Technical Analysis (Engineering) of NTSB Preliminary Report M/V Dali

It will be interesting to see what other ports mandate escort tugs for ships passing through bridges and passing by other critical infrastructure.

It should be all of them.


I’m starting to get the impression that the ship electrical systems that run the main engine oil pumps, fuel pumps, hydraulic pumps, cooling water pumps, steering etc. are too complex and prone to failure to be trustworthy.

Maybe it’s time for a return to mechanically driven critical systems for the main engine?

Or a dedicated main engine driven generator with battery back up for these critical systems?


Hear … hear.

1 Like

The undervoltage protection and the UV coil are 2 different things. The sole purpose of the UV is to ‘enable’ the breaker to close when there is voltage ‘upstream’ of the breaker. Similar to a generator breaker. This UV coil is not the primary means to trip the breaker due to a fault. That is usually done with separate coil (shunt coil) with the requisite wire break monitoring, etc. The UV coil may be supplemental - but not the primary means to trip the hair trigger.

Trips will include (among other faults), under/over voltage, under/voltage frequency, high current with different settings per the TCC, etc.

All the breakers on the high tension or voltage bus board (HV) will have this installed such that they drop out upon loss of voltage. Includes the reefer breakers (10-14 in number), BT, HR1/2, etc.

As soon as voltage is restored to the HV (or HT) bus, these breakers are ‘armed’ or ‘engerized’ or ‘ready’ to close. Closing will be on a programmable sequential start program - with the HR1/2 as the priority and then the rest.

Was around for the change from steam to slow speed diesel. As a deck officer the enormous amount of single point failures that were being added to our major systems was by far our biggest concern.


The VDR will have all of that information once they publish it.


As I mentioned it earlier, kVA and kW have possibly been mixed for the generators and also for the same physical diesel engine there are various ratings. So we finally don’t know exactly what the different kW and kVA ratings refer to and in addition there are values for the maximum load in use, a.s.o.
Your EDG (Emergency Diesel Generator) rating estimation is better than mine.
Also, like you, I don’t expect any diesel engine driven fire pump on a container ship.

Please correct me if I’m wrong but based on the report it seems that the DGs didn’t even stop running, during all blackouts several DGs were still running. The power failures during the incident voyage were exclusively caused by the wrong breakers being opened at the wrong moment.

Isolating “TR1” by opening “HR1” and “LR1” could have been caused by the differential protection of transformer “TR1”, that caused the first blackout (whole 440 V 60 Hz supply excepted possibly the reefer supply) but both “DG3” and “DG4” were most probably still supplying the 6.6 kV busbar, as in case of frequency and/or voltage fault the corresponding generator protection would have tripped which was not the case as both breaker “DGR3” and breaker “DGR4” remained closed.
Transformer “TR1” was not defective as it was brought online later as breakers “HR1” and LR1" were closed manually, which ended the first blackout.
Possibly “HR1” and “LR2” trips are linked, if so, any fault requiring “HR1” or "HR2 " to trip will trip both “HR1” and “LR1” together.

The second blackout occurred as both “DG3” and “DG4” tripped by opening “DGR3” and “DGR4” which caused a total loss of the 6.6 kV 60 Hz supply (including the supply of the transformers for the reefer containers) and therefore a total loss of the 440 V 60 Hz supply, excepted the 440 V 60 Hz loads powered by the emergency diesel generator (“EG”), which include also some 220 V 60 Hz loads.

As the NTSB preliminay report doesn’t mention that the (auxililary) diesel engines “DG3” and “DG4” were normally stopped or tripped, we can assume that they were still running with generators disconnected from the 6.6 kV busbars. Also it’s unlikely that they’d have produced black smoke while probably idling at around 720 RPM.

“DG2” which was in stand-by mode started automatically after the 2nd blackout. It is not formally mentioned if its generator breaker “DGR2” was closed automatically but logically it should be the case, especially as the 6.6 kV busbar war dead (without voltage) at that time and therefore no sync procedure was required.

The NTSB preliminary report omits to mention that at some unspecified time between the beginning of the 2nd blackout and when breaker “DGR2” was closed, breaker “HR1” and/or breaker “LR1” must have been opened (how and for which reasons is not formally known), if both “HR1” and “LR1” would have been closed when “DGR2” was closed, the 440 V 60 Hz bus would have been immediately powered which would have ended the 2nd blackout.This is not the case as both breakers “HR2” and “LR2” had to be closed manually somehat later, which put transformer “TR2” back online and ended the 2nd blackout.

