Introduction To Basic Vibration Analysis: By: Jack D. Peters

Introduction to
Basic Vibration
Analysis
By: Jack D. Peters
Connection Technology Center, Inc
7939 Rae Boulevard
Victor, New York 14564
www.ctconline.com
Eastman Kodak
1976 – 2004
9 Senior Engineer for Vibration

Monitoring of Photographic Film
and Paper Manufacturing
Machines world wide.
9 24/7 Permanent Monitoring using

dynamic signal analyzers and
proprietary statistical process
control software generating
quality alerts and alarms.
Jack D. Peters - CTC 2

Vibration Institute
Category IV Analyst
9 ISO 18436-2
9 Instructor
9 Past Chairman CNY
www.cnyvi.com

Connection Technology Center
2004 – Present

Data Collection
Auto Pwr Spec 1

0.2
G
rms
Magnitude
0
3 Hz 2.5 kHz
Portable
Route Based Permanent
Continuous On-line

Portable Data Collectors

Portable Data Collectors
9 Route Based 9 Data Analysis

9 Frequency 9 History
Spectrum 9 Trending
9 Time Waveform 9 Download to PC
9 Orbits 9 Alarms
9 Balancing 9 “Smart” algorithms
9 Alignment

On-Line Systems (permanent)
24/7
Rack mounted systems with

large numbers of channels.

Dynamic Signal Analyzers
“Test & Measurement”
Large PC driven solutions
with multiple channels and
windows based software.
Smaller portable units

with 2 – 4 channel
inputs and firmware
operating systems.

Tape Recorders
“Insurance Policy”
Multi-channel
digital audio tape
recorders.
For the
Measurement that
can’t get away !
What’s This ?
1
0.0002
inch
Peak
Magnitude
0
0 Hz 100 Hz

FFT, Frequency Spectrum, Power
Spectrum
1
0.0002
inch
Peak
Magnitude
0
0 Hz 100 Hz

Scaling X & Y
1
0.0002
inch
Peak
Y Magnitude
0
0 Hz 100 Hz
X

Scaling X & Y
1
A 0.0002
inch
M Peak
P
L
Magnitude
I
T
U
D 0
0 Hz 100 Hz
E FREQUENCY

Scaling X & Y
1
H 0.0002
inch
o Peak
w
B
a
Magnitude
d
i
s
i 0
t 0 Hz 100 Hz
What is it

What’s That ?
1
0.0004
inch
Real
-0.0004
0 s 7.996094 s

Time Waveform
1
0.0004
inch
Real
-0.0004
0 s 7.996094 s

Scaling X & Y
1
0.0004
inch
Y Real
-0.0004
0 s 7.996094 s

Scaling X & Y
1
A 0.0004
inch
M
P
L
I Real
T
U
D -0.0004
0 s 7.996094 s
E TIME

Scaling X & Y
1
H 0.0004
o inch
B
a
Real
d
i
s
i -0.0004
t 0 s 7.996094 s
What is it

The X Scale
What is it ?

Single Frequency
X:55 Hz Y:706.8129 mV
Pwr Spec 1
1
V 55 Hz
rms
Magnitude
0
0 Hz 100 Hz
X:27.00806 ms Y:3.579427 mV
dX:18.18848 ms dY:2.449082 mV
Time 1
1
V
8.82 ms
Real
-1
0 s 62.46948 ms

Frequency & Time
fHz = 1/tSec
tSec = 1/fHz

Frequency & Time
FT = 1
If: F = 1/T and T = 1/F

Then: FT = 1

Concept !
FT = 1
If: F increases
Then: t decreases
If: T increases
Then: f decreases

Single Frequency
X:55 Hz Y:706.8129 mV
Pwr Spec 1
1
V 55 Hz
rms
Magnitude
0
0 Hz 100 Hz
X:27.00806 ms Y:3.579427 mV
dX:18.18848 ms dY:2.449082 mV
Time 1
1
V
8.82 ms
Real
-1
0 s 62.46948 ms

Multiple Frequencies
X:55 Hz Y:706.8129 mV
Pwr Spec 1
1
0 Hz 100 Hz
X:78 Hz Y:706.9236 mV
Pwr Spec 1
1
0 Hz 100 Hz
X:21 Hz Y:706.7825 mV
Pwr Spec 1
1
0 Hz 100 Hz
X:42 Hz Y:706.9266 mV
Pwr Spec 1
1
0 Hz 100 Hz

Multiple Time Waveforms
Time 55 1
1
V
0 s 62.46948 ms
Time 78 1
1
V
0 s 62.46948 ms
Time 21 1
1
V
0 s 62.46948 ms
Time 42 1
1
V
0 s 62.46948 ms

Real Life Time Waveform
55 + 78 + 21 + 42 = Trouble !
TIME 1
4
V
Real
-4
0 s 62.46948 ms

FFT Capabilities
TIME 1
4
V
Real
-4
0 s 62.46948 ms
X:21 Hz Y:706.7825 mV
X:42 Hz Y:706.9266 mV
X:55 Hz Y:706.8129 mV
X:78 Hz Y:706.9236 mV
FREQUENCY 1
1
V
rms
0 Hz 100 Hz

The Most Copied Slide in the History of
Vibration Analysis !
Amplitude
Inp
ut
Tim en cy
e qu
F re
Tim
eW
ave m
for
m p ectru
S

Lines or Bins
1
The FFT always 0.0002
inch
has a defined Peak This spectrum
number of lines has 800 lines, or
or Bins. the X scale is
100, 200, 400, Magnitude broken down into
800, 1600, and 800 bins.
3200 lines are
common
choices. 0
0 Hz 100 Hz

