Part 91 operators have a lot of flexibility in their operations not available to commercial flights conducted under Parts 135 or 121. Whenever persons or property is being carried for compensation, different rules apply. One of them involves minimum weather requirements for takeoff under IFR.
The non-commercial Part 91 operator, however, has no such restrictions. We can blast off into any weather conditions we want without needing to meet a visibility minimum or having an alternate airport nearby in case of a problem developing shortly after takeoff. With that flexibility, of course, comes some responsibilities.
For example, it’s usually a good idea to forego a takeoff if the airport’s weather is below minimums for the lowest available approach. If some other operational concern is present—weight, for example, or inoperative equipment—making it necessary to burn off fuel before landing or making it impossible to execute the approach, maybe we should wait for better weather or change the aircraft’s loading.
Regardless, the low-visibility takeoff is something we learn about in our instrument training and, hopefully, practice. In addition to ensuring we can reliably track the runway centerline until liftoff, we’re concerned with a smooth transition from visual flight to flying the gauges. We’re also concerned about establishing and maintaining a positive rate of climb and doing all the other things necessary during an instrument departure in low IMC.
Low weather, by itself, isn’t a reason for an experienced and prepared pilot to cancel a takeoff, especially if better conditions are nearby or just a couple of hundred feet above us. But above all, a low-visibility takeoff should be flown much like any other, especially when it comes to establishing the initial climb.
On February 17, 2010, at about 0754 Pacific time, a Cessna 310R was substantially damaged when it impacted multiple residential structures and terrain following an in-flight collision with power lines and a power line tower shortly after takeoff from the Palo Alto Airport (PAO), East Palo Alto, Calif. The instrument-rated commercial pilot and his two passengers were killed. Instrument conditions prevailed and an IFR flight plan was filed. At 0742, PAO’s weather observation included variable wind at five knots, one-eighth of a mile visibility with fog and a vertical visibility of 100 feet.
The pilot was issued his IFR clearance at 0741, which cleared the pilot to his destination via a right turn to a heading of 060 degrees within one mile of PAO, then via vectors. The flight was cleared to climb and maintain 3000 feet. At 0751, ATC advised the flight’s IFR release was obtained and stated, “the runway is not visible, so [takeoff is] at your own risk.” Subsequently, the pilot advised he was beginning his takeoff roll on Runway 31. No further radio communications were heard from the pilot.
A witness walking on a levee near the accident site observed an airplane “suddenly” emerge from the fog to her left. She watched the airplane fly in a level or slightly nose-up attitude from her left to her right at a low altitude until it impacted power lines shortly thereafter.
The first identified point of contact (FIPC) was an electrical tower located about 0.41 nm northwest of the departure runway. The airplane struck the tower and power lines at about 40 to 50 feet agl. Debris from the aircraft was spread throughout the approximate 897-foot long wreckage path, which was oriented on a magnetic heading of about 237 degrees.
The main wreckage was adjacent to a residential structure, remaining partially on the residence driveway, yard, sidewalk and street. The main wreckage consisted of remains of the fuselage, cockpit structure, empennage, horizontal stabilizer, elevators, vertical stabilizer, rudder, right wing and right engine, all of which exhibited severe thermal damage.
Both engines subsequently were examined at their manufacturer’s facilities. There were no pre-impact mechanical anomalies that would have precluded normal operation. Both propellers also were examined by their manufacturer, which determined they were under power at the time of impact.
Local law enforcement provided recordings from an array of acoustic sensors—dubbed “Shotspotter”—deployed in high-crime areas and triggered by impulsive events such as explosions or gunshots. The triangulated recordings, coupled with ATC radar data, were plotted and a flight path was interpolated in one-second intervals. The flight path showed an approximate 45-degree left turn shortly after departure, leading to the FPIC.
A sound spectrum study was performed to determine the engine speed from recorded audio from four sensors on the ground at the time of the accident. Because the aircraft had a groundspeed of more than 200 knots at the time of the accident, the Doppler frequency shift had to be calculated. The study determined that both engines were operating near their full power.
The National Transportation Safety Board determined the probable cause(s) of this accident to include: “The pilot’s failure follow the standard instrument departure as instructed, and his failure to attain a sufficient altitude to maintain clearance from power lines during takeoff in instrument meteorological conditions.” Given the evidence and circumstances, there’s not much to add. But there are lots of questions.
The primary question is why did the pilot fail to gain any altitude over the half mile or so of the flight’s duration? Even with full fuel and three aboard, a late-model 310 with both engines producing full power should effortlessly achieve at least a 1000-fpm climb shortly after a sea-level takeoff—PAO’s field elevation is a lofty seven feet.
Another question is why did the pilot fail to turn right to the cleared heading of 060? In fact, the NTSB’s reconstructed flight path depicts the aircraft actually turning left, away from the desired heading. The aircraft first began colliding with obstructions within a half mile of the departure runway, Runway 31. And the clearance specifically stated the turn to 060 degrees should be completed within one mile of the departure airport.
It’s possible the pilot somehow became distracted shortly after liftoff. We’ve all been there. But a more likely scenario might involve well-known and predictable gyro errors. See the sidebar on the opposite page for more.