NTSB Factual Report

HISTORY OF FLIGHT

On August 5, 1997, about 0945 central daylight time, an amateur built Adventurer 333, N323VA, suffered an in-flight separation of both wings and the empennage prior to impacting the terrain, in a swamp, about five miles northwest of Fond Du Lac, Wisconsin. The airplane was destroyed. The airline transport rated pilot sustained fatal injuries. The 14 CFR Part 91 personal flight was operating in visual meteorological conditions. No flight plan was on file. The flight departed Oshkosh, Wisconsin, about 0930. The destination was unknown.

The airplane had been on display at the Experimental Aircraft Association's annual convention in Oshkosh, Wisconsin, about 14 miles north of the accident site, for several days prior to the accident flight. It was determined that no flights of the airplane had occurred in the intervening period between the arrival at Oshkosh and the accident flight. The flight initiated on the day of the accident was for the purpose of returning the airplane to Arkansas.

Several witnesses to the flight of the accident airplane were interviewed. None of those witnesses indicated that they either observed the airplane prior to or at the time of the in-flight separation, nor to the impact in the swamp. The witness statements are attached as an addendum to this report, and included in the Fond du Lac County Sheriff's report. Some witnesses described seeing the airplane descend in pieces and heard the engine "sputtering." At least one witness said she thought the pieces might be the wings.

PERSONNEL INFORMATION

The pilot, born September 27, 1928, was the holder of an Airline Transport Pilot Certificate. His most recent biennial flight review was on February 28, 1997. He was the holder of a third class medical certificate issued March 15, 1996, with the limitation, "Airman must possess lenses corrected for near and distant vision to exercise privileges of this certificate."

AIRCRAFT INFORMATION

The airplane was an amateur built amphibian aircraft. The airframe logbook indicated that that airplane was completed and received certification on February 23, 1997. The logbook did not indicate the time in service; however an interview with a family member of the pilot/owner indicated a total time between 19 and 40 hours, at the time of the accident. The airplane had been involved in a landing mishap on its initial flight, sustaining damage to the left landing gear and wing. At that time, the pilot indicated the off airport landing was due to an engine malfunction.

WRECKAGE AND IMPACT INFORMATION

The wreckage of the airplane was located in the Eldorado Marsh, in Fond du Lac County, near Fond Du Lac, Wisconsin. The accident scene was accessible on the surface by airboat only, due to the swamp and remote location of the wreckage. From the air the debris trail was oriented on a southerly heading. The first identifiable item found was the horizontal stabilizer with the elevator attached. The right and left wings, and the fuselage followed in that order. The engine and propeller remained with the fuselage. The fuselage was inverted and oriented on a southerly heading. The debris trail was approximately 1/8 mile long. The wreckage was located in the swamp and was in an area of water (swamp) covered by dense vegetation. The wreckage was removed by helicopter and examined at a remote location.

The portion of the vertical stabilizer extending above the horizontal stabilizer and the complete rudder was never located. The horizontal stabilizer with the elevator attached was located and had separated from the vertical stabilizer attachment. Both wings had separated in flight. The spars showed distortion and failure in a downward direction. The lift struts failed at their attach points and were bent inward and upward. The fuselage was crushed during the ground impact sequence and the engine, which was found adjacent to the fuselage, was attached to it by cables and wiring only. The propeller was manufactured of composite material. There was no obvious leading edge damage; however, several blades were broken from the pusher engine/propeller impacting the terrain with the fuselage.

During the examination indications of fuel were found in the water around the wreckage and a strong smell of fuel permeated the wreckage on recovery. Due to the extensive damage and in-flight separation of flight controls, continuity of the controls was not possible. No cable driven controls gave an indication of failure other than could be associated with an in-flight separation of the controls to which they were attached. An examination of the engine and propeller failed to reveal any pre-existing anomalies.

MEDICAL AND PATHOLOGICAL INFORMATION

A post-mortem examination of the pilot was conducted by the Milwaukee County Medical Examiner, in Milwaukee, Wisconsin, on August 6, 1997. No pre-existing anomalies were found during the autopsy. Specimens were obtained from the pilot and submitted for toxicological examination and were found to be negative for those drugs screened.