LRF
The Lowest Resolvable Frequency is determined by:
Frequency Span / Number of Analyzer Lines
The frequency span is calculated as the ending

frequency minus the starting frequency.
The number of analyzer lines depends on the analyzer
and how the operator has set it up.
Example: 0 - 400 Hz using 800 lines

Answer = (400 - 0) / 800 = 0.5 Hz / Line

Bandwidth
The Bandwidth can be defined by:
(Frequency Span / Analyzer Lines) Window Function
Uniform Window Function = 1.0
Note: More discussion
later on window functions
Hanning Window Function = 1.5 for the analyzer !
Flat Top Window Function = 3.8
Example: 0 - 400 Hz using 800 Lines & Hanning Window

Answer = (400 / 800) 1.5 = 0.75 Hz / Line

Resolution
The frequency resolution is defined in the following
manner:
2 (Frequency Span / Analyzer Lines) Window Function
or
Resolution = 2 (Bandwidth)
Example: 0 - 400 Hz using 800 Lines & Hanning Window

Answer = 2 (400 / 800) 1.5 = 1.5 Hz / Line

Using Resolution
The student wishes to measure two frequency

disturbances that are very close together.
Frequency #1 = 29.5 Hz.
Frequency #2 = 30 Hz.
The instructor suggests a hanning window and 800 lines.
What frequency span is required to accurately measure
these two frequency disturbances ?

Using Resolution
Resolution = 30 - 29.5 = 0.5 Hz / Line

Resolution = 2 (Bandwidth)
BW = (Frequency Span / Analyzer Lines) Window Function
Resolution = 2 (Frequency Span / 800) 1.5
0.5 = 2 (Frequency Span / 800) 1.5
0.5 = 3 (Frequency Span) / 800
400 = 3 (Frequency Span)
133 Hz = Frequency Span

Data Sampling Time
Data sampling time is the amount of time required to take
one record or sample of data. It is dependent on the
frequency span and the number of analyzer lines being used.
TSample = Nlines / Fspan
Using 400 lines with a 800 Hz frequency span will require:

400 / 800 = 0.5 seconds

Average & Overlap
9 Average - On
TR#1 TR#2 TR#3
9 Overlap Percent - 50%
FFT#1 FFT#2 FFT#3
9 Overlap is the amount
of old data that is used 0% Overlap
50% Overlap
TR#1
How long will it take for 10 TR#2
averages at 75% overlap TR#3
using a 800 line analyzer and FFT#1
a 200 Hz frequency span?
FFT#2
FFT#3

75% Overlap ?
9 10 Averages Average #1 = 800 / 200
9 75% Overlap Average #1 = 4 seconds
9 800 Lines
Average #2 - #10 = (4 x 0.25)
9 200 Hz
Average #2 - #10 = 1 second
each
Total time = 4 + (1 x 9)
Total time = 13 seconds

Filter Windows
9 Window filters are applied to the time
waveform data to simulate data that
starts and stops at zero.
9 They will cause errors in the time
waveform and frequency spectrum.
9 We still like window filters !

Window Comparisons

Filter Windows
9 Hanning (Frequency) Hanning 16% Amplitude Error
9 Flat Top (Amplitude)
9 Uniform (No Window)
9 Force Exponential
Force/Expo Set-up
(Frequency Response) Flat Top 1% Amplitude Error
Window functions courtesy of Agilent

“The Fundamentals of Signal Analysis”
Application Note AN 243

Filter Windows
9 Use the Hanning Window for normal
vibration monitoring (Frequency)
9 Use the Flat Top Window for calibration
and accuracy (Amplitude)
9 Use the Uniform Window for bump
testing and resonance checks (No
Window)

The Y Scale
How bad is it ?

Amplitude
Acceleration = g’s rms. or peak
Velocity = inch/s rms. or peak
Displacement = mils peak to peak

Note: 1 mil = 0.001 inches

Pk-Pk (Peak - Peak)
X:55 Hz Y:1.999169 V
Pwr Spec 1
The Peak - Peak value V
2
is expressed from the Pk-Pk

Magnitude
peak to peak amplitude. 0

0 Hz 100 Hz
The peak to peak value X:22.43042 ms

dX:9.094238 ms
Y:-993.8563 mV
dY:1.994871 V
Time 1
is measured in the time 1
V
waveform. Real
-1
0 s 62.46948 ms
Peak - Peak. = 2 V

Pk (Peak)
The time wave has not
changed. The Peak X:55 Hz Y:999.5843 mV
Pwr Spec 1
value is expressed V
1
from zero to the largest Peak

Magnitude
positive or negative 0
0 Hz 100 Hz
peak amplitude. X:27.00806 ms Y:3.579427 mV
dX:4.516602 ms dY:997.4356 mV
The peak value is 1

Time 1
V
measured in the time Real
waveform. -1
0 s 62.46948 ms
Peak. = 1 V

RMS (Root Mean Square)
The time wave has
not changed. X:55 Hz Y:706.8129 mV
Pwr Spec 1
The rms. value is 1
V rms. = 707 mV
rms
expressed from zero Magnitude
to 70.7% of the peak 0

0 Hz 100 Hz
amplitude for a single X:27.00806 ms Y:3.579427 mV
dX:2.288818 ms dY:709.1976 mV
frequency. Time 1
1
V
The rms. value is Real
calculated for the -1

0 s 62.46948 ms
spectrum.