Therefore I assume that:

  1. At any time during the whole incident voyage at least one DG (Diesel Generator, any one of “DG1”, “DG2”, “DG3” or “DG4”) that was able to supply enough power to allow the ME (Main Engine) to run was “very probably” operating correctly even if not necessarily connected (i.e. ready to supply power any time).
  2. The first blackout was due to the transformer “TR1” going offline as both breaker “HR1” and “LR1” were opened for unknown reasons.
    Neither generators “DG3” nor “DG4” stopped, both remained connected to the 6.6 kV busbars, which means that with a very high probability both voltage and frequency were within their respective tolerances (if not the corresponding genset controller would have opended the corresponding generator breaker).
  3. As the crew manually closed “TR1” and “LR1” which brought transformer “TR1” back online the 1st blackout ended.
  4. Both generators “DG3” and “DG4” tripped (breaker “DGR3” and breaker “DGR4” were both opened) for unspecified reasons which caused the 2nd blackout.
  5. Not mentioned in report: An some unspecified time, breaker “TR1” and/or breaker “LR1” was opened. Also it can be supposed that both “DG3” and “DG4” didn’t stop as again nothing is mentioned in the report.
  6. Generator 2 (“DG2”) which was in stand-by mode automatically started as power was lost due to the 2nd blackout, generator breaker “DGR2” automatically closed, restoring power to the 6.6 kV busbar.
    It can be assumed that “DGR2” was closed automatically though not formally mentioned in the report as there is no mention about it having been closed manually.
    As both transformers “TR1” and “TR2” were still offline, the main 440 V 60 Hz busbars remained dead (not energized).
  7. Breakers “HR2” and “LR2” were closed manually, bringing back online transformer “TR2” and therefore supplying the main 440 V 60 Hz busbars which ended the 2nd blackout.
  8. The report does not mention anything about the transformers powering the outlets for the reefer containes. If those loads played a role cannot be assessed as there is no information available.
  9. It can be assumed that as during the whole incident voyage at least 2 (“DGR3” and “DGR4”) and after the automatic start of “DGR2” even 3 diesel generators were runing and available.
  10. Both blackouts were caused by breakers being opened for unknown reasons.
  11. As generators were always running, a fast and appropriate reaction of the crew could have reduced both blackouts to a few seconds but it’s not realistic to expect that in such a situation.
  12. It’s formally unknown if there was an attempt to start the ME (Main Engine), the report mentions only that it was not restarted.
  13. The black smoke could have been generated by trying to restart the ME (Main Engine) as there is no obvious reason which could explain why the generators would produce such smoke for such a long time. I don’t see any other explanations and don’t expect the boiler to have caused it but maybe I’m wrong.
    If a DG engine would be overloaded so far to massively generate black smoke, which typically can happen when suddenly applying a heavy load, either the engine recovers very quickly (only producing intense black smoke a few seconds) or the generator trips as the voltage and frequency will drop below the alarm limits before the engine stalls so I don’t expect any DG to produce such black smoke more than a few seconds when starting or during quick massive load changes.
  14. The causes of the breaker trips are unknown. That first a transformer was tripped and than brougt back online successfuly shortly after, followed by two simultaneous generator trips seems sort of odd.
    As the breaker trip causes are currently unknown, it’s impossible to know if the trips were initiated due to true faults or if they were spurious.
    The settings of the protection devices need to be checked and also tests should be performed either in situ or with simulated PT (Potential Transformer) and CT (Current Transformer) signals to verifiy if trip thresholds have not drifted and if timeout delays are respected. Also other devices which can trip breakers should be checked.
    Overall I rather suspect a problem with the protection systems and/or the power management system (including simple wiring problems to firmware issues like stack overflows and other rare oddities which may suddenly affect a device after years of uninterrupted service) though usually most issues are related to commissionning (wrong parameters, wiring erros,…).

I have said before that the wind effects on container ships is very variable due to the distribution of the deck load. I would guess, and it is a guess that a NNE wind acting on the port quarter would be inclined to induce a turn to port.
Or push the vessel bodily to starboard. But at 8 kts I would not think it would have had much effect. In my experience you only start to have problems with the wind at slower speeds which is why sometimes it’s better to keep a bit of speed on.
Speculation of course.

