NTSB Factual Report

HISTORY OF FLIGHT

On September 26, 1996, about 2314 central daylight time, a Piper PA-28-180, N8803J, registered to Guilford Air Service, crashed while on approach to the Birmingham International Airport, Birmingham Alabama. Visual meteorological conditions prevailed at the time and an IFR Flight plan was filed for the 14 CFR Part 91 personal flight. The airplane was substantially damaged and the commercial-rated pilot and pilot-rated front seat passenger were seriously injured. A rear seat passenger was fatally injured. The flight originated about 1953 eastern daylight time from the Air Harbor Airport, Greensboro, North Carolina.

The pilot-in-command stated that the fuel tanks were full and the flight departed VFR then about 3 minutes after takeoff, he obtained his IFR clearance. He has no recollection of the second half of the flight.

According to a summary of communications with the Greensboro, North Carolina Air Traffic Control Tower (ATCT) at 1956, the pilot called and requested his IFR clearance. The pilot was advised to proceed northbound and radar vectors were given to the BZM V222 airway from the Barretts Mountain VOR, then as filed in the flight plan. After departure the flight was cleared to climb to the IFR requested altitude of 8,000 feet then to 10,000 feet at the request of the pilot. The flight continued and descended to 8,000 feet then while in contact with the Atlanta Air Route Traffic Control Center, at about the midpoint of the flight, the pilot requested a descent to 6,000 feet, which was granted. Air Traffic Control was transferred to Birmingham Approach Control and at 2245, the pilot was advised by the approach controller to expect the straight in localizer approach to runway 23 which was acknowledged by the pilot. The flight was vectored to the final approach course and at 2254, the pilot was advised to descend and maintain 4,000 feet. At 2302.24, the flight was vectored to intercept the localizer course and at 2307.57, he was advised to descend and maintain 2,500 feet. The controller asked if the flight was able to maintain 120 [knots] which the pilot acknowledged. The pilot was then advised to contact the tower controller and at 2311.41, the pilot advised the controller "and zero three juliett lost engine, ah emergency ill try and put it down on this road down here." The controller acknowledged the statement and the pilot then advised that he would try to land on a highway. There was no further communication with the accident pilot.

According to the pilot-rated passenger in the left front seat, he verified before the flight departed that each fuel tank was filled to the top of the filler neck area. He stated that the pilot-in-command performed the fuel calculations for the flight which was planned for 10,000 feet. The passenger stated that he performed the takeoff but the pilot-in-command took over the controls when the flight entered to clouds which was the main portion of the flight.

According to witnesses near the accident site, both observed the airplane flying eastbound then saw a white flash of light when the airplane collided with power lines. The airplane then descended left wing low and impacted a vehicle then the ground. One witness arrived at the accident site less than 1 minute after the accident and did not smell or observe any fuel leaking. Both did not recall hearing the engine operating during the descent.

The pilot-in-command further stated that during the preflight planning, he used 8.0-9.0 gallons-per-hour for his fuel consumption calculations. This was based on his last 6 months of flying the airplane on training flights. He did not perform fuel consumption calculations using the performance charts in the owners handbook. The flight altitude varied between 8,000 and 9,000 feet and he stated that he typically leans the fuel/air ratio by pulling the mixture control until the engine begins to run rough then he enrichens slightly the fuel/air ratio.

PERSONNEL INFORMATION

Information pertaining to the first pilot seated in the right front seat is contained in the First Pilot information blocks on page 3 of the Factual Report-Aviation. Additionally, review of the pilot-in-command's pilot logbook revealed that from July 5, 1996, excluding the accident flight, he had flown the accident airplane on 17 cross-country flights lasting a total of about 24 hours. Information pertaining to the pilot-rated passenger seated in the left front seat is contained in Supplement E.