Suffix Comparison
X:27.00806 ms Y:3.579427 mV X:55 Hz Y:706.8129 mV
dX:2.288818 ms dY:709.1976 m Pwr Spec 1
Time 1 2
1 V
V rms
RMS Real
Magnitude
-1 0
0 s 62.46948 ms 0 Hz 100 Hz
X:27.00806 ms Y:3.579427 mV X:55 Hz Y:999.5843 mV

dX:4.516602 ms dY:997.4356 m Pwr Spec 1
Time 1 2
1 V
Peak V
Real
Peak
Magnitude
-1 0
0 s 62.46948 ms 0 Hz 100 Hz
X:22.43042 ms Y:-993.8563 mV X:55 Hz Y:1.999169 V

dX:9.094238 ms dY:1.994871 V Pwr Spec 1
Time 1 2
1 V
Peak - Peak V
Real
Pk-Pk
Magnitude
-1 0
0 s 62.46948 ms 0 Hz 100 Hz

Changing Suffixes
Many times it is necessary to change between suffixes.
Pk-Pk / 2 = Peak
Peak x 0.707 = RMS (Peak / 1.414 = RMS)
RMS / 0.707 = Peak (RMS x 1.414 = Peak)
Peak x 2 = Pk-Pk

Converting the Unit Suffix
x2 Peak - Peak ÷2
Peak Peak
÷ 0.707 RMS x 0.707

Engineering Units (EU)
Engineering units are used to give meaning to the
amplitude of the measurement.
Instead of the default “volts”, it is possible to incorporate
a unit proportional to volts that will have greater
meaning to the user.
Examples: 100 mV / g 20 mV / Pa
1 V / in/s 200 mV / mil
50 mV / psi 10 mV / fpm
33 mV / % 10 mV / V
EU’s the Hard Way
Sometimes we forget to use EU’s, or just don’t understand how

to set up the analyzer. The measurement is in volts!
There is no immediate need to panic if ????
You know what the EU is for the sensor you are using.
Example: An accelerometer outputs 100 mV / g and there is a
10 mV peak in the frequency spectrum.
What is the amplitude in g’s ?
Answer = 10 mV / 100 mV = 0.1 g

The Big Three EU’s
Acceleration
Velocity
Displacement
Converting the Big 3
In many cases we are confronted with Acceleration,

Velocity, or Displacement, but are not happy with it.
Maybe we have taken the measurement in
acceleration, but the model calls for displacement.
Maybe we have taken the data in displacement, but
the manufacturer quoted the equipment
specifications in velocity.
How do we change between these EU’s ?

Changing Units
Velocity = 2π f D
Acceleration = 2π f V or (2π f)2 D
f = frequency, cycles/sec or Hz
D = displacement, inches (mm)
V = velocity, in./sec. (mm/sec)
A = acceleration, in/sec2 (mm/sec2)
(divide by 386.1 in/sec2/g to obtain acceleration in g’s)
(divide by 9807 mm/sec2/g to obtain acceleration in g’s)
386.1 What ?
1g = 32.2 feet/second2
32.2 feet 12 inches

second2 X
foot
2
386.1 inches/second
g

Doing the Math Units
0.5g x 386.1 inches
second2
g
2π x 25 cycles
There is a 0.5 g second
vibration at 25 Hz.
0.5g x 386.1 inches x 1 second
What is the velocity ? g second2 2π x 25 cycles
0.5 x 386.1 inches

2π x 25 cycles second
cycle
1.23 inches/second

Radians, Degrees, or Time
π
2
3600 = 2π Radians
900
3600 / 2π Radians
57.3250 / Radian
π 1800 00 0
3600 2π
2700
3π
Jack D. Peters - CTC
2 61
π
2
900
π 1800 00 0
3600 2π
2700
3π
Jack D. Peters - CTC
2 62
π
2
900
0 00 1800 π 3600 2π
Period (seconds/cycle)
2700
3π
2
Changing Units:
Acceleration to Velocity
A
= 2 πf V
A
=V
2πf
Changing Units:
Velocity to Displacement
V
= 2 πf D
V
=D
2 πf
Converting the Unit Measure
÷ 386.1 Acceleration (g’s) x 386.1
Acceleration Acceleration
(inch/s2) (inch/s2)
x 2(Pi)f ÷ 2(Pi)f
Velocity Velocity
(inch/s) (inch/s)
x 2(Pi)f Displacement (inch) ÷ 2(Pi)f

Converting the SI Unit
Measure
÷ 9807 Acceleration (g’s) x 9807
Acceleration Acceleration
(mm/s2) (mm/s2)
x 2(Pi)f ÷ 2(Pi)f
Velocity Velocity
(mm/s) (mm/s)
x 2(Pi)f Displacement (mm) ÷ 2(Pi)f

Acceleration - Velocity
Example: Find the equivalent peak velocity for a 25 Hz

vibration at 7 mg RMS ?
= (g x 386.1) / (2 Pi x F)
= (0.007 x 386.1) / (6.28 x 25)
= 0.017 inches / second RMS
Answer = 0.017 x 1.414 = 0.024 inches / second Pk

Velocity - Displacement
Example: Find the equivalent pk-pk displacement for a

25 Hz vibration at 0.024 in/s Pk ?
= Velocity / (2 Pi x F)
= 0.024 / (6.28 x 25)
= 0.000153 inches Pk
Answer = 0.000153 x 2 = 0.000306 inches Pk-Pk

Acceleration - Displacement
Example: Find the equivalent Pk-Pk displacement for a

