NTSB Factual Report

HISTORY OF FLIGHT

On July 12, 2002, about 1332 eastern daylight time, an American General AA-5, N6024L, registered to a private individual, operating as a Title 14 CFR Part 91 personal flight, nosed over while ditching in the Florida Everglades, in the vicinity of Boca Raton, Florida, following an emergency descent, following a loss of engine power. Visual meteorological conditions prevailed and no flight plan was filed. The aircraft received minor damage, and both private-rated pilots were fatally injured. The flight departed from Palm Beach County Glades Airport, Pahokee, Florida, the same day, about 1300.

Examination of the Fort Lauderdale Executive (FXE) Airport tower radio communication information revealed that at 1330 an occupant of N6024L had communicated on the FXE tower frequency stating that he was, "declaring an emergency" and was "17 nautical miles northwest", but did not state the nature of the emergency. The information showed that in response to the radio communication transmission, the tower controller responded, "say request", and in response the individual stated, "I'm at 400 feet, descending at 400 feet per minute, and 80 miles per hour." No other transmissions were reportedly heard from occupants of the aircraft.

Radar data obtained from FAA Miami Approach Control showed that the accident airplane's last radar position was verified at 13:32:06 at an altitude of 500 feet, on a heading of 156 degrees, and at a speed of 66 knots. At that time the airplane's transponder code was shown as 7700 with the aircraft on a bearing of 333 degrees and 23.64 NM from the radar facility. The airplane was found inverted, in about 5 feet of water, in the Loxahatchee National Wildlife Refuge, in Palm Beach County, Florida.

PERSONNEL INFORMATION

There were two occupants in the accident airplane. The owner of N6024L was found in the right seat, and a review of records showed that he held an FAA private pilot certificate with airplane single engine land and instrument airplane ratings. Records also showed that he held an FAA third-class medical certificate, issued on August 10, 2000, with the limitation, "Holder must wear corrective lenses." His personal flight log book indicated that he completed a biennial flight review on November 14, 2000, and an instrument proficiency check on May 26, 2002. It also showed that he had acquired about 1,805 flight hours, of which about 12 flight hours were flown in the preceding 6 months.

The second occupant of the airplane was found in the left seat, and records showed that he held an FAA private pilot certificate with airplane single engine land and instrument airplane ratings. Information obtained from the FAA showed that he held a third-class medical certificate, issued on February 2, 2002, and it contained no limitations or restrictions. Additional information was not obtained by the NTSB regarding this pilot's flight experience, but according to FAA records, his last application for a medical certificate was on February 2, 2002, at which time he reported having accumulated a total of 600 flight hours, 40 of which took place within the previous 6 months.

AIRCRAFT INFORMATION

The accident airplane is a 1972 American General AA-5, serial number AA5-0124. It was equipped with a Lycoming O-320-E2G engine, whose serial number was L-5792-27A. According to the airplane's maintenance records, the engine was overhauled on January 31, 1979, and installed on N6024L on February 24, 1979, at a total flight time of 1919.79 hours on the airplane. The engine had accumulated 1326.7 hours of service time during the 23 years following the overhaul and installation. The number 2 cylinder was changed in November 1991, and Numbers 1 and 3 cylinders were changed in December 1994. Logbook entries indicate that no repairs were performed on the No. 4 cylinder. The aircraft had undergone an annual/100 hour inspection on September 5, 2001, at a tachometer time of 3219.62 hours. During the inspection, the cylinder compression readings taken were: (1) 68/80 psi, (2) 75/80 psi, (3) 73/80 psi, and (4) 73/80 psi. The aircraft had flown 26.87 hours between the date of the annual/100 hour inspection and the date of the accident. The airplane was also equipped with a two-bladed McCauley propeller, serial number E18969.

METEOROLOGICAL INFORMATION

Information obtained from the National Weather Service Miami WSR-88D doppler radar showed that at 1327, approximately the same time as the accident, there was convective activity, specifically, some moderate thunderstorm echoes west of the accident site. The information also showed that the thunderstorms moved into the area of where the accident occurred at about 1700.

Special surface observation taken at the Fort Lauderdale Executive Airport at 1340, showed winds to be calm, visibility 10 statute miles, scattered clouds at 9,000 feet, temperature 88 degrees F, dewpoint temperature 72 degrees F, and the altimeter setting was 29.99 inches Hg.

