NTSB Factual Report

HISTORY OF FLIGHT

On April 27, 1996, at 1420 central daylight time, a Piper PA-28-151, N7845C, was destroyed when it impacted the ground in an uncontrolled descent following an in-flight fire near Vera, Oklahoma. The flight instructor, two student pilots, and one passenger were all fatally injured. The aircraft was registered to a private owner and operated by Kiamana Aviation under Title 14 CFR Part 91. Visual meteorological conditions prevailed for the cross country instructional flight that originated from Siloam Springs, Arkansas, approximately 50 minutes before the accident. A VFR flight plan had been filed.

The owner of the flight school reported to the investigator-in-charge that the airplane left Ponca City, Oklahoma, at approximately 0930 that morning for an estimated 1 hour and 20 minute flight to Siloam Springs, Arkansas. The owner further reported that the school's policy was to fully fuel all aircraft as soon as they land; therefore, he believes that "this airplane took off fully fueled."

According to witnesses in Siloam Springs, the airplane was on the ground for approximately 3 hours while the 4 occupants attended a regional "Spring Fly In." Records at the Smith Field FBO, near Siloam Springs, Arkansas, indicate that no fuel was purchased for the airplane on that date. A weather report was obtained and a flight plan was filed with Jonesboro FSS for the return trip to Ponca City.

The airplane departed Siloam Springs at 1330, climbed to 4,500 feet MSL, and the pilot called Jonesboro FSS to activate the flight plan. The pilot reported to Jonesboro FSS that they were presently in light to moderate turbulence and inquired if the weather at Ponca City was "still clear." Jonesboro FSS replied that Ponca City was 5,000 feet scattered, 15,000 feet overcast, 12 miles visibility, and the wind was 180 degrees for 13 knots.

The airplane was next seen near Vera, Oklahoma, by a witness, who works in the aviation industry. He reported that "I observed a small white aircraft with blue markings flying at a low altitude. Since I work in the aviation industry, I was somewhat startled to notice it was flying just above the tree tops." He observed the airplane for approximately "5 to 10 seconds before it disappeared beyond the trees, about 3/4 of a mile from him. I would estimate this to be about 1 mile from the site of the crash."

The investigator-in-charge queried the Tulsa approach control radar facility for a possible radar track of the accident airplane. They reported that they could not locate a transponder signal or a primary return signal that could be correlated with the airplane. Further questioning of the radar facility revealed that their equipment could not see below 500 feet AGL along the projected flight path of the accident airplane.

Witnesses near the accident site "heard a small engine plane" and their attention was drawn to it because "the engine tone changed pitch." They observed the airplane at very low altitude "banking side to side with white smoke trailing from it." The witnesses further stated that they saw the white smoke turn to black smoke and one witness saw "orange flame coming out the left side of the engine compartment." The aircraft was observed to roll left and impact the ground inverted, in a nearly vertical orientation.

PERSONNEL INFORMATION

The flight school reported to the investigator-in-charge that the instructor pilot had approximately 400 hours of flight time. FAA records indicate that the instructor pilot started his flying career by obtaining his student certificate on April 1, 1994, and he became a flight instructor on September 23, 1995.

The student pilot, who was in the left seat at the time of the accident, had received her student certificate 25 days prior to the accident. It was estimated from interviews with her friends that she had received approximately 11 hours of flight training prior to the accident. The second student pilot, who flew the first leg of the instructional cross country flight from Ponca City to Siloam Springs, had received his student certificate from the FAA on February 16, 1996.

AIRCRAFT INFORMATION

The accident airplane was built in 1975 and had accrued approximately 7,049 hours of flying time. The last annual inspection was performed 3 days before the accident at a tachometer time of 7,039.72 hours. The inspection checklist for the most recent annual inspection showed that the muffler and attachment clamps had been inspected. A top overhaul of the engine was performed on November 27, 1995, at a tachometer reading of 6,951.03. The engine's front exhaust muffler was replaced during the previous annual inspection on February 14, 1995, at a tachometer time of 6,781.21 hours. See attached copies of pertinent maintenance records.

The muffler system consists of two parallel mufflers with a heat exchanger shroud around them (see attached diagram). The right side of the muffler has two intake ports protruding from it where the number one exhaust stack and the number three exhaust stack are respectively inserted. The intake ports and the exhaust stacks each have a hole in them for attachment and alignment of the exhaust stacks to the muffler. A special clamp with a pin in it (see photo No.7) was designed to fit around the muffler intake port in such a manner as to allow the pin to penetrate the hole in the muffler intake port and the inserted exhaust stack. This securely attaches the muffler to the exhaust stacks and does not allow for any movement between them.

The left side of the muffler also has protruding intake ports for the purpose of receiving the number two exhaust stack and the number four exhaust stack respectively. The only difference is that there are no holes provided for securing them together and no clamps are needed for their attachment. They are designed to help hold the muffler in place and allow for expansion and contraction (see photo No.17).

METEOROLOGICAL INFORMATION

The student pilot called for weather at 0830 and was informed that there was an Airmet over eastern Oklahoma for occasional moderate turbulence below 6,000 feet. The student pilot called again for weather at approximately 1300 and was told that a Flight Precaution had been issued for the same region for occasional moderate turbulence from the surface to 15,000 feet. The pilot was informed shortly after takeoff from Siloam Springs, Arkansas, by Jonesboro FSS that the Flight Precaution had been updated to include moderate to occasionally severe turbulence from the surface to 14,000 feet.

