MARINE SAFETY INVESTIGATION

REPORT 160

Independent investigation into the lifeboat incident onboard the Philippines flag bulk carrier

at Abbot Point, Queenslandon 6 August 2000

Washington Trader

Department of Transport and Regional Services

Australian Transport Safety Bureau

Navigation Act 1912Navigation (Marine Casualty) Regulations

investigation intothe lifeboat incident on board the Philippines flag bulk carrier

Washington Traderat Abbot Point, Queensland on 6 August 2000

Report No. 160

December 2002

ISSN 1447-087XISBN 1 877071 21 8

Investigations into marine casualties occurring within the Commonwealth's jurisdiction are conductedunder the provisions of the Navigation (Marine Casualty) Regulations, made pursuant to subsections425 (1) (ea) and 425 (1AAA) of the Navigation Act 1912. The Regulations provide discretionarypowers to the Inspector to investigate incidents as defined by the Regulations. Where an investigationis undertaken, the Inspector must submit a report to the Executive Director of the Australian TransportSafety Bureau (ATSB).

It is ATSB policy to publish such reports in full as an educational tool to increase awareness of thecauses of marine incidents so as to improve safety at sea and enhance the protection of the marineenvironment.

To increase the value of the safety material presented in this report, readers are encouraged to copy orreprint the material, in part or in whole, for further distribution, but should acknowledge the source.Additional copies of the report can be downloaded from the Bureau’s website www.atsb.gov.au

Australian Transport Safety BureauPO Box 967Civic Square ACT 2608 AUSTRALIA

Phone: 02 6274 64781800 621 372

Fax: 02 6274 6699E-mail: marine@atsb.gov.au

ii

iii

Contents

Incident summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .1

Sources of Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2

Narrative . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

Washington Trader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

Lifeboats . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3

The incident . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

Comment and analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

Lifeboat inspection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

Initial observations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

The after hook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

The forward hook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

The operation unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

Further examination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

Other evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

The incident . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

Pac Monarch incident . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

The after hook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .12

Automatic gripes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

The forward hook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .13

On-load release design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

Hook orientation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

The retaining latch . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15

Hydrostatic interlock . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .16

Indicator lights . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .17

Operation and maintenance of the NS-Hook release system . . . . . . . . . . . . . . . . . . . . . . . . . . .17

Crew training . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .18

Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .19

Recommendations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .21

Submissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .23

Washington Trader . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .25

iv

Figures

1. Washington Trader alongside the Abbot Point coal loader . . . . . . . . . . . . . . . . . . . . . . . . . . . . .v

2. Lifeboat release system schematic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4

3. Hook unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

4. Operation unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .5

5. Pressure switch and alarm light control box . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .6

6. Washington Trader’s starboard lifeboat . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .7

7. Damage to the stern of Washington Trader’s port lifeboat . . . . . . . . . . . . . . . . . . . . . . . . . . . . .9

8. The after hook as found . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

9. The forward hook as found . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .10

10. Operating unit as found . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .11

11. Washington Trader: Events and causal factors chart . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .20

v

FIGURE 1:Washington Trader alongside the Abbot Point coal loader

vi

IncidentSummaryAt 2036 on 5 August 2000, the Philippines flag,panamax bulk carrier, Washington Traderarrived at the anchorage at the Abbot Point coalloader in Queensland. The ship was in ballastand intending to load a full cargo of coal forexport to Japan.

While at anchor on 6 August, the masterdecided to take the opportunity to conduct anemergency steering and fire drill, followed byan abandon ship drill using the ship’s totallyenclosed lifeboats. At 1600, the crewcommenced the abandon ship drill. As there hadbeen a full lifeboat drill some seven weekspreviously, the drill involved a boat muster,followed by the lowering of the unmannedlifeboats to deck level.

By 1612, the boat muster had been completedand both lifeboats prepared for lowering. Themaster ordered the starboard lifeboat to belowered first and this was completed withoutincident.

The master then ordered the port boat to belowered. The mate, who was the officer incharge of the port boat, started lowering the boatby operating the brake release lever on the davitwinch. When the boat was approximatelyhalfway to the deck, the master saw the afterend of the boat swing and jerk twice followedby the after fall detaching from its on-loadrelease hook. When the after fall detached, theboat’s stern fell and swung forward and the boatwas seen to jerk twice more followed by theforward on-load release hook releasing its fall.The boat then fell stern first approximately 15 m to the water below.

After the incident, the lifeboat was brought backalongside. There was obvious damage to thestern of the lifeboat and a jacobs ladder wasrigged to allow the mate to enter the boat toinspect the damage. He found that the boat wasflooded and severely damaged at the stern withthe canopy broken and set-in. He also found thatthe after deck around the on-load release hookhad been damaged by the impact.

After initial unsuccessful attempts to recover thelifeboat, it was secured alongside overnight. Itwas recovered early the next morning using theauxiliary lifting shackles on the on-load releasehook units. The lifeboat was re-stowed in itsdavit and secured with additional lashing.

1

Sources ofinformationThe master and crew of Washington Trader

Nishi-Nippon F.R.P. Shipbuilding Company Pty

Australian Maritime Safety Authority

Transportation Safety Board of Canada

Bureau Veritas

ReferencesSafety of Life at Sea Convention, (SOLAS)1974 and subsequent amendments, InternationalMaritime Organization

2

Narrative

Washington TraderWashington Trader is a Philippines flag bulkcarrier of 74 228 deadweight tonnes at itssummer draught of 13.821 m. The vessel isowned by St Vincent Shipping Company and ismanaged by Victoria Ship Management Inc. ofthe Philippines. It is classed, I 3/3 E1 BulkCarrier ALT2 ESP3 Deep Sea4 , with BureauVeritas.

Washington Trader was built in 1999 by SaseboHeavy Industries Company of Sasebo in Japan.The ship has an overall length of 225 m, amoulded breadth of 32.2 m and a mouldeddepth of 19.2 m. Propulsive power is providedby a 6-cylinder B&W 6S60MC single acting,direct reversing 2-stroke diesel engine, of 8 826 kW. The main engine drives a single fixedpitch propeller, which gives the ship a servicespeed of 14.5 knots.

The ship is of standard bulk carrier design withseven cargo holds located forward of theaccommodation superstructure.

Washington Trader has a crew of 20 comprisedof a master and three mates, chief and threeengineers, boatswain and five deck ratings, fourengine room ratings, a cook and a messman.

