NTSB Factual Report

HISTORY OF FLIGHT

On September 27, 2010, about 1054 mountain standard time, a Beechcraft F33A, N1600W, collided with a building during landing at the Phoenix Deer Valley Airport, Phoenix, Arizona. The pilot/owner was operating the airplane under the provisions of 14 Code of Federal Regulations (CFR) Part 91. The private pilot sustained fatal injuries. The airplane sustained substantial damage to the wings, fuselage, and empennage from impact damage and a post-crash fire. The local personal flight departed Phoenix Deer Valley Airport about 0938. Visual meteorological conditions prevailed, and no flight plan had been filed.

The pilot's family reported that he flew weekly to maintain proficiency, which was the purpose of this flight.

The airplane had a Garmin GPSMAP 196 portable global positioning satellite system (GPS) unit installed. A National Transportation Safety Board (NTSB) specialist downloaded the recorded information and prepared a factual report, which is in the public docket for this accident. The data indicated that the flight lasted about 1 hour 16 minutes. The pilot departed to the east, turned north, and climbed to a peak recorded altitude of 8,123 feet mean sea level (msl) at 1020:34. He made a right turn to a southerly heading at 1021:13, and began descending. He turned to the southwest for a downwind entry to the north of runway 07L.

A review of recorded air traffic control tower (ATCT) transmissions revealed that the pilot reported inbound for landing at 4,000 feet. He was cleared into the traffic pattern for a left downwind for runway 07L. There were several airplanes ahead of him, and he asked the ATCT to call his turn to left base for him. After his traffic called a 3 mile final, the ATCT advised him to turn, and he acknowledged.

While on final, the ATCT advised the pilot that he appeared to be low. He responded that he was experiencing engine difficulties, and would try to make the runway. Witnesses reported that the airplane was very low, and the engine was sputtering and backfiring. Several witnesses, including a couple of certified flight instructors, stated that the airplane's wings were rocking, and the nose attitude was high. One witness looking head-on at the airplane noted that the nose pitched down just prior to the airplane colliding with a building. The airplane and building caught fire; one witness reported that the building's sprinkler system activated, and an alarm sounded.

PERSONNEL INFORMATION

A review of Federal Aviation Administration (FAA) airman records revealed that the 72-year-old pilot possessed a private pilot certificate with ratings for airplane single-engine land, single-engine sea, multiengine land, and instrument airplane. The FAA issued the pilot a third-class medical certificate on February 24, 2010. It had the limitations that the pilot must wear and possess corrective lenses for near and distant vision.

On an insurance application dated September 3, 2010, the pilot reported that he had a total flight time of 1,872 hours as pilot-in-command. He had an estimated 647 hours in this make and model. He completed a biennial proficiency check in August 2009.

AIRCRAFT INFORMATION

The airplane was a Beechcraft F33A, serial number CE-380. A review of the airplane's logbooks revealed that the airplane had an annual inspection on May 1, 2010, at a total airframe time of 6,950.4 hours. The tachometer read 702.9 at the last inspection.

The engine was a Continental Motors, Inc. (CMI), IO-520-BA(8), serial number 280845-R. Total time recorded on the factory remanufactured engine was 702.9 hours at the last annual inspection.

COMMUNICATIONS

The pilot was in contact with the Phoenix Deer Valley airport traffic control tower (ATCT).

AIRPORT INFORMATION

The Airport/ Facility Directory, Southwest U. S., indicated that runway 07L was 4,508 feet long and 75 feet wide. The runway surface was asphalt. The parallel runway 07R was 8,197 feet long and 100 feet wide; the displaced threshold was 917 feet long.

WRECKAGE AND IMPACT INFORMATION

Investigators from the NTSB, FAA, Beechcraft, and CMI, examined the wreckage at the accident scene. The first identified point of contact (FIPC) was a ground scar. The debris path was along a magnetic heading of 052 degrees; the orientation of the fuselage was 052 degrees. Detailed on-site examination notes are in the public docket.

The airplane came to rest inside an industrial building lobby. The fuselage, one propeller blade, empennage, most of the right wing, most of the left wing with the aileron attached, and landing gear were contained within the office space.

