<![CDATA[Cavitation is one of the main concern on centrifugal pump faults that could cause component damages up to production failure in the industries. It is essential to detect and diagnose the fault as early as possible to prevent a catasthropic failure. Cavitation on sentrifugal pump could be caused by many factor, one of them are caused by the pump operating speed. This paper presents a method that able to detect cavitation by monitoring the vibrations level of the pump based on statistical analysis of time domain. This method is known as vibration monitoring technique that is undoubtedly the most effective technique to detect rotational machinery faults. The cavitation simulated on the test rig by varying the operating speed at 1000 RPM, 1200 RPM, 1400 RPM, â¦, 2600 RPM and by varying the size of suction valve opening. The cavitation phenomena are measured and indicated by magnitude of vibration level changes in stastical parameter such as Probability Density Function (PDF), Variance, Standard Deviation, Root Mean Square (RMS), Peak Value, Crest Factors and Kurtosis. The results show that PDF, Variance, Standard Deviation and RMS are proved to be able to detect cavitation caused by the pump operating speed variation. However, parameter such as Peak Value, Crest Factor and Kurtosis show low sensitivity and not suitable for the cavitation detection purposes.]]>
Copyrights © 2017
