CESSNA 177RG — Lake Placid, NY

What happened

The purpose of the flight was to photograph the accident airplane while airborne for a magazine article. There were two airplanes that made up the flight. The lead airplane (a Beech A36) took off first from the departure airport’s only runway with a photographer onboard and the aft right door removed. The accident airplane took off about 700 ft behind the Beech. The owner was to fly the accident airplane during the takeoff and climb-out, and after joining up in formation the pilot-rated passenger was to take over the controls and fly the airplane during the formation photo shoot.

During the taxi to the runway, the accident airplane’s engine was running when the Beech pulled up next to it. The engine then shut off. About 5 seconds later, the engine restarted. During the takeoff roll, the engine sounded to a witness as if the propeller was set for climb and not takeoff; he then heard the engine surge. During the initial climb, it sounded to the witness as if it were not running at full power. The accident airplane then made a gentle left turn while it was 300 to 400 ft above ground level (agl) to join up with the Beech. After closing to about 1,000 ft away from the Beech, the accident airplane suddenly entered a hard right turn back toward the airport.

During the turn, the pilot of the Beech heard the pilot-rated passenger transmit on the common traffic advisory frequency something similar to, “we have a problem and we’re returning to the airport.” The airplane then struck an embankment in a right-wing, nose-low attitude about 440 ft from the approach end of the runway and about 250 ft left of the centerline. The pilot and pilot-rated passenger were fatally injured, and the airplane was substantially damaged.

An airplane- and glider-rated pilot also observed the accident from the opposite side of the airport while pushing back a glider after landing on grass near the runway. A short while after the Beech took off, he saw the accident airplane take off, noting that it took longer (an estimated 100 additional ft) compared to the Beech. When the accident airplane was “just barely off” he saw “white smoke” briefly come out its exhaust pipes before smoke stopped being emitted. One or two seconds later, the accident airplane turned slowly to the left about 20° to 30° and was “barely going up.” It continued in this direction for about 1 mile, then turned right “to come back to the airport.” The airplane at this time was about 300 to 400 ft in altitude on a right base leg for the approach end of the runway “but was always turning right” and not on a square base leg. When the airplane’s heading was 80° to 90° off the runway heading and it was still 300 to 400 ft agl, the “nose dropped down” and the airplane continued to turn right heading for the runway threshold. The witness did not remember hearing the engine and postulated that it might be because he was too far away. The airplane then disappeared short of the runway threshold on what appeared to be the runway centerline.

Both the accident pilot and pilot-rated passenger had accrued thousands of hours of flight time. However, the pilot had only about 10 hours of flight experience in the accident airplane make and model; it could not be determined if the pilot-rated passenger had any flight experience in the accident airplane make and model because no pilot logbooks were recovered or provided.

Review of images captured by the photographer on board the Beech showed no evidence of any open doors, smoke, or liquids leaking from the accident airplane. The photographs also indicated that the flaps were partially extended and that, just before impact, the landing gear may have been in transit and a nose-up pitch input was being applied.

A review of maintenance records did not reveal any evidence of discrepancies or abnormalities with the airplane, propeller, or engine. However, examination of the airplane, propeller, and engine did reveal some discrepancies. The hydraulic lifters had not been replaced during the last engine overhaul as required by the engine manufacturer; however, examination of the plungers per the engine manufacturer’s inspection guidelines did not reveal any anomalies or evidence of a leaking plunger in any of the lifters. Two of the hydraulic lifters exhibited a slower bleed-down rate during testing; however, this would have had little effect on valve timing at higher engine rpms.

Additionally, the extension portion of the fuel reservoir drain control was found to be improperly installed, 90° forward on the arm assembly with the tang over the aft side of the arm. Despite this condition, it could still be moved slightly even though the actuating cables for both drain valves were bent due to impact damage. None of the discrepancies found were likely to have resulted in any preimpact failures or malfunctions that would have affected normal operation.

Around the time of the accident, the density altitude was about 2,758 ft above mean sea level (msl). The airplane would have had a takeoff distance that was about 37% longer than normal with an approximate 28% decrease in the rate of climb. Given the conditions, the pilot could have reasonably anticipated that the airplane would have used more of the runway length during the takeoff and would have had a more sluggish climb rate than normal.

Two video cameras mounted on a nearby residential building recorded the airplane just before impact as it passed --right to left--in a right-banked, descending turn. Spectrum analysis of a sound consistent with the airplane’s engine that was recorded on one of the videos was used to estimate the engine speed was 2,125 rpm (±45 rpm) during a 7-second period that ended 2 seconds before ground impact. Based on this analysis, the engine was likely operating at that time.

