Chapter 3

VIBRATION ANALYSIS OVERVIEW

Vibration theory and vibration profile, or signature, analyses are complex subjects that are the topic of many textbooks. The purpose of this chapter is to provide enough theory to allow the concept of vibration profiles and their analyses to be understood before beginning the more in-depth discussions in the later sections of this module.

THEORETICAL VIBRATION PROFILES

A vibration is a periodic motion or one that repeats itself after a certain interval of time. This time interval is referred to as the period of the vibration, T. A plot, or profile, of a vibration is shown in Figure 3.1, which shows the period, T, and the maxi1 mum displacement or amplitude, X,. The inverse of the period, - is called the T frequency, A of the vibration, which can be expressed in units of cycles per second (cps) or Hertz (Hz). A harmonic function is the simplest type of periodic motion and is shown in Figure 3.2, which is the harmonic function for the small oscillations of a simple pendulum. Such a relationship can be expressed by the equation:

X = X, sin(ot)
where
X = Vibration displacement (thousandths of an inch, or mils) X, = Maximum displacement or amplitude (mils) o = Circular frequency (radians per second)
t
= Time (seconds).

Vibration Analysis Overview

X
-eJ

t W

Q cn .-

0 0

fi

Figure 3.1 Periodic motionfor bearing pedestal of a steam turbine.

f Figure 3.2 Small oscillations o a simple pendulum, harmonicfunction.

Vibration Fundamentals

ACTUAL VIBRATION PROFILES

The process of vibration analysis requires the gathering of complex machine data, which must then be deciphered. As opposed to the simple theoretical vibration curves shown in Figures 3.1 and 3.2 above, the profile for a piece of equipment is extremely complex. This is true because there are usually many sources of vibration. Each source generates its own curve, but these are essentially added and displayed as a composite profile. These profiles can be displayed in two formats: time domain and frequency domain.
Time Domain

Vibration data plotted as amplitude versus time is referred to as a time-domain data profile. Some simple examples are shown in Figures 3.1 and 3.2. An example of the complexity of these type of data for an actual piece of industrial machinery is shown in Figure 3.3. Time-domain plots must be used for all linear and reciprocating motion machinery. They are useful in the overall analysis of machine-trains to study changes in operating conditions. However, time-domain data are difficult to use. Because all of the vibration data in this type of plot are added to represent the total displacement at any given time, it is difficult to determine the contribution of any particular vibration source. The French physicist and mathematician Jean Fourier determined that nonharmonic data functions such as the time-domain vibration profile are the mathematical sum of

3.00

- .00
h .

6-3.00 E 3. 3.00
.P

.00
-3.00I

0.01

0.02
Time (Sec)

0.03

0.04

0.05

Figure 3.3 Example o a typicaltime-domain vibmtion profile for apiece o machinery. f f

Vibration Analysis Overview

Figure 3.4 Discrete (harmonic)and total (nonharmonic)time-domain vibration curves.

simple harmonic functions. The dashed-line curves in Figure 3.4 represent discrete harmonic components of the total, or summed, nonharmonic curve represented by the solid line. These type of data, which are routinely taken during the life of a machine, are directly comparable to historical data taken at exactly the same running speed and load. However, this is not practical because of variations in day-to-day plant operations and changes in running speed. This significantly affects the profile and makes it impossible to compare historical data.
Frequency Domain

From a practical standpoint, simple harmonic vibration functions are related to the circular frequencies of the rotating or moving components. Therefore, these frequencies are some multiple of the basic running speed of the machine-train, which is expressed in revolutions per minute (rpm) or cycles per minute (cpm). Determining

10

Vibration Fundamentals

OS&NJi80

L!%

:XI3

50

100

150 200 FREQUENCY IN Hi!

250

300

xi0

Figure 3.5 Typicalfrequency-domain vibration signature.

these frequencies is the first basic step in analyzing the operating condition of the machine-train. Frequency-domain data are obtained by converting time-domain data using a mathematical technique referred to as a fast Fourier transform (FIT). FFT allows each vibration component of a complex machine-train spectrum to be shown as a discrete frequency peak. The frequency-domain amplitude can be the displacement per unit time related to a particular frequency, which is plotted as the Y-axis against frequency as the X-axis. This is opposed to the time-domain spectrum, which sums the velocities of all frequencies and plots the sum as the Y-axis against time as the X-axis. An example of a frequency-domain plot or vibration signature is shown in Figure 3.5. Frequency-domain data are required for equipment operating at more than one running speed and all rotating applications. Because the X-axis of the spectrum is frequency normalized to the running speed, a change in running speed will not affect the plot. A vibration component that is present at one running speed will still be found in the same location on the plot for another running speed after the normalization, although the amplitude may be different.

Vibration Analysis Overview hterprefation of Vibration Data

11

The key to using vibration signature analysis for predictive maintenance, diagnostic, and other applications is the ability to differentiate between normal and abnormal vibration profiles. Many vibrations are normal for a piece of rotating or moving machinery. Examples of these are normal rotations of shafts and other rotors, contact with bearings, gear-mesh, etc. However, specific problems with machinery generate abnormal, yet identifiable, vibrations. Examples of these are loose bolts, misaligned shafts, worn bearings, leaks, and incipient metal fatigue. Predictive maintenance utilizing vibration signature analysis is based on the following facts, which form the basis for the methods used to identify and quantify the root causes of failure: 1. All common machinery problems and failure modes have distinct vibration frequency components that can be isolated and identified. 2. A frequency-domain vibration signature is generally used for the analysis because it is comprised of discrete peaks, each representing a specific vibration source. 3. There is a cause, referred to as a forcing function, for every frequency component in a machine-trains vibration signature. 4. When the signature of a machine is compared over time, it will repeat until some event changes the vibration pattern (i.e., the amplitude of each distinct vibration component will remain constant until there is a change in the operating dynamics of the machine-train). While an increase or a decrease in amplitude may indicate degradation of the machine-train, this is not always the case. Variations in load, operating practices, and a variety of other normal changes also generate a change in the amplitude of one or more frequency components within the vibration signature. In addition, it is important to note that a lower amplitude does not necessarily indicate an improvement in the mechanical condition of the machine-train. Therefore, it is important that the source of all amplitude variations be clearly understood.

VIBRATION MEASURING EQUIPMENT

Vibration data are obtained by the following procedure: (1) Mount a transducer onto the machinery at various locations, typically machine housing and bearing caps, and (2) use a portable data-gathering device, referred to as a vibration monitor or analyzer, to connect to the transducer to obtain vibration readings.
Transducer

The transducer most commonly used to obtain vibration measurements is an accelerometer. It incorporates piezoelectric (i.e., pressure-sensitive) films to convert mechanical energy into electrical signals. The device generally incorporates a weight

12

Vibration Fundamentals

suspended between two piezoelectric films. The weight moves in response to vibration and squeezes the piezoelectric films, which sends an electrical signal each time the weight squeezes it.
Portable VibrationAnalyzer

The portable vibration analyzer incorporates a microprocessor that allows it to convert the electrical signal mathematically to acceleration per unit time, perform a FFT, and store the data. It can be programmed to generate alarms and displays of the data. The data stored by the analyzer can be downloaded to a personal or a more powerful computer to perform more sophisticated analyses, data storage and retrieval, and report generation.