The Marine Casualty Investigation Board (MCIB), the Irish government agency for investigating maritime accidents and incidents, published its report on the fire that broke out onboard the Ferry “Frazer Tintern” midway between Ballyhack, Co. Wexford and Passage East, Co. Waterford.
The incident
On the morning of 5 August 2021 at approximately 07.45 hrs a deck crewmember of the ferry “Frazer Tintern” reported to the Master of the vessel that he smelled diesel fumes coming from the No.1/No.4 engine compartment. Upon further visual investigation, a diesel fuel leak was discovered where a fuel return line had become disconnected on the No.1 engine.
The engine was shut down and a phone call made to the company’s marine engineer to report the situation. As the engineer was unavailable at the time to attend the problem, the decision was taken to call out a local marine mechanic who had previously carried out repairs on the vessel.
The leak was repaired by the mechanic by refitting the return line and securing it with a hose clip and verified by the Master who ran up the engine to check the repair. The Master had worked on the vessel since its arrival in Ireland and was familiar with operating the machinery. The area around the No.1 engine compartment was cleaned by the crew with detergent and deemed free of diesel residue, as was the deck plating adjacent to the engine compartment. The vessel resumed normal service at approximately 08.20 hrs. The engine compartment was monitored hourly by the crew and there were no further reportable events for the remainder of that morning shift. There is no record of this incident being mentioned to the crew taking over for the evening shift at 13.00 hrs.
On the evening of 5 August at approximately 18.05 hrs while sailing from Ballyhack to Passage East, a strong smell of diesel fumes was noted by the Master and simultaneously by one of the deck crewmembers who called this over the radio to the wheelhouse. The same crewmember informed the Master that he was going to investigate the source of the diesel fumes by approaching the starboard machinery space access.
When the crewmember arrived at the mesh gate on the starboard side leading to the compartment of No.1/No.4 engines, he discovered smoke and flames pouring from the compartment and immediately informed the Master of the source and location of the fire. The Master immediately shut down No.1 engine and switched off the engine room fans. The fire flaps and fan shutdown for the starboard machinery space were inaccessible due to the location of the fire.
The two deck crewmembers accessed portable fire extinguishers and discharged them at the fire. The fire was knocked back and fire hoses were run out to provide boundary cooling, while the Master continued to navigate the vessel towards Passage East slipway. The machinery space fire suppression system was not operated.
The passengers were summoned to the muster station and instructed to don lifejackets that were handed out by crewmembers. The vessel docked at Passage East slipway where all passengers and vehicles were safely disembarked.
The vessel was moored up and the remaining engines were shut down. The three crewmembers then carried out a visual inspection of the engine compartment after the remaining smoke had dispersed and confirmed that the fire was fully extinguished.
The Master contacted the company marine engineer and informed him of the event and status of the vessel. The crew remained onboard the vessel until the arrival of the company marine engineer. The Port of Waterford authorities were also informed of the event. At no time during or after the event was a distress/Pan-Pan call made on Very High Frequency (VHF) radio.
The incident was reported to the MSO and the vessel was inspected prior to and on completion of repairs at New Ross, Co. Wexford Boat Yard.
Analysis
The weather on the day of the incident was warm with a maximum air temperature of around 20 Celsius (see Appendix 7.6 Met Éireann Report). The starboard side of the vessel being predominantly south facing is also subject to solar heating effect through the steel superstructure. The area around No.1 engine is the hottest area in the machinery space due to being the exit point for airflow through the area. The exhaust from No.2 generator which was also running at the time of the incident is routed past No.1 main engine. This would have further increased the temperature in the area. It is likely that gas oil spraying in this environment could be at, or close to, its flashpoint where it will ignite easily. The temperature in this area would have had a contributary influence on starting the fire.
The nature of the airflow through the machinery space could have led to a high degree of hot swirling air flow in the vicinity of No.1 main engine. This would have assisted atomisation of any diesel fuel leaking in the area contributing to starting the fire.
The ignition source of the fire may have been arcing, causing sparks in the engine mounted alternator, but more likely to have been the hot exhaust manifold or turbocharger casing of No.1 engine that was not lagged or enclosed. The exposed hot surfaces of the manifold and turbocharger would have provided an ignition source for the fuel.
The return fuel lines on the main engines on the vessel were seen to be fixed using three distinct methods. The original fixing by the manufacturer was by use of spring clips, but other pipes were attached using cable ties or hose (jubilee) clips. The repair to the fuel line on the morning of the incident was made by fitting a hose clip. This may have contributed to return fuel line failure.
The extensive damage caused by the fire in a very short period indicated that a considerable amount of fuel was being released to support the fire while it was active. The fact that the fire died off quickly after No. 1 engine was shut down indicates that the fuel to the fire was being fed by mechanical means from the engine. Shutting off the fuel removed one of the sides of the fire triangle, the other two sides being the heat and air necessary for a fire to continue. Shutting the fire flaps would have restricted air flow and helped to contain the fire. Operating the quick closing fuel valves would have also stopped the fuel but would have also shut down the other main engines. The considerable amount of fuel being released was likely to have been from the fuel lift pump. Should the fuel return line be blocked off in any way the pressure regulating valve will become inoperative resulting in a significant increase in fuel pressure on the return side of the system, consequently leading to a pipe failure.
The fire suppression system for the machinery space was not operated as the Master did not want to lose all propulsion at the time and deemed it safer to complete the short run and land the passengers at Passage East, Co. Waterford.
Conclusions
The fire was most likely caused by a return line fuel leak on No.1 main engine providing fuel to the area. The volume and pressure of the fuel was greatly increased by the fuel return line being blocked or shut off. The ambient high temperature and swirling air flow in the vicinity assisted in the atomisation of the fuel.
The fuel may have been ignited by arcing of the No.1 main engine alternator. It was more likely to have been from fuel spraying onto hot surfaces such as the engine exhaust manifold or turbocharger casing.
It is hard to rule out the No. 1 engine fuel leak on the morning of the event having an association with the fire. The repair was carried out by using a hose clip to re-attach the return fuel line rather than a spring clip as used by the manufacturer. If the hose clip was over tightened it may have damaged the hose resulting in failure later in the day. Return fuel lines on this engine would have been new with the engine in 2016.
Due to the airflow from No. 1 engine access door, the fire spread onto the car deck in an area open to the public. This also prevented access to the port side fire flaps and fan stops. Although the door has been modified to close on activation of the fire alarm, in this instance the fire would have spread to the public space prior to the door closing.
The operators have stated that prior to the incident they carried out fire drills. There are no records of these being carried out or what was involved in the drills. Post incident, drills are carried out on a weekly basis with at least one of each drill (Man Overboard/Fire in Accommodation, Car Deck or Engine Room/Abandon Ship) carried out per month.
The operators have also put in place a Domestic SMS to ensure daily checks are carried out and a smooth handover between shifts has been introduced.
The Master of the vessel should have notified the Coast Guard of the fire by a VHF Pan-Pan call as soon as it was discovered.
EXPLORE MORE at mcib’s accident report
