HISTORY OF FLIGHT Use your browsers 'back' function to return to synopsisReturn to Query Page
On March 17, 1996, about 1237 eastern standard time, a float equipped Cessna U206G, N9983Z, registered to and operated by Key West Seaplane Service, Inc., crashed shortly after takeoff from the Gulf of Mexico Seaplane Base, Key West, Florida. Visual meteorological conditions prevailed at the time and a DVFR flight plan was filed and activated for the 14 CFR Part 135 on-demand, non-scheduled, domestic, passenger flight. The airplane was destroyed and the commercial-rated pilot and four passengers were fatally injured. One passenger was seriously injured. The flight originated from the Seaplane Base about 3 minutes earlier.
According to company personnel, the pilot fueled the airplane before departure and the passengers were questioned about their weights before boarding the airplane. The seating information is provided by the Director of Maintenance (D.O.M.) for the company who witnessed the loading of and departure of the airplane. The right front seat was occupied by a passenger who was later determined to be a single-engine land rated pilot. The pilot was observed to taxi about 1-2 miles away from the seaplane base and the takeoff was initiated from a point abeam the Garrison Bight Channel just north of the Fleming Key. The flight departed into the wind and was about 60 feet above the water about abeam the Salt Pond Keys. The D.O.M. stated that he then diverted his attention and he further stated that he watched the takeoff because the engine was near the time required to be overhauled.
According to radar data from the Key West Naval Air Station, four secondary radar returns each 5 seconds apart were recorded. The airplane was first observed at 1236.52 while at 100 feet, heading 174 degrees and on the 106 degree radial and 2.4 nautical miles from the Key West Very High Frequency OmniDirectional Range Station (VOR). The next secondary radar return indicated that the airplane was at 200 feet on a heading of 131 degrees. The third secondary radar return indicated that the airplane was at 100 feet on a heading of 177 degrees. The final radar return indicated that the airplane was at 100 feet; on the 111 degree radial and 2.6 nautical miles from the VOR; heading was not available. Calculations determined that the approximate indicated airspeed was 54 knots using the change in time and distance between the radar targets and 6 knots of wind from 120 degrees recorded at the Key West International Airport at 1240. The airplane Pilot's Operating Handbook indicates that the indicated stall speed at gross weight with the flaps retracted and at a 45-angle of bank is 54 knots.
Review of a transcript of communications with the Key West International Airport, Air Traffic Control Tower (ATCT), revealed that the pilot contacted the tower and stated that the flight was departing Key West "I'd like to go Cow Key Channel (which is over water) along the shoreline westbound to the Tortugas." The controller advised the pilot of the wind direction and velocity and also provided the altimeter setting, and advised the pilot to advise when the flight was airborne. The pilot acknowledged the clearance then advised that the flight was airborne. The controller acknowledged the pilot transmission and initially cleared the flight to proceed southbound then advised the pilot to make a right turn to pass behind another airplane that was preparing to depart from the airport.
Numerous witnesses near the accident site observed the airplane flying towards the island and all stated that when the flight was near beach, the airplane began banking to the right which increased to about a nearly wings vertical angle of bank. One witness observed the airplane begin a climbing right turn before the airplane pitched nose down. The airplane was then observed to pitch nose down and impact the water nose and right wing low. The airplane then rolled inverted and came to rest about 20 yards from a seawall in about 6 feet of water with the left float out of the water. Witnesses and CFR personnel near the crash site jumped into the water and helped to recover the occupants.
Information pertaining to the pilot-in-command is contained in the NTSB Factual Report-Aviation and Supplement U. Additionally, review of the company records revealed that he began his initial ground training with the company on November 15, 1995. His initial flight training with the company was completed 3 days later, and his airman competency/proficiency check flight was accomplished with an FAA operations inspector on November 21, 1995. Review of his pilot logbook revealed a break in flying from 1991 to 1993. The last entry in his logbook was the FAA checkride entry which occurred in November 1995. The two entries before that were dated in 1993 and were flown in the accident airplane. According to FAA personnel, the pilot did not have any previous accidents or incidents and he did not have any enforcement action pertaining to violation of any FAR'S. He was prescribed medication to control transient essential hypertension; however, he indicated on his last medical application dated November 15, 1995, that he was not taking any medication. On that application he indicated that he had flown 0 hours in the previous 6 months and his last medical was in 1993.
Information pertaining to the pilot-rated passenger in the right front seat is contained in Supplement E.