At the time of the incident, the master ofWashington Trader held a foreign-going master’scertificate of competency issued in thePhilippines, and had 20 years experience at sea,the last two in command. The mate at the time

of the incident held a foreign going mate’scertificate of competency, also issued in thePhilippines, and had two years experience aschief officer. Both the master and the mate hadbeen on the ship since the delivery voyage inJanuary, 2000.

LifeboatsWashington Trader is equipped with two 25-mantotally enclosed lifeboats. The lifeboats are typeCML-19 boats constructed by Nishi-NipponF.R.P. Shipbuilding Company of Yamaguchi,Japan. The lifeboats are stowed in davits locatedon the port and starboard sides of ‘A’ deck, thefirst deck of accommodation above the maindeck.

The lifeboats are constructed of fibre-reinforcedplastic and each boat is 5.3 m in length, has abreadth of 2.3 m and a depth of 1.0 m. Theunladen weight of the boats is 1 670 kg with afully laden design weight of 3 870 kg.

The internal configuration of the lifeboats istypical of many modern totally enclosed boats.The coxswain’s thwart is located at the stern ofthe boat and is raised to allow all-round visionfrom a small ‘conning’ bubble in the top of thecanopy. All of the boat’s controls are accessiblefrom this position, including the davit winchbrake remote release cable and the on-load hookrelease lever. Seating for the rest of the crew isprovided around the periphery of the boat withsome seating also provided along the centre-lineforward of the coxswain’s position. There arehatches located midships on the inboard side ofeach boat for boarding the crew and additionalhatches at the forward and after ends of thecabin to allow the crew access to the on-loadrelease hooks.

3

1 Vessel class, division I ships are to meet the BV rules requirements for construction and scantlings of the hull and essential componentsrelating to propulsion and safety, as applicable. Rating fraction 3/3 is assigned to ships the condition of which is considered satisfactory to BV. Equipment symbol E, placed after the rating fraction indicates that the ship’s anchors and chain cables meet the applicable requirements of the BV rules.

2 ALT notation for ships designed in such a way that certain cargo spaces may be empty at a draught up to and including the scantling draught.

3 Enhanced Survey Programme.4 Notation assigned to vessels which are capable of deep sea navigation in any area and at any period of the year.

Propulsive power is provided by a DaihatsuCLMD-30 4-stroke diesel engine, which givesthe boat a fully laden speed of 6.1 knots.

Washington Trader’s lifeboats are fitted with‘NS-Hook Release Mechanism’ on-load fallrelease systems designed and manufactured bythe boat builder, Nishi-Nippon F.R.P.Shipbuilding Company. The main componentsof each system are the forward and after hookunits, the operation unit located on the afterbulkhead behind the coxswain’s seat and apressure switch and alarm light control boxmounted on the port bulkhead adjacent to thecoxswain’s position. Flexible operating cablesconnect the operating unit to the two hook units(figure 2). The system is designed to beoperable by one person with simultaneousrelease of both hooks occurring when therelease handle is actuated.

The main components of each hook unit are thehook, side plates, blocker and blocking knob,cable reset lever, latch and auxiliary liftingshackle (figure 3). When the operating lever isactuated the operating cable moves the cablereset lever down until its quadrant clears thenotch on the back of the blocker. The blocker isthen free to rotate and release the toe of thehook which allows the hook to rotate to the

open position. Resetting the hook unit is a two-handed operation it involves;

• rotating the hook to the closed position,

• rotating the blocking knob to move theblocker into position to lock the toe of thehook, and

• then moving the cable reset lever up until theblocker is locked in position.

The blocking knob is fitted with an indicatorwhich the operator must align with acorresponding mark on the hook side plateduring the resetting operation. Similarly, the endof the cable reset lever is painted orange tocorrespond with reference marks painted on theside plates. These marks indicate when the cablereset lever is in the fully reset position. Thelatch is fitted to the hook opening to retain thesuspension ring within the closed hook. It isswivel mounted to allow the suspension ring tobe placed on the hook and retained duringlifeboat recovery and may be manually trippedto disconnect the fall from the hook duringlaunching if required.

The main working components of the operationunit (figure 4) are the release handle, the cableconnection arms connected to each hook unitoperating cable, the interlock and the solenoid

4

Alarm light/control box

After hook unit

Electrical cables

Operation unit

Flexible operating cables

Pressure switch

Forward hook unit

FIGURE 2:Lifeboat release system schematic

which drives it. Actuation of the release handle(when the interlock is not in the blockingposition) raises each cable connection arm withthe resulting motion transmitted to each hookunit via the operating cables. There is a safetypin to lock the release handle in the stowedposition.

The solenoid, three limit switches within theoperation unit (one on each cable connection

arm and the third on the interlock) and the waterpressure switch are wired to the pressure switchand alarm light control panel.

The pressure switch and alarm light controlpanel indicate the status of the release systemwith red (LR) and green lights (LG) andcontrols the hydrostatic interlocking (figure 5).The limit switches on each cable connectionarm (LS1 and LS2) are wired in series with the

5

Hook

Latch

Operating cable

Stay

Side plates (only one visible)

Auxiliary lifting shackle

Blocker (blocking knob is notshown)

Cable reset lever

FIGURE 3:Hook unit

Release handle

FIGURE 4:Operation unit

Safety pin

Cable connectionarms (only one

visible)

Limit switches(only one shown)

Operating cables (only one visible)

Limit switch

Perspex safety cover

Interlock

Solenoid

water pressure switch (PRS) and are used in thecontrol circuit to energise the relay (R). Therelay contacts in turn control the power supplyto the red light and the solenoid.

Under normal circumstances if the power switch(S) is on, the red light will be lit if either of thehooks are not in the reset position or the boat isnot waterborne. When the boat is afloat(assuming the hooks are in the reset position)the relay will be energised. The relay contactsupplying power to the red light will open (thelight will go out) and the second relay contactwill close and energise the solenoid. Theenergised solenoid moves the interlock to theunblocking position. At this time the interlocklimit switch (LS3) will close and light the greenlight to indicate to the operator that it is safe tooperate the release lever. Once the hooks aretripped, LS1 and LS2 will open to de-energisethe relay which in turn will de-energise thesolenoid and relight the red lamp. The interlockwill be returned to the blocking position by thespring and when LS3 is opened the green lightwill be extinguished.

When the hooks are reset with the lifeboat stillafloat, the control system will return to the samestate it was in before the release lever wasoperated ie. LS1, LS2 and PRS closed, relay Rand the solenoid energised, and the green lightlit. At this time the green light indicates to the

operator that the hooks are reset and that it issafe to lift the lifeboat.