WRECKAGE AND IMPACT INFORMATION

The airplane was found inverted, in an open field, with the nose approximately 70 degrees below the horizon. Both propeller blades remained attached to the crankshaft. One blade was bent back approximately 45 degrees from the hub, the other was bent back approximately 20 degrees and exhibited torsional twisting. Both blades had chordwise scratching. The engine was buried in the ground up to the level of the accessory case. The rear of the engine was covered by the fire wall and the remains of the instrument panel. The majority of the airplane was severely burned except for the right wing which displayed longitudinal crushing across the entire span. All structural components were accounted for, but, due to extent of the impact damage and post-crash fire, it was not possible to determine control continuity. The engine crankshaft was rotated and continuity to the accessory gears was confirmed.

The cabin entry door was found separated from the fuselage and not collocated with associated aircraft structural components. The cabin doors' upper latch was found in the unlatched position. Normal operation of this latch requires a positive engagement to secure it in the latched position and a positive disengagement to move it to the unlatched position.

One piece of wreckage was identified as the left lower portion of the engine cowling where it attaches to the firewall. This piece was located approximately 10 feet from the left side of the buried engine and on the edge of the post-crash burn area. There was an 8 inch by 18 inch hole burned through this cowling which had an area around it where no soot deposits were found (soot will not adhere above 700 degrees Fahrenheit). Adjacent to this hole, there was a burned hole on the engine firewall which measured 2 inches by 4 inches.

The lower left rubber engine mount was totally consumed by fire (see photo No.4), whereas the other three rubber mounts sustained minimal damage. The left magneto was found significantly more damaged by fire then the right magneto. The oil cooler was damaged by fire and the bottom of the oil cooler was found ruptured open.

Pieces of windshield plexiglas were found outside of the post-crash fire area. These pieces revealed heavy sooting and some melting on the interior side of the plexiglas.

The airplane's fire extinguishing bottle was found severely burnt and damaged in the wreckage debris. The plastic safety tie and seal for the fire extinguisher were located during wreckage recovery in an unburned condition and the safety tie was broken.

MEDICAL AND PATHOLOGICAL INFORMATION

Autopsies and toxicological tests were ordered and performed. The autopsies were performed by the Office of the Chief Medical Examiner at Tulsa, Oklahoma, on April 28 and 29, 1996. Toxicological test results were negative. Carbon monoxide analysis could not be performed due to a lack of suitable specimens.

FIRE

The number three exhaust stack was still attached to the exhaust muffler by an automotive type clamp. The number one exhaust stack had separated from the muffler and an automotive type attachment clamp was still present on the muffler intake port. The manufactured attachment/alignment holes were present (see photos No.5 and No.6) on the exhaust stacks and on the muffler intake ports, but were not being utilized for their intended purpose.

On the left side of the muffler, the number two muffler intake port was crushed flat (see photos No.8 and No.9) and it also had an automotive type clamp around it. There was no automotive clamp present on the number four muffler intake port, but there was a circumferential line which demarked varying degrees of oxidation (see photo No.10) and unoxidized multiple scratches parallel to the longitudinal axis of the intake port.

The number two exhaust stack was found separated from the muffler, but attached to the engine (see photo No.2) and minimally damaged(see photo No.11). The end of number four exhaust stack was partially closed due to an inward crushing of one of its' walls and this matched an inward crushing found on the number four muffler intake port. Additionally, the number four exhaust stack end had multiple circumferential wear rings (see photo No.10) where it attaches to the muffler intake port. Examination of both of these exhaust stack ends revealed thinning in the cross section and extensive scale along the inner diameter surface adjacent to their ends (see enclosed metallurgist's factual report).

The FAA Advisory Circular 43.13-1A, 1972, states, "The exhaust system often operates at red-hot temperatures of 1000 degrees F. or more; therefore, parts such as ignition leads, hoses, fuel lines, and flexible air ducts, should be protected from radiation and convection heating by heat shields or adequate clearance."

The fuel supply line to the carburetor connects to a t-fitting which in turn is connected to the left side of the carburetor (see photo No.12). The fuel supply line was protected with a firesleeve. According to the manufacturer of the firesleeve, "the purpose of this firesleeve is not to increase the service temperature of a hose line, but to protect the hose from direct fire long enough to allow appropriate action to be taken." The manufacturer further states that the operational temperature range for the fuel line is -40 degrees F to 300 degrees F and for the firesleeve it is -65 degrees F to 500 degrees F. The t-fitting was located approximately 8 inches aft of the number two exhaust stack connection point to the muffler.

The fuel supply line was found separated from its' aluminum sleeve socket which was still connected to the t-fitting. The seamless synthetic rubber inner portion of the fuel supply line was absent (see photo No.13). The aluminum sleeve socket showed little discoloration and little deformation. The manufacturer of the fuel supply line states that, "fuel supply lines are very flexible when new, but after they have been installed for a relatively short period of time, they take a "set", they become very rigid."

The t-fitting has a second fuel line (see photos No.12 and No.14) attached to it which leads to the cockpit fuel pressure gauge. According to the manufacturer of this fuel line, it has an operational temperature range of -40 degrees F to 250 degrees F and no firesleeve was found on the fuel line. The aluminum sleeve socket of the fuel line showed considerable deformation and exhibited a melting pattern which was consistent with high velocity air flow. This aluminum sleeve socket was located approximately 3 inches aft of the junction of the fuel supply line to the carburetor and its' aluminum sleeve socket (see photo No.12).

ADDITIONAL DATA

The airplane was released to the owner's representative.

NTSB Probable Cause

improper installation by unknown person(s) of incorrect exhaust system clamps, which led to an exhaust system disconnect and a subsequent engine compartment fire.