At 2100 on Monday, 6 August, the small general cargo vessel ANL Purpose was at the position 18 23.6' S, 152 54.3' E making good 11 knots on a course of 174 (T) on passage from Lae, New Guinea, to Melbourne. The weather was fine. The nearest land was Marion Reef, 45 nautical miles to the south west and Lihou Reef, some 76 nautical miles to the north west. The master was on watch on the bridge and the chief engineer was in his cabin, when an engine alarm annunciated.

Upon investigating, the chief engineer heard loud noises coming from the main engine, which he promptly stopped. After opening the port crankcase door on no. 4 unit, he saw that both the piston and the cylinder liner had shattered into numerous small pieces and the engine sump was full of debris. There was also significant damage to the engine block. The chief engineer realised that it would not be possible for the ship's staff to carry out repairs sufficient to restart the main engine, even on five cylinders and, at 2215, he advised the Master accordingly.

The master informed the vessel's owner and AusSAR (the Australian Search and Rescue Co-ordination Centre) of the vessel's predicament. ANL Purpose was drifting, in a mainly north-westerly direction, at a speed of approximately 1.3 knots. It was, however, in no immediate danger.

At 1654 on 8 August, the Townsville-based tug Giru made a rendezvous with ANL Purpose which had, by that time, drifted to a position of 1759.1' S, 15215.1' E, and passed a tow to the ship. The tow, at a speed of 7 to 71/2knots, proceeded towards Brisbane in good weather with light south-easterly winds.

On 10 August, in the shelter of Saumarez Reef, the tow was taken over by the Brisbane-based tug Bulimba. The good weather held for the remainder of the tow to Brisbane and, on 12 August, ANL Purpose arrived at the repair wharf of Forgacs Cairncross shipyard in the Brisbane River.

The investigation concluded that:

• The proximate cause of the failure was the partial seizure and consequent break-up of the piston in no. 4 unit. The partial seizure was brought on by an obstruction to piston cooling oil flow caused by the axial movement of the bottom-end bearing shells, which may have been incorrectly fitted.
• The piston in no.4 unit was suffering from significant fatigue cracking, as were those in nos.1 and 2 units, and the additional loading from the partial seizure was sufficient to cause the break-up of the piston.
• The pistons were those originally fitted sixteen years earlier and there was no evidence that they had ever been replaced or crack-detected during that time. The engine manufacturers had issued service bulletins relating to the maintenance and crackdetecting of the later two-piece type of pistons, but not for the original, one-piece type.
• The standard of maintenance records kept aboard the vessel was less than adequate.

The report recommends that the engine manufacturer consider issuing a service bulletin covering the crack-detecting of any of the earlier, one-piece, pistons which may still be in service.

Conclusions

• The ANL Purpose main engine failure occurred due to the partial seizure and consequent fracture and break-up of the piston from the number-four cylinder assembly.
• First indication of engine failure was the detection of elevated crankcase oil mist levels due to increased friction within the cylinder.
• The presence of pre-existing fatigue cracking within the body of the number-four piston had contributed to the failure.
• Two other pistons from the engine were also cracked in a similar manner.
• In-service cracking of the pistons was a known design issue and the engine manufacturer had published enhanced inspection requirements for pistons that have accrued more that 24,000 operating hours. A later service letter published service time limits for the main components of the VASA 32 engines. At the time of the failure, the pistons from the ANL Purpose had a service time well in excess of the published limits for piston replacement.
• The connecting rod big-end bearing shells had migrated forward of their normal positions and had sustained damage as a result of contact with the crankshaft journal radius.
• Movement of the bearing shells had partially obstructed the oil flow pathway between crankshaft and connecting rod / piston. At a constant feed pressure, this would reduce the total volume of oil flowing through the number-four assembly.
• Control of the piston crown operating temperatures requires that a positive flow of oil be maintained. Any reduction of oil flow volume would be expected to result in an increase in piston temperatures.
• An increase in piston temperatures will result in the physical expansion of the piston body, reducing the bore clearances and increasing the risk of binding and seizure. Evidence of binding and partial seizure was found on fragments of the piston skirt and cylinder bore from the failed engine, although it could not be ascertained at what stage of the failure the damage was sustained.
• Binding within the cylinder bores will elevate the loads being transmitted through the piston body. Under these conditions, the presence of cracking or other defects may predispose the component to failure in the manner observed within the main engine of ANL Purpose.