NTSB Factual Report

HISTORY OF FLIGHT

On September 6, 1997, about 1415 central daylight time, a Cessna U206B float plane, N206KY, was destroyed when it nosed over during a touch and go landing at Buckhorn Bay, Gilbertsville, Kentucky, while on a 14 CFR Part 91 sightseeing flight. The certificated airline transport pilot and one passenger received minor injuries. Two other passengers were fatally injured. Visual meteorological conditions prevailed and no flight plan was filed. The flight originated from Kuttawa Harbor Seaplane Base, Kuttawa, Kentucky, about 1250.

The pilot was interviewed on September 8, 1997. Present at the interview were the NTSB Investigator In Charge (IIC), and an attorney who represented the pilot. The pilot reported that the purpose of the flight was for sightseeing. During the sightseeing flight, the passengers asked to fly over a camp ground near Gilbertsville, Kentucky. Once the flight arrived near Gilbertsville, the pilot performed an uneventful touch and go landing at Buckhorn Bay. The pilot then performed a second touch and go with 10 degrees of flaps at a touch down speed of 65 to 75 miles per hour and a vertical descent between 100 and 200 feet per minute. During the touch down, the nose was pitched up between 6 to 8 degrees, and the floats contacted the water about mid section. A few seconds after contact with the water, the airplane nosed over. The pilot stated that he did not experience any problems with the engine during the flight.

A witness located in a boat in Buckhorn Bay approximately 4/10 of a mile from the accident stated; "The wind and water were calm as the plane touched down in the mouth of Buckhorn Bay, slightly west of the midline, approximately 3 [to] 4 tenths of a mile from my location across from Moors swim area. What appeared to be a smooth landing suddenly turned disastrous as the plane abruptly flipped forward turning [one half] revolution tail over nose resting on its back and sinking in the process."

Other witnesses stated that the winds were calm, there didn't appear to be any obstructions in the water, at the time of the accident.

WRECKAGE DESCRIPTION

The wreckage was examined by the Safety Board IIC, on September 8 and 9, 1997. The examination revealed that all major components of the airplane were accounted for. During the examination, flight control continuity was confirmed from the elevator, rudder, and water rudders through the cockpit controls. The right wing was separated and the left wing was removed during the recovery; however flight control continuity was confirmed from the ailerons to the control cables at the wing root. Examination of the Whipline manufactured model 3900 floats revealed that the right float, bottom outboard skin, had separated from the keel. The skin was compressed upward and aft from the front of the float about 8 feet aft. There was no evidence of scratches or impact marks on the separated skin of the right float.

On November 20, 1997, the engine test ran satisfactorily at the Teledyne Continental Motors, Mobile, Alabama, under the supervision of an Federal Aviation Administration Inspector (FAA).

MAINTENANCE

A review of the maintenance records revealed that the outboard bottom skin, sister keel brace, and sister keel, were replaced on the right float, May 15, 1997. The airplane had incurred 48 hours total flight time since the repair.

PILOT INFORMATION

The pilot held an Airline Transport Pilot Certificate with ratings for single engine land and sea, and multi-engine land. He was issued a Second Class medical certificate by the FAA on November 21, 1996. The pilot's most recent flight review was on September 6, 1996, in a Piper J-3. He had a total flight experience of 23,400 hours, of which 12,000 were as pilot in command, and 62 hours were in make and model incurred in the pervious 30 days.

ADDITIONAL INFORMATION

During a telephone interview with a mechanic on September 11, 1997, he stated that he performed the repair, and that the repair was accomplished according to the float manufacturer's repair manual and Advisory Circular 43.13. Also, when the mechanic contacted the float manufacturer regarding the repair, the manufacturer recommended to replace the float; however, the mechanic elected to proceed with the repair.

The service manual for repairing float hull skins, revision dated January 15,1997, was the latest for the Whipline model 3730 amphibious float with electro power pack and the model 3900 floats.

The manual stated:

"All outside hull skins are bonded to corner extrusions, chines, keel, etc., with a special heat pressure 3M adhesive. This bond adheres skins to inside of all extrusions. If skin bond must be broken from extrusion for a large distance, heating extrusion with a propane torch until bond starts to loosen and slide apart is the simplest method. CAUTION must be taken not to heat and loosen bonds not needing replacement. Skins may be reattached to extrusions by method shown in figures 10, 11 and 12. If skin bond must be broken from extrusion for a large distance, heating extrusion with a propane torch until bond starts to loosen and slide apart is the simplest method." A review of figures 10, 11 and 12 revealed that the outboard bottom skin was reattached through riveting the new skin to the keel, or to preexisting skin remaining in the keel.

TEST AND RESEARCH

On September 17, 1997, the floats were further examined by the Safety Board IIC and a Board Metallurgist, at Trident Engineering Associates, Inc., Annapolis, Maryland. Examination of the floats revealed that the outboard bottom skin of the right float had separated from the keel. There was no evidence of preexisting skin in the keel or riveting of the new skin to the keel.

In the NTSB Metallurgist's factual report, it stated the following:

"The right float right forward bottom skin panel was reportedly replaced about 50 hours before the accident. Examination of the keel extrusion slot for this panel (in the area where the panel had pulled out of the slot) revealed multiple nicks and gouges, consistent with insertion of some type of sharp tool along the bond line between the original panel and the slot. The repair manual for the floats allows the forward bottom skin panels to be replaced using three slightly different methods. One of the methods calls for complete removal of the original panel from the keel extrusion slot, installation of a replacement panel and shim into the slot, and riveting of the panel to the keel extrusion (through the slot) along the entire length of the panel. This type of riveting was not found on the keel extrusion. The other methods of replacing the panel call for cutting the panel near the keel extrusion leaving the bonded portion attached to the keel extrusion. No portion of the original panel was found in the keel extrusion slot."

WRECKAGE RELEASE

On September 9, 1997, the wreckage was released in part to a representative of the Avemco Insurance Company. The floats and engine were retained for further examination. On September 17, the floats were released to the insurance representative. On November 20, 1997, the engine was released to the insurance representative.

NTSB Probable Cause

The improper repair of a float by maintenance personnel which resulted in the total failure of the float.