You are on page 1of 178

Techncial Associates Field Dynamic Balancing

Copyright 2001 Techncial Associates of Charlotte, P.C.


Section Subject Page
1. INTRODUCTION TO DYNAMIC BALANCING......................................................1-1
2. TYPES OF UNBALANCE........................................................................................2-1
A. Static Unbalance............................................................................................. 2-2
B. Couple Unbalance.......................................................................................... 2-3
C. Quasi-Static Unbalance .................................................................................. 2-4
D. Dynamic Unbalance ....................................................................................... 2-5
3. TYPES OF BALANCE PROBLEMS........................................................................3-1
A. Rigid Vs Flexible Rotors.......... 3-1
B. Critical Speeds.......... 3-4
4. HOW TO ENSURE THE DOMINANT PROBLEM IS UNBALANCE. ... 4-1
A. Review of Typical Spectra and Phase Behaviors for Common
Machinery Problems...................................................................................4-1
1. Mass Unbalance....................................................................................... 4-1
2. Eccentric Rotor.......................................................................................... 4-1
3. Bent Shaft ................................................................................................. 4-3
4. Misalignment............................................................................................ 4-3
5. Resonance................................................................................................ 4-3
6. Mechanical Looseness/Weakness ........................................................... 4-4
B. Summary of Phase Relationships for Various Machinery........................4-4
1. Force (or Static) Unbalance....................................................................... 4-4
2. Couple Unbalance.................................................................................... 4-4
3. Dynamic Unbalance ................................................................................. 4-4
4. Angular Misalignment............................................................................... 4-6
5. Parallel Misalignment............................................................................... 4-6
6. Bent Shaft ................................................................................................. 4-6
7. Resonance................................................................................................ 4-6
8. Rotor Rub.................................................................................................. 4-6
9. Mechanical Looseness/Weakness Due to Base/Frame
Problems or Loose Hold Down Bolts ....................................................... 4-6
10. Mechanical Looseness Due to a Cracked Frame.................................... 4-6
C. Summary of Normal Unbalance Symptoms ..............................................4-6
1. Special Characteristics ............................................................................. 4-6
2. Centrifugal Force Due to Unbalance......................................................... 4-6
3. Unbalance Force Directivity...................................................................... 4-7
4. Radial/Axial Vibration Comparison .......................................................... 4-7
5. Overhung Rotor Unbalance Directivity...................................................... 4-7
6. Steadiness & Repeatability of Phase Due To Unbalance ......................... 4-8
7. Resonant Amplitude Magnification........................................................... 4-8
8. Phase Behavior For Dominant Static,Couple & Dynamic Unbalance ....... 4-8
TABLE OF CONTENTS AND SEMINAR AGENDA
Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
Techncial Associates Field Dynamic Balancing
Section Subject Page
5. CAUSES OF UNBALANCE ....................................................................................5-1
A. Assembly Errors ............................................................................................. 5-1
B. Casting Blow Holes......................................................................................... 5-1
C. Fabrication Tolerance Problems...................................................................... 5-1
D. Key Length Problems ...................................................................................... 5-1
E. Rotational Distortion ........................................................................................ 5-3
F. Deposit Buildup or Erosion .............................................................................. 5-3
G. Unsymmetrical Design.................................................................................... 5-3
6. WHY DYNAMIC BALANCING IS IMPORTANT ....................................................6-1
7. UNITS OF EXPRESSING UNBALANCE................................................................7-1
8. VECTORS...............................................................................................................8-1
9. DYNAMIC FIELD BALANCING TECHNIQUES.....................................................9-
A. Recommended Trial Weight Size.................................................................... 9-
B. How a Strobe-Lit Mark On a Rotor Moves When a Trial Weight is Added........ 9-
C. Single-Plane Balancing Using a Strobe Light And a Swept-Filter Analyzer .... 9-
D. Single-Plane Method of Balancing.................................................................. 9-
E. Balancing in One Run ..................................................................................... 9-
F. Two-Plane Balancing Techniques .................................................................... 9-
G. Cross-Effects ................................................................................................... 9-
H. Single-Plane Method For Two-Plane Balancing.............................................. 9-