Washington Trader’s lifeboat davits consist ofthe frame, forward and aft cradles, boardingplatform, suspension blocks and falls and agripe system (figure 6). In the housed positioneach lifeboat is suspended in the cradles by thesuspension blocks with the fall wires runningfrom each suspension block, via a series ofsheaves, to the davit winch. Lashing unitsadjacent to the boarding platform are used tosecure each cradle to the davit frame when theyare in the stowed position. The cradles arepinned at the base to the davit frame to allowthem to pivot; outboard initially when the boatis lowered, and inboard at the final stage ofraising the boat. The davits are also equippedwith a wire-operated winch brake remoterelease. The remote release allows an operatorinside the lifeboat to lower it from the stowedposition. The davit and lifeboat are designed sothat normal boarding of the crew takes placewith the boat in the stowed position.

A gripe system is used to secure each lifeboat inits cradles when stowed for a sea passage. Itconsists of two plastic coated yellow wires, oneat each end of the lifeboat, which are passedaround bollards fitted to the boat’s bow andstern. The lower end of each wire is fixed to thebase of the davit frame and a ring on the upper

6

FIGURE 5:Pressure switch and alarm light control box

Circuit diagram

Control box

end of each wire is secured by a rotating hookmechanism (lashing lever) attached to each davitcradle. A turnbuckle allows the tension of eachgripe to be adjusted. The gripes must be fittedmanually when the lifeboats are stowed but areautomatically released by the rotating hookmechanism when the boats are lowered.

On Washington Trader each crew member isallocated a position in either the port orstarboard boat. The mate is the designatedofficer in charge of the port boat and the secondmate is the officer in charge of the starboardboat.

The incident At 2036 on 5 August 2000, Washington Traderarrived at the anchorage at Abbot Point north ofBowen in Queensland. The ship had arrivedafter completing an uneventful passage fromKawasaki in Japan to load a cargo of coal at theAbbot Point coal loader.

On 6 August, the day following the ship’s arrivalat the anchorage, the master made the decisionto conduct some safety drills. There was asouth-easterly wind of approximately 25 knotsblowing through the anchorage and a 1–1.5 msea. The ship was lying at anchor on a headingof approximately 120°.

At 1500 an emergency steering drill wasconducted followed by a fire drill and then an‘abandon ship’ drill. The ‘abandon ship’ drillcommenced at approximately 1600 and involvedpreparing and lowering the empty lifeboats todeck level only, as both lifeboats had beenlowered to the water and taken away by the crewon the previous lifeboat drill about seven weeksearlier.

Initially the crew were mustered in life jacketsat their boat stations and the roll was called. Themaster was stationed on the bridge. After all thecrew had been accounted for, the mate andsecond mate checked the life jackets and attireof the crew at their respective boat stations andthen prepared the lifeboats for lowering. Atapproximately 1612, both officers advised themaster that they were ready to lower thelifeboats. The master then gave the order tolower the starboard boat. The starboard boat wassubsequently lowered to deck level by thesecond mate without incident.

At 1615, the master ordered the port lifeboat tobe lowered. After giving the order, the mastermade his way to the port bridge wing to overseethe lowering of the boat. The mate startedlowering the boat by operating the brake releaselever on the davit winch while standing on deck.When the boat was approximately halfway to

7

FIGURE 6:Washington Trader’s starboard lifeboat

After gripe

deck level, the master, from his vantage point onthe bridge wing, observed the aft section of thelifeboat seem to swing followed by a jerk, then asecond jerk and then the after suspension ringcame out of its hook. Once the after fall haddetached, the master saw the stern of thelifeboat drop and swing forward with theforward hook still attached. The lifeboatappeared to jerk twice more before the forwardhook opened, releasing its suspension ring. Afterthe forward fall detached, the boat fell stern-firstto the water some 15 m below.

The master went to the boat deck and saw thatthe lifeboat’s forward painter was taut with theboat floating at the ship’s stern. He ordered thecrew to pull the boat forward using the painter.Once the boat was pulled in, the crew observedthat the stern of the lifeboat was damaged. Themaster ordered the boat pulled further forwardalongside number six hold and secured by theforward and aft painters. A jacobs ladder wasrigged and the mate went down to inspect theboat. The mate reported that the boat wasflooded and the after section of the canopy wasdamaged and set in. He also indicated that thedeck around the after hook mounting wasdamaged and that he felt that the boat could notbe lifted using this hook.

The master considered the problem ofrecovering the boat and decided to use the ship’sprovisions crane with a sling secured around thestern of the lifeboat. The lifeboat was hauled tothe starboard side of the ship and the ship’smain engine was prepared for manoeuvring to

make a leeway while recovering the boat. Fromabout 1720, attempts were made to recover thelifeboat. After some problems securing a linearound the stern of boat, the master rememberedthat the after hook is secured to the keel of theboat by a strong steel stay. He thought that itmay still be possible to lift the boat using thelifeboat davit attached to the after hook despitethe damage to the after deck. The matesubsequently checked and found that the stayattaching the hook to the keel of the boatappeared to be undamaged.

At 1829 the master decided to abandon theattempt to recover the lifeboat for the day as thelight was failing and the wind was becomingstronger. The lifeboat was subsequently re-secured, on fenders, on the port side of numbersix hold with painters fore and aft. A watch wasposted overnight to ensure that the lifeboat didnot sink.

The master called the crew at dawn the nextmorning and by 0600 he was manoeuvring theship to make a leeway to recover the lifeboat.The lifeboat was brought under its davit withthe mate aboard and he attached the falls to theauxiliary lifting shackles of the lifeboat’sforward and aft hook units. The davit winch wasthen used to lift the boat clear of the water.When the lifeboat had been hoisted to thestowed position, the gripes were resecured andan additional 20 mm synthetic rope was used tolash the boat into the davit cradle. The recoveryof the port lifeboat was completed by 0900 on 7 August.

8

Comment andanalysis

Lifeboat inspection

Initial observationsWashington Trader arrived alongside the AbbotPoint coal loader on the evening of 8 August2000. An investigator from the AustralianTransport Safety Bureau (ATSB) attended theship and conducted an initial inspection of theport lifeboat.

Externally, all of the damage to the lifeboat wasconfined to the stern (figure 7). The canopy atthe stern was extensively damaged with several

large cracks and small sections missing. It wasset-in in way of the after hatch which washanging ajar from its damaged frame. The afterdeck was extensively damaged with much of thedeck around the on-load hook broken up. Therewere several large cracks in the hull at thetransom. The rudder/propeller nozzle had beenbroken away from the rudder stock and keelblock, both of which had been deformed by theimpact.

Inside the boat the damage was limited to thefixtures on the after bulkhead with the mountingbrackets for the hook release operation unittwisted. The engine exhaust skin fitting on thetransom was also damaged.