WRECKAGE AND IMPACT INFORMATION

N6204L was found in about 5 feet of water in an area accessible only by boat or helicopter. The airplane had come to rest on a heading of about 310 degrees, approximately 400 yards north of Levy 39 which defined the southern boundary of the Loxahatchee National Wildlife Refuge. The wreckage was located 4.1 miles northwest of the Loxahatchee Road entrance gate, at coordinates, 26 degrees 22.852 minutes North latitude, 080 degrees 22.241 minutes West longitude. The airplane had nosed over upon impact, and it lay inverted on a heading paralleling the road atop Levy 39.

Examination of the scene revealed that all structural components of the airplane were located in the immediate vicinity. The airplane wreckage was recovered from the water using a helicopter with an external sling. When examined, the flight controls were found to be intact, and there was control continuity for roll, pitch and yaw. The throttle was observed to be about 1.5 inches aft, the electric fuel pump switch was in the "off" position, the magnetos were on "both", the mixture control was at idle-cutoff, the primer knob was in and locked, and the carburetor heat was off. The fuel selector was on "left main". The lever controlling the flaps showed an extension of 5 degrees, but when verified, the flaps were found to have been fully retracted. Elevator trim was set very near to the neutral position. A transponder code of 7700 had been entered. Approximately 15.5 gallons of fuel were drained from the left wing tank and 13.0 gallons from the right wing tank. Examination of the fuel as well as the last refueling source revealed no evidence of contamination. The fuel system was intact, and fuel was found to be present throughout. The engine was saturated with water, as a result of being immersed, but oil was also present within the engine crankcase and sump. The engine accessory components were found to be secure. Initial attempts to rotate the propeller revealed restricted movement, so the airplane was then removed from the levy for a detailed examination.

Follow-on examination of the engine revealed that the interior of the No. 4 cylinder assembly had incurred mechanical damage. Both the top of the piston and the cylinder head exhibited multiple indentations, and a hole was found to be present in the top of the piston. The exhaust valve tulip was found to be separated from the stem. The oil pressure screen was inspected and found to contain a small amount of aluminum particles. The oil suction screen was free of debris, but an accumulation of sludge was found inside the plug. The oil pump gears were found to be intact. The No. 4 cylinder assembly was removed to inspect and retrieve broken parts, and during disassembly, fragments of the exhaust valve tulip were found within the combustion chamber and in the exhaust system. The fragments varied both in size and shape and approximately 50 percent of the tulip was not recovered. The majority of the valve stem remained in the guide and was not removed.

Various steel components within the power section exhibited corrosion. Corrosion was found on the crankshaft, camshaft, and accessory gears. The bearing surfaces within the power section exhibited evidence of dirt/particle embedment. The valve train components, camshaft, and hydraulic lifters were inspected and were found to have remained intact. The NTSB retained the parts pertaining to the No. 4 cylinder assembly, to include the piston, exhaust valve fragments, along with the hydraulic lifter assemblies and the center main bearing for laboratory analysis

The exhaust valve stem was examined for evidence of exhaust valve guide material transfer, and no guide transfer material was found. The stem fragment contained black/gray oxidation discoloration and erosion all around the stem wall between the fracture and a position approximately 2.55 inches from the fracture towards the stem tip. The total length of the fractured exhaust stem from tip to fracture was measured to be approximately 3.65 inches. Erosion was most severe between the fracture and 0.45 inch from the fracture, and moderate erosion damage was noted in the area between 0.45 inches and 2.25 inches from the fractured region. Beginning erosion damage was noted in the area between 2.25 and 2.55 inches from the fractured region. A small portion of the stem, near the tip, did not appear to contain erosion damage. The outer diameter in this area of the stem measured 0.4925 inch. For explanatory purposes, referred to as the upper portion of the exhaust valve stem, an area of the stem near to the valve head, referred to as the lower portion. The outer diameter on the lower portion of the stem, which did contain erosion damage, its measurement was 0.4869 inch.

According to information obtained from Lycoming Engines, the specified diameter for a new exhaust valve stem on a parallel valve head cylinder is 0.4935 inch to 0.4945 inch and the service minimum diameter permitted is 0.4915 inch. Visual examination of the open end of the exhaust port for the number 4 cylinder revealed that the lower portion of the exhaust valve guide was cracked over a distance of approximately 8 millimeters; an unmarked arrow in figure 5 indicates the cracked area. The inside diameter of the exhaust valve guide lower portion beyond the cracked region measured 0.5160 inch. The inside diameter of the exhaust valve guide upper portion measured 0.5030 inch. The diameter specified by the manufacturer for a new guide is 0.4985 inch to 0.4995 inch. The exhaust valve guide is expected to wear under normal service conditions, leading to an increase in the inside diameter. An increase in inside diameter up to 0.5145 inch is permitted with up to 1,500 hours of flight time.

