Transcript:
On the 26th of March 2024, at around 0129 hours, the container ship Dolly collided with one of the pillars of the Francis Scott Key Bridge in Baltimore, causing the bridge to collapse. Based on the reports these past few days, this tragic incident has already caused at least six fatalities and, according to some analysts, could potentially incur total losses between two to four billion dollars.
By now, I’m pretty sure a lot of you have already watched various videos, commentaries, and reactions about this incident. We have seen footage of the ship moments before it hit the bridge. We already know that a blackout occurred, which led to the loss of propulsion. I’m pretty sure a lot of you have been wondering as to what could have caused this failure, which in turn cascaded into this catastrophe. We’ll dive into that, but before we do, just a disclaimer: the actual sequence of events in the engine room has not yet been made public. We only saw a bit of the timeline from what was recovered from the Voyage data recorder. So, I’ll analyze this based only on what I see from this footage, and I’ll try to make sense of it based on my knowledge and experience as a chief marine engineer. It’s just like doing an incident investigation and root cause analysis, which we, as the onboard management team, typically do whenever incidents happen on board.
Let’s take a look at the CCTV footage and check out the timestamp so we can analyze the sequence of events. So here we can see the vessel as it was approaching the bridge. As per the video’s timestamp, the blackout occurred at 32 seconds past 0124. There’s a slight difference in the time between the footage and the transcription from the VDR, so if I happen to mention a different time from what is shown on the CCTV, that just means I’m referring to a different source. Anyway, when a blackout occurs on board a ship, all of the machinery stops inside the engine room, the noise quickly dies down, except for the very loud and sometimes panic-inducing audible alarms, which will be persistent for about a minute or so as abnormal. Also, the engine room will suddenly go dark.
Now, as a ship’s engineer, those are the two things that you don’t want in the engine room, especially during maneuvering: darkness and silence because that means the ship is dead on the water. Whenever I’m on board a ship, when I’m sleeping, I actually get awakened if it suddenly becomes quiet. The noise and vibration from the running engines are actually the normal background. Now isn’t there a backup generator or something? Well, yes, but it’s a little bit more complicated than that. First, I need to explain a little bit about the marine power plant.
So basically, we have the main engine, which is the huge engine that provides propulsion power; in short, it’s the engine that spins the propeller and causes the ship to move. Now, the thing is, the main engine requires multiple auxiliary machinery to be running, like fuel pumps, lube oil pumps, cooling water pumps, and others, before it can be operated. And those pumps, those systems, they need electricity to function. For that, we have the generator engines, which provide electricity to the entire ship. For the purpose of this video, I’ll refer to them as the big generators. Typically, there are three generator engines on board, but for a container ship of this size, it probably even has four. I’m not sure for this ship, but what’s for sure is that the total electrical load requirement of the entire ship can be easily handled by at least two generators, with one extra generator on standby, ready to take over in case of failure. So there is usually no shortage of redundancies when it comes to electric power.
And then there is the emergency generator. It automatically starts in case of a blackout. SOLAS requires it to automatically start within 45 seconds of a power failure. It is located outside of the engine room and has its own fuel tank. However, it is driven by a smaller engine, and it’s dedicated to provide power to only a select group of machinery and lighting, all of which are powered through the emergency switchboard, which automatically disconnects from the main switchboard in case of a blackout. Now the equipment connected to the emergency switchboard serves the purpose of restarting the bigger generators, maintaining the operation of radar and navigation equipment, emergency and navigation lights, and of course, firefighting. The emergency generator doesn’t generate enough power to restart the main engine and restore propulsion. However, it provides power to the ship’s steering gear. Yes, once the emergency generator kicks in, the ship should be able to use the rudder, you know, to steer the ship. But more on that later.
And then there are also a few batteries, which automatically connect to the radios and other small emergency equipment, a few battery-operated lights, as well as UPS for computer-based devices like the EIS. Okay, so now that we have the basic idea of a marine power plant, let’s look at the footage again. As we have seen earlier, the blackout occurred at 32 seconds past 0124. So, as I mentioned, the emergency generator should kick in within 45 seconds as per SOLAS requirements. Here we see the electric power was restored at 31 seconds past 0125. That’s 59 seconds. Wait a minute, does that mean the emergency generator did not run and automatically connect on time?
Well, there were signs that it didn’t go online within the required time. If you look closely, for the entire duration that the lights were out, the nav lights were also out, and they never came on until all the other lights went on. The navigation lights are powered through the emergency switchboard, which is supposed to be powered by the emergency generator during blackouts. Also, according to the initial transcript of the timeline, the VDR stopped recording the ship’s data, except for the audio in the wheelhouse. It resumed a little over a minute later. Those are telltale signs that even the emergency power didn’t kick in within 45 seconds. What we’re seeing here, when the lights came back on, was probably just the emergency generators. I might be wrong, but I’m very familiar with the sequence when restoring power after a blackout, so I say there’s a big chance that that’s the case.
