NTSB Factual Report

HISTORY OF FLIGHT

On August 5, 2010, about 1714 Pacific daylight time, a Cessna 172M, N9730H, was substantially damaged when it struck trees shortly after takeoff from Roche Harbor Airport (WA09), Roche Harbor, Washington. The certificated flight instructor and one passenger received minor injuries, and one passenger was uninjured. The flight was operated under the provisions of Title 14 Code of Federal Regulations (CFR) Part 91. Visual meteorological conditions prevailed, and no flight plan was filed for the personal flight.

According to the pilot, the airplane belonged to the Whidbey Island Navy Flying Club (WINFC) and was based at Whidbey Island Naval Air Station (NUW), Oak Harbor, Washington, located about 27 miles southeast of the accident airport. The accident flight was to be the return leg of a round trip flight from NUW to WA09 and back, with a stop for dining at WA09. The two passengers, who were husband and wife, won the flight in a raffle. Until the day of the flight, the pilot and passengers were unknown to one another.

The flight from NUW to WA09 was uneventful, and the preflight and run-up for the return leg were normal. The pilot elected to depart on runway 7, since the windsock indicated approximately 20 degrees crosswind from the right, at about 5 knots. Since a hill topped with trees began about 1,000 feet beyond the east end of the runway, the pilot decided to conduct what he referred to as a short field takeoff, and set the flaps to between 10 and 15 degrees.

In his verbal recount to the National Transportation Safety Board (NTSB) investigator, the pilot stated that he planned to lift off at an indicated airspeed of 55, but the airplane became airborne at slightly slower speed. He then leveled the wings to accelerate before beginning the climb, and when he passed the end of the runway he started the climb. At that point, he observed that the airspeed indicator (ASI) registered 60 knots instead of the value of 67 that he expected. The pilot reported that the airplane then, "got a negative gust" and "wobbled a bit," and he adjusted the pitch attitude to continue the climb. The pilot observed that the airspeed had decreased to approximately 60 to 55 knots, and he then heard a "very distinct pop," which he associated with the engine. He told the passengers that there was a problem. The pilot thought that the engine speed was slightly lower than normal, and observed that the tachometer indicated no more than approximately 2,400 rpm. He reported that the airplane was approximately 150 feet above ground level at that time.

The pilot realized the airplane was not going to clear the hill and trees just beyond the departure end of the runway, and he started looking for a place to put down. He observed that the airplane was losing airspeed, and that the ASI indicated about 55 to 50 knots. The pilot stated that the stall warning horn sounded momentarily at least one time. Since the pilot did not observe any suitable landing areas, he focused on managing the airspeed. He stated that just before the airplane struck the trees, he extended the flaps further, with the expectation of "ballooning" the airplane over the trees. The airplane struck trees, and came to rest on the ground about 1,500 feet beyond the departure end of the runway. The pilot and both passengers extricated themselves from the airplane.

PERSONNEL INFORMATION

According to information provided by WINFC, the pilot was the WINFC Chief Pilot. He held multiple Federal Aviation Administration (FAA) certificates and ratings, including flight instructor and airline transport pilot. He had a total flight experience of about 21,600 hours, which included over 500 hours in the accident airplane make and model. The pilot's most recent FAA third-class medical certificate was issued in October 2008, and his most recent flight review was completed in September 2008. He indicated that he had flown into WA09 about 10 times prior to the accident.

AIRPLANE INFORMATION

According to FAA records, the airplane was manufactured in 1975, and was first registered to the WINFC in 1993. The airplane was equipped with a Lycoming O-320 series piston engine, and a fixed-pitch propeller. Supplemental type certificates for the use of automotive gasoline in the airplane and engine were issued in 2005. The airspeed indicator registered in knots on the outer scale, and mph on the inner scale. A representative of WINFC estimated the airplane weight at the time of the accident to be about 100 pounds below the maximum certificated gross weight of 2,300 pounds.

