Catastrophe in the balance
Preventing critical equipment from failing while at sea can be a matter of life and death.
According to the latest figures published by the UK Government's Marine Accident Investigation Branch (MAIB), approximately 23 per cent of all accidents involving merchant vessels with a gross tonnage of over 100gt were due to machinery failure; while for passenger and other commercial vessels with a gross tonnage of under 100gt this figure rose to almost 35 per cent.
The MAIB definition of what constitutes a marine accident includes, in extreme cases, loss of a vessel, injury to crew or passengers, and loss of life. The definition of machinery failure includes many different factors, from burst pressure pipes and hoses to the movement of cargo and the collapse of hoists or hatch covers.
The key issue, however, is what causes machinery to fail in the first place. Again, the reasons can be varied, but as the Japanese Ministry of Land, Transport and Tourism found in surveys of the maritime industry, factors such as insufficient maintenance, inspection or handling of equipment were the cause of around 60 per cent of marine machinery failures. To this figure can be added problems caused both by incorrect lubrication and poor machine installation, and, in particular, misalignment and balancing of shafts.
Failure of operating machinery often arises from components where there is relative movement between loaded surfaces; typically, these components include couplings, gears, bearings, pistons and seals, where the combination of movement and load creates an environment within which damage or wear can occur.
Examples include marine propulsion systems, which must be completely dependable when a vessel is at sea. Gearboxes are a crucial part of a propulsion system and are subjected to particularly high mechanical loads. Similarly, at a component level, products such as bearings have to withstand extreme loads for sustained periods. For example, the thrust and radial bearings used in the azimuthing thrusters found in tugs or harbour vessels are typically subject to constant 85kN radial loads, while at full forward speeds a load of around 170kN can be applied for up to 75 per cent of the time the thrusters are in operation.
It should also be noted that the shaft bearings and bearing housings used in applications such as thrusters have to tolerate extremes of temperature, as they are normally surrounded by cold water but are then rapidly brought up to full operating speed creating both rapid expansion and the potential for thermal shock.
Depending on the type and size of vessel, there can, in addition to the main engine and propulsion systems, be between 75 and 150 auxiliary machines on board, ranging from cargo pumps and engine room fans, to refrigeration plants and turbochargers. In each case there are moving and wearing parts, many of which are subject to high loads and have the potential to fail, either slowly over time or catastrophically.
Fortunately, it is possible to detect these problems as they begin to develop and while the machinery is still within its normal operating tolerances. The principal techniques used are vibration and temperature monitoring, together with analysis of lubricating oils to detect changes in oil condition caused by water and fuel contamination, or by metallic content or oxidisation.
The latest sensors or accelerometers used for vibration monitoring, such as those supplied by SKF, generally use piezoelectric or piezoceramic technology. This provides a robust and reliable method of measuring both high and low frequencies, with low hysteresis characteristics and excellent levels of accuracy over a wide temperature range; additionally, these sensors can be packaged in a compact stainless steel sensor housing sealed against moisture, dust, oils and other contaminants.
In simple terms, these devices function by using a piezoelectric crystal element bonded to a mass so that when the sensor is subject to an accelerative force the mass compresses the crystal; this causes it to produce an electrical signal that is proportional to the level of force applied. The signal is then amplified and conditioned using in-built electronics that create an output signal that is suitable for use by higher level data acquisition or control systems.
Accelerometers are generally mounted in a number of key locations on the equipment to be monitored with output data either being read periodically using sophisticated hand-held data collectors, for immediate analysis or subsequent downloading, or being routed via switch boxes to a centralised or higher level system for continuous monitoring.
Importance of people
The technology described above has been in use for some time in on-shore applications where it is accepted that investment in condition monitoring systems can reduce the need for routine maintenance, extend the operating life of equipment
and cut overall operating costs. Although the shipping industry has been slower to adopt modern condition monitoring techniques, it has now begun to implement much of the new technology.
Just as importantly, there is a growing recognition among ship operators that to benefit from modern condition monitoring technology and techniques, they must be used and implemented correctly with effective and ongoing training. This is especially important in the marine sector where, unlike many industries where plant or factory operations remain largely static, operating conditions are subject to a far wider range of variables; data readings taken while the vessel is under ballast conditions may be different from those gathered during loaded passage, while readings taken in port will almost certainly be different from those taken while the vessel is at sea, especially in heavy weather.
Crew training is, however, just one aspect of an effective condition monitoring programme. Equally crucial is the understanding that condition monitoring can eliminate much of the often unnecessary maintenance that is routinely carried out by ships' crews as part of their duties.
