Curiosity is a much underused tool for improving safety. From the commissioning of the 93m chemical tanker Key Bora in 2005 no-one wondered why the astern response of its controllable pitch propeller, CPP, was four times slower than its forward response, it was accepted with a shrug as just one of the quirks of this particular vessel. It had not gone unnoticed, it had just gone unquestioned until she rammed a jetty in Hull putting a 90cm hole in her bulbous bow just above the waterline.
It is a good example of how something Not Quite Right, NQR, can lead to a close call and when both go unremarked sooner or later there will be a hit. In the old days of naval warfare the first shot rarely hit the target, it would either overshoot or under shoot the target. A range adjustment would be made and a second shot fired. If that didn’t hit the target it still enabled the gun crew to get a more accurate range, to bracket it, and the next shot would hit the target. A wise commander on the target vessel would take avoiding action to prevent the aggressor bracketing his vessel.
NQRs and near-misses are rather like those naval battles. If you don’t take avoiding action they will get you sooner or later.
During the Key Bora sea trials conducted in August 2005, the CPP system was commissioned and tested by local service agents for ZF Marine. manufacturer of the CPP, in the presence of Bureau Veritas surveyors. The full range of ahead and astern movements in the constant engine speed mode was tested. The response times to attain the various pitch positions were not recorded; BV’s rules for classification of ships did not require the response times to be recorded.
On 23 December after the contact incident further tests were carried out on the system by electrical technicians in the presence of the master, chief engineer and a BV surveyor. These tests did not include a check of the time delay between pitch demand and response. Following these tests, BV concluded: Occasional survey of the machinery installation carried out to verify the operation of the CPP. Extensive testing carried out from all remote locations, back up controls and locally and no defect could be identified at this time. The system was additionally checked by electrical technitian [sic] from the shipyard who confirmed the system operated as required. No further action deemed necessary at this time.
In fact, the day after the incident MAIB inspectors and a surveyor from BV witnessed manoeuvring operations from the bridge while Key Bora shifted to a repair berth. During this time, despite the pitch control lever being set to 70% astern for over 75 seconds, the pitch achieved did not
exceed 50%. Tests to compare the response times for ahead and astern pitch were carried out with the clutch disengaged. The time taken to achieve 50% ahead pitch was 5 seconds, whereas it took 21 seconds to obtain a 50% astern response. When the system was set to backup control, the response time for both 50% ahead and astern was 3 seconds.
When the ZF Marine service agent, Berg, saw video of the tests it concluded that the system appeared defective and may require its technician to carry out repairs followed by a sea trial. These repairs did not take place.
There was no performance standard against anyone could judge the system’s response.
Since Key Bora‘s officers did not have the password with would have enabled them to change the settings on the CPP’s programmable logic controller, PLC, it is likely that the system had defective settings since its installation.
Since PLCs and programmable devices hae been involved in incidents and subject to warnings over the years a curious mind will ask what does one do if something goes wrong. Key Bora had a backup system that would enable the CPP to be directly controlled through a joystick with instructions clearly marked on the console. However, the vessel’s safety management system contained no requirement for deck officers to
familiarise themselves with this mode of control or to conduct propulsion failure drills, and no such drills or familiarisation had been completed.
From 2005 onwards no-one had been curious about the odd behaviour of the CPP, that it’s behaviour was NQR.
20 December 2013 was not the first time Key Bora had been to Hull under pilotage. Following a visit in November 2010, the pilot on that occasion had made an entry on the port’s Pilot and Vessel Information System PAVIS stating that the vessel had very slow astern power. However, the port did not require pilot to test the engines before manoeuvring.
A week before the incident the vessel was in Rotterdam Key Bora’s master was replaced by a new master who had not worked on the vessel before. During the handover from his predecessor, the new master was informed that the astern response of the CPP was very slow, and while
still in Rotterdam he experienced a near collision due to this slow astern response. No-one was hurt, no damage was done so he did not inform the vessel’s managers of the incident.
In a sense it was a ranging shot, one that missed but which indicated that avoiding action should be taken be taken. Since the company did not require the new master to be familiar with the backup he can, perhaps, not be faulted for not being curious about it.
Key Bora’s SMS 12 contained a pre-arrival checklist that required the engines to be tested. The requirement to test and prepare the engines for manoeuvring was also reinforced in the ICS 13 publication ‘Bridge Procedures Guide (Fourth Edition)’. However, as the vessel had anchored briefly before picking up the pilot, engine movements were deemed to have been tested. ABP did not require the port’s pilots to test the propulsion system on vessels the size of Key Bora before commencing pilotage. The pilot knew about the slow astern reponse and the master had verified it during the master-pilot exchange.
Around 18:29:25 as the vessel passed north of the Hook Buoy, the CPP pitch was set to zero and the pilot asked the master to transfer the manoeuvring controls to the port wing console. The vessel’s speed over the ground at the time was 4.4kts 6 , which included a tidal stream of 2.2kts from the stern. The master and pilot were on the bridge and the third officer was in charge of the forward mooring station.
The starboard anchor was ready to let go.
There another NQR lurking in the background: The vessel was fitted with a Phontech CI 3100 talkback system, with an audio output of 10 watts and with repeater stations forward and aft. It was not used on the day of the accident as it was defective. At the anchor station were equipped with UHF radios of 1- 4 watts adjustable output.
Shortly after shifting control to the port wing the pilot ordered full astern. The master placed the pitch control to full astern, and the vessel continued to move ahead while the propeller pitch indicator was seen to move very slowly astern. It was too late to learn how to use the emergency backup system with its far faster response.
When the vessel’s bow was 40m from the western approach jetty for the Alexandra dock, the pilot realised that an impact was inevitable so he asked the master to drop the starboard anchor and applied maximum starboard thrust on the bow thruster in an attempt to maintain control of the vessel. The master instructed the third mate, by UHF radio, to drop the anchor. He had to repeat this order five times before the third mate acknowledged him and dropped the anchor because of the noise on deck.
Whether using the far louder talkback system, and thus quicker dropping of the anchor would have avoided is uncertain.
At about 18:38:47 Key Bora’s bow made heavy contact with the jetty. Following the impact, the anchor was recovered and Key Bora was manoeuvred away from the jetty and into the open lock gate. The astern response remained very slow, but the vessel was successfully berthed at the designated berth.
There are numerous take-aways from this story but the one MAC would suggest first is ‘be curious’. When something is NQR find out why. An SMS that does not require the master’s familiarity with the emergency backup for a vessel with CPPs is NQR. A CPP that isn’t resonding as it should is NQR. So when something is NQR be nosey, be curious.
Practice curiosity, it really is a useful safety tool.