Narrative:

I experienced an on-airport forced landing due to an engine failure; with no injuries and no damage to the aircraft or airport property. The flight was entirely normal and uneventful through takeoff; cruise; descent; pattern entry; and turn onto final approach. My description will therefore commence with once established on final approach. Note that I had already completed the descent checklist during initial descent before entering the pattern and gumps check on downwind. During each of these checks I verified mixture rich (but leaned slightly one index finger knuckle to account for high density altitude); fuel shutoff in (meaning full in so fuel is not cutoff); and prime off. This is all in addition to the standard descent and approach checklist items. Because I already knew that the katana can lose an engine if power is reduced to idle and the fuel pump is off; I double-checked the pump setting and mixture; as per my habits; not only on descent but in every leg of the pattern including final. I also always start my approach high to account for any engine failure on short final. That proved to be a lifesaver. After rolling out on the centerline I dropped full flaps and trimmed for 60 knots. I reduced power to slightly above idle and let the aircraft descend slightly faster vertically down to the glideslope. Slightly above the glideslope I advanced the throttle slightly to anticipate intercepting the glideslope; intending to increase power about 100 RPM to maintain the glideslope. Up to this point the approach was stabilized and entirely normal. There were no gusts or turbulence and the approach was right on centerline. Upon advancing the throttle; there simply was no response from the engine. I immediately advanced the throttle smoothly to about 75% of full travel; and the engine responded with a slight increase of RPM; perhaps 100-150 at most. I glanced down quickly to check for fuel pump on. It was. I checked the mixture setting and it was right at 1 index finger knuckle lean of full rich; right where I had put it. I then checked the fuel shutoff by physically pressing on it; and it was full in; meaning fuel flow was on. Finally I glanced at the prime switch and it was off. I didn't verify ignition switch or master switch position until after landing; but after full stop I verified that the ignition switch was still on both and the master switch (both sides) was on. There simply was nothing I could find in my technique or procedures that would have caused an engine failure on short final. Once I completed my internal scan for possible causes of the power loss; I focused on flying the airplane. I shoved the nose down to maintain 60 kts; and then realized that I was too far below the glideslope to make the runway. I could also see that the approach light array was a significant barrier. So I decided to depart the centerline early and aim for the grass to the immediate west of [the runway]. This only required a slight bank to the right. I could see that it would be close to clear the hill immediately above the road running east/west just before the approach end; so I decided to raise the flaps to half setting to stretch the glide. I wouldn't have done so had we been lower; but I decided early and high enough that the risk was minimal. I also knew that this aircraft's flaps retract slowly; and with its large wings and low drag profile I was likely to stretch the glide slightly; which was all I needed. It worked. As the flaps retracted I shoved the nose down some more and I could feel the aircraft accelerate. As we approached the ground I resisted the almost overwhelming temptation to haul back on the stick and maintained 60 kts as long as possible. Somewhere along the way; I don't recall exactly when; I broadcast a mayday and then focused on the landing. I flared the aircraft but leveled off slightly high. We landed firmly but the touchdown was actually less firm than some landings I did as a student pilot. I did note the airspeed during the flare as my eyes were outside; but I'd say we had slowed to 45-50 kts at touchdown. I don't recall hearing the stall warning horn; so I did a good job maintaining speed to avoid spinning-in. I thought the aircraft was starting to roll to the right during the flare and I thought for a moment that I had stalled; but then we were down and rolling out. I realized later that I had touched down on an incline from left to right which created an illusion of banking to the right during the landing. Once the mains touched the nose wheel came down immediately even though I had the stick well back; and we were down. There was no bounce. I only had to make minor rudder adjustments to track straight during the rollout. The surface was perfect for an emergency landing - firm enough to allow good directional control but soft enough to help cushion the landing. I used only minor brake pressure to bring the aircraft to a stop. Our location was perhaps 50 feet west of [the runway]; short of the approach end and almost abeam the approach light array. Had we landed on the centerline we probably would have crashed into the array; with significant damage to the airport and aircraft and probably significant injuries. Once the aircraft was stopped I immediately checked my passenger; who reported he was fine with no injuries. I checked myself; also with no injuries; then commenced shutdown procedures except I left the master and avionics on. After about 30 seconds I called the tower to report no damage; no injuries. The tower responded that help was on the way. I then completed shutdown procedures and we exited the aircraft. I did a quick walk-around and noted no gas leaking; and no apparent damage to the aircraft. An airport operations car arrived soon after; followed by emergency responders. I called the flying club and notified them of the incident; and then filled out paperwork with the airport operations person. The emergency responders left and a tug arrived to tow the aircraft to flying club maintenance; once NTSB approved moving the aircraft. My passenger and I were dropped off at the flying club after which I returned the aircraft book and we left. I flew this aircraft with my primary flight training CFI; for my FAA flight review and annual flight review; combined. During the run up prior to departure for this flight; I leaned the mixture slowly to max RPM as I had been trained previously. Upon reaching max RPM the engine stabilized and then suddenly stopped! [The instructor] thought that I had leaned the mixture too far; and suggested setting the mixture full rich for run up and takeoff. So we did that; restarted the engine; and no further issues were noted during the rest of the flight. But I was surprised since in all previous training I had received emphasized leaning the mixture for highest RPM; which typically was about halfway of mixture lever full travel. I never mechanically set the mixture to halfway; I always set it based on delivered RPM. During the incident flight today; I set the mixture slightly lean of full rich both for departure and arrival (one index finger knuckle length); to account for high density altitude. This was a mixture setting considerably richer than max RPM but slightly leaner than full rich. I felt that was the safest thing to do since I didn't feel comfortable going full rich at high density altitude; but I was satisfied that leaning to max RPM might not have enough margin.note that I have flown three other katana tail numbers and did my first solo in the katana; with over 31 hours in type. I had always set the mixture in these aircraft; as trained; for best RPM and never had any problems as long as the pump was on; including during extensive touch and go operations. In all of these aircraft best RPM was typically about halfway of mixture level travel. I also flew these aircraft in high density altitude operations as like today; and during the winter during extreme cold.root cause has not yet been determined as the aircraft is still being examined by flying club maintenance. I am filing this report because the katana may have a previously unknown sensitivity to mixture setting at high density altitude. It is already well known that the katana can lose an engine if power is reduced to idle while the fuel pump is off. In my case the pump was on; so either the pump failed or there is some other issue. It was definitely not pilot error.