AIRCRAFT INFORMATION

Information pertaining to the airplane is contained in the Aircraft Information Blocks on page 2 of the Factual Report-Aviation. Review of the aircraft logbooks revealed that there was no entry that indicated that the right rear seat outboard restraint attach cable had been replaced. The carburetor that was installed had been overhauled last on June 12, 1967, and was installed on January 6, 1968, at a tachometer time of 1277.0 hours. The engine had been overhauled in November 1976, at a tachometer time of 3,352 hours. According to a non-mandatory service bulletin dated November 18, 1991, the carburetor is required to be overhauled when the engine is overhauled or 10 years following overhaul of the unit or placement of the unit in service, whichever occurs first. The carburetor had been installed 28 years 8 months at the time of the accident. The tachometer at the accident was noted to indicate 4,866.50.

METEOROLOGICAL INFORMATION

Visual meteorological conditions prevailed at the time of the accident. Additional weather information is contained on pages 3 and 4 of the Factual Report-Aviation. The pilot obtained a standard weather briefing via telephone and during that briefing was advised that the winds aloft at 9,000 feet from the departure airport area, midpoint of the flight, and near the destination were from 240 degrees at 20 knots, 200 degrees at 29 knots, and 200 degrees at 28 knots respectively. He was also advised that the winds aloft at 6,000 feet at the midpoint of the flight were from 180 degrees at 31 knots. The flight was vectored at one point for weather avoidance.

COMMUNICATIONS

The pilot was in contact with the Birmingham International Airport Air Traffic Control Tower at the time of the accident. A transcript of communications is an attachment to this report.

AERODROME INFORMATION

The pilot was cleared to execute the Localizer Approach to runway 23. The crash site was located 3.25 nautical miles and 067 degrees from the center of the airport.

WRECKAGE AND IMPACT INFORMATION

Examination of the accident site revealed that the airplane first collided with the top and middle of unmarked power lines suspended 49.5 and 48.5 feet above the surface of a road. The airplane then impacted the ground on a heading of 110 degrees, 65 feet from the point of contact with the power line and also impacted a vehicle at that time. The airplane came to rest upright on a heading of 338 degrees in the eastbound lanes of the road. All components necessary to sustain flight were attached or partially attached to the airframe. An approximate 2 foot by 2 foot stain was noted on the ground beneath the left wing fuel tank area. The forward and main spars of that wing were observed to be failed with evidence of overload failure. No signs of fuel leakage was noted on the top or bottom skins of the right wing fuel tank aft of either the filler cap or from the sump drain. No sign of fuel leakage was noted aft of the left wing fuel cap or in the cockpit or cabin area. Additionally, no fuel leakage was noted inside either of the wings. Both fuel caps were observed to be in place. Examination of the left wing revealed it was displaced up at about a 45-degree angle and chordwise crushing was noted of the leading edge skin. The right wing fuel tank which was intact was drained and found to contain 8 ounces of 100 Low Lead fuel. Examination of the elevator, rudder, and aileron flight controls revealed no evidence of preimpact failure or malfunction. The flaps were determined to be extended 25 degrees and the fuel selector was found in the off position.

The engine was placed in a test cell and impact damage to components which consisted of the No. 2 cylinder valve cover, starter, and the No. 2 cylinder bottom spark plug were replaced for the attempted engine run. Before starting the engine an auxiliary fuel pump was activated and fuel seepage was noted from the carburetor. The carburetor was lightly tapped with a mallet and the leakage stopped. The engine was then started and operated for 18 minutes. During that time when checked, each magneto drop was within limits though the difference was 70 rpm. With the installed propeller the maximum rpm at with the mixture full rich was 2,300 rpm and black smoke was noted from the exhaust. The fuel/air ratio was leaned and with the same propeller installed, the rpm increased to 2,350 rpm. With the fuel/air ratio leaned the difference in magneto drops decreased to within 30 rpm of each other. Following completion of the engine run the carburetor was removed and opened for inspection which revealed that the float height was improper and the material on the end of the needle valve was worn. The carburetor was then sent to another facility for bench testing. See tests and research section of this report.

