When a mate was loading vacuum gas oil on a cold winter evening an alarm was heard, indicating high pressure in one of the two tanks which were being filled. After the alarm had been noted, the watch seaman was sent out to check whether the pressure relief valve was working. Since he saw gas streaming out of it, the seaman assumed that the valve was working, which he told to the mate.
At the same time the mate changed over to filling some of the other tanks and, when these were full, switched back to filling the tanks in which the pressure alarm had sounded. He asked the seaman to close the manual valve on the tanks when loading was completed. Another alarm for high pressure sounded. The mate started to open the tanks which he had recently finished loading, but quickly realised that the manual valve had been closed, and so he closed the tanks. Once again he sent out the seaman to check the pressure relief valve. Again, the seaman reported that he saw hot vapour coming out.

The mate assumed that he pressure sensor was not operating correctly, which he had experienced once before. Soon afterwards there was an explosion and violent shaking. As a result of hot oil rushing into the cold air it vaporised and the fire alarm was activated. The mate contacted the loading terminal ashore requesting that they stop loading and at the same time he closed all valves and manifolds. The starboard tank had exploded as a result of excess pressure. The bulkhead by the starboard tank buckled and ruptured, as did the tank top to the pipe tunnel and the tank top to the double bottom. When the explosion took place both tanks were 66% full.
One major course of this accident was ice formation in the air vent pipe. Since the ice was not completely sealed some of the gas could pass, which the seaman saw coming out of the pressure relief valve. The vent pipes should drain automatically, but did not. There were no injuries and no environmental damage, but eight tons of steel needed to be replaced.
During the winter at least one other tank exploded in a Swedish ship also due to ice formation in an S tube for ventilation. The Swedish Transport Agency recommends that all pipes are checked for sufficient draining.
SFu dnr 060502 TSS 2010-536