Information pertaining to the airplane is contained in the NTSB Factual Report-Aviation and Supplements A and B. Review of the engine maintenance records revealed that the engine had accumulated 1,612.1 hours since rebuild by the engine manufacturer. The right magneto had accumulated 1,503 hours since new at the time of the accident and had been reinstalled on the engine on December 12, 1995. The left magneto had accumulated 302 hours since new at the time of the accident and was installed on October 10, 1995. The engine manufacturer recommends overhaul every 1,700 hours and the magneto manufacturer recommends overhaul not to exceed the engine overhaul time. Additionally, review of the inspection checklist for the last 100-hour inspection which was completed 2 days before the accident revealed that the mechanic did not initial that he had checked the landing gear group which requires in part inspection of the bottom of the floats for leaking around rivets. Review of the discrepancy and repair list for that inspection revealed that he did document repair to a loose pump tube for one of the compartments for the right float and the water rudders were rigged.
Air traffic control communications with the Key West International Airport Control Tower were recorded and a transcript is an attachment to this report.
WRECKAGE AND IMPACT INFORMATION
Examination of the crash site revealed that the airplane was inverted in about 6 feet of water about 20 yards from a seawall. The right wing and engine assembly were separated from the airframe. During recovery of the airplane six inflatable type packaged life preservers were recovered. The airplane was taken to a nearby facility for further examination.
Examination of the elevator and rudder flight controls revealed control cable continuity from the cockpit to the control surfaces. The right wing aileron flight control cables were found to exhibit evidence of overload failure. Approximately 5 feet of the right float was observed to be displaced upward and the entire length of the float was observed to exhibit evidence of impact damage. Water was found in all float compartments. Examination of the aft bottom section of the right float revealed numerous damaged rivets which join the keel to the bottom skin panel near the aft bulkhead. The keel was pulled away from the skin panel and corrosion was observed between the keel and the skin. Examination of the left wing revealed chordwise crushing of the leading edge. The right horizontal stabilizer was displaced downward at about a 45-degree angle. The left float exhibited minor damage and about 4 gallons of water was pumped from the forward bay compartment using a hand pump found in the wreckage. The remainder of the float only contained residual water. The flaps were determined to be retracted and the fuel selector was positioned to the "right" tank position. The ignition key which was not damaged was in the ignition switch and was found in the "off" position. The pilot's control column was displaced to the right and both control horns were not broken. The co-pilot's control column was separated but both control horns were not broken. The mixture, propeller, and throttle controls were full forward in the cockpit. The auxiliary fuel pump switch was found in the "off" position. Cursory examination of the engine revealed fuel at the fuel lines at the inlet and the outlet of the engine driven fuel pump. The outer shielding of the No. 1 cylinder top ignition lead was observed to be cut to the center electrode at an area where a clamp was installed to secure the ignition lead to the cylinder.
The engine was recovered and installed on a test stand for an attempted run. Both magnetos were tight against the accessory case and were properly timed to the engine. The servo fuel injector was free of obstructions and an initial attempt to start the engine was unsuccessful. The magnetos were removed from the engine and examined which revealed that the right magneto point gap was greater than the manufacturers limits. Both magnetos were dried with air and placed on a test bench and the left was observed to operate throughout the entire rpm range. The right magneto was observed to fire intermittently throughout the rpm range with the point gap as found at .018 inch. The points were lightly dressed and the magneto again operated intermittently throughout the rpm range. The point gap was then adjusted to within the manufacturers limits and the magneto was returned to the test bench and operated normally throughout the entire rpm range. Replacement magnetos and an ignition harness were then installed on the engine and it was started and operated to near full rpm using the engine driven fuel pump. A differential compression test of cylinder Nos.1-6 was accomplished after the engine run which revealed 74/80, 73/80, 74/80, 77/80, 74/80, and 69/80 respectively. The engine and accident magnetos with ignition harness were then sent to the engine manufacturer's facility for another engine run but corrosion damage to the magnetos prevented operation of the engine. Replacement magnetos were then installed on the engine and the point gaps were set as found on the accident magnetos during the accident investigation. Also, the fuel manifold valve was replaced due to corrosion. The engine was then started and operated to 2,836 rpm though damage to the No. 1 cylinder top ignition lead was noted. Also, during the magneto check of the right magneto, the rpm decrease was 156 rpm greater that during the left magneto check. The ignition harness was replaced and the engine restarted and operated to 2,862 rpm. The accident magnetos were disassembled by a representative of the manufacturer in the presence of an NTSB investigator.