52 Hz vibration at 15 mg RMS ?
= (g x 386.1) / (2 Pi x F)2
= (0.015 x 386.1) / (6.28 x 52)2
= 0.000054 inches RMS
Answer = (0.000054 x 1.414) 2 = 0.000154 inches Pk-Pk

Sensors
d
ee
Sp
Displacement
Frequency

Accelerometers
9 IEPE 9 Charge Mode
¾ Electronics inside ¾ Charge Amplifier
¾ Industrial ¾ Test & Measurement

Industrial Requirements and
Applications
¾ Requirements
9 Functionality
9 Durability
9 Affordability
¾ Applications
9 Trending
9 Alarming
9 Diagnostics
¾ Remember
9 One sensor does not
fit all applications
9 Fit, Form & Function

Accelerometer Advantages
9 Measures casing vibration
9 Measures absolute motion
9 Can integrate to Velocity output
9 Easy to mount
9 Large range of frequency response
9 Available in many configurations

Accelerometer Disadvantages
9 Does not measure shaft vibration
9 Sensitive to mounting techniques and
surface conditions
9 Difficult to perform calibration check
9 Double integration to displacement
often causes low frequency noise
9 One accelerometer does not fit all
applications

Mass & Charge
Relative movement
between post &
mass creates shear
in ceramic
producing charge.
Mass
Ceramic/Quartz
Post

Accelerometer Parameters
Performance Suited for Application
9 Sensitivity (mV/g)
9 Frequency Response
9 Dynamic Range

Typical Frequency Response
Amplification Isolation
Region Region
The natural frequency is Phase between
Transmission Region excited causing gain sensor & machine
The usable frequency range of the accelrometer around resonance is shifted by 180
Amplitude
based on acceptable amplitude limits degrees and signal

rolls off to zero
+/- 3dB
+/- 10%
+/- 5%
Frequency

Mounting the Accelerometer

Realistic Mounting
Stud
Vibration Institute, Basic Bees Wax

In the real world,
Machinery Vibrations &
Machinery Vibration Analysis Adhesive mounting might
Magnet not be as good as
Hand Held the manufacturer
had in the lab !
What about
paint, rust,
grease, oil, etc?
100 1,000 10,000

Frequency, Hz

Sensitivity, Range & Application
Sensitivity Range Output Application

A 10 mV/g accelerometer will have a
10 mV/g +/- 500 g +/- 5 VAC dynamic range of +/- 500 g’s, and a
dynamic output of +/- 5 volts AC.
They are typically used for machinery
50 mV/g +/- 100 g +/- 5 VAC that is generating high amplitude
vibrations. With the large dynamic
range, they are much less likely to
100 mV/g +/- 50 g +/- 5 VAC become saturated as a result of the
high amplitude vibrations.
500 mV/g +/- 10 g +/- 5 VAC


They are typically used for general
50 mV/g +/- 100 g +/- 5 VAC purpose machinery measurements,
and are sometimes offered as standard
sensors for data collectors.
100 mV/g +/- 50 g +/- 5 VAC
500 mV/g +/- 10 g +/- 5 VAC


Approximately 90% of all vibration
50 mV/g +/- 100 g +/- 5 VAC analysis and data collection is
accomplished with a 100 mV/g
accelerometer.
100 mV/g +/- 50 g +/- 5 VAC Some sensors are also available with
a +/- 80g dynamic range for measuring
larger signal amplitudes.
500 mV/g +/- 10 g +/- 5 VAC


This high output sensor is typically
50 mV/g +/- 100 g +/- 5 VAC used for low speed equipment, low
frequency measurements, and low
amplitude analysis.
100 mV/g +/- 50 g +/- 5 VAC The high output provides a much
better signal to noise ratio for low
amplitude signals.
500 mV/g +/- 10 g +/- 5 VAC

Mounting Location
Vertical
¾Load Zone
9Axial
¾Radial
9Vertical Horizontal
9Horizontal Axial
Vibration Institute, Basic Machinery

Vibrations & Machinery Vibration Analysis

Mounting Location
¾Load Zone
9Axial
¾Radial
9Vertical
9Horizontal

Accelerometer Alarms
Machine Condition Velocity Limit
rms peak
Acceptance of new or repaired equipment < 0.08 < 0.16

Unrestricted operation (normal) < 0.12 < 0.24
Surveillance 0.12 - 0.28 0.24 - 0.7
Unsuitable for Operation > 0.28 > 0.7
Note #1: The rms velocity (in/sec) is the band power or

band energy calculated in the frequency spectrum.
Note #2: The peak velocity (in/sec) is the largest positive
or negative peak measured in the time waveform.

Velocity Sensors
9 Self Generating – no power
supply required
9 Magnet inside coil generates
velocity proportional to
vibration
9 Spring mass system
9 10 Hz. to 1000 Hz.
9 Phase change
9 Directional mounting
9 Large & Heavy
9 500 mV/inch/sec

PiezoVelocity Sensors
9 Remember everything that you just learned about an
accelerometer
9 The output of the accelerometer has been integrated
to velocity
9 100 mV/inch/sec

Proximity Probes, Cables, &
Drivers
9 Overview
9 Technical
Background
9 Technical
Specifications
9 Applications

5 Meter & 9 Meter Systems
Proximity Probe
Typical lengths: 0.5 and 1.0 meters

Extension Cable
Probe Length + Extension Cable Length

must equal 5 or 9 meters in system length

Driver
Electronics tuned for 5 or 9 meter systems

Application
9 Measure Displacement
9 Vibration in plain bearing applications
9 Non Contact sensing of the shaft
9 Ideal for measuring:

9 Shaft vibration
9 Shaft centerline position
9 Shaft axial position (Thrust Bearing)
9 Rod drop
9 Speed (key phaser)

Common Applications
9Compressors 9 Plain Bearings

9Steam Turbines 9 Journal Bearings
9Pumps 9 Fluid Film Bearings
9Fans 9 Babbitt Bearings
9Blowers 9 Sleeve Bearings
9Generators 9 Tilting Pad Bearings
9Gear Boxes 9 Recip’s (cross head)

Displacement Probes -
Advantages
9 Non-contact
9 Measure relative shaft vibration
9 Measure shaft centerline position (DC gap)
9 Measure axial position (Thrust)
9 Flat frequency response dc – 10KHz
9 Simple calibration
9 Suitable for harsh environments

Displacement Probes -
Disadvantages
9 Probe can move (vibrate)

9 Doesn’t work on all metals
9 Plated shafts may give
false measurement Plated shaft is
9 Measurement is affected by round, but core
material is not..
scratches & tool marks in shaft
9 Available system lengths (probe, cable & driver)
5 meter or 9 meter are standard
9 Must have relief at sensing tip from surrounding
metal (counter bore)

Technical Background
Driver • The tip of the probe emits a radio

frequency signal into the surrounding
Cable area as a magnetic field
Probe
• As a conductive target intercepts the
magnetic field, eddy currents are
generated on the surface of the target,
and power is drained from the radio
Target frequency signal

Driver
• Power varies with target movement
in the radio frequency field creating
Cable a variation in the output voltage of
the driver
Probe - A small DC voltage indicates that the target is
close to the probe tip
- A large DC voltage indicates that the target is
far away from the probe tip
- The variation of DC voltage is the AC dynamic
Target signal indicating the vibration (displacement)

Sensitivity, Range, Response
Driver
Cable
Typical non-contact
displacement sensor for
measuring shaft vibration on
a sleeve or journal bearing.
Probe
Sensitivity 200 mV/mil
(8 V/mm)
Eddy Currents Dynamic Range 10 – 90 mils
(.25 – 2.3 mm)
Shaft
Frequency Response DC – 10 kHz
Journal/Sleeve

Linearity
Gap Gap Output
mils mm VDC
10 0.25 -2.01
20 0.51 -3.97
30 0.76 -5.90
40 1.02 -7.88
50 1.27 -9.87
60 1.52 -11.80
200 mV/mil x 50 mils = 70 1.78 -13.80
80 2.03 -15.69
-10.00 VDC
90 2.29 -17.66
100 2.54 -19.58

Materials & Output Values
9 Typical Calibration Examples

9 200 mv/mil (8 V/mm) 9 Copper 380 mV/mil
9 Depends on probe, 9 Aluminum 370 mV/mil
9 Brass 330 mV/mil
cable (length), and
9 Tungsten Carbide 290 mV/mil
driver. 9 Stainless Steel 250 mV/mil
9 Target material varies 9 Steel 4140, 4340 200 mV/mil
output. Based on typical output sensitivity

of 200 mV/mil. (8 V/mm)

It’s a Harsh World Out There!

Driver to Driven

API Standard 670 American Petroleum Institute
4th Edition, December 01, 2000
(Typical Information) www.techstreet.com $168.00 USD/Copy
Industry Standard for Proximity Probes

Vertical (Y) Horizontal (X)
Probe orientation Gap

based on facing Shaft
Driver to Driven
Journal/Sleeve

DC Gap with Dynamic AC
Time Record 1 DC Gap

-9.75
V
A negative voltage
level proportional to
30 mV p-p VAC the gap spacing
Dynamic
Dynamic AC
≈ -10.00 Real
Varying DC voltage
VDC
simulates dynamic
Gap AC voltage for
vibration output
-10.25 30 mV/(200 mV/mil)

0 s 79.96092 ms = 0.15 mil’s p-p

Looking at Orbits with a Scope
Vertical for Horizontal for

Amplitude Time Base
Y X

The Orbit Display

Rolling the Scope
Machine
Vertical
Machine
Horizontal
X
Y

Modern Instrumentation & Orbits
Modern instrumentation can

compensate for the location of the X
and Y probes providing a true machine
vertical and horizontal measurement.

Proximity Probe Alarms
Machine Condition Allowable R/C
< 3,600 RPM < 10,000 RPM
Normal 0.3 0.2

Surveillance 0.3 - 0.5 0.2 - 0.4
Planned Shutdown 0.5 0.4
Unsuitable for Operation 0.7 0.6
Note #1: R is the relative displacement of the shaft

measured by either probe in mils peak-peak.
Note #2: C is the diametrical clearance (difference
between shaft OD and journal ID) measured in mils.