TESTS AND RESEARCH

After the initial examination of the accident airplane six other similar airplanes were located in various phases of construction. Each of these airplanes was examined individually. Two of the airplanes were being built from plans and parts, while the other four were being constructed from "rapid build" kits. None of these airplanes differed in design from the accident airplane. The accident airplane was constructed from a "rapid build" kit.

A structural engineer of the NTSB's "Office of Aviation Safety" assessed components from the accident airplane from a composite construction evaluation aspect. His "Composite Chairman's Factual Report" is attached as an addendum to this report.

The structural engineer's examination included the horizontal stabilizer and portions of the vertical stabilizer and rudder hinge attachment areas. The top attachment area was examined without the matching vertical component and the bottom attachment area was examined in concert with the associated vertical stabilizer. The top and bottom attachment areas on the horizontal stabilizer were smooth and clean, and did not show signs of failure of the structure. The bonding areas were found clean with little resin or other adhesive in the respective attachment areas. The fiberglass cloth was easily seen and had a dry, fibrous appearance. The matching bonding surface had the fillet strips attached. The strips were smooth, had paint adhering to the outside surfaces, and had resin in and around the fiberglass. The attachment areas did not show any signs similar to extreme vibratory oscillations nor evidence of bird strikes.

The attachment areas of the lower vertical stabilizer to the horizontal stabilizer were examined and revealed that the horizontal stabilizer was attached to the vertical with multiple layers of fiberglass with resin impregnated into the glass forming a fillet on the outside of the left and right sides of the vertical stabilizer. No fillet material was found on the inside of the stabilizer. Adventurer 333 plans were examined and revealed that the original plans/kit had the builder attaching the horizontal stabilizer to the vertical. (The accident airplane had the horizontal attached at the kit manufacturer's facility). In the original plans, the builder was instructed to use the fillet attachment procedure on the insides as well as the outsides of the vertical stabilizer. The accident airplane has two lower and two upper fillets, all on the outside of the stabilizer.

The examination showed that material in the surfaces conformed to the requirements for properly processed fiberglass cloth. The resin used in the composite materials was similar to polyester resin. According to "Composite Aircraft Design" by Martin Hollmann, "... polyester resins are primarily designed for low strength applications where low cost, ease of producibility, low viscosity, and good wetting out are needed."

During the examination of the six other airplanes under construction the builders were asked about any problems they encountered with the product they received from the kit supplier/manufacturer. The main complaint centered on "quality control" issues. There were several complaints about the fiberglass showing a tendency to delaminate in multiple areas from the foam core. In all such cases the builder was required to inject polyester (or epoxy) into the areas to correct and stabilize the faults. One builder said. "It seemed that the more the unit was handled or moved, and stressed, the more skin tended to delaminate."

Examination of existing problems pointed out by builders indicated the presence of parting wax compound in the area where adhesion of parts should have taken place. Specifically on one airplane, the prop-wash deflector shield had peeled free of the wing. On another airplane in the attachment area between the horizontal and vertical stabilizer, the fiberglass tape had loosened and it was easy to peel the tape from both surfaces. In both of these instances there was evidence of the presence of the parting wax compound. Likewise both of these airplanes were of the "rapid build" kits which were supplied by the kit supplier/manufacturer.

One particular airplane revealed misalignment of the horizontal stabilizer where it was attached to the vertical stabilizer. The horizontal stabilizer was installed "out of plumb," as it was canted to right side of the fuselage. This misalignment was visible to the naked eye. This airplane was also a "rapid build" kit supplied by the kit supplier/manufacturer.

ADDITIONAL INFORMATION

A party to the investigation was the Federal Aviation Administration, Flight Standards District Office, Milwaukee, Wisconsin.

The wreckage was released to a representative of the pilot/owner on January 23, 1998.

NTSB Probable Cause

The kit manufacturer's use of inadequate material in construction of the airplane, his inadequate application of quality controls resulting in the debonding of the horizontal stabilizer. Factors were the total loss of the horizontal stabilizer, the elevator, the rudder, both wings, and a partial loss of the vertical stabilizer, and wing struts.