Google
 

Original NASA ASRS Text

Title: The pilot experienced an engine failure in a Katana DA20-C1 Eclipse and successfully landed short of the runway. The engine did not respond to throttle increase after turning final.

Narrative: I experienced an on-airport forced landing due to an engine failure; with no injuries and no damage to the aircraft or airport property. The flight was entirely normal and uneventful through takeoff; cruise; descent; pattern entry; and turn onto final approach. My description will therefore commence with once established on final approach. Note that I had already completed the descent checklist during initial descent before entering the pattern and GUMPS check on downwind. During each of these checks I verified mixture RICH (but leaned slightly one index finger knuckle to account for high density altitude); fuel shutoff IN (meaning full in so fuel is NOT cutoff); and prime OFF. This is all in addition to the standard descent and approach checklist items. Because I already knew that the Katana can lose an engine if power is reduced to idle and the fuel pump is OFF; I double-checked the pump setting and mixture; as per my habits; not only on descent but in every leg of the pattern including final. I also always start my approach high to account for any engine failure on short final. That proved to be a lifesaver. After rolling out on the centerline I dropped full flaps and trimmed for 60 knots. I reduced power to slightly above idle and let the aircraft descend slightly faster vertically down to the glideslope. Slightly above the glideslope I advanced the throttle slightly to anticipate intercepting the glideslope; intending to increase power about 100 RPM to maintain the glideslope. Up to this point the approach was stabilized and entirely normal. There were no gusts or turbulence and the approach was right on centerline. Upon advancing the throttle; there simply was no response from the engine. I immediately advanced the throttle smoothly to about 75% of full travel; and the engine responded with a slight increase of RPM; perhaps 100-150 at most. I glanced down quickly to check for fuel pump ON. It was. I checked the mixture setting and it was right at 1 index finger knuckle lean of full rich; right where I had put it. I then checked the fuel shutoff by physically pressing on it; and it was full IN; meaning fuel flow was on. Finally I glanced at the prime switch and it was OFF. I didn't verify ignition switch or master switch position until after landing; but after full stop I verified that the ignition switch was still on BOTH and the master switch (both sides) was ON. There simply was nothing I could find in my technique or procedures that would have caused an engine failure on short final. Once I completed my internal scan for possible causes of the power loss; I focused on flying the airplane. I shoved the nose down to maintain 60 Kts; and then realized that I was too far below the glideslope to make the runway. I could also see that the approach light array was a significant barrier. So I decided to depart the centerline early and aim for the grass to the immediate west of [the runway]. This only required a slight bank to the right. I could see that it would be close to clear the hill immediately above the road running east/west just before the approach end; so I decided to raise the flaps to half setting to stretch the glide. I wouldn't have done so had we been lower; but I decided early and high enough that the risk was minimal. I also knew that this aircraft's flaps retract slowly; and with its large wings and low drag profile I was likely to stretch the glide slightly; which was all I needed. It worked. As the flaps retracted I shoved the nose down some more and I could feel the aircraft accelerate. As we approached the ground I resisted the almost overwhelming temptation to haul back on the stick and maintained 60 kts as long as possible. Somewhere along the way; I don't recall exactly when; I broadcast a mayday and then focused on the landing. I flared the aircraft but leveled off slightly high. We landed firmly but the touchdown was actually less firm than some landings I did as a student pilot. I did note the airspeed during the flare as my eyes were outside; but I'd say we had slowed to 45-50 kts at touchdown. I don't recall hearing the stall warning horn; so I did a good job maintaining speed to avoid spinning-in. I thought