N1893M – Cessna (Textron Aviation) 337G – 2006-09-02 NTSB Accident Report

HISTORY OF FLIGHT

On September 2, 2006, about 0923 mountain standard time, a Cessna 337G, N1893M, collided with wires during the initial climb out from Ernest A. Love Field, Prescott, Arizona. The owner was operating the airplane under the provisions of 14 Code of Federal Regulations (CFR) Part 91. The private pilot, an airline transport pilot rated passenger, and another passenger were killed. A post-crash fire destroyed the airplane. The local personal flight departed about 2 minutes earlier. Visual meteorological conditions prevailed, and no flight plan had been filed.

An operations technician at the airport reported that the airplane taxied to his facility for fuel. The female pilot informed him that they had not flown for a while, and were going to practice touch-and-goes at Seligman, Arizona, and possibly Williams, Arizona. He was standing on the patio at his facility (less than 1,000 feet from the departure end of runway 21L), and observed N1893M depart on 21L. The airplane was about 50 feet above ground level (agl) as it flew past the departure threshold. The airplane flew over the golf course at the end of the airport boundary, and never gained altitude. He went back inside his building, and the power went off.

Another operations technician stated that he topped off the airplane's fuel tanks with 89.6 gallons of fuel. The older of two males asked him to fill both tanks 1-2 inches before the top. During the fueling process, the older gentleman had indicated that they were going sightseeing around Arizona. The technician stated that the gentleman checked the left tank after he fueled it, and asked him to add more fuel. The gentleman then secured the fuel cap cover. The same process occurred on the right wing. However, the technician noted that as he added more fuel to the right wing, gas started coming out of the vent below the wing. No fuel spilled over the wing. The gentleman checked the right wing, and pointed out to the female that a seal or o-ring was missing from the gas cap, and that could possible explain the leakage. The gentleman secured the fuel cap cover.

The older gentleman went inside to pay. Upon returning to the airplane, the technician observed the older adults in the front seats, and the young male in the back. He noted that fuel was still leaking out the vent. While driving to fuel another airplane, he heard the female talking to ground control.

A recording of the air traffic control tower transmissions indicated that a female requested takeoff clearance with a turn to the right for a touch-and-go on the right runway followed by a departure to the northwest. After takeoff, the controller noted that the airplane was low. He asked if she needed any assistance. A male responded that they did not. The controller then cleared the pilot for right closed traffic for runway 21R when able, and the male responded "93 Mike." That was the last transmission from the airplane.

A witness, who was a pilot, was traveling north on Willow Creek Road. As he passed Pioneer Parkway, he observed an airplane on a westerly heading that just departed Love Field. The airplane was flying low and slow. He thought that this airplane was too low and too slow, and that it should be pulling up for altitude. The airplane passed over Willow Creek Road approximately 100 feet in front of his vehicle. He thought that the airplane was in trouble with such a slow air speed and low altitude. He pulled over to the shoulder of the road. The airplane continued on a westerly heading, and he could clearly see that it was just clearing tree top height. He thought that the airplane began a power on stall. He said that the nose was high, the tail was low, and it went into a total stall (no forward movement). It then made a hard right banking turn towards the north and the ground; a large plum of black smoke erupted at impact. He was dialing 911 on his cell phone when he observed the airplane at tree top level before it stalled.

Another witness, who was also a pilot, was at a convenience store near the airport. He observed the takeoff. He stated that he saw both engines running, and the landing gear was down. As it proceeded over highway 89, he observed that it was not climbing. He estimated that it traveled about 1,000 yards, and began a gradual descent. Due to his concerns, he entered his vehicle, and went on Willow Creek Road as he tried to follow the airplane. He saw it turn right toward the north. The landing gear was still down, and it was still slowly descending. He thought that both engines were still operating; however, he could not hear them. He observed the airplane collide with the first of two sets of power lines. It almost immediately nosed over, and crashed, which caused a large explosion.

PERSONNEL INFORMATION

Pilot

A review of Federal Aviation Administration (FAA) airman records revealed that the pilot held a private pilot certificate with ratings for airplane single engine land, multiengine land, and instrument airplane. The pilot had a first-class medical certificate issued in July 2005. It had the limitation that the pilot must wear corrective lenses.