A close examination of the on-load releasesystem was conducted on the morning of 9 August. The master and mate stated that theon-load release system had not been adjusted,

9

FIGURE 7:Damage to the stern of Washington Trader’s port lifeboat

reset, or handled in any way between the time ofthe incident and the investigator’s inspection.

The after hook The after hook was found to be in an un-trippedor closed position (figure 8). The cable resetlever on the after hook was in the reset positionalthough the end of the lever was approximately20 mm lower than the painted reference markson the hook side plates. The retaining latch onthe hook was damaged with both of its sidemembers bent outboard. The lug at the top ofthe outboard side member was missing. Therewere a number of witness marks (scores ordamage) on the latch and the suspension ringwhich were consistent with the ring having beenforced past the retaining latch with the hook in aclosed position.

The forward hook The forward hook was found to be in a trippedor open position (figure 9). The cable reset leverwas in a partially reset position with the end ofthe lever approximately 25 mm above the fullytripped position. There was no damage evident

to any part of the forward hook mechanismincluding the suspension ring or retaining latch.The position of the cable reset lever indicatedthat the forward hook may not have been fullyreset after the previous boat drill, or that thehook may have been forced open when the fullweight of the boat was transferred to theforward fall.

The operation unit

The release handle of the operation unit was inthe un-tripped position with a piece of twinefastened around the handle and the safety pin(figure 10). The pivoting safety pin was slightlybent adjacent to the eye at its pinned end andthus could not be seated in the notch of theopposite operation unit side plate. The afterhook cable connection arm was found to be inthe reset position with the forward hook cableconnection arm in a tripped position, ie. theafter arm was down and forward arm was up.The hydrostatic interlock lever was disengageddue to the tripped position of the forward cableconnection arm. There was no sign of damage toany component of the operation unit other than

10

FIGURE 8:The after hook as found

FIGURE 9:The forward hook as found

some deformation of the mounting bracketsattaching it to the after cabin bulkhead.

The pressure switch and alarm light control boxalso showed signs of impact damage with themetal box deformed and both indicator lightlenses broken.

Further examinationSeveral tests were performed on the on-loadrelease system after the initial inspection. Thetests were performed in-situ with the lifeboat inthe stowed position. Initially, both hooks werereset; the forward cable reset lever was found tobe quite stiff and a bar was required to move thelever up to its fully reset position, the after cablereset lever however, was found to be easilymoved under light finger pressure. Thehydrostatic interlock lever was then manuallytripped and the release handle moved to releasethe hooks. The system was found to operatesatisfactorily with both hooks tripping almostsimultaneously. This process of resetting andtripping the hooks was repeated several times.The forward cable reset lever remained stiffdespite the repeated movements. The forward

hook was tested next by moving the cable resetlever to the same partially reset position ‘asfound’ with the hook closed. A load was thenapplied to the hook with the intention ofbreaking the hook open. The hook remainedclosed under the limited load.

The whole system showed slight surfacecorrosion on all galvanised components andsigns of lubrication on all moving parts.

After completing the inspection of the portlifeboat, the starboard lifeboat on-load releasesystem was checked. Both release hooks werefound to be fully reset. The hydrostatic interlockwas found to be fully engaged with the safetypin correctly seated in its notch in the side plateof the operation unit. The presence of lashingtwine fastened around the release handle andsafety pin, in the same fashion as the portlifeboat, was noted.

Other evidence

The master and mate were interviewed andprovided detailed accounts of the incident.Copies of relevant ship’s documents wereobtained including the lifeboat davit and on-loadrelease instruction manuals, lifeboat certificatesof survey, and lifeboat maintenance records.Records of past drills and on-board safetytraining were also obtained and provided anindication of the crew’s level of training inrespect of the lifeboats.

On 15 August 2000, Nishi-Nippon F.R.P.Shipbuilding Company, the manufacturer of thelifeboat and on-load release system, werecontacted by the Australian Transport SafetyBureau. They were provided with details of theincident and were asked to supply the designspecifications of the on-load release system. Acopy of the design specifications wassubsequently provided to the investigation.

The incidentAlthough it was not possible to definitelyascertain what had occurred, considering the

11

FIGURE 10:Operating unit as found

statements of the witnesses and the physicalevidence presented in the form of the releasesystem as found, the sequence of events that ledto Washington Trader’s port lifeboat detachingfrom its falls and falling to the water appears tobe:

• The lifeboat tilted forward and/or bouncedduring lowering causing the after suspensionring to slide or bounce to the point of theafter hook and then force its way past theretaining latch.

• When the after fall detached, the stern of theboat fell to leave the boat suspended only bythe forward fall and hook.

• The full weight of the swinging lifeboat wasapplied to the forward hook, which openedshortly after this large transient load wasapplied, causing the lifeboat to fall stern-firstto the water.

Pac Monarch incidentOn 26 October, approximately 10 weeks afterthe Washington Trader incident, the port lifeboaton board the Bahamas registered bulk carrierPac Monarch released prematurely from itsfalls. The lifeboat fell to the water from a heightof 15 m with four crewmembers aboard. Therewere three fatalities and one serious injury. Theship was in the port of Vancouver at the timeand the Transportation Safety Board (TSB) ofCanada investigated the incident.

Pac Monarch is a sister ship to WashingtonTrader and was also constructed by SaseboHeavy Industries Company, in 2000. Thelifeboats and davits fitted to Pac Monarch areidentical to those on Washington Trader.

The circumstances of the Pac Monarch incidentare very similar to those on Washington Trader.The lifeboat was in the process of being loweredand, just after the davit hit the stops at decklevel, the after fall released from its hookfollowed by the forward fall a short time later.

The after fall retaining latch was found to be ina similar condition to that on Washington Traderindicating that the means by which the after fallhad released was probably very similar.

Unfortunately there were no eye-witnesses tothe hooks releasing on Pac Monarch, but thelone survivor recalls a ‘different kind of jerkafter the davits hit the stops’. The TSB tookpossession of the lifeboat and conducted anumber of tests on the boat and release system.These included a load test of the on-load releasehooks to ascertain how the forward hookresponded when subjected to the load of theswinging boat.

The TSB testing also established, from thedamage to Pac Monarch’s after hook retaininglatch and suspension ring, that the ring exited itshook at an angle between 45 and 60 degrees.This would indicate, as in the case ofWashington Trader, that the lifeboat either: tiltedforward and the suspension ring slid off its hookor bounced in such a fashion that the suspensionring was at an angle at the instant it was forcedpast the latch.

