Load sharing imbalance caused loss of position

A DP2 pipelay vessel was set up on auto DP engaged in pipelay operations, inside the 500m zone of a fixed platform. The DP design concept of the vessel was such that there were two redundant groups that could only operate as a common system with bus tiebreakers closed.

The vessel was set up with power generation and thruster supplies equally distributed across each redundant group, as shown in the below example:

The bus ties between the redundant groups were closed creating a common power and propulsion system and DG1, 2 & 3 were online, DG4 on standby. The weather was typical for the area: wind speed 4kts, current 3.0kts and wave height 1.5m.

The engineer on watch noticed a developing load imbalance across the online engines. DG1 and 3 dropped load quite rapidly, leading to them both being tripped on reverse power protection, leaving only DG2 online. The PMS automatically instigated a standby start of DG4. However, before DG4 was up to speed and synchronised, DG2 tripped on overload, leading to a blackout. DG4 connected to the switchboard as part of the blackout recovery process and the engineers worked to bring DG1 and 3 back online.

The DPO regained control of the vessel 35m away from the original position. The pipe being laid remained connected to the vessel but was buckled. The decision was made to abandon the pipe and move the vessel to a safer location for detailed investigation.

An investigation found that the DG2 governor had experienced a mechanical failure, leading to excessive fuel supplied to the engine. The engine therefore ‘grabbed’ load from the otherwise healthy online machines forcing them to trip on reverse power. The generator protection and power management systems had no way of identifying the faulty machine although the PMS did generate load imbalance alarms prior to blackout.

The standard operating procedure for the DP system was dictated by the vessel’s DP operations manual and the vessel was set up for DP operations accordingly. The vessel had two redundant groups and the redundancy concept was such that they could only be operated as a common system (bus ties closed).

Review of the DP system failure modes and effects analysis (FMEA) showed that it failed to consider the specific failure modes associated with closed bus tie operations despite the analysis stating that it was ‘industry compliant’.

This case study demonstrates the importance of the FMEA. The FMEA is a key component to the input of DP operation, activity specific procedures, annual DP trials, and exercises and drills.

Therefore, it is vital that this analysis is kept up-to-date and current, considering all possible DP operating modes of the vessel. Although stating compliance with a variety of IMCA Guidance, this well-known failure mode had not been captured.