According to information provided by the operator, the airplane had a total time in service of approximately 12,440 hours. The engine had a total time in service of approximately 2,820 hours, and had accumulated approximately 900 hours since its most recent overhaul. The most recent 100-hour inspection of the airplane was completed on June 24, 2010, and the most recent annual inspection was completed in February 2010. The airplane had accumulated approximately 135 hours in service since the most recent annual inspection, and approximately 35 hours in service since the most recent 100-hour inspection.

METEOROLOGICAL INFORMATION

No weather broadcast or recording facilities were located at WA09. There were at least two windsocks at WA09; the one closest to airplane during the run-up was located about 410 feet east of the runway 7 threshold, and about 80 feet south of the runway centerline. That windsock was situated about 200 feet northwest of a stand of trees. The next closest windsock was approximately 2,400 feet east of the runway 7 threshold. Commercially available recorded weather information for the town of Roche Harbor included winds of 5 knots from 270 degrees, and a temperature of about 25 degrees C.

About the time of the accident, the recorded weather at an airport 5 miles north of WA09 included winds of about 13 knots from 160 degrees. The recorded weather at an airport 7 miles southeast of WA09 included calm winds, and a temperature of about 21 degrees C. The recorded weather at an airport 10 miles northeast of WA09 included winds of about 11 knots from 220 degrees, and a temperature about 20 degrees C.

AIRPORT INFORMATION

According to commercially available information, WA09 was a private airport with a single runway, and required pilots to have prior permission to land. The runway dimensions were listed as 3,593 feet by 30 feet, and the surface was cited as "asphalt, in fair condition." The published obstacle information for runway 7 was stated as "160 ft. tree, 1136 ft. from runway, 75 ft. left of centerline." Although no FAA data regarding the runway slope was available, commercially available topographic data indicated that the runway 7 threshold elevation was 87 feet above mean sea level (msl), and the runway 25 threshold elevation was 156 feet msl, for an elevation difference of 69 feet.

WRECKAGE AND IMPACT INFORMATION

According to information provided by law enforcement and FAA personnel, the airplane struck trees, descended to the ground, and came to rest in a steep nose-down attitude. The fuselage was buckled at the tailcone and in the cabin area. The left wing was displaced aft, crumpled, and partially separated from the fuselage. The nose, right wing, and horizontal stabilizer were crumpled in the aft direction. Both wing flaps remained extended equally, and the cockpit flap position indicator reading registered approximately 22 degrees. One blade of the propeller was bent forward, and one was bent aft; both exhibited some chordwise scoring. Recovery personnel reported that tree branch segments with cut ends were found in the immediate vicinity of the wreckage. The terrain elevation at the accident location was approximately 180 feet msl, and the trees were approximately 50 feet tall.

TESTS AND RESEARCH

Engine

Post accident examination of the engine was conducted by an FAA inspector about 5 weeks after the accident. The airplane’s cowling was removed and the engine was visually inspected; no anomalies were noted. According to the dipstick, the engine contained 7 quarts of oil. The spark plugs were removed and examined, and no anomalies were observed. Fuel was found in the fuel strainer.

To facilitate an engine run, the spark plugs were reinstalled, and the accident propeller was removed and replaced with a straight but non-airworthy propeller. A supplemental fuel tank was attached and plumbed into the fuel line for the airplane's right fuel tank. The engine was successfully started. A small amount of oil smoke was emitted on startup, the oil pressure was observed to be in the green arc on the oil pressure gauge, and the ammeter indicated a charging condition. Due to the non-airworthy propeller and other safety concerns, the engine was not run above 1,000 rpm.

Another examination of the engine by NTSB personnel was conducted 10 days after the FAA examination; that examination also did not reveal any anomalies. All ignition leads and spark plugs were examined and measured for electrical resistance; all measurements and component conditions were unremarkable. The magnetos were removed and disassembled; all components were secure and intact, and no evidence of damage, carbon tracking or shorting was found.

