In its 2017 Safety Digest report, the UK MAIB published its findings into the sinking of an unmanned survey vessel due to a not fully closed, defective butterfly valve that was not fully closed, providing a description of what happened and the lessons learnt in an attempt to prevent similar accidents from occurring.
A fisheries survey vessel sank while alongside a marina berth. The vessel, a 17m long aluminium catamaran, had two independent engine rooms, each located in the port and starboard side hulls. The vessel was unmanned at the time of the accident.
When the vessel was salvaged it was noted that there was water leaking into its port engine room, through a ship side valve connected to the port main engine exhaust system. The vessel had undergone maintenance, carried out by contractors in the week leading up to the foundering. The crew had reported water ingress and exhaust gas leaks in the port side engine room. Contractors had attended and had identified a defective section of corrugated exhaust pipe connected to the port main engine turbo charger. The decision was taken to remove the defective section of pipe and to fabricate a new one at the contractor’s workshop.
The vessel had a single butterfly ship side valve that was connected to the exhaust piping by a short section of rubber hose. The butterfly valve was closed by the contractors and the defective section of corrugated exhaust pipe removed. The remaining exhaust pipe, still connected to the butterfly valve by the rubber hose, was secured by a rope that kept the open end of the pipe above the waterline. There was no water ingress at this time.
Crew from the vessel attended on the day prior to the foundering and nothing amiss was noted.
Following the vessel’s recovery, the affected ship side valve was checked in situ and it was found that it could not be fully closed. The valve was removed from the vessel; on inspection it was noted that the valve body was damaged. As found, it would not have been possible to fully close the valve; this was not apparent to the contractors or the vessel’s crew prior to the accident. The water ingress and subsequent flooding occurred as either the rope, holding the remaining section of exhaust pipe above the waterline, failed, or the short section of rubber hose connected to the valve lost its seal as it was lifted by the rope.
In order for this work to be successful the ship side valve needed to be fully closed. It was assumed that the valve was fully closed and no additional checks were made. Best practice, when working on any ship side valve and associated pipework is to make sure the valve is fitted with a suitable blank prior to any pipework being removed.
A single ship side valve is a single point of failure. Any associated works on a system connected to such a single point of failure must be fully risk assessed and appropriate control measures put in place prior to maintenance being undertaken to vessels afloat.
The vessel was unmanned, except for brief periods, between the pipework being removed and the vessel foundering. There was no alarm or monitoring system put in place to alert the crew to the water ingress and the problem was noticed only when the vessel began to sink. It would have been prudent to maintain some form of watch, or at least to have attended the vessel at frequent intervals during the period that the pipework was removed.