Although a calculated weight and balance for the flight was not found, the airplane’s weight was estimated using information from the pilot’s operating handbook (POH) and balance form found in the airplane’s maintenance records. With an estimated fuel load of 40 gallons (out of a 60-gallon usable fuel capacity), plus the reported weight of the occupants, the airplane’s total weight would have been about 292 lbs below its maximum gross weight and the center of gravity (CG) would have been about 1 inch forward of the CG limit. Assuming a fuel load of 60 gallons, the airplane’s total weight would have been 172 lbs below its maximum gross weight and the CG would have been about 1.2 inches forward of the CG limit.

The airplane was equipped with a portable GPS receiver that captured accident flight data. An airplane performance study using the recovered GPS data showed that after takeoff, a little more than 1,140 ft from the runway threshold, the airplane’s total energy (potential plus kinetic) reached a peak and then began to decrease, consistent with a decrease in engine power. The airplane then began to transition from rolling left to rolling right, and the lift coefficient began to increase sharply.

The lift coefficient then exceeded the airplane’s expected maximum lift coefficient (which was calculated based on data from the POH), which was indicative of an aerodynamic stall. The roll angle then reached a maximum of 39°, the airspeed reached a minimum of 52 kts calibrated airspeed (KCAS), and the altitude reached a maximum of 1,929 ft msl, followed by a sharp drop at -1,300 ft/min; all of these parameters showed behavior consistent with an aerodynamic stall.

Several days before the accident, the pilot asked a flight instructor if he would sit in the right seat of the accident airplane to accompany him on a practice flight to prepare for the photo flight. The flight instructor agreed, having already flown the airplane several times. The flight instructor read off the checklist for the pilot and they discussed what the pilot in command (PIC) would do in an emergency during the takeoff roll and departure. Upon takeoff from the same departure airport, when about 200 ft in the air, they heard a noise and then felt a draft. The flight instructor turned around and saw that the baggage door was wide open. The flight was roughly 500 to 800 ft above the ground. The pilot “immediately made a very steep bank to the left to turn back to the runway’s approach end” and landed.

Review of the GPS data from the practice flight revealed that the airplane had lifted off from the runway and began a sharp roll to the left, reaching a roll angle of 40° before rolling back to the right. The airplane then reached a roll angle of 27° as it aligned with the runway. During the left turn, the track angle changed by about 9.6° per second, compared with a standard-rate turn of 3° per second. The airplane reached a maximum altitude of about 400 ft agl during the left turn before it descended and landed. Analysis of this previous air turnback to land revealed that it was similar in nature to the aggressive air turnback observed during the accident flight.

Operation of an airplane outside the approved CG limits will result in control difficulty. The primary concern in balancing an aircraft is the fore and aft location of the CG along the longitudinal axis. If the CG is displaced too far forward on the longitudinal axis, a nose-heavy condition will result which can cause problems in controlling and raising the nose, especially in takeoff and landing. During landing, one of the most critical phases of flight, exceeding the forward CG limit often results in decreased performance, higher stalling speeds, and higher control forces. In extreme cases, a CG location beyond the forward limit may result in nose heaviness, making it difficult or impossible to flare for landing.

Limits for the location of the CG are established and published by the manufacturer and are often established at a location that is determined by the landing characteristics of an airplane. Manufacturers place the forward CG limit as far rearward as possible to aid pilots in avoiding damage when landing and to assure that sufficient elevator/control deflection is available at minimum airspeed. The actual location of the CG can be altered by many factors controlled by the pilot. Placement of baggage and cargo items, etc. determines the CG location; for the accident flight, the pilot could have added ballast in the baggage compartment to obtain a favorable balance. However, no evidence of added ballast was discovered despite the charts and graphs provided in the POH to enable pilots to make weight and balance computations.

The pilot’s failure to perform weight and balance calculations led to the airplane taking off outside its forward CG limit, which likely degraded the controllability of the airplane. The subsequent partial loss of power during the takeoff and climb and along with the higher-than-normal density altitude, likely reduced the airplane’s climb performance substantially. The pilot’s subsequent aggressive use of the flight controls to turn back to the airport ultimately resulted in the airplane exceeding its critical angle of attack and entering an aerodynamic stall at an altitude from which a safe recovery by the pilot was not possible.

An editorial "what led to it / how to avoid it" analysis for this accident is generated separately and will appear here.