I. Vector Calculations For Two-Plane Balancing.................................................. 9-
J . Rotor Balancing By Static Couple Derivation.................................................. 9-
K. Single-Plane Balancing With Remote Phase And A Data Collector................. 9-
L. Taking Phase Readings With A Data Collector................................................. 9-
M. Single-Plane Balancing Using A Data Collector ............................................. 9-
N. Two-Plane Balancing Using A Data Collector.................................................. 9-
O. Overhung Rotors ............................................................................................. 9-
P. Multi-Plane Balancing .................................................................................... 9-
Q. Splitting Balance Correction Weights.............................................................. 9-
R. Combining Balance Correction Weights Using Vectors .................................. 9-
S. Effect of Angular Measurement Errors of Potential Unbalance Reduction ....... 9-
1. Effect of Phase Angle Measurement Errors By Instruments ...................... 9-
2. Effect of Angular Measurement Errors When Attaching
Balance Correction Weights .................................................................... 9-
TABLE OF CONTENTS AND SEMINAR AGENDA
Field Dynamic Balancing
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
Section Subject Page
10. Balancing Machines - Soft-Bearing Vs Hard-Bearing Machines ...................10-
A. Soft-Bearing Machine................................................................................... 10-
B. Hard-Bearing Machine ................................................................................ 10-
11. Recommended Vibration And Balance Tolerances..........................................11-
A. Vibration Tolerances..................................................................................... 11-
1. Recommended Overall Vibration Specifications.......................................... 11-
2. Synopsis Of Spectral Alarm Band Specifications ......................................... 11-
B. Balance Tolerances On Allowable Residual Unbalance................................. 11-
1. ISO 1940 Balance Quality Grades................................................................ 11-
a. Application of Tolerances to Single-Plane Problems.......................... 11-
b. Application of Tolerances to Two-Plane Problems.............................. 11-
c. Application of Tolerances to Special Rotor Geometries ...................... 11-
C. How to Determine Residual Unbalance Remaining in a Rotor After
Balancing .................................................................................................... 11-
D. Comparison of ISO 1940 With API and Navy Balance Specifications .......... 11-
APPENDIX A Balancing Terminology
APPENDIX B Weight Removal Charts
APPENDIX C Conversion Chart for Converting Inches of Flat Stock # 1020 Steel to
Ounces of Weight
APPENDIX D Three-Point Method of Balancing
TABLE OF CONTENTS AND SEMINAR AGENDA
Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
Techncial Associates Field Dynamic Balancing
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
Copyright 2001 Techncial Associates of Charlotte, P.C.
Techncial Associates Field Dynamic Balancing
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
Copyright 2001 Techncial Associates of Charlotte, P.C.
Techncial Associates Field Dynamic Balancing
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
RECOMMENDED PERIODICALS FOR THOSE INTERESTED IN
PREDICTIVE MAINTENANCE
1. Sound and Vibration Magazine
P.O. Box 40416
Bay Village, OH 44140
Mr. J ack Mowry, Editor and Publisher
Phone: 216-835-0101
Fax : 216-835-9303
Terms: Normally free for bona fide qualified personnel concentrating in the Sound and
Vibration Analysis/Plant Engineering Technologies. Non-qualified personnel -
$25/per year within the U.S.
Comments: This is a monthly publication that normally will include approximately 4-6 issues
per year devoted to Predictive Maintenance. Their Predictive Maintenance articles
are usually practical and in good depth; normally contain real meat for the
PPM vibration analyst. Sound and Vibration has been published for over 25
years.
2. Vibrations Magazine
The Vibration Institute
6262 South Kingery Hwy, Suite 212
Willowbrook, IL 60514
Institute Director - Dr. Ronald Eshleman
Phone: 630-654-2254
Fax : 630-654-2271
Terms: Vibrations Magazine is sent to Vibration Institute members as part of their annual
fee, (approx. $45 per year). It is available for subscription to non-members at
$55/per year; $60/foreign.
This is a quarterly publication of the Vibration Institute. Always contains very practical and useful
Predictive Maintenance Articles and Case Histories. Well worth the small investment.
Comments: Yearly Vibration Institute fee includes reduced proceedings for that year if desired
for the National Conference normally held in J une. They normally meet once per
year at a fee of about $675/per person, ($600/person for Institute members)
including conference proceedings notes and mini-seminar papers. All of the
papers presented, as well as mini-courses, at the meeting are filled with meat for
the Predictive Maintenance Vibration Analyst. Vibrations Magazine was first
published in 1985 although the Institute has been in existence since approximately
1972, with their first annual meeting in 1977. The Vibration Institute has several
chapters located around the United States which normally meet on a quarterly
basis. The Carolinas' Vibration Institute Chapter normally meets in Greenville, SC;
Charleston, SC; Columbia, SC; Charlotte, NC; Raleigh, NC; and in the Winston
Salem, NC areas. For Institute membership information, please contact: Dr. Ron
Eshleman at 630-654-2254. When doing so, be sure to ask what regional chapter
is located to your area. Membership fees for the Annual Meeting Proceedings are
$30/per year (normal cost is approx. $60/per year for proceedings if annual
meeting is not attended). Please tell Ron that we recommended you joining the
Vibration Institute when you call or write to him.