An investigator from the ATSB revisitedWashington Trader at the Hay Point coal loaderin Mackay, Queensland on its next voyage toAustralia. A series of tests were performed onthe port lifeboat (which had been replaced sincethe incident) in an effort to understand how theafter suspension ring detached from its hook.The tests involved repeated lowering of thelifeboat from the davit head under the similarconditions to those which existed at the time ofthe incident. The tests were video-taped for laterexamination.

The after hookAt the time of the incident on 6 August, themaster said that he saw the after end of the boatswing and then jerk, twice, before the after falldetached. The lifeboat was approximatelyhalfway to the deck level at the time. There is

12

little doubt that it was these motions of the boatthat caused the after suspension ring to bounceor slide and exit the closed hook. For thelifeboat to have swung and jerked in this fashionthere must have been abnormal interaction withthe davit during the lowering process.

A number of hypothesis were formed about thecause of the lifeboat’s observed motions at thetime of the incident. These included:

• possible interaction between the hooks at thehead of each cradle and the fall suspensionblocks,

• the lifeboat impacting on some part of thedavit or the ship during the lowering,

• a momentary ‘hang up’ caused by some partof the lifeboat or davit fouling,

• or the impact when the cradles reached theirstops at deck level.

These hypothesis were tested at Hay Point and itwas concluded that the most likely explanationfor the observed motion was related to theaction of the automatic gripe system.

Automatic gripesWashington Trader’s lifeboat davits have beendesigned to facilitate the rapid launching of astowed boat. Initial preparation of the davit forlaunching a boat consists of removing thelocking pin from the davit winch brake lever,then disengaging the two cradle stoppers locatedon the embarkation platform.

Release of the gripes is automatic, and occurswhen the lashing levers which connect the topend of each gripe to the davit cradle, are allowedto turn out as the cradle moves away from thedavit frame. This progressively slackens thegripe in the initial stage of lowering, until thelashing levers swing free and the gripes are fullyreleased. If the forward and after gripes are notreleased by the lashing levers simultaneously, or

a gripe wire fouls on the boat as it falls awayfrom its lashing lever, there is a possibility thatthe one end of the lifeboat may be ‘hung up’momentarily.

The lifeboat testing conducted at Hay Pointshowed that the gripe wires had a propensity to‘take a turn’ around the lifeboat’s bollardsduring lowering. During several of the tests thetop of the wire, fitted with a heavy steel ring,fell rapidly after being released from the lashinglever, swinging around the point where the wirewas led over the lifeboat bollard. It did notcomplete a full turn around the bollard duringthe tests, and thus did not foul and hang up thestern of the boat, but the potential for such anevent to occur was clearly demonstrated. If theafter gripe wire had fouled in this fashion at thetime of the incident the after end of the boatwould have been hung up and allowed the aftersuspension ring to slide/bounce to the openingof the hook. When the weight of the boat cameback on the fall, the ring would have beenpulled through the retaining latch. Such a hang-up would account for the master’s observation ofthe aft end of the boat seeming to swing andjerk when the lifeboat was approximatelyhalfway to the deck at the time of the incident.

The forward hookThe forward hook, as found after the incident,was open with the painted end of the cable resetlever approximately 25 mm above the fullytripped position. Full travel of the lever from thetripped position to the fully reset positioninvolves a movement of approximately 140 mm.Providing the forward cable reset lever had notbeen moved after the incident, the position ofthe lever indicates two possibilities. Either thehook had not been correctly reset after theprevious lifeboat drill and/or the transient loadon the hook at the time of the incident was largeenough to force the hook open, moving thecable reset lever to the partially reset position inthe process.

13

The International Maritime Organization’sSafety of Life at Sea Convention (SOLAS)Chapter III, Regulation 41, section 7.6.4requires, in respect of on-load releasemechanisms, that:

The mechanism shall be designed with a factor ofsafety of 6 based on the ultimate strength of thematerials used, assuming the mass of the lifeboatis equally distributed between the falls.

The design calculations provided by NishiNippon F.R.P Shipbuilding Company indicatedthat the safety factor for each component of theon-load release system was in excess of six. Aspart of the Nippon Kaiji Kyokai type approvalprocess a prototype of the mechanism was alsotested and the safety factor of six verified underload.

An analysis of the design of the on-load releasemechanism revealed that it was unlikely that thecable reset lever could have been moved, fromthe fully reset position to the position foundafter the incident, by the full weight of theswinging lifeboat. The direction of the forceapplied on the cable reset lever quadrant by theblocker does not induce a tripping motion in thelever regardless of its magnitude if the point ofcontact is in alignment with the centre line ofthe cable reset lever pin. In addition, thehydrostatic interlock effectively bars the travelof the operating cable, and thus the cable resetlever, when the system is fully reset. The TSBconducted static load tests on the hooksrecovered from Pac Monarch’s lifeboat andthese tests also showed that the hook would nothave opened under load from the fully resetposition.

In submission Nishi-Nippon F.R.P. Shipbuildingstated:

We have reviewed the reset procedure for therelease system that should be followed during therecovery operation for the released boat, and, as aresult, we insist that the hook will notaccidentally open unless the release system isactuated, provided that the interlock system has

been inspected as instructed by the lifeboatoperating manual.

In light of this analysis, it is likely that the hookwas not fully reset the previous time the portlifeboat was lowered to the water and the releasesystem actuated on June 7, 2000. There mayhave been a number of reasons why the hookwas not properly reset on this occasion.

Resetting each hook unit is, because of thelimited access to the forward and aft decks, aone-man operation. In the final stage ofresetting, the cable reset lever must be moved,without any mechanical advantage assisting,against any resistance in the length of theoperating cable and the operation and hook unitmechanisms. Washington Trader’s forward hookunit cable reset lever was found to be very stiff.When resetting the hook after the incident,considerable force was required to move thelever through its full range of movement.Resetting of the lever was made even moredifficult because the bar at the bottom of theretainer latch obstructs access to the lever in onesection of travel. The force opposing themovement of the lever was due mainly tofriction in the operating cable. The operatingcables are fitted with oil seals at each end, andso it is unlikely that the cable friction was aresult of the boat’s immersion in seawater.Further evidence of past difficulty resetting theforward hook was found in the form of a short,bent, extension pipe in the bottom of thelifeboat. The investigator noted that this pipewas just the right size and shape to fit over theend of the cable reset lever to assist in resettingit. This difficulty in moving the cable reset levermay have been why it appears that the forwardhook was not fully reset during the previouslifeboat drill.

