Vibrations Monitoring as a Predictive Maintenance Tool for Reciprocating Engines
During the operational life of mechanical systems maintenance costs
constitute a major portion of total expenses. Reciprocating engines are
no exception. Accordingly, an adequate maintenance program can reduce these
costs over the life time of a system. In recent years predictive maintenance
programs have been implemented worldwide, replacing older maintenance philosophies
and realizing the task of reducing maintenance costs. In heavy industry,
where maintenance costs can represent 40% of the overall production costs,
the application of a predictive maintenance program can reduce these costs
by more than 50%. In contrast to run-to-failure and preventive maintenance
approaches, which are commonly implemented for reciprocating engines, predictive
maintenance is a condition-driven program.
| |
| |
To achieve its goals, a predictive maintenance program requires
diverse monitoring and diagnostic techniques. For machines and other equipment
with moving parts, the vibration monitoring method is undoubtedly the commonest
and most reliable technique. The concept behind the method is that every
machine with moving parts vibrates in response to the excitations acting
upon its components. Variations in the excitation forces, the machine's
components or its integrity will change the vibration pattern. Consequently,
periodic monitoring of the machine's vibration signature can provide useful
information regarding its components and the excitations acting upon them.
| |
| |
The successful usage of the vibration analysis method for monitoring industrial and other equipment prompted the adoption of this method for monitoring and analyzing reciprocating engines. During a series of experiments the vibrations of various engines were measured while the engines were running under normal and abnormal conditions. Optimal locations and directions of the acceleration transducers relative to the engine-blocks as well as the required sampling duration and frequency were identified. The consistency and reliability of the measurements were validated by comparisons of data sets that were accumulated at different times and locations of the engines. Because of the cluttered vibration signature of the engine, that results from the complicated processes involved in its operation together with its mechanical complexity, a few earlier attempts to apply this method failed to produce practical or conclusive findings. Therefore, the challenging objective of Engines-PdM was to introduce a unique diagnostic scheme that can deduce accurate information about the engine's condition from its vibration signature. The designated on-line diagnostic procedure which is currently being developed enables the ability to accurately analyze the monitored vibrations. Application of the procedure allowed to define a normal operation envelop and to establish criteria for distinguishing between normal and abnormal states. Furthermore, the resulting vibration signatures can be used to identify the source of the malfunction. During the course of the study the method successfully detected malfunctions such as damaged engine mountings, misfiring, deteriorated fuel injectors, fouled spark plugs, worn spark plugs, improper spark timing, off-timing valves, partially blocked inlet and outlet manifolds, unbalanced crankshaft and mechanical disorders. The method is currently being applied to an all-new series of diesel engines. |