The carburetor was removed and disassembled; the fuel inlet screen was clean, the throttle valve moved properly, and no contamination was found in the float bowl. The float was intact and did not contain any fuel. The nozzle, needle valve and venturi all exhibited normal characteristics, condition and security. The gascolator was removed; the screen was inspected and found to be clean. The engine drive train was manually rotated, continuity was established for all rocker arm and valve assemblies, and "thumb compression" was obtained in all cylinders.

ADDITIONAL INFORMATION

Takeoff Procedures and Performance Information

The "Normal Procedures" section of the airplane manufacturer's Pilot’s Operating Handbook (POH) appropriate for the accident airplane serial number contained two different takeoff checklists, one for "normal take-off" and one for "maximum performance take-off." The normal takeoff checklist specified that the nose wheel was to be lifted at 55 knots indicated air speed (KIAS), and that the climb was to be conducted at 70 to 80 KIAS. The maximum performance takeoff checklist specified that the elevator was to be used to maintain a "slight tail low" attitude during the takeoff roll, and that the climb was to be conducted at 59 KIAS "until all obstacles are cleared."

Both checklists stated that the wing flaps should be "UP" for takeoff. The POH amplified procedures stated that "Normal and obstacle clearance take-offs are performed with wing flaps up," and that "the use of 10 [degrees] flaps will shorten the ground run approximately 10 [per cent], but this advantage is lost in climb to a 50-foot obstacle." The POH stated that the 10 degree takeoff flap setting "is reserved for minimum ground runs or for takeoff from soft or rough fields." It also stated that if 10 degrees of flaps "are used for minimum ground runs, it is preferable to leave them extended…in the climb to the obstacle," and that an "obstacle clearance speed of 55 KIAS" was to be used.

The operator developed a checklist specific to the accident airplane for use by its pilots, and copies were provided to the NTSB. Comparison of the operator's checklist with the airplane manufacturer's POH checklist revealed several significant differences. The operator's checklist included three sections entitled "Before Takeoff, After Takeoff, and Cruise," while the POH included four sections "Before Takeoff, Takeoff, Enroute Climb, and Cruise" for the same flight regime. The operator's "Before Takeoff and After Takeoff" sections contained a total of 15 line items, while the POH "Before Takeoff, Takeoff and Enroute Climb" sections contained a total of 28 line items. The operator's checklist did not address or differentiate between a normal takeoff and a maximum performance takeoff. The only reference to the use of flaps for takeoff in the operator's checklist was located in the "Before Takeoff" checklist, which stated "Flaps – SET." Subsequent to the accident, the operator compared each self-developed checklist to its respective POH counterpart for each airplane in its fleet; no other discrepancies were identified.

Exact POH based performance calculations for the takeoff roll and distance to climb over a 50-foot obstacle, which accounted for all the actual conditions of the flight, could not be accomplished because the actual conditions were either not addressed by, or exceeded the parameter ranges of, the manufacturer's data. The POH data did not account for flap settings other than retracted, tailwinds greater than 10 knots, or sloped runways. The actual conditions were 10 to 15 degrees of flaps, a tailwind of about 5 to 11 knots, and an upsloping runway. Each of these conditions would have an adverse effect on takeoff and climb performance, and their effect would be cumulative in combination.

Zero wind calculations for takeoff ground roll and 50-foot obstacle clearance distances, based on the available POH data, a gross weight of 2,200 pounds, and a sea level pressure altitude, resulted in values of 810 feet and 1,430 feet respectively for a temperature of 20 degrees C, and 840 feet and 1,480 feet respectively, for a temperature of 25 degrees C. Calculations for the same conditions, but with an 8-knot tailwind, resulted in values of 1,134 feet and 2,002 feet respectively, for a temperature of 20 degrees C, and 1,176 feet and 2,072 feet respectively, for a temperature of 25 degrees C. These values did not account for the actual flap setting or the runway slope.

NTSB Probable Cause

The pilot’s failure to attain an adequate climb airspeed due to his use of flaps for takeoff, and his selection of a runway, which resulted in an uphill, and possibly downwind, takeoff and subsequent collision with a known obstacle. Contributing to the accident was the operator's cockpit checklist that did not accurately reflect the manufacturer's takeoff procedures.