Much of this work involves dismantling machinery to check for wear or internal damage. In many instances, however, equipment is reassembled without change or with an automatic replacement of components even though the existing parts were still operating within tolerance and without wear. Interestingly, it has been shown that many machine failures occur shortly after maintenance has been carried out due to incorrect reassembly or other errors.
Minimising maintenance requirements through effective condition monitoring can therefore reduce failures and extend the operating life of machinery. Ultimately, this can improve the profitability of both vessel owners and managers. Indeed, this is being acknowledged by many of the industry bodies and insurers who are increasingly accepting vibration trend analyses, performance monitoring and evidence of other non-intrusive condition monitoring methods as the basis for inspection
The move from a traditional time-based maintenance regime to one where effective data collection and analysis can be used to predict when and where machine maintenance will be required also opens up the potential for new strategies that can further improve equipment performance and reduce operating costs.
These strategies are again based on techniques developed by SKF in on-shore applications and use a broader approach to the challenge of on-board condition monitoring, for both individual vessels and for marine fleets. A solution of this type, which takes into account operational criteria, allows maintenance issues to be assessed, analysed and managed simultaneously. In doing so, managers are able to minimise costs through maximising the efficiency of maintenance procedures, while achieving consistently high levels of machine uptime and availability.
In essence, the approach championed by companies such as SKF for a condition-based maintenance programme can be broken down into: assessment and mapping, data collection, analysis, correction, verification, and improvement.
The first stage, assessment and mapping, begins with a detailed mapping of each ship's critical machinery and the establishment of measuring points and trend values. It is important to include operational staff such as chief and ship engineers in these discussions, as they will have the hands-on experience and knowledge of failed machines and items that need planned maintenance. The critical nature of each machine is typically determined by looking at the complexity of its maintenance requirements and the impact on the operation of the vessel and the business if it should fail.
Data collection is then carried out using the techniques described previously, with a combination of portable analysers, fixed online systems and remote wireless or satellite connections being used to gather and communicate critical vessel information.
Once machine reliability data has been collected, it needs to be analysed. This can be done by a ship's engineers or, increasingly, be handled by a remote monitoring centre that is able to analyse data in real-time and then advise the ship's crew if remedial action needs to be taken.
Work requests can be submitted via a computerised maintenance management system and be combined with other pre-determined planned and corrective maintenance activities. In most instances where machinery is operating within normal limits, the role of the remote technicians is to produce customised reports detailing identified potential problems, recommending appropriate actions, and facilitating the scheduling of maintenance procedures.
There are a number of powerful software tools available to improve data collection and analysis, and which can be run on both hand-held and centralised computer systems. For example, SKF Results Reporter is a data management and reporting
tool that can be used on-board and on-shore by most fleets and organisations. The system is available in modular packages, allowing each operator to select a combination that meet their business needs.
The final stages of the condition based maintenance programme involve effective use of the feedback from the data analysis process. This might involve repairs or modifications to machine systems such as scheduled replacement of bearings or other wearing parts or the re-alignment or re-balancing of shafts or inter-connected systems.
This process of work control relies heavily on the priorities and structure determined during stages one and two, allowing maintenance activity to be planned in detail and scheduled, with tasks being prioritised according to timescales, man-hours required, data feedback, and staff competences. Effective planning at this stage, combined with good spares management, well-defined job plans and trained staff, allows the most effective use of resources.
Once remedial work is complete it is then important to continue to monitor conditions to successfully identify areas
for further improvement in terms of machine performance, energy efficiency or output. Essentially, this becomes a continuous process of machine reliability improvement using a database of statistics and readings gathered over time.
It should be noted that ongoing data gathering should not just be restricted to vibration monitoring, as other techniques provide valuable information about machine conditions. These techniques include temperature measurement of machine hotspots using infrared detectors, ultrasonic measurement to detect pressure or vacuum leaks, and oil condition monitoring to detect contamination caused by wearing machine parts or the ingress of moisture or fuels.
The bottom line
In today's increasingly competitive global market, where ship operators are under many different pressures, the move from traditional and largely reactive maintenance regimes to one where vessel managers and engineers are in control of machine maintenance can have a considerable impact on operating margins.
In particular, replacing the potential problems and costs of routine machine overhaul with a system where repairs can be carefully planned to fit in with business operations, and where components are only being replaced when necessary, can considerably reduce labour and parts costs.
Just as importantly, a condition based maintenance programme can improve the operating life of each vessel and the safety of its crew and passengers.