Gene Vogel
EASA Pump & Vibration Specialist

Many EASA service centers conduct vibration-based machine condition monitoring programs for their customers. This is a valuable revenue source that aligns the service center as a team member with the customer to provide uptime on their machinery and control maintenance costs. It also puts some of the responsibility for predicting machine failures on service center personnel who may be managing vibration data for hundreds of machines in multiple customer facilities. Given that vibration data is complex, composed of amplitude, frequency and phase, that can be a daunting task.

It is generally impractical to conduct a proper vibration analysis on every machine for every route data collection event. Even simple four-bearing machines will have 3 to 12 measurement points with spectra data and waveform data on each point. Some measurement points will have multiple parameters recorded including velocity, acceleration and some special bearing band parameters. 

It is important to be able to use the proper software tools to screen data from routine data collection routes, identifying those machines that likely have developing faults. Then valuable analysis time is spent on machines that may have incipient faults, which if identified can allow repairs to avoid more extensive machine damage and downtime.

The first line of screening may be an overall vibration amplitude level which is known to be inordinately high. But overall vibration alarms will miss many incipient faults. Commonly, acceptable vibration levels from imbalance, flow turbulence on pumps and fans and background vibration will mask low amplitude vibration from bearing and other critical machine components. By the time the fault becomes evident in overall vibration levels, the fault is far advanced or a failure has already occurred.

There are several techniques for a more detailed screening of vibration data. The most common are frequency band alarms and enveloping. In addition, rule base “intelligent” algorithms can be implemented to screen for known fault patterns. 

A service center technician or engineer responsible for the success of a vibration based condition monitoring program should have a good understanding of these data screening techniques.

Also discussed in this article:

• Band alarms
• Enveloping or spectra alarms
• Rule based "intelligent" alarms
• Trending
• Concerns

LOGIN TO DOWNLOAD THE ARTICLE

AVAILABLE IN SPANISH

Categories: Balancing, Vibration

Tags: Vibration Balancing Predictive maintenance Condition monitoring

Rate this article:

5.0

Print