However, it’s also possible that they managed to restart the big generators and put them on load. That could also happen even without waiting for the emergency generator to kick in. Anyway, a few seconds later, after the lights came back on, you can see black smoke coming out from the stack. Initially, I thought this was from the main engine. I thought they were trying to do crash astern maneuvers, and the black smoke was the result of a poor fuel-to-air ratio because they canceled the load limits of the main engine, you know, to quickly increase the speed. However, it was only 13 seconds after the lights went on. I doubt they managed to restart the main engine that fast. There are a lot of things to reset before the main engine can be ready for operation. Well, it’s not impossible, but I seriously doubt they were able to. It’s more likely that the black smoke was coming from the big generators.
Anyway, assuming they were able to start the big generators and put them on load, this means they should have had the capability to restart the main engine. The problem is, at a speed of around 8 knots, it is very likely that the propeller was still spinning in the ahead direction due to momentum. Running the main engine in crash astern direction would be a bit tricky, and there will definitely be a few start failures before they get the engine running. It’s a matter of technique, but I’ve done it many times as part of a testing procedure years prior to the ship entering port. It should work as long as there’s nothing wrong with your main engine or fuel.
However, if we look at the AIS tracking, I don’t think they were able to run the main engine at all. Here’s why: by this time, the steering gear should be functional, even if they were still in emergency power mode. One of the steering gear motors would have already been functioning because it’s connected to the emergency switchboard. We can also see from the VDR timeline transcription that various rudder commands were given by the pilot at around this time. So that means the rudder was responding to the helm commands. Okay, so the steering gear was working at this time.
However, assuming they had use of the rudder, why didn’t the ship turn to safety? This kind of confirms that they weren’t able to restart the main engine because without propulsion power, the rudder won’t be able to turn a ship of this size effectively. To explain it simply, the rudder makes the ship turn by deflecting the water stream generated by the propeller. The force of that deflection creates a turning moment or torque around the ship’s center of gravity. So that, in combination with inertia, causes the ship to begin turning in the direction determined by the rudder position. The fact that they weren’t able to turn the ship is an indication that there was no propulsion.
So why was there black smoke coming from the stack? Where did it come from? Since the smoke appeared around 13 seconds after the lights went on, we can presume that it came from the big generators. I’m ruling out the auxiliary boiler automatically firing in this case, because it takes about a minute of pre-purging before it could fire. Presumably, they were trying to restore the main electrical power, which is logical because that’s the only way for them to be able to restart the main engine. But normally, there shouldn’t be any black smoke. In US waters, ships are required to use LSMGO or low sulfur marine gas oil, so it’s relatively cleaner compared to the black stuff that we normally use when the ship is out in the open sea. And even when ships use heavy fuel, there still isn’t supposed to be any black smoke under normal operating conditions. So why was there black smoke? This could possibly be one of the clues in finding out the root cause of the blackout.
Whenever ships are maneuvering into or out of ports, two generators are usually in service and at least one is on standby. Typically, one generator is capable of handling the essential electric loads needed to operate the main engine, but for safety, two of them are put on service to ensure that there is more than enough power available and that blackouts won’t occur while maneuvering. So, in case one of the in-service generator trips, the one on standby should be able to start and take over. But in case it doesn’t, the preferential trip will be activated on the main switchboard. This will shut down the non-essential machinery, you know, the ones not needed for maneuvering, like refrigeration and air conditioning, purifiers, blowers, and various other machinery. So, this preferential trip should be able to reduce the load enough that one generator is more than sufficient to maneuver the ship.
Now, for a ship to go on blackout while two generators are running means they both malfunctioned at the same time, or almost the same time. The two most probable reasons for this to happen are either fuel-related or switchboard-related. Either reason may explain why the standby generator didn’t automatically start and take on load. A switchboard malfunction, although possible, is quite improbable. For this, it is still possible, though, but it will involve a bit of, well, a lot of malfunctioning parts, or even human intervention, like someone doing electrical repairs on the main switchboard, which is a little bit unthinkable while the ship is maneuvering. I’m not ruling that out, though. It’s bad practice, but I’ve seen it happen before. It could explain why they weren’t able to bring the generators back on load, but it can’t explain the black smoke.