Basic on-site tests showed that the forward hookwould support weight when the cable reset leverwas only partially reset with the implication ofthe design being that the full weight of the boatcould be taken on a partially reset hook.However, with the cable reset lever in a partially

14

reset position, the hook blocker is held closedby a relatively small area of contact with thequadrant on the cable reset lever. With this typeof limited contact it is possible that a highdynamic load placed on the hook would forcethe hook blocker to the open position movingthe cable reset lever downward in the process.

On-load release design The design of the NS-Hook on-load releasesystem complies with the requirements ofSOLAS and has been approved by a number ofclassification societies including Bureau Veritasand Nippon Kaiji Kyokai (Class NK).Nevertheless, the design has some aspects whichmay have contributed to the incident on boardWashington Trader. Some aspects of the hookunits, hydrostatic interlock and the systemindicator lights exhibit design inadequacies,difficulties or ambiguities for the lifeboatoperators.

Hook orientation Both the forward and after on-load hook releaseunits fitted to Washington Trader’s lifeboats areidentical and are mounted so that each hookthroat faces aft. If Washington Trader’s afterhook unit had been mounted with the throat ofthe hook facing forward, there would have beensignificantly less chance that the suspensionring would have been freed from the hook as theboat tilted forward and/or bounced when beinglowered on August 6. In addition, operating thehook would be simplified, as in the currentorientation, the cable reset lever is located at theafter end of the hook unit and must bemanipulated by the operator while almostunseen.

There is no standard with regard to theorientation of lifeboat release hooks. Variouslifeboat manufacturers fit their on-load releasehooks in various ways. Only both hooksmounted with their openings facing inboardprovides for maximum fall ‘failure safety’. Withthis hook orientation, if one fall shouldprematurely release, the remaining fall

suspension ring will still be positively located inthe throat of its hook to take the weight of thevertical boat. With other methods of mounting,there will be cases where the remaining fall willbe free to pull through the retaining latch or linkstopper, and off the hook, as the boat falls to thevertical.

SOLAS requires that lifeboats fitted to vesselsover 20 000 tonnes must be capable of beinglaunched with the ship making 5 knots headwayin calm water. The design rationale for fittingthe hooks with both openings facing aft, in thecase of Washington Trader, appears to be toallow both hooks to take the weight of thelifeboat if it is being towed astern of the davitwhen the ship is making headway. A secondaryconsideration may be to allow the suspensionrings to move freely out and away from the hookunits, when the hooks are released with the boatmaking forward way, minimising the risk offouling the falls during release.

In their submission Nishi-Nippon F.R.PShipbuilding stated that they had orientated theafter hook with its opening aft ‘considering thelaunching process at 5 kt speed.’

The retaining latchThe design of the retaining latch on WashingtonTrader’s after on-load release hook wassignificant in this incident. The damage to thelatch and the marks on the suspension ring showthat the ring was forced past the latch to detachthe after fall. A latch of stronger design fitted toclose the opening of what is a relatively shallowhook may have prevented the ring from comingoff the hook.

Retaining latches, or link stoppers, are fitted toon-load release hooks with the intention oftrapping the fall suspension rings inside thehooks. This is necessary to ensure that the ringsare retained in the hooks when the boat iswaterborne either prior to releasing the hooks oron completion of a lifeboat exercise, when thesuspension rings are placed in the reset hooksprior to lifting the boat. Under normal circum-

15

stances when the boat is suspended from thefalls, there will be no weight on the latches asthe suspension rings are located well within thethroat of the hook.

On Washington Trader’s lifeboats, the relativelylight construction of the hook latches wouldindicate that they are only designed to guard thesuspension rings when they are loose in thehook. They are not designed to restrain asuspension ring under load. When the aftersuspension ring slid or jumped to the point ofthe hook during the incident, the lightlyconstructed latch would have presented minimalresistance to the heavily loaded suspension ring(of 20 mm cross-sectional diameter) as it exitedthe hook.

As fitted, no part of the latch makes contactwith the hook. There is a clearance of more than10 mm between the latch lugs and the point ofthe hook and thus the hook opening is notcompletely barred by the latch. In addition, theplacement of the latch, so that its side membersare well outside the line of the point of theclosed hook, means the side clearance betweenthe side member of the latch and the hook isapproximately 15 mm. The combination of theseclearances means that a bouncing or slidingsuspension ring, at an angle, may be passedthrough the clearance between the latch and thepoint of the hook with only the light sidemember on one side providing any significantrestraint. The damage to Washington Trader’s aftlatch would indicate that this was probably whatoccurred in the incident. If the latch had beenof stronger construction with a continuoushorizontal top member with a smaller clearancefrom the tip of the hook, or bearing on theinside the hook, it would have prevented theloaded suspension ring from exiting the hook.

In their submission Nishi-Nippon F.R.PShipbuilding stated with respect to the retaininglatch:

Note that the latch is not designed to bear theworking load on the hook but is intended toprevent the non-loaded suspension ring from

being disengaged from the hook. To be able toavoid a rare accidental situation, your suggestion(increased mechanical strength, smallerclearance) will be helpful.

Hydrostatic interlock

An aspect of Washington Trader’s on-loadrelease design that raises concern is theoperation of the hydrostatic interlock. Theinterlock is rendered useless if both hookmechanisms are not fully reset. In addition, theinterlock does not engage to lock the hooks inthe fully reset position until the lifeboat is liftedclear of the water and the water pressure switchde-energises the solenoid.

The action of the interlock, when engaged, is tophysically block the upward tripping travel ofthe cable connection arms and thereby also lockthe release handle. When resetting the hooks,both cable connection arms must be in the fullyreset position to allow the interlock to move tothe blocking position when the boat is raisedclear of the water. If either of the cableconnection arms is not fully reset, or theoperating cables are out of adjustment, then theinterlock will be prevented from moving to theblocking position and the thus the release handlewill still be free to move to trip the hooks. Onlya small clearance between the interlock and thecable connection arms is allowed in the designof the release mechanism for systemmaladjustment or ‘lost motion’ in the operatingcables or hook mechanisms.

During the time between resetting the hooks andraising the lifeboat from the water there is adanger that the cable reset levers (unrestrainedby the action of the interlock on the cableconnection arms) will fall from the fully resetposition. With no load on the hooks, there is noforce transmitted by the blockers to thequadrants on the cable reset levers to preventthem from moving downward. At this time theonly thing restraining the levers is the friction inthe operating cable. Once the boat is lifted clearof the water, if one of the cable reset levers hasmoved down, the interlock will not engage.

16

If the hydrostatic interlock is not fully engaged,the on-load release system may be tripped usingthe release handle at any time with neither ofthe hook mechanisms in a ‘locked’ state. In thiscondition the lifeboat is prone to accidentalrelease.