The engine with two of the three propeller blades attached was displaced from the airframe; it went through a steel door and frame into a second room.

The first identified point of contact was a ground scar along a 052-degrees magnetic bearing that headed toward the left front edge of the building when facing the entrance. The scar continued across a red curb that was 10 feet from the office's entry point; the red curb exhibited scuff marks.

There was a scrape mark that angled up 38 degrees along the right side of the building's entrance. The bottom of the scrape started about 8 feet high.

Above the center of the entrance were two areas that had chunks of concrete broken off, and emanating from these disturbed areas were two sets of paint transfer marks. One mark was at the 1 o'clock position, and was similar in appearance to a propeller blade. The paint was black from its base to the end except for a small outboard portion that was red. Another paint transfer mark was at the 11 o'clock position; it was red and about the same distance from the base as the other red mark.

The portion of the left wing outboard of the aileron actuating rod was along the outside of the north facing side of the building.

The fuel selector valve was selected to the left tank.

The ignition switch was in the BOTH position. The auxiliary fuel pump switch was not observed. The flap position switch was in the OFF position.

The control cables for the rudder and elevators remained attached, and control continuity was established from the control surfaces to the buckled center cabin area.

The forward seats were intact and remained attached to their respective seat rails. The forward seat stops were not visible. The seats contained lap belts; the shoulder harnesses were hanging from the headliner, above and aft of the forward seats. The seat belts were not buckled, and the FAA inspector reported that first responders did not have to disconnect them to extricate the pilot.

MEDICAL AND PATHOLOGICAL INFORMATION

The Maricopa County Medical Examiner completed an autopsy, and determined that the cause of death was smoke inhalation. The FAA Bioaeronautical Science Research Laboratory, Oklahoma City, Oklahoma, performed toxicological testing of specimens of the pilot.

Analysis of the specimens contained findings of 41 percent for carbon monoxide detected in blood, and no cyanide detected in blood. There was no ethanol detected in vitreous.

The report contained the following findings for tested drugs: 67.87 (ug/ml, ug/g) acetaminophen detected in urine.

TESTS AND RESEARCH

Investigators examined the wreckage at Air Transport, Phoenix, on September 29, 2010. A complete report of the airframe and engine examination is in the public docket.

Engine

Investigators slung the engine from a hoist, and removed the top spark plugs. They rotated the crankshaft with a tool in an accessory drive gear. The crankshaft rotated freely through 360 degrees. The valves moved approximately the same amount of lift except for the exhaust valve for cylinder number two, which did not move at all. Cylinder number two sustained crush damage, had missing material, and a broken exhaust rocker arm. Investigators obtained thumb compression on all cylinders except cylinder number two. The fuel pump shaft rotated freely, and the gears in the accessory case turned freely.

A borescope inspection revealed no mechanical deformation on the valves, cylinder walls, or internal cylinder head. The combustion chambers and piston heads had a layer of white and yellow colored deposits. The numbers one, three, and five cylinder sparkplug electrode areas had white deposits.

Investigators manually rotated the magnetos, and both magnetos produced spark at all posts. The impulse couplings engaged.

Throttle Body/Fuel Metering Unit

CMI made the throttle body (part number 628528) and fuel metering (part number 629904-2) unit. The mixture control lever sustained damage; it was fractured at the control cable attachment. The fracture surface was jagged and angular. When manually manipulated, the mixture control lever slipped on the mixture control shaft, and moved independently of the shaft. Investigators removed the bronze control lever, and observed smearing of material at the beveled area of the lever. The mixture control shaft had some smeared bronze material within the beveled splines.

The throttle lever sustained damage, and was cracked near the throttle cable attachment area. The throttle control lever moved independently from the throttle control shaft. The throttle body and fuel control were disassembled, and the control levers were removed. The throttle lever's beveled area had smearing and deformation of material; the throttle control shaft had smeared bronze in the beveled spline area (with much more bronze material being noted in the splines of the throttle shaft than the splines of the mixture shaft).