the aircraft was starting to roll to the right during the flare and I thought for a moment that I had stalled; but then we were down and rolling out. I realized later that I had touched down on an incline from left to right which created an illusion of banking to the right during the landing. Once the mains touched the nose wheel came down immediately even though I had the stick well back; and we were down. There was no bounce. I only had to make minor rudder adjustments to track straight during the rollout. The surface was perfect for an emergency landing - firm enough to allow good directional control but soft enough to help cushion the landing. I used only minor brake pressure to bring the aircraft to a stop. Our location was perhaps 50 feet west of [the runway]; short of the approach end and almost abeam the approach light array. Had we landed on the centerline we probably would have crashed into the array; with significant damage to the airport and aircraft and probably significant injuries. Once the aircraft was stopped I immediately checked my passenger; who reported he was fine with no injuries. I checked myself; also with no injuries; then commenced shutdown procedures except I left the master and avionics on. After about 30 seconds I called the Tower to report no damage; no injuries. The Tower responded that help was on the way. I then completed shutdown procedures and we exited the aircraft. I did a quick walk-around and noted no gas leaking; and no apparent damage to the aircraft. An airport operations car arrived soon after; followed by emergency responders. I called the flying club and notified them of the incident; and then filled out paperwork with the airport operations person. The emergency responders left and a tug arrived to tow the aircraft to flying club maintenance; once NTSB approved moving the aircraft. My passenger and I were dropped off at the flying club after which I returned the aircraft book and we left. I flew this aircraft with my primary flight training CFI; for my FAA flight review and annual flight review; combined. During the run up prior to departure for this flight; I leaned the mixture slowly to max RPM as I had been trained previously. Upon reaching max RPM the engine stabilized and then suddenly stopped! [The Instructor] thought that I had leaned the mixture too far; and suggested setting the mixture full rich for run up and takeoff. So we did that; restarted the engine; and no further issues were noted during the rest of the flight. But I was surprised since in all previous training I had received emphasized leaning the mixture for highest RPM; which typically was about halfway of mixture lever full travel. I never mechanically set the mixture to halfway; I always set it based on delivered RPM. During the incident flight today; I set the mixture slightly lean of full rich both for departure and arrival (one index finger knuckle length); to account for high density altitude. This was a mixture setting considerably richer than max RPM but slightly leaner than full rich. I felt that was the safest thing to do since I didn't feel comfortable going full rich at high density altitude; but I was satisfied that leaning to max RPM might not have enough margin.Note that I have flown three other Katana tail numbers and did my first solo in the Katana; with over 31 hours in type. I had always set the mixture in these aircraft; as trained; for best RPM and never had any problems as long as the pump was ON; including during extensive touch and go operations. In all of these aircraft best RPM was typically about halfway of mixture level travel. I also flew these aircraft in high density altitude operations as like today; and during the winter during extreme cold.Root cause has not yet been determined as the aircraft is still being examined by flying club maintenance. I am filing this report because the Katana may have a previously unknown sensitivity to mixture setting at high density altitude. It is already well known that the Katana can lose an engine if power is reduced to idle while the fuel pump is off. In my case the pump was on; so either the pump failed or there is some other issue. It was definitely not pilot error.

Data retrieved from NASA's ASRS site and automatically converted to unabbreviated mixed upper/lowercase text. This report is for informational purposes with no guarantee of accuracy. See NASA's ASRS site for official report.