel flow suggest that engine power may have been recovering as the pilot leaned the mixture. This would be consistent with a complete loss of engine power due to an overly rich fuel-air mixture. Although leaning would potentially have restored a proper fuel-air mixture and allowed the engine to regain power, the pilot’s decision to shut down the engine precluded this possibility.

A postaccident examination of the right turbocharger assembly revealed that it appeared intact and was unremarkable. However, the examination of the left turbocharger revealed that the left turbocharger compressor vanes were curled with corresponding scraping damage to the housing inlet consistent with contact while the vanes were rotating. The turbine vanes appeared intact with no scraping damage noted. The compressor and turbine vanes rotated freely; however, the shaft exhibited a slight amount of axial play. The curled blade tips and scraping damage to the housing inlet suggest that the damage occurred when the engine was operating. The engine was not operating during ground impact since the pilot had shut down the engine. Had the turbocharger hit sufficiently hard to make contact marks, it is likely that the impact marks would be static in nature, with single impact marks for each blade, and not showing rotation like the scraping marks. Therefore, it is likely that the turbocharger failed in flight.

The cause of the initial reductions in engine power were likely the result of an inadvertent opening of the alternate air door and intentional power adjustments made by the pilot when leveling at cruise altitudes. The subsequent failure of the left turbocharger after reaching FL230 likely resulted in a loss of engine power as a result the failure combined with the rapid reduction of manifold pressure and rich mixture setting. However, the pilot’s action of establishing a full rich mixture and activating the fuel pump on high boost may have contributed to an overly rich fuel-air mixture and a complete loss of engine power. There was no evidence of an engine fire or oil leak and the source of the cabin smoke was not determined, although it may have been associated with the rotational damage of the left turbocharger. Under the circumstances, the pilot’s decision to shut down the engine and deploy the airframe parachute was reasonable.

Text from the NTSB. Open the NTSB docket for CEN22LA176 for the full record.