Absolute Shaft Displacement
Vertical (Y) Horizontal (X)

Velocity
1. Measure the
vertical shaft
displacement. Vertical Measures
2. Measure the D = 2.85 milsp-p @1650

vertical casing V = 0.24 IPSpk @ 2110
velocity.
3. Include phase 3600 RPM

Graphical Addition
Vertical Measures 900

D = 2.85 milsp-p @1650
3.86 milsp-p @
V = 0.24 IPSp @ 2110 1520
Velocity leads 1.27 milsp-p @
displacement by 900 1210
1800 00
2110 - 900 = 1210 2.85 milsp-p @
1650
Dp-p = 2[0.24/(2πf)]
Dp-p = 2[0.24/(6.28x60)]
D = 1.27 milsp-p @ 1210 2700

Mathematical Addition
D = 2.85 milsp-p @1650 900

D = y2 + x2
D = 1.27 milsp-p @ 1210
D = 1.832 + (-3.40)2
y = 2.85 milsp-p x sin 1650
D = 3.86 milsp-p
y = 0.74 milsp-p
y = 1.27 milsp-p x sin 1210 3.86 milsp-p
1.83 milsp-p
y = 1.09 mils p-p @ 1520
1800 -3.4 milsp-p
00
y = 0.74 + 1.09 = 1.83 milsp-p
x = 2.85 milsp-p x cos 1650 900 + acos 1.83/3.86
x = -2.75 milsp-p 900 + 620 = 1520
x = 1.27 milsp-p x cos 1210
x = -0.65 milsp-p
x = - 2.75 + - 0.65 = - 3.40 milsp-p 2700

Shaft Centerline
Bore Dia.
Shaft
On CCW
Dia.
Centers Rotation
Diametrical CW
Zero
Clearance Rotation
RPM

Shaft Centerline @ Zero RPM
Diametrical
Clearance = 8 mils
Dia. Clr. / 2 = Radial Clr. = 4 mils

Plotting Shaft Position Change
Y -450 X +450
0
At Running Speed
CCW Rotation
Y = -1 mil
X = +2 mils
Shaft Change =
2.24 mils @ 71.60

Attitude Angle & Eccentricity Ratio
Shaft Attitude Angle
Y -450 X +450 320

Typical 200 to 500
Eccentricity Ratio =
Rad Clr 4
3.8 s
mils 3.85 mils/ 4.0 mils = .96

mi
5
l
0 Typical >.7 < 1.0

32
0 = On Centers
1 = Contact

Axial Shaft Position (Thrust)
Shaft Two axial oriented probes are

used for redundancy to monitor
the axial movement of the shaft
or thrust collar.

Natural Frequency
9 A result of the Mass (m) and
Stiffness (k) of the machine design
9 Resonance occurs when a natural
frequency is excited by a force
9 Critical speed occurs when the
machine speed matches the natural
frequency and creates resonance

Natural Frequency
X:164.0625 ms Y:1.379613 G
dX:554.6875 ms dY:-729.2974 mG Time Waveform
Time Record 1 TIME1.63
2
G
Real
-2
0 s 8 s
X:109.125 Hz Y:214.7374 mG Frequency Spectrum

Auto Pwr Spec 1 HZ1.63
0.3
G
rms
Real
0
50.00001 Hz 150 Hz

fn = [1/(2π)] k/m
¿ INCREASE the ¿ INCREASE the

stiffness ( k ) mass ( m )
¿ INCREASE the À DECREASE the

frequency (f) frequency ( f )

Natural Frequency fN = [1/(2 π)] √ k/m
10 lbs. 30 lbs. 50 lbs. 95 lbs.

Pull Strength
Frequency Response ≈ 2000 Hz.
k/m ≈ k/m ≈ k/m ≈ k/m

Bump Testing Set-up
UNIFORM WINDOW
9 Take your time – Bump around
9 Do not over range or clip the input signal
9 800 – 1600 lines of resolution
9 Try some different frequency spans
9 Only 1 bump for each time record
9 About 4 averages (depends on noise)

Why the Uniform Window ?

Bump It ! Two Responses !
X:23.4375 ms Y:1.63297 G Time Waveform
dX:76.17188 ms dY:-1.36474 G
2
G
Real
-2
0 s 1 s

X:65.5 Hz Y:12.23725 mG
X:70.75 Hz Y:8.475402 mG
0.015
G
rms
Real
0
0 Hz 100 Hz

Mental Health Check !
dX:76.17188 ms dY:-1.36474 G
76.17 msec/5 = 15.23 msec

2
G
Real
-2 F = 1/0.01523 sec = 65.64 Hz

0 s 1 s

X:65.5 Hz Y:12.23725 mG
X:70.75 Hz Y:8.475402 mG
0.015
rms
G
65.5 Hz
Real
0
0 Hz 100 Hz

Using the Time Waveform
X:99.60938 ms Y:268.2297 mG
2
G A0 = 1.633 G F = 1/0.01523 sec = 65.64 Hz
An = 0.268 G
Real
n = 5 cycles
-2 LN = natural log
0 s 1 s
Log decrement = (1/n)[LN(A0/An)] = (1/5)[LN(1.633/0.268)] = 0.36

Damping ratio = Log dec/2Pi = 0.36/2Pi = 0.36/6.28 = 0.057
Amplification factor = 1/(2*Damping) = 1/(2*0.057) = 8.68

Using the Spectrum
X:65.5 Hz Y:12.23725 mG
X:70.75 Hz Y:8.475402 mG
0.015 F = 65.5 Hz
G
rms f1 = 58.75 Hz f2 = 70.75 Hz
Real
0
-3dB
0 Hz 100 Hz
Find the –3dB points = AF * .707 = 12.24 mG * .707 = 8.65 mG

Find the frequencies at the –3dB points (f1 and f2)
Amplification factor = F/ (f2 - f1) = 65.5/(70.75 – 58.75) = 5.46

Bump Testing Summary
¾ Take your time ¾ Do a mental health
¾ Choose your check
weapon ¾ Calculate the
¾ Bump around amplification factor
¾ Uniform Window ¾ Change the mass
¾ Look at the time ¾ Change the stiffness
waveform
¾ Look at the ¾ Add damping
frequency spectrum ¾ Bump around