SOLAS 1997, Chapter III, 41, 7.6.2.2 contains aclarification with regard to the operation oflifeboat on-load release system interlocks:

To prevent an accidental release during recoveryof the boat, the mechanical protection (interlock)should only engage once the release mechanismis properly and completely reset.

The interlock system fitted to WashingtonTrader’s lifeboat on-load release system seemsto comply with the SOLAS requirement;however it only engages when the system isfully reset and the lifeboat is lifted clear of thewater.

The resetting instructions require that a visualcheck of the cable connection arms and theposition of the interlock is performed to ensurethat the interlock is engaged prior to lifting theboat. In addition, the status of the indicatorlights will indicate whether the hydrostaticinterlock is engaged. However, for an inexpe-rienced crew, the complexity of resetting thesystem with multiple checks may be confusing.

Indicator lightsThe red and green lights on Washington Trader’slifeboat release system pressure switch andalarm light control box are multi-indicative.Each lamp indicates to the operator twodifferent things depending on the systemcondition and lifeboat operation beingperformed. When the lifeboat is being launchedthe red light indicates that the boat is not yetwaterborne and that it is unsafe to operate therelease lever. When the hooks have been trippedwith the lifeboat floating, the red light indicatesthat the hooks are tripped. When launching thelifeboat the green light indicates that the boat iswaterborne so the release lever may be operated.It also indicates to the operator, once the hooks

have been reset, that it is safe to lift the boatfrom the water. This multiple indication functionof the lights may have been a source ofconfusion for the crew of Washington Trader.

Accepted safe practice when designing criticalcontrol and instrumentation indication systemsis to have a single light indicating a singlecondition. Multi-indication is bad design from ahuman performance perspective as it relies onthe operator having not only a completeunderstanding of the operation of the system butalso thorough situational awareness, in this case,when the crew are conducting a lifeboatoperation. For an inexperienced crew recoveringa lifeboat in possibly adverse sea conditions itwould be easy to confuse the meaning of theindicator lights.

During the previous boat drill, the releasesystem indicator lights would have shown thatthe forward fall was not fully reset– if this wasindeed the case. The red indicator light shouldhave remained illuminated in the time betweenresetting the hooks and raising the boat, ratherthan the green light coming on to indicate thatthe hooks had been properly reset and that itwas safe to hoist. Whether the crew checked theindicator lights, or misinterpreted their meaning,is a matter for conjecture. After the boat hadcleared the water, the red light should have beenlit in any case, as a result of the open waterpressure switch, and from this time, the crewwould have had no indication from the lightsthat the forward hook was not fully reset.

Operation and Maintenance ofthe ‘NS-Hook’ release systemThe manufacturer’s operation manual for theNS-Hook release system contains detailedinstructions for operating and maintaining therelease system. The operating instructionsprovide adequate explanation for the operationof the system together with appropriatewarnings. Included in the instruction manual isa detailed periodic inspection and maintenanceschedule.

17

The operating instructions include detailedrelease and resetting instructions including aflow chart for releasing the hooks. A copy of theflow chart was mounted on the bulkhead insidethe lifeboat, however there were no instructionsinside the boat for resetting the release system.There were no additional ship-specific operatingprocedures for the on-load release system and,when asked, the master referred solely to themanufacturer’s instruction manual.

The operating manual contains maintenanceinstructions for the release system and theshipboard records indicate that the lifeboatrelease gear was being checked as permanufacturer’s instructions. The ship’s recordswere in the form of completed checklists, a wallchart showing weekly safety equipment checks,and in the ship’s maintenance recording system.

Included in the weekly checks on the releasesystem is a check of the cable reset levers oneach lifeboat hook unit. The ship’s recordsindicate that these checks were performed oneach lifeboat on the days of 10, 17, 24, 29 June,and 8, 15, 22, 30 July. That is, the status of therelease system on the port lifeboat including theposition of the forward cable reset lever waschecked, according to the maintenance records,on eight separate occasions since the last timethe system was actuated on June 7. Given thatthe forward cable reset lever was, in fact, notreset properly on June 7, and that themaintenance records are accurate, then thecrew’s subsequent checks failed to detect theincorrect position of the cable reset lever. This

may have been the result of a poor knowledge ofthe system, or carelessness on the part of thecrew performing the checks.

Crew TrainingWashington Trader’s safety management systemincludes a standard form for recording thetraining and education of the crew. Records ofthe drills conducted since the delivery of theship in January 2000, and the training receivedby the crew on these occasions, were recordedin this format.

Since the delivery of the ship, there had been nochange of the crew. Any training conducted overthe previous eight months had involved the crewon board at the time of the incident. The boatdrills conducted since January consisted of six‘abandon ship’ drills, where both boats had beenlowered to deck level, and four lifeboat drills,where one boat had been lowered to the waterand released. The port and starboard lifeboatshad been released from their hooks and takenaway twice each by their respective crews priorto the incident.

The crew had received instruction on theoperation of on-load release systems duringsome of the ‘abandon ship’ drills but the portlifeboat crew had actually reset the system ononly one occasion prior to the lifeboat drillconducted on 7 June. It is likely that the crewmember resetting the forward hook on thisoccasion was not completely conversant withthe operation, given such limited hands-onexperience.

18

Conclusions These conclusions identify the different factorscontributing to the incident and should not beread as apportioning blame or liability to anyparticular individual or organisation.

Based on the evidence available, the followingfactors are considered to have contributed to theincident involving Washington Trader’s portlifeboat on 6 August 2000:

1. The lifeboat bounced and/or tilted forwardduring lowering, probably as a result of theafter gripe wire fouling on the boat, whichled to the after suspension ring sliding and/orbouncing out of the closed hook.

2. When the after fall was released the lifeboat’sstern fell with the boat suspended only by theforward fall. Almost immediately, theforward on-load release hook opened andreleased the forward fall.

3. The forward hook was not fully reset whenthe previous lifeboat drill was conducted on7 June 2000.

4. Lifeboat checks after 7 June 2000 did notdetect that the hook was only partially reset.

5. Reasons why the forward hook was not resetfully during the previous drill include:

• the relative complexity of resetting thesystem,

• the cable reset lever was stiff to operate,

• poor knowledge of the operation of theon-load release system,

• ambiguous system indicator lights.

6. The retaining latch on the after hook waspoorly designed as it was neither ofsufficient strength nor fitted in a way whichwould provide an adequate safeguard toprevent the suspension ring exiting the hookunder load.

7. The orientation of the after hook increasedthe likelihood of such an incident occurringand in addition made resetting the hookdifficult.