The base metal on the exhaust valve stem was examined with a scanning electron microscope with energy dispersive spectroscopy (EDS) to determine whether the exhaust valves were pre-Lycoming Service Bulletin 240, which suggested that old exhaust valves made from martensitic stainless steal, be replaced with the new valves made with a nickel based superalloy. EDS spectra from the stem showed major peaks of iron and chromium. In addition, small peaks for the lower atomic element silicon and the metallic element nickel were also generated. The presence of iron, chromium, and nickel is consistent with a stainless steel, and inconsistent with a nickel-base alloy.

MEDICAL AND PATHOLOGICAL INFORMATION

According to the Coral Springs Fire Rescue divers who responded to the accident, the underwater visibility in the area where the accident had occurred was zero. They further stated that by "feel", they were able to determine that the airplane canopy was closed, and had not been broken. They said that they had difficulty finding the canopy latch, and in an effort to remove the victims they used a pry-bar to remove the glass from the canopy frame. The divers said that they found the pilot/owner in the right/front seat with the seat belt and harness secured, and the left seat occupant was found with the seat belt and harness unsecured.

Postmortem examinations of the left pilot seat occupant, as well as the right pilot seat occupant were performed by a medical examiner with the District Medical Examiner's Office, District 15, West Palm Beach, Florida, and the cause of death in both pilots was deemed to be due to drowning.

Wuesthoff Reference Laboratory, Melbourne, Florida, performed postmortem toxicology tests on specimens obtained from the owner/private-rated pilot. The specimens were tested for cannabinoids, cocaine metab, opiates, benzo, barbiturates, and tricyclics, and none were found to exist. Wuesthoff Reference Laboratory also performed postmortem toxicology tests on specimens obtained from the other private-rated pilot. The specimens were tested for cannabinoids, cocaine metab, opiates, benzo, propoxyphene, methadone, PCP, methaqual, barbiturates, amph/meth and salicylates, and none were found to exist.

The FAA Toxicology and Accident Research Laboratory, Oklahoma City, Oklahoma, performed toxicology studies on samples from both pilots. The samples were tested for carbon monoxide, cyanide, volatiles, and drugs, and for both pilots none of the substances were found to exist.

TEST AND RESEARCH

Reportedly, this engine’s last overhaul had been 23 years before the accident. At the time of the accident, the engine had reportedly accumulated 1246.49 hours time in service. Lycoming Service Instruction No. 1009AQ indicates that the O-320 Series engine should be overhauled after 2,000 hours time in service. Engines that do not accumulate the hourly time between overhauls are recommended to be overhauled in the twelfth (12th) year. In addition, Lycoming Mandatory Service Bulletin No. 388B addresses the subject: Procedure to Determine Valve and Guide Condition, and a review of maintenance records for the subject engine showed that on July 2, 1999, subject engine had been given a "Lycoming 400-hour valve inspection."

Lycoming Service Bulletin (SB) 388B, dated May 13, 1992, describes two methods for determining exhaust valve and guide condition. SB 388B suggested inspections at a maximum interval of 400-hours time in service for airplane engines. In the first method, the clearance between the exhaust valve stem and the exhaust valve guide is measured. The difference between the outer diameter of the exhaust valve stem and the inner diameter of the exhaust valve guide should be 0.0020 inch to 0.0038 inch. For the No. 4 cylinder in the accident engine, this difference is between 0.0236 inch and 0.0259 inch. According to SB 388B, Failure to comply with the provisions of this publication could result in engine failure due to excessive carbon build up between the valve guide and valve stem resulting in sticking exhaust valves; or broken exhaust valves which result from excessive wear (bell-mouthing) of the exhaust valve guide. The hardness of the No. 4 cylinder exhaust valve stem varied from 33 HRC to 43 HRC; the hardness increased moving from the lower portion of the stem to the upper portion. The specified hardness range for the exhaust valve stem is 70 to 77 HRA (39 to 52 HRC) for the part. In 1983, Lycoming changed the part number for the exhaust valve in this engine. Service Bulletin 240 suggests that the old exhaust valves, made from martensitic stainless steel, should be replaced with the new part, made from a nickel-base superalloy, at the time of overhaul.

ADDITIONAL INFORMATION

The NTSB released the airplane wreckage and all components that had been retained by the NTSB were released on February 4, 2005, to Mr. Les Sychek, Adjuster, AIG Aviation.

This report was modified on March 29, 2005.

NTSB Probable Cause

The owner/operator's failure to perform an overhaul or comply with a service bulletin 240 which would have probably precluded the continued installation of the number four cylinder exhaust valve and its subsequent failure.