Now, the fuel-related reason seems to be more plausible in this case. Simply put, the fuel supply was suddenly cut off. I don’t think it would be something like clogged filters because that would happen gradually, and there are numerous failsafes for that, and it is easy enough to rectify. A sudden stoppage of the fuel supply would explain why both generators would shut down at the same time, as they share a common fuel line, and also prevent the standby generator from starting automatically. But since the ship was presumably using LSMGO, they might have been using the small flushing pump instead of the dedicated fuel oil booster pumps for the generators. The small pumps might not have had enough pressure, you know, to sustain the load on the generators. Or maybe someone mistakenly closed a valve or accidentally activated the quick closing valves or maybe even forgot to refill the fuel tank. It could happen. I’ve actually seen that before. There are many maybes, but the thing is, at some point, the fuel supply got cut off.
However, a simple closed valve is easy enough to rectify, and it shouldn’t have been difficult, you know, to restart everything from there. Now, there’s also the angle of contaminated fuel. Yes, that could also happen. In fact, the black smoke seems to be an indication that this might have been the case. There’s a possibility that they change over to a different tank, which had bad fuel or contaminated fuel, which could have caused the generators to stop at the same time. Or, and this one is a bit controversial, they tried to change over to heavy fuel oil too early, and something went wrong. That could also explain why the standby generator didn’t start and why there was black smoke afterward. Now, we can’t know for sure what specifically caused it, and I’m not accusing anybody of doing anything. I’m just trying to make sense of things because this could happen to any seafarer. It could happen to me.
Anyway, at around 0126, the pilot called for tugs in the vicinity through VHF. Now, looking at the footage, after over 1 minute since the light went on, it went dark again, but strangely enough, the navigation lights didn’t go out this time. This might have been an attempt to put the big generators on load, although black smoke was still coming out of the stack again. This is not normal as it indicates a poor fuel-to-air ratio or incomplete combustion, so it’s really strange since, prior to the blackout, there was no black smoke. So, at this point, the pilot gave the order to drop the port anchor since the ship was not responding to the rudder movements. As we can see, it didn’t have much effect.
At around 25 seconds after 0127, the pilot issued a radio call through VHF, reporting that the vessel had lost all power approaching the Key Bridge. So, since the lights went out again, we are assuming that even the emergency generator was, for some reason, disconnected. That means they don’t have any rudder control anymore because there’s no electricity supplying the steering gear motors. Anyway, after about 30 seconds, the lights went on again. Looking at the vessel tracking, the ship barely lost any speed and was already veering to starboard. At this point, collision was imminent.
At first, I actually thought the main engine was running astern at this point because the ship suddenly turned to starboard. This phenomenon is known as propeller walk, and the ship turned to starboard because it had a right-handed propeller. But, of course, after a thorough review of the footage, we know by now that they weren’t able to restart the main engine, which brought us to this fateful moment. At around 0129, the container ship Dolly collided with the Key Bridge at a speed of around seven knots, which caused the bridge to collapse.
After a major incident like this, we can easily assume that the crew morale is very low. The captain, being the overall in command of the ship, is definitely in a very stressful situation. But since the immediate cause was a blackout, the chief engineer would be up to his neck in crap and probably feeling depressed right now. You see, even if they were not the ones directly responsible for anything, it’s command responsibility. Unless there’s undeniable evidence that someone else did something wrong, the captain will always be in the spotlight. If it’s engine related, then the chief engineer will be too. Both of them will spend the next few months writing up statements, explaining what happened, how it happened, why it happened, and who did it. Presumably, they will be brought to court, so that’s another gloomy prospect that they won’t be glad to look forward to.
Looking at this tragic incident, we can expect that the reaction of ship owners will be to send their crew for additional training. But the thing is, there are already training courses using simulators that address this particular scenario. In fact, I used to handle the engine side of those scenarios back when I was teaching in maritime training centers. To be honest, what happened in Baltimore is almost exactly what happens every time we run those simulations, wherein a blackout occurs in an area with heavy traffic. It almost always ends up in a collision. It’s basically Kobayashi Maru, a no-win scenario, but without Captain Kirk cheating. Because realistically, if something like that happens, there won’t always be enough time to reset all the systems and restore power and propulsion. Our technology just isn’t there yet. The best we could do is to train the crew to retain their presence of mind and try to minimize the damage by acting promptly when faced with a similar situation.
Could they have done things differently to avoid a catastrophic outcome? Maybe. If the port regulations required tug assistance all the way out of the harbor, would that have made a difference? Definitely. But could have, would have, should have… One thing is for sure, though: prevention is better than a cure. No matter how much we want to spin this, we cannot deny that there were lapses on board the ship. I know it’s a hard pill to swallow, but it is what it is. Sometimes, even if you have been very meticulous in your work, things have a way of going wrong. There will be times when you will find yourself between a rock and a hard place, and the choices that you make will determine if you’ll come out of it unscathed or not come out at all. Thank you for watching, and see you on the next one.