Improperly Installed Fuel Line Caused Engine Room Fire on APL Boxship

This article is from back in May of this year, didn’t see it until now.

As a first attempt, the crew sent in an attack team with a hose to try and quell the fire, but they quickly determined that it was “too much” to fight and retreated. The captain ordered all ventilation hatches to the space closed and the fixed firefighting CO2 system activated.

Crew deserves a lot of credit here. Bad fire, could have been much worse. Seems like often times situations likes this end with the crew abandoning the vessel.

Full report here:

Forum thread at the time here.

This is from the NTSB report:

Following the fire, the team leader for the salvor’s marine firefighting team, who had 48 years of experience in the field, stated that he thought that the vessel’s crewmembers were “very well trained,” and that that their decision, preparation, and activation of the ship’s fixed CO2 system to extinguish the fire “was just textbook perfect.”

1 Like

I’ve heard rumor that the operating company didn’t feel the same as Flag and NTSB and there may have been some personnel changes at the top after this fire. Anyone know specifics?

A few comments. And I’m not nitpicking the crew of this vessel. They put the fire out without casualties and that’s all that counts. But I have studied the history of shipboard fires, and I teach marine firefighting and run a large simulator, so I have an interest in the subject. There are common threads that run through case histories. A few come up again in this NTSB report:

The first and second engineers headed to their emergency squad locker at safety
storeroom B. Once there, the second engineer opened the closest engine room door
which was a few feet away…and saw flames… Realizing the severity of the fire, he and
the first engineer donned firefighter outfits and SCBAs, then attempted to use ABC
portable fire extinguishers to quell the fire through the engine room door.

Opening a door to a compartment involved in fire, before suiting-up, is Russian roulette. You might win, but when you lose, you lose big. Example: the F/V Galaxy fire, where opening a door by unprotected personnel led to casualties.

To escape the smoke that was filling the bridge and the alarms whose noise
prevented clear communication, the captain transferred navigational control of the
vessel to the starboard bridge wing station, where he was able to control the vessel and
communicate, primarily using UHF radio, with the chief mate and emergency squad

Classic marine firefighting: smoke clears out all the interior spaces lickety-split, driving captain and crew out on deck about as fast as they can put their pants on. This should be treated as the rule and not the exception, and drills and training should reflect this.

…the second engineer and the two ABs from emergency squad 1, who all had donned firefighter outfits and SCBAs, made two separate entries… The attack team advanced into the space about 6 feet, but they realized the fire was “too much."…decided to prepare for a CO2 release into the space…

Mounting a direct attack before using the fixed system has led to casualties. But I can’t recall an example where using the fixed system before a direct attack led to anyone getting hurt. So the question is, Why aren’t marine firefighters taught to set off the fixed system first, and then afterwards consider direct attack? On vessels with emergency fire pumps setting off the fixed system is usually the safest, quickest, and most efficient route.

There are exceptions of course. What may be prudent at sea may not be prudent in a narrow strait. But the fixed system will accomplish a fraction of the time, without water damage, what a direct attack can take hours to accomplish.

Again, I’m not nitpicking this crew. They got the fire out without casualties, so kudos to them.


Great comments and reflection. We can all be “Armchair” Fire Fighters or “Monday Morning Quarterbacks”. Actually, many of us in the marine industry often “hotwash” and critique past events; hoping to learn something from these unfortunate events.

Anyone who has participated in the operational side or command side of actual fires and emergencies would remind you that rule number one is the preservation of human life then extinguishment of the fire while maintaining control of the vessel and minimizing damage.

While I agree with you that more often than not- that the safest response is to trip the fuel supply, close the fuel valves, secure ventilation and close fire dampers all while getting a Muster- then dump a flooding type fixed system- Remember that this takes time. More often than not it also reduces control of the vessel, produces a momentary blackout condition (that’s provided the EDG comes on the line). Unless it presents a risk to the crew, I would most likely try to get the fire out in the incipient stage… And also do remember that with many of the ship fixed systems- flooding the engine spaces is a one shot deal.

From the photo- it looks like a B&W MC or MC-C Engine. (just read the report- I am correct). My last liner service job had the identical /return leak off tubing coming from each of the cylinder fuel pumps- they were a frequent nuisance. Let’s see- The lines and fittings- while in the OEM diagrams- were shipyard furnished items at new build and were custom fabricated and fitted on a cylinder by cylinder basis. They were so much of a nuisance- that we had six straight sections of three varying lengths- with the fittings attached at both ends- we would bend the pipe in the shop and fit it as needed.

But there’s a missing piece here- all of this series of engine that I have sailed on- have drain “gutters” and catchment leads which are all piped centrally to the aft end of the engine to a leak detection system which alarms once a steady flow of oil enters it- that should have alarmed first. Granted- an open fuel tubing line (the size I remember is 19 mm) spits out a lot of diesel quickly, and we have all been remiss at least once in our career in getting insulation blankets back on that pain in the butt exhaust lead from the exhaust valve to the exhaust receiver (ours was metal sheathing- clamped and closed and was ALWAYS put back on after repairs)…

Again, kudos to these brave crewmembers- they did a great job…

1 Like

That’s a good question.

As near as I can tell, a lot of the curriculum for dealing with shipboard emergencies comes from the Navy, specifically WW2 Navy.