1x (Running Speed)
9 Mass Unbalance 1X
9 Critical Speed 1X
9 Misalignment 1x, 2x, 3x
9 Looseness 1X, 2X, 3X, 4X, 5X, ….Nx
9 Runout 1X

1x Mass Unbalance
X:30 Hz
X:60 Hz
Y:584.5464 minch
Y:88.18431 minch
fT = 1 ?
FREQ 1
0.7
inch
1x 1600 Lines
rms
Magnitude
2x
0
0 Hz 100 Hz
TIME 1
1.5
inch
Real
-1.5
0 s 15.99609 s

1x Mass Unbalance
Ft = 1 ?
FREQ 1
0.7
inch
1600 Lines
rms
Magnitude
0
0 Hz 6.4 kHz
TIME 1
1.5
inch
Real
-1.5
0 s 249.939 ms

FT ≠ 1 !
1x Mass Unbalance
But it makes a
nice set of plots
X:30 Hz
X:60 Hz
Y:584.5464 minch
Y:88.18431 minch
to analyze !
FREQ 1
0.7
inch
1x
rms
Magnitude
2x
0
0 Hz 100 Hz
TIME 1
1.5
inch
Primarily 1x
Real
-1.5
0 s 249.939 ms

1x, 2x, 3x Misalignment
1x
2x
1x
2x

1x, 2x, 3x Misalignment
1x 2x
Angular Offset
Misalignment Misalignment
Look for a 1800 phase shift across the coupling in

axial vibration measurements. Be careful with the way
you mount the accelerometer. Don’t create the 1800
phase shift by flipping the accelerometer around.
Rolling Element Bearings
9 Rolling element bearings will not generate
frequencies that are even multiples of
running speed. They are non-synchronous.
9 They often generate low amplitudes
9 They have stages of failure starting with
high frequency stress waves deteriorating
to low frequency components.
9 When the vibration gets better – shut the
machine off immediately!

Rolling Element Bearing Frequencies
“Inner Race Rotates”
FTF = (Hz/2)[1-(B/P)cosCA]
BPFO = (N/2)Hz[1-(B/P)cosCA]
BPFI = (N/2)Hz[1+(B/P)cosCA]
BSF = (PHz/2B){1-[(B/P)cosCA]2}
Where:
Hz. = rotor speed in cps
N = number of rolling elements
Inner race and shaft rotate.
B = ball diameter
Outer race is held or fixed. P = pitch diameter
CA = contact angle
Rolling Element Bearing Frequencies
“Outer Race Rotates”
FTF = (Hz/2)[1+(B/P)cosCA]
BPFO = (N/2)Hz[1+(B/P)cosCA]
No BPFI = (N/2)Hz[1-(B/P)cosCA]
Rotation
BSF = (PHz/2B){1-[(B/P)cosCA]2}
Where:
Hz. = rotor speed in cps
N = number of rolling elements
Inner race and shaft fixed.
B = ball diameter
Outer race rotates. P = pitch diameter
CA = contact angle
Approximate Rolling Element
Bearing Frequencies
“Approximate Calculations when Inner Race Rotates”
Ball Pass Frequency Outer Race

BPFO = .41 x number of rolling elements x speed
Ball Pass Frequency Inner Race

BPFI = .59 x number of rolling elements x speed
Ball Spin Frequency

BSF = .22 x number of rolling elements x speed
Fundamental Train Frequency (Cage Frequency)

FTF = .41 x speed

Rolling Element Bearings (BPFI)
9 - CENTER ROLL
532E044D -MIH MOTOR INBOARD HORIZONTAL
1.2
Route Spectrum
21-Feb-04 08:37:46
SKF 6326 OVERALL= 5.20 V-AN

PK = 2.13
LOAD = 100.0
0.9 7.66 FTF RPM = 1174. (19.57 Hz)
PK Velocity in mm/Sec
43.01 BSF
0.6
61.31 BPFO
95.26 BPFI
0.3
0
Freq: 589.03
0 300 600 900 1200
Ordr: 30.10
Frequency in Hz Spec: .289
Dfrq: 94.91

Rolling Element Bearings (BPFI)
9 - CENTER ROLL
532E044D -MIH MOTOR INBOARD HORIZONTAL
20
Route Waveform
21-Feb-04 08:37:46
15
RMS = 3.52
LOAD = 100.0
RPM = 1506. (25.09 Hz)
10 CF ALARM
PK(+) = 17.23
PK(-) = 17.94
CRESTF= 5.10
Acceleration in G-s
PK ALARM
0
-5
PK ALARM
Angel Fish
-10
Impacts create
CF ALARM
Resonance of
-15
Inner Ring
-20
0 50 100 150 200
Time in mSecs

ft = 1 ?
t is very small
F is very high
F max
ft = 1 ?
t is longer
f is lower
F max
ft = 1 ?
T is really long
f is really low
F max
Rolling Element Bearings ?
As the frequency gets lower
bad things are happening !



