Primary finding
Probable cause
The failure of the water heater relay and the overpressure relief valve to operate properly. In addition, the design of the relief valve was inadequate in that it allowed the o-ring to migrate from its normal position preventing the valve from opening.
Investigator assessment
Analysis narrative
The water heater in the forward galley exploded. Examination revealed that the contacts of the water heater relay, which, when energized, provides electrical power to the 3 heater elements, were rough, ragged, discolored, and showed excessive erosion. According to the manufacturer, when a relay exceeds its lifetime, it is possible for erosion and wear on the surface to cause the contacts to stick in the closed position. In addition, the pressure relief valve, which is designed to crack at 65 psi, was tested to 300 psi without cracking. Upon disassembly, the o-ring was found to have migrated from its normal location into the lower valve body where it lodged between the inside diameter and the head of the poppet. In this condition, the o-ring presented a completed flow blockage and prevented the pressure relief valve from opening.
Source record
Factual narrative
HISTORY OF FLIGHT On October 3, 1993, about 1555 eastern daylight time, 9Y-THR, a McDonnell Douglas MD-83, experienced an explosion in the forward galley during climb to cruise altitude. The airplane was registered to and operating as British West Indies Airlines [BWIA] flight No. 431, scheduled, international, passenger service from Miami, Florida, to Barbados. The airplane was substantially damaged, 1 flight attendant received serious injuries, 1 flight attendant received minor injuries, 82 persons were not injured. Visual meteorological conditions prevailed at the time and an IFR flight plan was filed. The flight had originated from Miami, Florida, about 1525 and landed back at Miami, Florida, around 1625 the same day. Examination of the airplane revealed that a water heater, Nordskog model, had over pressurized and exploded. The entire forward galley and the cabin wall to the cockpit was destroyed. The water heater was removed and examined at the NTSB laboratory. All failures noted were consistent with overload. The over pressure valve was examined and the "O" ring had migrated blocking the valve "Pop Off" function. The electrical contacts to the heater element were found failed in a way that allowed constant power to the element. Details of these examinations are included in this report as attachments. -