No personal flight records were located for the pilot. The National Transportation Safety Board investigator-in-charge (IIC) obtained the aeronautical experience listed in this report from a review of the FAA airmen medical records on file in the Airman and Medical Records Center located in Oklahoma City, Oklahoma. The pilot reported on the July 2005 medical application that she had a total time of 1,000 hours with 5 hours logged in the last 6 months.

Pilot Rated Passenger

A review of FAA airman records revealed that the passenger held an airline transport pilot certificate with a rating for airplane single engine land. He had a commercial pilot certificate with a rating for airplane multiengine land, and a private pilot certificate with a glider rating. He also held a type rating in the C/L-300. He had a second-class medical certificate issued in July 2006. It had no limitations or waivers.

A review of logbook excerpts provided to the IIC indicated that the pilot rated passenger had a total flight time of 3,438 hours as of the last entry dated June 24, 2006.

AIRCRAFT INFORMATION

The airplane was a Cessna 337G, serial number 33701493. It had two engines in a puller/pusher centerline thrust configuration. The airplane's logbooks contained an entry for an annual inspection on October 18, 2005. The annual entry noted a total airframe time of 2,403.8 hours at an hour meter time of 1,665.7. The front engine logbook noted installation of a zero time hour meter at a total airframe time of 1,036 hours.

The front engine was a Teledyne Continental Motors (TCM) factory rebuilt IO-360-G2, serial number 224804R. The rebuild date was September 26, 1979, and it was installed on the airplane on November 12, 1979. Total time recorded on the engine at the last annual inspection was 1,665.7 hours. The manufacturer's recommended time between overhauls is 1,500 hours or 12 years.

The front propeller was a McCauley model D2AF34C306-B/78CAA-0, serial number 860112. At the last annual, the recorded time since overhaul was 125.5 hours.

The rear engine was a TCM Factory Rebuilt IO-360-GB2, serial number 236607R. The rebuild date was December 21, 1981. The engine was installed on February 4, 1982, at an hour meter time of 219.7 hours. The last annual inspection noted an hour meter time of 1,665.7, but indicated that the time since rebuild was 1,146.0 hours.

The rear propeller was a McCauley model D2AF34C307/L78CBA-2, serial number 724144. The logbook contained an entry for an overhaul on October 18, 2000. At the last annual, the recorded time since overhaul was 45.8 hours.

METEOROLOGICAL INFORMATION

The closest official weather observation station was (KPRC). The elevation of the weather observation station was 5,045 feet mean sea level (msl). An aviation routine weather report (METAR) for KPRC was issued at 0853. It stated: winds calm; visibility 10 miles; skies clear; temperature 23/73 degrees Celsius/Fahrenheit; dew point 8/46 degrees Celsius/Fahrenheit; altimeter 30.15 inches of mercury.

COMMUNICATIONS

The airplane was in contact with the Prescott airport traffic control tower (ATCT).

AIRPORT INFORMATION

The Airport/ Facility Directory, Southwest U. S., indicated that runway 21L was 7,550 feet long and 150 feet wide. The runway surface was asphalt. The field elevation was 5,045 feet.

WRECKAGE AND IMPACT INFORMATION

Investigators from the Safety Board, the FAA, and Cessna Aircraft Company examined the wreckage at the accident scene.

The first identified point of contact (FIPC) was the top set of wires on a powerline that separated and fell to the ground. The wires were on a magnetic bearing of 341 degrees. The debris path was along an approximate magnetic heading of 308 degrees, and the main wreckage was about 320 feet from the FIPC. A support pole, about 1-foot in diameter, that was 21 feet to the right of the FIPC, had a scrape on its eastern face. At the base of this support structure was an outboard piece of the right wing, which had a semicircular impression in its leading edge that was about 1-foot across. Another set of powerlines was 72 feet west of the first set. One strand of the second set of wires separated and fell to the ground.