Gear Mesh
9 Number of Teeth x Speed of the Shaft
it is mounted on.
9 Sidebands around gear mesh will be
spaced at the shaft speed the gear is
mounted on.
9 Typically the vibration will be in the
axial direction

Gear Mesh & Shaft Speeds
1776 RPM
25T 29T
(29.6 Hz)
16.09 Hz (965.2 RPM)
46T 3.13 Hz (187.9 RPM)

149T
Shaft Speeds Gear Mesh
Inter Speed = 29.6(25/46) = 16.09 Hz GMH = 29.6 x 25 = 740 Hz

16.09 x 60 = 965.2 CPM 740 x 60 = 44,400 CPM
Output Speed = 16.09(29/149) = 3.13 Hz GML = 16.09 x 29 = 466.6 Hz
3.13 x 60 = 187.9 CPM 466.6 x 60 =27,996 CPM

Gear Mesh with Sidebands of Shaft
Speed
X:30.59605 Hz Y:31.80463 mpsi

X:31.82788 Hz Y:89.65971 mpsi
X:33.05971 Hz Y:25.62417 mpsi
Gear Mesh = 31.828 Hz
1
0.1
psi Sideband spacing = 1.232 Hz
rms 1.232 Hz x 60 = 73.9 CPM
Magnitude
73.9 RPM = Shaft Speed
0
20 Hz
Zoom Window 40 Hz

Fans
9 Blade Pass
9 Number of Blades x Speed of the Shaft
the rotor is mounted on.
9 Look at the damper and duct work for
flow and restrictions.
9 Blade clearance, discharge angle, wear
& tear
9 Unbalance, misalignment, bearings

Pumps
9 Vane Pass
9 Number of Vanes x Speed of the Shaft the rotor is mounted on.
9 Look at the input and output pressures
9 Vane clearance, discharge angle, wear & tear
9 Recirculation
9 Random noise in FFT & Time Waveform
9 Axial shuttling, High back pressure, Low flow rate
9 Fluid being forced back into pump
9 Cavitation
9 Random noise in the FFT & Time Waveform
9 Audible noise, Low back pressure, High flow rate
9 Air entrained in fluid
9 Unbalance, misalignment, bearings

Motors (synchronous)
9 Synchronous Speed
9 (2 x Line Frequency)/number of poles
9 Stator
9 2 x Line Frequency and Multiples
9 Rotor
9 Sidebands Around Running Speed =
Slip Frequency x Number of Poles
with Multiples
9 Unbalance, Misalignment, Bearings

Bibliography
9 Eisenmann, Robert Sr. & Eisenmann, Robert Jr., Machinery
Malfunction Diagnosis and Correction, ISBN 0-13-240946-1
9 Eshleman, Ronald L., Basic Machinery Vibrations, ISBN 0-
9669500-0-3
9 Vibration Institute, Basic Machinery Vibrations & Machinery
Vibration Analysis
9 LaRocque, Thomas, Vibration Analysis Design, Selection,
Mounting, and Installation, Application Note, Connection
Technology Center
9 Agilent Technologies, The Fundamentals of Signal Analysis,
Application note 243
9 Agilent Technologies, Effective Machinery Measurements using
Dynamic Signal Analyzers, Application note 243-1

Thank You !
You can find technical papers on
this and other subjects at
www.ctconline.com
in the “Technical Resources” section

Introduction To Basic Vibration Analysis: By: Jack D. Peters

Загружено:

Сведения о документе

Исходное описание:

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Introduction To Basic Vibration Analysis: By: Jack D. Peters

Загружено:

Авторское право:

Доступные форматы

Introduction to

9 Senior Engineer for Vibration

9 24/7 Permanent Monitoring using

Jack D. Peters - CTC 2

9 Past Chairman CNY

Jack D. Peters - CTC 3

Jack D. Peters - CTC 4

Auto Pwr Spec 1

Jack D. Peters - CTC 6

Jack D. Peters - CTC 7

9 Route Based 9 Data Analysis

Jack D. Peters - CTC 8

Rack mounted systems with

Jack D. Peters - CTC 9

Smaller portable units

Jack D. Peters - CTC 10

Jack D. Peters - CTC 12

Jack D. Peters - CTC 13

Jack D. Peters - CTC 14

Jack D. Peters - CTC 15

Jack D. Peters - CTC 16

Jack D. Peters - CTC 17

Jack D. Peters - CTC 18

Jack D. Peters - CTC 19

Jack D. Peters - CTC 20

Jack D. Peters - CTC 21

Jack D. Peters - CTC 22

Jack D. Peters - CTC 23

Jack D. Peters - CTC 24

If: F = 1/T and T = 1/F

Jack D. Peters - CTC 25

Jack D. Peters - CTC 26

Jack D. Peters - CTC 27

Jack D. Peters - CTC 28

Jack D. Peters - CTC 29

Jack D. Peters - CTC 30

Jack D. Peters - CTC 31

Jack D. Peters - CTC 32

Jack D. Peters - CTC 33

The frequency span is calculated as the ending

Example: 0 - 400 Hz using 800 lines

Jack D. Peters - CTC 34

Flat Top Window Function = 3.8

Example: 0 - 400 Hz using 800 Lines & Hanning Window

Jack D. Peters - CTC 35

Example: 0 - 400 Hz using 800 Lines & Hanning Window

Jack D. Peters - CTC 36

The student wishes to measure two frequency

Jack D. Peters - CTC 37

Resolution = 30 - 29.5 = 0.5 Hz / Line

Jack D. Peters - CTC 38

TSample = Nlines / Fspan

Using 400 lines with a 800 Hz frequency span will require:

Jack D. Peters - CTC 39

Jack D. Peters - CTC 40

Jack D. Peters - CTC 41

Jack D. Peters - CTC 42

Jack D. Peters - CTC 43

Window functions courtesy of Agilent

Jack D. Peters - CTC 44

Jack D. Peters - CTC 45

Jack D. Peters - CTC 46