Disassembly of the left and right magnetos revealed no evidence of preimpact failure or malfunction. The contact end of each carbon brush in both magnetos were observed to be flush against the end of each distributor gear shaft. A circular gouge was noted on each magneto's secondary contact tab which were out of position by 3/16 inch. The gouge on the secondary tab mated with each magneto's distributor gear shaft. Arching and misfiring was not noted inside each magneto.
The propeller was removed from the engine was sent to the manufacturer's facility for examination under the control of an FAA inspector. Examination revealed that two of the three blades exhibited forward bending and also decrease pitch twisting. The propeller governor was bench tested at another facility and in the as found position of the control arm, found to operate at 1,700 rpm. The low pitch rpm stop was determined to set 50 rpm greater than specified. There were no further discrepancies to the governor.
MEDICAL AND PATHOLOGICAL
Post-mortem examinations of the pilot and passengers were performed by R.J. Nelms, Jr., M.D., Medical Examiner, and David B. Wheeler, D.O. Associate Medical Examiner, District Sixteen, Monroe County, Florida. The cause of death for the pilot was listed as drowning and multiple injuries due to blunt trauma. The cause of death for the right front seat passenger was listed as multiple injuries due to blunt trauma. The cause of death for the left middle seat occupant was listed as multiple injuries due to blunt trauma. The cause of death for the right middle seat occupant was listed as salt water drowning complicating cerebral contusion from aircraft crash. The cause of death for an occupant in one of the rear seats was listed as salt water drowning complicating aircraft crash. The sole surviving occupant seated in one of the rear seat positions was rescued from the airplane and first transported to a local hospital for treatment of his injuries, then transported to a hospital in Miami for continued treatment of head and leg injuries. According to the Key West Police Department personnel, the injured sustained a skull fracture and a bone in the lower left leg was broken.
Toxicological testing of specimens of the pilot was performed by the FAA Accident and Research Laboratory. The results were negative for cyanide, volatiles, and drugs. Verapamil, and Norverapamil were both detected in the blood and urine.
According to the Physician's Desk Reference, verapamil HCI is indicated for the management of essential hypertension.
According to FAA certification personnel with the Civil Aeromedical Institute (CAMI), a pilot may continue to fly while taking verapamil HCI.
Additional persons participating in the investigation from UNISON were Messrs. Brad Mottier and Harry Fenton.
Testing of a fuel sample taken from the operator's facility that fueled the airplane before takeoff revealed no evidence of contaminants.
Interviews with passengers who had flown in the accident airplane with the accident pilot the day before the accident revealed that the pilot banked the airplane without warning to a wings nearly vertical position while flying over the ocean to give the passengers a view of aquatic life. Further interviews with passengers who had flown with the accident pilot earlier in the month revealed that the pilot advised them on the takeoff run to lean forward in their seats. The pilot aborted the takeoff when the flight would not become airborne and taxied to the ramp where cargo was offloaded. The pilot again attempted to takeoff and the flight was again unable to become airborne. The pilot aborted the takeoff, returned to the ramp, and off loaded the passengers.
A witness who was on a boat about 22 nautical miles west of Key West near the Markasa Keys, on the day of the accident between 0900 and 1000 hours, stated he observed a green and tan colored airplane flying eastbound about 20-30 feet above the water. The airplane then changed direction completing a 360- degree turn with a bank angle of nearly 90 degrees.
The airplane was green with gold stripes and on the day of the accident, the accident pilot flew the airplane on two round trip flights to the Dry Tortugas, at 0800 and 1000 hours. According to the chief pilot for the company, each flight leg is approximately .6 hour.
Review of the load manifest for the flight which listed the passengers and pilots weights, the fuel load, and the baggage weight revealed discrepancies. Weight and balance calculations using verified weights of the pilot and passengers, the estimated fuel load, and the weight of the water pumped from the left forward float compartment revealed that the airplane was within weight and balance limits specified by the airplane manufacturer.
Review of the pilot's operating handbook (POH) revealed that the stall indicated airspeed at 3,500 pounds (gross weight floatplane) with the flaps retracted and 0-degree angle of bank is about 45 knots with a most rearward C.G. The stall indicated airspeed increases to about 48 knots with a 30-degree angle of bank and up to 54 knots at a 45-degree angle of bank. At 60-degrees angle of bank the stall indicated airspeed is about 64 knots with a most rearward C.G. and at 3,500 pounds. The aft C.G. limit for all weights is 47.4 inches and the C.G. at the time of the accident was calculated to be about 46.36 inches. The wreckage minus the retained engine and propeller was released to Mr. Charles Maynard, on March 27, 1996. The engine and propeller were released to Mr. Maynard on May 7, 1996.