JP Instruments (JPI) EDM-800 Unit

The airplane had a JPI EDM-800 unit installed that recorded various engine parameters. An NTSB specialist downloaded the recorded information and prepared a factual report, which is in the public docket for this accident. The date/time data was manually adjusted by the operator. The time was not accurately adjusted, so the specialist provided time as seconds elapsed since device power-up. The data indicated that the engine parameters were correlated for the majority of the flight. The data was presented both graphically (time in seconds) and in a spreadsheet (time in minutes:seconds).

At time zero, the revolutions per minute (rpm) were 1,077, fuel flow was 2.9 gallons per hour (gph), and manifold pressure (MAP) was 14 inches of mercury (inHG).

About 850 seconds (14:18), the rpm began increasing, and reached about 2,700 rpm. Fuel flow was about 27 gph, and MAP was about 27 inHG. At 1398 seconds (23:18), the rpm reduced to approximately 2,500 rpm, fuel flow decreased to about 19 gph, and MAP to 22 inHG.

At 1530 seconds (25:30); the fuel flow reduced to 12 gph. The EGT reached its peak for the cruise portion of the flight (approximately 1,500 degrees F, with slight variations between cylinders). The MAP, fuel flow, EGT, and cylinder head temperatures (CHT) all remain essentially constant until 2640 seconds (44:00), when slight variations occurred. Correlated to the GPS data, the slight variations began as the flight reversed course and began the descent. As the airplane continued in a gradual descent, the MAP slowly increased, the outside air temperature (OAT) increased, and the fuel flow stayed relatively constant between 12.5 and 13.3 gph.

About 3600 seconds (60:00), the EGTs and CHTs began to increase as MAP increased and the fuel flow stayed the same. The EGTs reached their maximum temperatures (1,550 to 1,590 degrees) during the flight at 4368 seconds (72:48).

About 3900 (65:00), the rpm and fuel flow began decreasing.

Throughout the cruise portion of the flight, the JPI calculated horsepower had been about 60 percent. At 4490 seconds (74:50), engine horsepower was indicating about 55 percent. Correlated to the GPS data, the airplane was about 2,600 feet msl (1,100 feet above ground level), and beginning the base turn. The rpm started a gradual decrease, and the EGTs for all six cylinders became erratic. The EGTs for cylinders one, two, and six dropped slightly, and stayed at higher values than the other cylinders. The lowest CHT (cylinder number four) was 550 degrees different from the highest cylinder CHT (number two).

The data stopped recording at 4620 seconds (77:00). The MAP final value was near ambient conditions, rpm was about 1,850, and fuel flow was about 8.3 gph. No spikes were observed in the fuel flow.

ADDITIONAL INFORMATION

Pilot's Operating Handbook (POH)

The POH states to enrich the mixture as required during descents. Before landing, the POH indicates that the pilot should select the full rich position.

The emergency procedures section discusses loss of engine power. The POH tells the pilot to check fuel flow; if it is abnormally low, turn the auxiliary fuel pump on and lean as required. It says to turn the auxiliary fuel pump off if there is no improvement.

CMI Service Bulletin SB08-03

On March 14, 2008, CMI issued service bulletin (SB) SB08-3 regarding throttle and mixture control levers. It noted that two types of throttle and mixture control levers were in use in the field. The original style control levers were manufactured from a bronze material, and featured a non-machined chamfer on one side, which mated to the machined chamfer on the throttle and mixture control shafts. Splines formed on the non-machined chamfer of the lever at installation. The control lever style in effect at the time of the SB were manufactured from stainless steel and featured a splined chamfer, which interlocked with the splined chamfer on the throttle and mixture control shafts. The SB provided inspection instructions at every 100-hour or annual inspection. The maintenance technician was to inspect the control lever for looseness, free play on the shaft, and proper installation.

On July 30, 2010, CMI issued revision A for SB08-03, which was effective at the time of the accident. This revision stated that all bronze material control levers must be replaced when removed for any reason. A review of the maintenance logbooks for the airframe and engine did not indicate that the levers had been removed. Part 91 operators are recommended, but not required, to implement service bulletins.

NTSB Probable Cause

A partial loss of engine power during approach for reasons that could not be determined because postaccident examination did not reveal any anomalies that would have precluded normal operation.