Page 96 - Failure Analysis Case Studies II

81

Fig. 2. Failed plastic tank, showing the bund wall in section (the front part was removed by the Fire Brigade). The jet
of caustic soda flew from the tank wall near the right-hand edge in the figure, just above the top of the bund wall. It
rose upward from a small crack in the weld, and landed some 4-5 m away from the exit point.

well as the panel plastic. They included DSC and FTIR to check material constitution, and simple
tensile tests to measure directly the strength of the welds. The experimental part of the investigation
was carried out by P.R. Lewis, the translation of the key standard (DVS 2205) covering such tanks
being carried out by G.W. Weidmann.

2. The accident

Failure of the tank occurred at about 18:45 h, and was discovered by the Production Director
at the factory. He was in the upstairs office at the factory, and was about to leave the premises at
the end of his working day, when he heard a bang or ‘popping’ sound coming from the factory
floor. It was followed by the sound of ‘rain’, and looking down through an internal window, he
saw ajet spray of fluid issuing from the lower part of the recently filled black storage tank, situated
in a further corner of the factory (Fig. 2). He observed that the jet was issuing in the form of an
“. . .8 or 9 inch wide stream.. .” from the side of the tank facing a storage bay. The jet curved
upwards, so easily clearing the top of the adjacent bund wall, and was about 4-5 m long. The end
of the jet was hitting storage containers, and flowing down onto the factory floor.
The broken tank was simply a static storage container with a sight tube on one side, and pumping
facilities (Fig. 2). By 22:OO h that evening, the level of liquid in the tank had fallen substantially,
so lowering the hydrostatic pressure, decreasing the length of the jet stream, and causing it to fall
within the bund wall. The load of caustic soda which escaped freely into the factory represented