8. The design of the hydrostatic interlock wasinadequate as it did not engage once thehooks were in a fully reset position prior tothe lifeboat being lifted from the water. Itwas also rendered completely ineffective ifeither cable connection arm was not in thefully reset position.

19

20

FIGURE 11:Washington Trader: Events and causal factors chart

Cond

ition

s

Even

ts

Inci

dent

It is recommended that:

1. Nishi-Nippon F.R.P Shipbuilding Companyreview the design and construction of theirNS-Hook release system, in light of thesimilar incidents aboard Washington Traderand Pac monarch, with a view to minimisingthe risk of such an incident occuring again.

2. All ship operators should ensure thatshipboard safety management system classifylifeboat on-load release operations as‘critical’ with appropriate ship-specificoperating and maintenance procedures.

3. All ship operators should ensure thatpersonnel responsible for operating andmaintaining on-load release equipment areprovided with type-specific training.

21

Recommendations

22

SubmissionsUnder sub-regulation 16(3) of the Navigation(Marine Casualty) Regulations, if a report, orpart of a report, relates to a person’s affairs to amaterial extent, the Inspector must, if it isreasonable to do so, give that person a copy ofthe report or the relevant part of the report. Sub-regulation 16(4) provides that such a personmay provide written comments or informationrelating to the report.

The final draft of the report was sent toWashington Trader’s master at the time of theincident, Nishi-Nippon F.R.P. ShipbuildingCompany, Victoria Ship Management, BureauVeritas, the Transportation Safety Board,Canada, and the Australian Maritime SafetyAuthority.

Submissions were received from the president ofNishi-Nippon F.R.P. Shipbuilding Company andthe manager of Victoria Ship Management’smarine department. The Transportation SafetyBoard of Canada provided a copy of the draft oftheir Pac Monarch investigation report. Somefactual material has been incorporated in thisreport which was drawn from the Pac Monarchinvestigation.

The following are extracts from the submissionsfrom Nishi-Nippon F.R.P. ShipbuildingCompany and Victoria Ship Management whichare not included in the body of the report.

With regard to the conclusions of the reportNishi-Nippon F.R.P. Shipbuilding Companysubmitted:

Conclusions

5. Reasons

• Relative complexity

The resetting procedure will be completedafter completion of three consecutive steps onthe hook side and the final visual check on thelever side.

• Stiff to operate

Stiff cable that does not allow smoothoperation needs a maintenance work.

• Ambiguous

The indication with the lamp is intended toindicate whether or not the boat is afloat onthe water.

In addition, the resetting procedure isindicated based on the detected motion of armin the actuator (this arm is synchronized withthe reset lever of the hook). Therefore, theindication by the lamp does not mean thecompletion of resetting operation. Rememberthat the resetting process is complete onlyafter the status of the interlock is visuallychecked.

6. Retaining latch

Note that the latch is not designed to bear theworking load on the hook but is intended toprevent the non-loaded suspension ring frombeing disengaged from the hook.

7. Orientation of the after hook

Unlike the operation of the forward hook, theafter hook necessitates a work with the upperbody of the operator being exposed out of thehatch. This posture, however, is not demandingfor the operator - the after hook can be reliablyand readily reset this hook.

Supported by this fact, we have oriented thishook to the stern direction considering thelaunching process at 5 kt speed.

8. Design of the hydrostatic pressure detector

The hydrostatic pressure detector is intended toassist the operator in a sequence beginning withlowering and ending with releasing of thelifeboat. Note that the resetting sequence (whichbegins with recovering and ends with stowing ofthe lifeboat) is complete only after the operatorhas verified the interlock is in the engaged statein the final check process.

Victoria Ship Management submitted:

We impartially acknowledged your analysis, andconclusions mentioned in page 23 of your draft,with emphasis on items Nos. 3 – 8, the plausiblefactors that may have contributed to the statedincident.

23

Furthermore, we would like to hand on ourobservations that may be weighted as likelycontributing components leading to lifeboat fallaccidents:

a) The maker’s design on the releasing/re-settingmechanism is rather complicated. Thereshould have a modification on the maker’sview to devise components and operation inuncomplicated, sturdy and dependable mode -simple to the understanding of the officers andcrewmembers involved.

b) On the other hand, the ship personnel shouldmaintain fixed focus on safety during theactual lowering operation, and to be totallyaware of the consequence involved when anoversight is committed. A constant review ofthe procedures (disengaging and engaging) tothe whole team prior to exercise is essential inorder to avert mistakes.

c) These are lined in parallel to draft stipulatedin page 17, paragraph 2:

1) Gripe wires have a tendency to entangle or tomess up around the bollard or to any protrudingobject adjacent to the forward and after hookcomponents that may suspend the forward ofafter body of lifeboat, causing imbalance ormisalignment of davit while in progress oflifeboat lowering.

2) Unequal tension of wires when tightening theforward and after lashings duringstowing/securing of lifeboat into davit, may affectthe unlashing operation prior boat lowering, andmay contribute to the entanglement, if leftunchecked.

d) Improper procedure of stowing the lifeboat.The boat keel should be rested first to itsresting bar or plate before tightening thelashing wires. This means the whole weight ofthe boat rests on the bar, not on the boat falls.This may affect the smooth operation duringlowering of the boat.

e) Uneven rigging or adjustment of boat falleither forward or aft is another contributingfactor.

f) Abrupt release of brake lever during initiallowering resulting extreme twitch to the boat.This should be practiced and carefully tendedby the assigned person.

g) Uneven swing out of forward and aft davitarms, or delayed swing out of one of the arms(at the time of initial lowering of the boat)caused by temporary freezing of one of thedavit arm due to poor greasing maintenance,may cause excessive jerk to the boat whenthat is arm is suddenly freed.

24

Washington TraderName Washington Trader

IMO No. 9211602

Flag Philippines

Classification Society Bureau Veritas

Vessel type Bulk carrier

Owner St Vincent Shipping Co. Ltd

Year of build 1999

Builder Sasebo Heavy Industries Co. Ltd, Japan

Gross tonnage 38 928

Summer deadweight 74 228 tonnes

Length overall 225.00 m

Breadth, moulded 32.20 m

Draught (summer) 13.821 m

Engine B&W 6S60MC

Engine power 8 827 kW

Service speed 14.5 knots

Crew 20 (Filipino)

25

26

ww

w.a

tsb.

gov.

au18

00 6

21 3

72

Independent iinvestigation into the lifeboat incident on boardthe Philippine flag bulk carrier W

ashington Trader, at Abbot Point, Queensland on 6 August 2001

ISSN 1447-087XISBN 1 877071 21 8

Washington T. 12 .2002