This leads to poor practices on modern vessels. Another point is securing ventilation and all airflow and vents. Many a fire has been allowed to burn and build enough heat to allow a reflash post fixed system release because people were fiddling with vents. The attempt should be made, but delaying the decision to release the fixed system due to a vent closure failure is foolhardy. If you wait until there’s enough thermal mass in the engine room to cause a reflash retaining fixed agent will serve no purpose. On fire engine rooms will not remain gas tight, especially following a fixed release.

The school that the fixed system is a last resort option should be replaced by the visibility test.

1 Like

I agree. I think some mariners miss the point about how CO2 works. The mechanism is to flood the space with CO2 and drive out the air. No different than if you were flooding a tank with water. The purpose off flooding with CO2 is to reduce the ambient oxygen level to 15% or below by driving out the air. But, as with a tank and water, you need to vent the air, not keep it bottled up.

Practically speaking, there are always going to be air-leaks to the protected space. Which points out the fallacy of trying to close every possible door and vent before setting off the CO2. Close as many as you can, but the important thing is to get off the CO2 before the fire does a lot of damage, and hot enough to keep re-flashing. You can send firefighters with SCBAs out immediately after the CO2 release to double-check all closures were made, after the air has been vented.


Not that many E/R vents or doors on the SOLAS equipped ships I’m familiar with, three doors one escape scuttle and IIRC 4 E/R Vents.

The fire doors close automatically each time after someone passes, hook-backs are not allowed. The E/R vents are held open by pneumatic pressure and can be closed either remotely or locally at the vent.

Shutting down the fans and buttoning up the E/R can be done in about the time it takes a crew member to reach the fire control room or where ever the controls are located.

Are you sure?
Filling a closed environment with inert gas dilutes the combustible gas content to an acceptable level. Indeed, the gas/inert mix must be evacuated until the concentration is low enough.

Inside an enclosed environment, the fire burns the free oxygen; there is no need to evacuate it.
However, it is essential to prevent the aspiration of fresh air from outside, by replacing the burnt volume with inert gas.


This is assuming that all those things work. If things worked as intended all the time, they probably wouldn’t be on fucking fire.

The condition of SOLAS required firefighting equipment on a ship in international trade is going to get a lot of scrutiny with Port State Control, class and crew inspections. But you’re right it’s not 100% it’s gong to work when needed.

It’d be SOP to verify the vents and doors are closed before releasing the CO2. According to the report in this case the C/M checked every thing was closed while the crew was re-mustering after retreat from the engine room.

The initial entry by the 1 and 2nd A/E was done while all the ventilation was still running. Little chance of a backdraft at that point although they should still open the door while standing clear of the doors swing etc. as trained.

I suppose it could be argued that ventilation fans should remain running while the fire teams make entry. It’s not clear from the report if the vents had been closed or not at that point. Either way the guidance should be in the written fire plan. Otherwise I’d leave that debate to the experts.

As far as the direct / indirect attack debate… there’s never going to be a blanket rule to use the fixed system in every case. While there could be a debate as to where to draw the line It’s ultimately going to be up to the judgement of the crew actually on scene.

All I know is the design criteria and theory behind American fixed fire-fighting systems. It may be that SOLAS rules result in the same design criteria but based on a different theory. Here is the underlying design theory in the USA:

The CFRs design parameters for a fixed CO2 system specify that the amount of CO2 released must be sufficient to reduce the volume of O2 in the compartment to 15% or less, in order to reduce the O2 level on the atmosphere below the 16% level needed to support open flame. (That last part isn’t explicit in the CFRs but the ‘15%’ indicates the reasoning.)

The purpose of a Halon system is to impeded the chemical reaction of combustion, not reduce the O2 level, but it too does not diminish combustible gas content.

Releasing that amount of CO2 in a space increases atmospheric pressure. I can’t say how much but even small pressure changes in an enclosed space might cause structural damage. But in reality the vents associated with the engine room are tight but not completely gas-tight. Certainly, they aren’t on many any uninspected vessel. So the venting of over-pressure occurs naturally.

By the way, I’ve twice been on vessels where the engine room CO2 system was unintentionally set off while the vessel was underway. Quite exciting, and interesting to see what really occurs.

CO2 systems discharged into a reasonably sealed engine room are very effective, I’ve seen them deployed (often too late) but they work. Just have to make sure the space is cool before opening.
In the case being discussed I can understand the engineers thinking. They hoped to contain the situation and avoid loosing propulsion. Its every engineers biggest worry if in close quarters or bad weather. Considering that I think they did well with some minor caveats I am sure they are aware of now.

I don’t think anyone would consider not attempting to seal the space, but all too often in drills if you introduce a vent closure failure the response is to keep trying to close it or delay release until it can be sealed.

In my view it’s similar to flooding. 30 seconds to start dewatering. Then you move to damage control. People will stand there fiddling with a choked pump while the water forces them from the space when they could simply shove a mattress in the hole. Vent closure failure is similar.

1 Like

A great many engine room vents are closed simply by releasing a securing system and allowing the flap to close under gravity thus ensuring it highly unlikely that the engine room space be over pressurised.
I made a point of ensuring where a firefighting team wearing BA were entering a space they were carrying hoses not fire extinguishers.
The one engine room fire on a Ro-Ro that I had experience of, we extinguished the fire using two teams but it was a near thing, and we were close to a a decision to flood with CO2. An added complication was one of the crew had a heart attack during the fire. He survived.


The CE who was onboard at the time is now onboard a Matson ship as a relief CE. Don’t know about anyone else.

1 Like