MEDICAL AND PATHOLOGICAL

Toxicological analysis of specimens of the pilot was performed by the FAA Toxicology and Accident Research Laboratory. The results were negative for ethanol and tested drugs. Diazepam (.286 ug/ml) was detected in the blood. According to the hospital records, 10 mg valium was administered to the pilot during transport. Review of the Physicians Desk Reference under the chemical name valium revealed each ml contains 5 mg diazepam.

SURVIVAL ASPECTS

By design, each rear seat restraint is secured to the airframe. Examination of the right rear seat outboard seat belt assembly revealed the cable assembly was separated from the clevis. The clevis was attached to the attach fitting on the airframe but the swaged fitting was missing and was not located. The male/female sections of that lapbelt were secured with evidence of usage. Examination of the structural area of the right rear seat revealed no evidence of floor deformation or sidewall deformation. The seat back of the rear bench style rear seat was observed to be bent down and aft but the seat remained in the airplane. The cable assembly was retained for metallurgical examination. See tests and research section of this report. Additionally, examination of the right front seat outboard lapbelt assembly revealed that the webbing became unlaced from the attach triangle attached to the airframe. No determination could be made as to the reason for the separation.

TESTS AND RESEARCH

Metallurgical examination of the right rear seat cable assembly revealed no evidence of wear or corrosion. Optical examination of the cable strands revealed evidence of tensile overstress separations. The left rear outboard seat cable segment was removed from the airplane and pull tested to determine whether it met design criteria. A tension test was administered in which an in-line load of 2,000 pounds was applied and held. There was no evidence of slippage of the swaged fitting.

Additionally, the left rear seat male portion of the lapbelt assembly was also pull tested to determine whether it would unravel. The lapbelt performed as designed with no slippage noted.

A report of the examination of the carburetor is an attachment to this report. The economizer jet was improperly installed, in addition to the previous findings of the float height was incorrect, and the synthetic material on the float needle valve was worn. Bench test of the carburetor revealed that at 4 of the 6 test points the fuel flow in pounds per hour was greater than the maximum specified. According to representatives of the carburetor FAA-PMA facility that bench tested the carburetor, the excessive fuel flow in pounds per hour would be caused by the improper float height and the improper economizer jet setting. The representative also stated that the pilot could control this by leaning the fuel/air ratio. The pilot indicated in his IFR flight plan that the true airspeed would be 120 knots (KTAS). The performance charts found in the owners handbook for the airplane indicate true airspeeds in miles per hour. To obtain the 120 KTAS requires a TAS in miles per hour of about 138 which can be obtained flying at the planned altitude at about 70 percent power. At that power setting with the fuel/air ration leaned, the fuel burn is 9.3 gallons per hour. The performance charts do not indicate fuel used for engine start, taxi, climb.

ADDITIONAL DATA/INFORMATION

All seatbelts in the airplane were replaced on August 4, 1994. Examination of the seatbelts and the installation procedure instructions revealed all but the right outboard belt which separated, were properly installed.

Review of the airplane certification standards revealed the seat restraints were designed to withstand 9.0 G's forward.

According to the Birmingham, Alabama, Police Department Report, downed power lines were covering I-59 which caused three separate vehicular accidents with no injuries to the occupants of any of the vehicles. Additionally, the driver of the vehicle that the contacted the airplane was not injured.

The pilot-rated passenger told the airplane owner that during the latter portion of the flight, the pilot-in-command noted that the fuel gauges were indicating near empty and the flight had passed over two airports. When the pilot-in-command observed the city lights of the Birmingham, Alabama, area, he elected to continue the flight.

The wreckage minus the retained components was released to Mr. Bill Harwell on September 28, 1996. All retained components were released also to Mr. Bill Harwell on March 19, 1997.

NTSB Probable Cause

The pilot-in-command's improper in-flight planning/decision for electing to continue the flight after observing that the fuel gauges indicated empty. Contributing to the accident was an improperly adjusted economizer jet and an out of adjustment float. Also contributing to the accident was unsuitable terrain encountered during the forced landing.