Page 139 - Failure Analysis Case Studies II
P. 139

124
                                 RF*IN  --









































                         Fig. 9.  Port bilge keel showing bilge bracket: also showing longitudinal


          evidence is consistent with  the  fracture features,  and  that a  crack  of  -6ft  (-2m)  opened up,
          propagating downwards from the sheer strakeweb detail, and arrested  during the first incident.
          Subsequent propagation  upwards  to the gunwale probably occurred during the second incident,
          and the final separation of the two sections of the ship involved the joining of the arrested 6 ft crack
          and the crack propagated from the bilge keel. This process was accompanied by the deformation
          shown in Fig. 13.
            After the first reported  noises, two cracks 6 ft (- 2 m) long on the port side, and 6 in. (1 50 mm)
          on the starboard  side, had been visible - 12 ft (-4 m) below the deck in No. 3 tank. The bottom
          ends of these cracks could apparently be seen at times during the swell. At the time, the existence
          of a crack along the bottom shell plate had not been contemplated by the ship’s crew. However, the
          magnitude of the loss of oil indicated from the ullages taken suggested that the centre tank was also
          leaking. Together with the fact that water was reported to have been pumped from the No. 3 tanks
          before the occurrence of the second reported incident, this would point to the likelihood that the
          bottom plate fracture had occurred at, before, or just after, the reported cracks in the side plates. It
          is interesting to note that the No. 3 centre tank appeared much cleaner (of oil) than the wing tanks
          (Fig. 5). The loss of oil from the bottom shell fracture may well have been missed due to the presence
          of loose ice driven around the ship by the swell and the prevailing wind.
            It is difficult to determine the exact extent of the initial bottom shell fracture with regard to the
          longitudinals and centre girder, these will be described in detail later. As well as propagating up the
          port side the crack would appear to have run completely along the bottom plate and through the
          starboard bilge keel (parent plate material) as shown in Fig. 14, without stopping. Figure 14 also
          shows a region of crystalline appearance of the fracture on a small section where the residual oil
   134   135   136   137   138   139   140   141   142   143   144