Its been a while, so Ive forgotten plenty of things I learned when earning my multi-engine rating several years ago. As one result, Id be dangerous until and unless I obtained some remedial instruction in a conventional twin. Meanwhile, I have enough experience with night operations, complex airplanes and flying into unfamiliar airports to know I would not approach such an undertaking lightly, and wouldnt even consider it in a twin without a competent-in-type pilot sitting beside me, preferably in the left seat. My multi training involved flogging a well-worn Seneca I into submission and to commercial standards. After a few fits and starts, I got the hang of it and managed to stay ahead of the airplane most of the time. Toward the end of my training, on a very bumpy spring day, I found myself under the hood, in a holding pattern, with one turning and one “burning.” An ill-timed gust arrived-with full power on the good engine while in a gentle bank away from it-and the airplane decided this was a good time to roll over and play dead. After immediately retarding the good engines throttle and rolling back to wings-level without prodding from the instructor, I knew a lot more about how to handle flight at and below the airplanes minimum controllable airspeed. That unplanned VMCA demonstration told me all I needed to know about recovering from such a condition, and the need to keep up some speed and be thinking well ahead of the airplane during single-engine operations. It also instilled in me a lot of confidence and the realization I could get through the checkride. As with so many other completely safe operations nonetheless posing increased risk, we have to know when to change our plans and divert, when we can work around a problem-with the airplane or the weather-and when to close the throttles and turn things over to the insurance company. No one wants to ruin a perfectly good airplane, but sometimes thats the safest thing we can do. This month, we explore an accident involving a light twin and an apparent low-altitude roll-over that seems to pose more questions than it answers. BACKGROUND On April 2, 2008, at about 1929 Mountain time, a Beech 95-B55 Baron collided with terrain during landing at Benson, Ariz. The private pilot and one passenger were killed; the airplane sustained substantial damage. Visual conditions prevailed. The airport manager, after using the CTAF to advise the flight of the winds and suggest Runway 28 was preferred, observed the airplane on its first landing approach, which was high and fast. On its second approach, the airport manager subsequently reported the airplane was not as high or fast as on the first approach, but still was “hot.” He lost sight of the airplane as it passed the midfield point on the 4000-foot runway at about 10 feet agl. Then he heard what he described as a hard touchdown followed by increasing engine sounds. He saw a green (right wingtip) light arc to the left, heard a thud and the engine sounds stopped. INVESTIGATION The first identified point of contact (FIPC) was a ground scar 150 feet perpendicular to the runway edge and some 3100 feet from its approach end. The ground scar contained remnants of the rotating beacon mounted atop the vertical stabilizer. Flight control continuity was established for the ailerons, elevators and rudder, as well as for the elevator and rudder trim. The fuel selector valves were in the “on” position. The landing gear was extended while the flap handle was in the up position, and the flaps and flap actuators were retracted. The cockpit elevator trim indicator was in the green takeoff range. Each magneto switch was in the “Both” position, and the alternators and battery switches were in the “on” position. The left propeller exhibited relatively little or no rotational energy at impact, evidenced by the absence of blade twisting, harsh bending, leading edge damage, or clear indications of rotational scoring. A technician for the propeller manufacturer concluded that, if rotating, blade damage to the left propeller suggested impact at low power or no power. The same technician concluded blade damage to all three blades of the right propeller and the blade angle findings were consistent with significant power production at the time of impact. All damage was consistent with impact damage. After removing and replacing several damaged components, personnel from the engine manufacturer installed the left engine in a test cell. The engine started easily, and ran at various rpm between 1000 and 2700, responding to quick accelerations and decelerations promptly without hesitation, sputtering or stalling. Disassembly of the right engine did not reveal any pre-existing anomalies precluding normal operation or rated horsepower. The pilot was relatively inexperienced, with an estimated total time of 244 hours as of the last entry in his logbook two months earlier. He logged seven hours in the last 90 days, all in the accident airplane. He had an estimated 29 hours in this make and model. He received his multi-engine rating less than a year earlier, on June 17, 2007. Probable Cause The NTSB determined the probable cause of this accident to include: “The pilots misjudged speed and altitude during approach that led to a long landing and his subsequent failure to maintain control during an attempted go-around. Contributing to the accident were the dark night, the pilots low total night flight experience and low total time in the make and model airplane.” From the record, its not at all clear why the left engine wasnt making power at impact. Its entirely possible the pilot advanced the right engine alone, by mistake, thinking he had two throttles in his hand while initiating his last go-around. But the fact an experienced witness felt the airplane was high and hot on both attempts suggests the engine had been secured before beginning the first approach. Well never know for sure. Its tough enough to fly night approaches to an unfamiliar airport, no matter the equipment. Add a relatively inexperienced pilot and a complex airplane to the mix and the operation gets riskier. Going around is the right thing to do, presuming everythings working. But in this instance, something went wrong, providing clear evidence of the hazards present when attempting a single-engine go-around in a twin.