The airplane came to rest inverted. The orientation of the fuselage was 020 degrees; the orientation of the empennage was 320 degrees. The right wing was oriented with the tip pointing 220 degrees; the left wing tip was oriented 120 degrees. The front propeller separated and was east of the inverted left wing. Strands of the second powerline were strewn about the wreckage. Fire consumed most of the cabin, the inboard section of both wings, and the front sections of both tail booms.

The left aileron remained attached to the wing. One cable was continuous to the cockpit; the other separated at the cabin. The right aileron remained attached to the wing piece that separated at the powerline support structure at the FIPC. The right bellcrank separated from that wing piece. The pushrod separated from the bellcrank, but remained attached to the aileron. The Cessna investigator equated the flap actuator measurement to approximately the retracted position.

The left and right rudders remained attached to the vertical fins; investigators established continuity from the rudders to the crumpled cockpit area. The elevator remained attached to the horizontal stabilizer. Investigators established continuity from the elevator to the crumpled cockpit area. The elevator trim tab was slightly up.

The nose gear was down, and the doors were closed. The main gear position could not be determined due to mechanical and thermal damage.

MEDICAL AND PATHOLOGICAL INFORMATION

The Yavapai County Coroner completed autopsies on both pilots, and determined that the cause of death for both was multiple blunt force injuries. The FAA Bioaeronautical Sciences Research Laboratory, Oklahoma City, performed toxicological testing of specimens of both pilots.

Analysis of the specimens of both pilots contained no findings for carbon monoxide, cyanide, volatiles, and tested drugs.

TESTS AND RESEARCH

Investigators from the FAA, Cessna, and TCM examined the wreckage at Air Transport, Phoenix, Arizona, on September 5, 2006, under the supervision of the IIC.

Front Engine

The entire engine was charred and discolored. Investigators removed the engine, and placed it on a table. The oil sump partially melted. They removed the top spark plugs. None of the spark plugs sustained mechanical deformation, and the center electrodes had similar gaps and shape.

The fuel pump, starter, and vacuum pump separated; the ignition harness sustained thermal damage. The throttle, mixture, and propeller governor controls were all attached.

The right magneto separated, but produced spark at all posts when manually rotated. The left magneto remained attached. Upon removal, it produced spark at all posts when manually rotated.

A borescope inspection revealed no mechanical deformation on the valves, cylinder walls, or internal cylinder head.

Investigators manually rotated the crankshaft with a wrench. The crankshaft rotated freely, and the valves moved approximately the same amount of lift in firing order. The gears in the accessory case turned freely. They obtained thumb compression on all cylinders in firing order except for numbers two and four. Cylinders two and four had heavy fire damage and heat discoloration. Investigators heard air leaking around the rings on cylinders two and four during crankshaft rotation.

The number five cylinder had impact damage on the top of the cooling fins. All of the intake and exhaust pipes were crushed.

All of the injector nozzles were bent and covered with soot. The fuel pump drive coupling was intact. The interior was dry and discolored.

The oil screen had a light coating of metal particles on 10 percent of the screen. It was dry and discolored.

Front Propeller

The front propeller had leading edge gouges and chordwise scoring. The engine crankshaft flange fractured and separated, but remained attached to the propeller. The circumferential fracture surface was along an angular plane, and had a shear lip.

An investigator for the propeller manufacturer examined the propeller under the supervision of the FAA, and submitted a written report. He concluded that the propeller damage was the result of impact; there were no indications of any type of propeller failure. He could not determine the exact blade angle at impact. However, counterweight impression marks in the spinner shell indicated that the propeller was in the normal operating range at impact. Therefore, he concluded that the front propeller was rotating under conditions of power at impact, but he could not determine the exact amount of power.

Rear Engine

The rear engine and accessories sustained thermal damage.

Investigators removed the engine. They placed it on a table. As investigators turned the crankshaft to aid in removal of the engine driven fuel pump, they heard a sharp metallic sound. After removal of the fuel pump, they noted that the fuel pump did not turn, and the shaft had fractured and separated. The fracture surface was bright and shiny. The IIC retained the fuel pump drive shaft for further examination.

Investigators rotated the engine, but two valves did not move.

The IIC terminated the examination, and shipped the engine to the TCM factory.

