2011X06 Ruptured Tanks whilst Ballasting

 Ruptured Tanks whilst Ballasting
- UK.
- Official Report No. 7021offrep.

I had just finished writing the Official Report on the OSKO STAR for the November '94 edition when, by coincidence, the British MAIB Summary of Investigations No 2/94 appeared with a similar incident. This is reproduced below to emphasise the fact that this is not an uncommon occurrence.
MAIB reports can be obtained from:
5/7 Brunswick Place, Southampton, Hampshire SO15 2AN
Telephone +44 (0)703 395500

The vessel concerned was a 280,491 dwt VLCC and the accident occurred while the vessel was berthed at Immingham Terminal whilst ballasting 4 centre cargo tank. Although the vessel is installed with a segregated ballast system, segregated ballast alone was insufficient to prevent the maximum distance from the waterline to the cargo deck manifold, as required by the Terminal discharge facilities, being exceeded. In addition therefore it was necessary to load ballast into 4 centre cargo tank.

During ballasting of the cargo tank, this and two adjacent tanks ruptured. Fortunately there was no pollution, fire or injuries and no flammable or toxic gases detected on deck. However, damage to the vessel was significant and required a long stay in dry dock for extensive repairs. The rupture was caused by over-pressurisation of the tank. During ballasting the cargo tank vent isolating valve was shut leaving only the small orifice of the pressure vacuum valve (breather valve) for the tank to allow gases, displaced in the tank by incoming ballast, to freely discharge into the vent system.

Observations

Because of the circumstances of the closure of the cargo tank isolating valve, together with the non-realisation by crew members that over-pressurisation of the tank was taking place, rupture of the tank was inevitable. No reason has been given as to why the isolating valve was shut, but a number of operational factors indicate human error as a probable cause.

Comment

1. The vessel has 14 cargo tanks and 8 permanently segregated water ballast tanks. The engine room, accommodation and bridge are aft with the pump room immediately forward of the accommodation block. The inert gas, venting and cargo systems fully complied with SOLAS Convention requirements and were in a well maintained condition. The cargo/ballast and machinery control installations and the venting/inert gas systems are operated from the bridge console. However a number of system valves on the venting/inert gas arrangement, including the cargo tank vent isolating valves and their mandatory locking devices, are manually operated locally on deck. This is a typical arrangement where the cargo officer of the watch operates the remote control of the cargo and ballast operation from the bridge console, but it is dependant on the person on deck to open, close and lock the vent valves and make timely reports to the bridge of their current status.

Safe ballast and cargo movement is therefore dependent on a mixture of centralised control and local control of valve operation, a situation which compounds the risk of human error; incorrect operation of crucial valves in the system was a causative factor in this accident. If control and status indication of the cargo tank vent isolating valves was consistent with the remote control design of the rest of the system, the risk of inappropriate closure of these isolating valves would be reduced. There is clearly a need to review the utilisation of remote and local control of cargo and ballasting systems. Further, mandatory locking arrangements on the valves should be installed so that unauthorised operation is prevented by any means other than the correct key.

2. The investigation highlighted the need for consideration to be given to the design of tank venting systems, so that pressure relief devices fitted to prevent over-pressurisation of cargo and ballast tanks cannot be isolated from individual tanks. Closure of the vent isolating valve rendered safe pressure relief of 4 centre tank impossible. This situation was identical to that found on the MOBIL PETREL when this vessel suffered a similar incident in November 1989. The need to install a pressure relief device direct to the tank without an intermediate isolating valve is, in engineering terms, analogous to that of the boiler. A cargo and ballast tank, like a boiler, is a pressure vessel. A boiler however has its safety valve attached direct to the shell. The fitting of an isolating valve between the safety valve and the shell is unacceptable.

Pressure relief devices in vent outlets for loading, discharging and ballasting are designed on the basis of the maximum designed loading rate multiplied by the factor of 1.25. However during cargo and ballasting operations, cargo pumping rates may erroneously be increased and the maximum loading rate exceeded, therefore despite installation of pressure relief devices, over-pressurisation of the cargo tank is still a possibility. For this reason it is mandatory that the Master is provided with information regarding the maximum permissible loading rate for each cargo tank.

3. The findings in the investigation of this accident highlighted the need for continuous monitoring of the cargo tank pressure. This need was recognised as a result of the MOBIL PETREL investigation, when recommendations were made which included a proposal to fit mandatory pressure sensor measuring alarms to inform the cargo control station that the cargo and ballast pressures are approaching safe design limits. In both cases, early warning of impending over-pressurisation of the cargo tank would have prevented the accident. The proposal has been submitted for consideration to the Maritime Safety Committee of IMO. Use of pressure sensors and the requirement of not being able to isolate the pressure relief arrangement should be compulsory and not deemed an either/or option.

4. Because of operational requirements the vessel needed to ballast cargo tanks. MARPOL 73/78, Consolidated Edition 1991, Regulation 4, requires that ballast water should only be carried in cargo tanks in exceptional cases. These exceptional cases refer to tankers required to pass under a low bridge, or when local regulations require a specific draught for safe navigation . These categories do not appear to match the reason for ballasting 4 centre cargo tank of this vessel. This operation was required to ensure that the cargo could be safely discharged ashore by achieving the correct air draught. Ballasting of the tank was therefore done for operational reasons and not for safe navigation or safe passage under bridges. The regulation appears to be ambiguous and in need of clarification.

5. The vessel over-pressurised and ruptured cargo tanks during a ballasting operation. The published MAIB report on the MOBIL PETREL accident indicated the over-pressurisation of cargo tanks in tankers occurs frequently. Over-pressurisation of a cargo tank was a possible cause of the explosion and fire on the Cypriot registered tanker HAVEN in April 1992, resulting in the loss of 5 lives, sinking of the vessel and serious pollution. It was fortunate that the MOBIL PETREL and the accident at the Immingham Oil Terminal did not suffer a similar fate with the added consequential catastrophic impact on the respective terminals.