Serious engine room fire whilst divers in saturation

A member has reported a very serious incident in which there was a fire in the engine room of an offshore vessel. The incident occurred whilst the vessel was alongside, but there were divers in saturation at the time. The incident happened when a leak of diesel fuel, under pressure, ignited at the secondary duplex fuel filter on an engine. Only one engine -main engine 1 on the port side was running at the time, supplying all vessel systems.

During ‘routine’ maintenance work, the engineers became aware of flames near the exhaust or waste gate and at secondary fuel filter of the main engine that was running. An engineer attempted to get to an emergency stop location, but that moment the fire expanded explosively and within seconds engulfed the forward section of the engine in a ball of fire, and he was driven back, but managed to escape. Emergency procedures were followed including closing of valves, stopping of fans and engines and shutting of all watertight doors. The Hi-Fog Mist system was activated, and the fire was extinguished.

When the general fire alarm sounded, the Diving Superintendent was informed and all diving personnel were mustered. All the divers in saturation were transferred to the Hyperbaric Rescue Craft (HRC) and the transfer trunking was brought to surface. The seal on the HRC was confirmed. The dive system was switched to emergency power supply and the HRC made ready to launch. The crane was made ready to lift the HRC from the vessel if required. Once the situation was contained, the divers were able to return to the saturation complex to continue decompression. All divers completed their decompression without further issue.

Everyone escaped safely; no-one was injured. The fire had the potential to have caused multiple fatalities. There was serious fire damage to the port engine room and mezzanine deck level equipment and service wiring.

Our member wished to record that:

  • The conduct and actions of the Chief Engineer, 1st Engineer and both 2nd Engineers were carried out quickly, professionally and as safely as possible, their actions and the speed of their response greatly reduced the significant potential of this very serious incident;
  • The emergency response and swift actions of the Diving Superintendent, Diving Supervisors, Life Support Technicians and Divers demonstrated a professional and very competent response;
  • The Vessel Master, Chief Mate, Offshore Manager and Client Site Representative all contributed professionally and very competently during this incident.
Figure 1: Typical fuel filter of the sort that failed
Figure 1: Typical fuel filter of the sort that failed
Figure 2: Failed fuel filter unit
Figure 2: Failed fuel filter unit
Figure 3: Top of filter showing bolt hole and extruded O ring
Figure 3: Top of filter showing bolt hole and extruded O ring
Figure 4: Sheared bolt
Figure 4: Sheared bolt
Figure 5: Failed stud
Figure 5: Failed stud
Figure 6: Failed stud
Figure 6: Failed stud

Our member’s investigation noted the following:

  • With regard to the vessel and its crew:
    • The engine room personnel were very experienced, qualified and competent in the engineering processes and systems on-board the vessel
    • The vessel preventative maintenance system was well applied, actioned and documented and in many cases over and above scheduled requirements
    • The starboard engine room (undamaged) was found to be clean and very well maintained which provides an indication of the professionalism of the vessel engineers
    • The vessel had been in operation for four years. Fuel filter maintenance has been carried out in excess of stated requirements in order to cater for extended dynamic positioning (DP) operations.
  • With regard to the failed equipment that was found to be the cause of the fire:
    • The Bollfilter Duplex Filter units had been in service since the vessel was commissioned in 2010
    • Detailed analysis of the failed stud showed that: -One of the three studs broke by high cycle low stress fatigue -One stud contained a short fatigue crack -The primary cause of cracking was insufficient preload tension in the studs due to insufficient tightening of the nuts
    • The stud was manufactured from good quality creep resistant steel;
  • With regard to the fire itself:
    • Activation of the fuel quick closing valve(s) was undertaken but main engine 4 (which had been started as the burning engine was shut down) continued running;
  • The immediate cause of the incident was found to be:
    • One of the three filter housing cover stud bolts failed by high cycle, low stress fatigue as a result of insufficient pre-load tension in the studs due to insufficient tightening. This led to a loss of pressure integrity of the filter housing and the ‘O’ ring seal was extruded through the housing/lid interface. This allowed fuel at pressure to spray onto a turbo charger exhaust manifold, causing a fire;
  • The underlying causes of the incident were found to be:
    • Filter unit in line and in close proximity to the engine turbo charger exhaust (750mm)
    • Failed stud in direct line of engine turbo charger exhaust
    • No physical barrier between the fuel filter and the potential ignition source of the engine turbo charger exhaust. The cladding around the exhaust does not provide 100% cover
    • No local fuel shut off valve in close proximity to the forward engine space and areas where personnel may be present or at access/exit location
    • No local Hi-Fog Water Mist activation facility in the forward engine bay in the vicinity of the access way to the engine room
    • Fuel shut off valve was a considerable distance from the engine bay and the vicinity of the fuel supply;
  • The root causes of the incident were found to be:
    • Lack of stud bolt torque information available from either the vessel planned maintenance system or the filter manufacturers documentation
    • Vessel design: Fuel system location in close proximity to systems at auto-ignition temperatures
    • Vessel design: limited mitigation and recovery systems in the event of a fuel related event with personnel working in the engine room.

Our member took the following actions:

  • Immediate full visual inspection of all stud bolts in all other fuel filter units;
  • Replaced all studs in filter assemblies and tightened sufficiently to attain pre-load tension. Supplied information back to the fuel filter manufacturer such that torque settings were provided and included in the manual;
  • Fire retardant blankets were placed over remaining in-service filter housing in an attempt to prevent a similar event from occurring.

Our member took the following recommendations:

  • Fuel supply system:
    • Review the design of the secondary duplex fuel filter locations to assess if the filter housings could be re-located away from the vicinity of the turbo changer exhaust manifolds for each engine. If the filters were able to be re-located in compliance with technical considerations, relocate to a position away from the engine exhaust manifolds and other extreme heat sources
    • There were no physical barriers between the secondary duplex fuel filter housing and the engine turbo charger exhaust manifold. Provide physical barrier between the filter housing and all surfaces with a potential temperature at or above auto-ignition temperature to prevent fuel being sprayed towards the auto-ignition heat source;
  • Response to an engine room fire;
  • Provide for local fuel shut-off facility at the forward end of each engine if technically feasible. Position should be at, or close to the exit of the engine room:
    • Provide for Hi-Fog Water Mist activation in each access way of the port and starboard engine room exit area if technically feasible. Position should be at, or close to the exit of the engine room
    • Fuel Emergency Shut-off systems -Investigate further the action of the Quick Closing fuel shut-off valves, as the activation of the quick closing shut-off valves did not starve all engines of fuel. When activated, main engine 4 continued to run.