TCM personnel examined the engine under the supervision of the IIC at the factory in Mobile, Alabama, from January 31 to February 2, 2007. They submitted a written report, which contained the following observations.

The engine sustained mechanical and thermal damage. TCM personnel disassembled the engine.

Both magnetos sustained thermal damage, and could not be tested.

The throttle and metering assembly sustained thermal damage. The throttle was in a mid range position. The mixture control was in the 60 percent open position. The fuel manifold valve sustained thermal damage. The spring was in place, and the diaphragm melted into the filter screen. Injector nozzles 1, 2, and 6 were open; nozzles 3, 4, and 5 were plugged. The fuel pump exhibited extensive thermal damage. The fuel pump drive was sheared in two pieces at the shear section. The IIC retained the fuel pump drive pieces for further testing. The pump had thermal damage melting all of the rubber components and the spacer. The end cover and mixture control could not be removed because of thermal damage.

The oil sump sustained thermal damage, but contained no debris. The oil suction screen was unrestricted, and the oil pickup tube was not damaged. The oil pump exhibited light scoring. The drive gear was intact; however, the drive gear nut fractured and separated into three pieces. The IIC retained the nut pieces for further testing.

Inspection revealed no mechanical deformation on the valves, cylinder walls, or internal cylinder head. The inside of cylinders 1, 3, and 5 were wet and oily; cylinders 2, 4, and 6 were black and sooty. The faces of cylinder 1 and 2 contained semicircular impressions similar to a valve; these impressions contained combustion deposits. TCM personnel indicated that the combustion deposits were normal.

The crankshaft and counterweight assembly was undamaged and exhibited normal operating signatures. The connecting rod journals, main journals, and thrust surfaces were undamaged. TCM personnel indicated that they showed no signs of abnormal wear or lubrication distress. The crankshaft counterweight pins, plates and snap-rings were intact. The counterweights were undamaged; they had free and unrestricted movement on the hanger blades. The oil transfer passages were open and unrestricted. The oil transfer plug was tight and in position.

TCM personnel noted that the crankshaft main bearings exhibited no signs of lubrication distress. The crankshaft main bearings were intact and exhibited particle embedment. The connecting rod bearings exhibited no signs of lubrication distress. The connecting rod bearings were intact and exhibited particle embedment.

The cylinder bays were intact and undamaged. The main bearing support mating surfaces were intact and exhibited no signs of fretting. The main bearing support diameters were intact and exhibited no signs of bearing movement or bearing tang lock-slot elongation. The oil galleys and passages in the left and right crankcase halves were intact, clear and unrestricted.

The accessory gears had continuity. The teeth were undamaged.

Oil Pump Drive Gear Nut

A Safety Board Materials Laboratory specialist examined the oil pump drive gear nut. The specialist's factual report is part of the public docket. It fractured radially at three locations. The specialist noted that all fracture features were consistent with overstress fracture. The specialist observed no evidence of fatigue or other preexisting cracking, and the threads between the fractures were intact.

Engine Driven Fuel Pump

The specialist examined the fractured fuel pump drive shaft. The fracture appeared light gray, and had slight variations in surface topography that were symmetric about the center axis. Examination with a scanning electron microscope showed elongated dimples and smearing that was consistent with an overstress fracture under torsional loading. The specialist observed no evidence of fatigue or other preexisting cracking.

Rear Propeller

The propeller manufacturer noted that the rear propeller was not damaged except that the outboard half of each blade was missing. He attributed this to thermal damage. He determined that the rear propeller was operating at low or no power at impact.

Owner's Manual

The Owner's Manual describes normal takeoff procedures. It instructs the pilot to confirm that the rear engine is operating normally at the start of the takeoff run. It recommends that the pilot advance the rear engine throttle ahead of the front engine throttle. It tells the pilot to periodically monitor full throttle engine operation during the takeoff by glancing at the fuel flow and checking revolutions per minute (rpm) on the tachometer. It states that the pilot should discontinue the takeoff while sufficient runway still remains for braking to a safe stop if either indicator is below normal, the engine runs rough, or there is sluggish acceleration.

