On November 17, 1996 at 1700 central standard time (cst), a Beech 1900D, N841SK, being operated as Skyway Airlines flight 1267 executed an emergency landing at Fort Campbell, Kentucky. The flight crew reported smoke in the cockpit and declared an emergency during the flight. Both flight crewmembers donned their oxygen masks during the flight. After landing, the crew and passengers executed an emergency evacuation of the airplane. The two flight crewmembers and 14 passengers were uninjured in the incident. The 14 CFR Part 121 flight originated in Milwaukee, Wisconsin, with an intended destination of Indianapolis, Indiana. The flight was operating on an IFR flight plan. Use your browsers 'back' function to return to synopsisReturn to Query Page
The captain of the flight stated in a written report that he had just selected the windshield heat on. After the heat was turned on the copilot was the first to notice a strange smell. Both crewmembers reported that it smelled like tuna fish, and smelled the strongest closest to the instrument panel. The crew then reported that white smoke began developing from where the windshield meets the dash on the copilot's side. The captain realized that he had just turned the windshield heat on, so he then turned the windshield heat off. After the windshield heat was turned off the crew reported that the smoke stopped developing.
Post flight examination of the airplane revealed that electrical arcing was occurring on the copilots windshield at the low power heating element's terminal block. It was determined that the failure was a discontinuity at the low power braided wire. The braided wire is installed under a terminal block during manufacture. It was also determined that arcing was occurring between this discontinuity in the braid.
When the terminal block was removed the bonding agent near the center of the block was found to be gummy, however the bond between the block and the windshield was consider to be within the specifications required by the windshield manufacture. It was unclear if the gummy bonding agent was due to the electrical arcing which had occurred during and after the flight.
Following this incident two additional incidents of a windshields causing smoke in the cockpit were reported. The first incident occurred on a Beech 1900, serial number UE-37. The second occurred on a King Air 300, serial number FA-32. The failure of both windshields was very similar to the incident aircraft's windshield failure in this report. The total time on the UE-37 aircraft's windshield at failure was reported as 248 hours, the total time on the FA-32 aircraft's windshield was approximately 400 hours.
The airplane and windshield manufacturer began testing on numerous coupon samples following the UE-37 incident. Testing of the samples included testing the braided wire, at the low power terminal block. This testing revealed that as the samples were cycled, the braided wire began to deteriorate. As the braided wire deteriorated the resistance of the braid was found to fluctuate. One test sample began to smoke following approximately 420 hours of usage.
The manufacturer of the airplane reported that the chemicals used in bonding the terminal block had been changed in July 1995, due to Environmental Protection Agency (EPA) rules regarding the previous chemicals used. These new chemicals had incompatibilities with the braided wire and would cause the braid to deteriorate during usage.
The copilot's windshield on flight 1267 had been replaced due to Airworthiness Directive Number 96-15-01. The windshield had a total time in service of 493.8 hours and had been installed on September 7, 1996.