1 Like

May I suggest that you try and read some of the posts from the engineers (AV2013, KPChief, etc) and the experienced Pilots and Masters (Ausmariner, 244, 118, Texas Tanker, Tug Captain, Spo, etc ) starting from March 26. Everybody on this forum is now beyond the grief stage and more or less angry a totally avoidable incident can be allowed to happen. We are interested to get to the bottom of the root cause and ensure that the outcome of the lesson learnt are changes (not an exhaustive list) to design, procedures, training, etc are implemented in short order. Nobody on this forum is an all around expert on all aspects of ship operation and rely on each other and brainstorming possible causes. I am pleased to see the first of these measures mandating tug escort is already in place – at least in Baltimore.

From your posts, you are certainly a well informed electrical engineer. Some of your comments on engine operation, steering gear rules and setup (on your first post), etc needs some more research on your part on steel vessel regulations and rules.

In the report there were a couple of blackouts. One operator error with DG2 exhaust damper closed and the second with DG3 with low fuel pressure. Low fuel pressure?? WTF - Give us a break (vessel crew and NTSB). I think the cause of this incident is related to the passage blackouts – both 1 and 2. So what can cause (most likely) a momentary drop in fuel pressure? Auto Filter? Accumulator/hydrophore vessel on the low pressure fuel loop malfunction? Who knows – we have no idea if the fuel circuit on the vessel is a pressurized system. If this repeats (drop in fuel pressure) on the passage it may explain both blackouts. PR is very basic and almost seems it is by design such that theories from various sources including this forum may give them some ideas.

1 Like

The issues of blackouts and loss of propulsion have been studied and addressed in the offshore industry for a long time now, it is just a matter of using that knowledge and experience.

From USCG Safety Alert 11-22:

The Coast Guard strongly recommends that owners and operators of mobile offshore drilling units (MODUs) and vessels equipped with DP2 and DP3 systems:
• Use an open-bus configuration as the preferred mode of DP operation while performing critical activities. This eliminates the possibility that an unforeseen failure could propagate through a closed-bus and limits the blackout to the affected bus while the unaffected bus(es) would retain some, although diminished, position keeping capabilities.

Technically, Dali electric power plant could be operated in split mode, that is having two independent power generation and distribution systems, with both HV and LV bus ties open, and at least one generator connected to each HV switchboard. That is the easiest way to achieve redundancy and it could be used when sailing in congested waters.


There are still many steam LNGC around. Everything is electric except feed water pumps. And is virtually impossible to run the main boilers without the ship`s main automation system :man_facepalming:

From your observations I am guessing you are a Humber pilot. Did you do any sea time with Texaco Tankers.

Initial Sea time was with BP
And I was a Thames or more accurately a PLA Pilot
Did a bit of time with Texaco Tanker cadets at Warsah

Thanks for that. When I was with Texaco in early 70’s trainee Humber pilots completed their apprenticeship on the river. After obtaining their second mates ticket did their sea time for mates with us sailing as third mate.

1 Like

Times change.
At London we ditched that system quite a long time ago. After the 1988 pilotage act there were no more pilotage apprenticeships .
Some, myself included, said amen to that.
By the time I applied you had to have already an unrestricted masters certificate plus at least masters experience or 2 years as chief mate.
As the talent pool got smaller we did ditch the masters experience but still insisted on a substantial time as a senior officer.
Personally I served 7 years as master prior to joining the PLA

1 Like

I started as a trainee pilot in 1999 and retired in 2022


I agree and let us not forget that this was all over a period of only four minutes. Something else was at play.

It was our nightmare scenario as pilots.
If I was that way inclined it would keep me awake at night just thinking about it.
I know some of my former colleagues did worry about " The Big One" but I could not live like that.
During my tenure we did have some fairy large ones but nowhere near the scale of this one.
My heart goes out to the guys who were passengers on this one.

I must say that this split thread and its parent have been an absolute pleasure. A lot of very experienced and knowledgeable professionals from the engine and deck departments contributing civilly and constructively.

I have learnt a great deal and I thank you for that.

1 Like

And so do I
Thanks guys

1 Like