For maximum engine power, the Owner's Manual notes that the mixture should be adjusted during the initial takeoff roll to the fuel flow corresponding to the field elevation. There is a fuel flow placard adjacent to the fuel flow indicator. It points out that the power increase is significant above 3,000 feet, and the pilot should always use this procedure above 5,000 feet above sea level.

For normal takeoffs, the Owner's Manual indicates that use of 1/3 flaps reduces the takeoff distance by 10 percent compared to a flaps up takeoff. It then indicates that the pilot should level off as soon as any obstacles have been cleared to accelerate to a normal climb speed of 120 miles per hour (mph) while slowly retracting flaps.

For takeoff on long runways, the Owner's Manual recommends retracting the landing gear after reaching a point over the runway that a wheels-down forced landing onto that runway would be impractical. For short runways, it indicated that getting the gear up when safely airborne would get the airplane in a configuration that was favorable should an engine-out emergency occur.

Owner's Manual Emergency Procedures

Section III of the Owner's Manual discusses emergency procedures, and one section discusses engine failure during takeoff. This section points out that climb performance in the event of an engine out is optimal at the single engine best rate of climb speed. A blue line on the airspeed dial indicates this speed. In the event of engine failure, altitude can be maintained more easily at this speed while the propeller is being feathered. It recommends that the pilot obtain this speed as promptly as possible after takeoff. It notes that although 104 mph is the preferred speed, 90 mph may be used with obstacles immediately ahead. It recommends that a pilot should not continue a takeoff with an engine failure below the 90 mph gross weight obstacle clearance speed. At lower weights, 85 mph is acceptable.

One section deals with engine failure during takeoff. The airplane has centerline thrust, so there is no minimum single engine control speed as defined for a conventional twin engine airplane. It points out that the most critical time for an engine failure in a twin engine airplane is that 2 or 3 second period late in the takeoff run while the airplane is accelerating to the single engine best rate-of-climb speed.

The manual discussed general facts that the pilot should use as a guide if a loss of engine power occurs. It indicated that a windmilling propeller causes a severe drag penalty. It stated that climb or continued level flight may not be possible, depending on weight, altitude, and temperature. It noted that prompt identification of the inoperative engine and feathering the propeller is of the utmost importance if the takeoff is continued. Drag with the landing gear doors open and the gear partially extended is greater than drag with the gear fully extended. It noted a rate of climb penalty of 240 feet per minute (fpm) for the former and 110 fpm for the latter. Therefore, it said that since drag increases with the initiation of gear retraction, retraction should not be attempted unless adequate airspeed and altitude margins exist for sustained flight. It stated that this is especially important under conditions of weight, altitude, and temperature that result in little or no single-engine climb performance.

Owner's Manual Operational Data

Charts in Section VI of the manual provided data for takeoff and accelerate-stop distances; another chart provided single engine maximum rate-of-climb data.

The temperature was 32 degrees Fahrenheit above the chart's depicted 41 degrees. Investigators computed the distances using the least favorable condition of maximum gross weight and corrected for the temperature difference.

The chart's used a pressure altitude of 29.92. Adjusting for the difference between standard pressure and the conditions encountered of 30.15 inches of mercury barometric pressure, the calculated pressure altitude was slightly less than 5,000 feet. Investigators used the next highest data from the charts, which was the 5,000-foot data.

The takeoff and accelerate-stop distance charts specified using the 1/3 flaps on takeoff and from a hard surfaced runway. Investigators used the 0-knot headwind data.

The chart for takeoff data provided distances for a maximum of 85 mph indicated airspeed at 50 feet. Under these conditions, the takeoff ground run was 1,627 feet, and the distance to clear a 50-foot obstacle was 2,520 feet.

The chart for accelerate-stop distance provided information for airspeeds of 85, 90, and 95 mph. Investigators used the least favorable (highest) airspeed of 95 mph. They computed an accelerate-stop distance of 5,119 feet.

Another chart provided data for single engine maximum rate-of-climb. It depicted a gross weight climb speed of 103 mph with the front engine operating, the rear propeller feathered, mixture leaned as recommended, gear up, flaps up, full throttle, and 2,800 rpm. Under these conditions, the rate-of-climb would be 42 fpm.