R-0697-1
Copyright 2001 Techncial Associates of Charlotte, P.C.
Techncial Associates Field Dynamic Balancing
3. P/PM Technology Magazine
P.O. Box 1706
Minden, NV 89423-1706 (Pacific Coast Time)
Phone: 702-267-3970; 800-848-8324
Fax : 702-267-3941
Publisher- Mr. Ronald J ames; Assistant Publisher: Susan Estes
Terms: $42/per year for qualified USA subscribers, (individuals and establishments involved with
industrial plant and facilities maintenance; subscribers must be associated in engineering,
maintenance, purchasing or management capacity). $60/year for unqualified subscribers.
Comments: This is a bi-monthly magazine with articles about all facets of PPM Technologies,
including Vibration Analysis, Oil Analysis, Infrared Thermography, Ultrasonics, Steam
Trap Monitoring, Motor Current Signature Analysis, etc. These are normally good
practical articles. Also includes some cost savings information, although does not
necessarily include how these cost savings were truly determined. P/PM Technology
also hosts at least one major conference per year in various parts of the United States.
Intensive training courses in a variety of condition monitoring technologies will also be
offered in vibration analysis, root cause failure analysis, oil analysis, thermographic
analysis, ultrasonic analysis, etc..)
4. Maintenance Technology Magazine
1209 Dundee Ave., Suite 8
Elgin, IL 60120
Phone: 800-554-7470
Fax : 804-304-8603
Publisher: Arthur L. Rice
Terms: $95/per year for non-qualified people This is a monthly magazine that usually has at least
one article relating to Predictive Maintenance using vibration analysis within each issue. In
addition to vibration, it likewise always offers other articles covering the many other
technologies now within Predictive Maintenance.
5. Reliability Magazine
PO Box 856
Monteagle, TN 37356
Phone: 423-592-4848
Fax : 423-592-4849
Editor: Mr. J oseph L. Petersen
Terms: $49 per year in USA; $73 per year outside USA.
Comments: This bi-monthly magazine covers a wide variety of Condition Monitoring Technologies
including Vibration Analysis, Training, Alignment, Infrared Thermography, Balancing,
Lubrication Testing, CMMS and a unique category they entitle "Management Focus".
NOTE: In addition to these periodicals, many of the major predictive maintenance hardware and
software vendors put out periodic newsletters. Some of these in fact do include some real
meat in addition to their sales propaganda. We would recommend that you contact,
particularly the vendor supplying your predictive maintenance system for their newsletter.
Their newsletter will likewise advise you of updates in their current products.
R-0697-1
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
Copyright 2001 Techncial Associates of Charlotte, P.C.
Techncial Associates Field Dynamic Balancing
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
1-1
CHAPTER
1
INTRODUCTION TO DYNAMIC BALANCING INTRODUCTION TO DYNAMIC BALANCING
Unbalance has been found to be one of the most common causes of machinery
vibration, present to some degree on nearly all rotating machines. Vibration due to an
unbalance while the rotor is rotating is the result of a heavy spot located at a radius from
the mass centerline producing a centrifugal force. The amount of centrifugal force will
be the result of the weight of the heavy spot, the radius of such heavy spot and the
speed the rotor is rotating. Thus, unbalance can be described as centrifugal forces that
displace the rotors mass center from the rotors rotating center. Another way to state this
in general is unbalance is a condition, which exists when vibratory forces or motion is
applied to the bearings of a rotor as a result of centrifugal forces, particularly with
respect to a rigid rotor. How far and in what manner this displacement takes place will
be discussed later in this text.
In order to reduce the amount of forces generated by this imbalance there are several
factors that we will have to understand. Before a part can be balanced certain
conditions must be met.
1. The vibration must be due to unbalance. A complete vibration analysis
needs to be performed to make sure that unbalance is the primary cause
of the vibration forces.
2. We must be able to start and stop the rotor.
3. We must be able to add or remove weight.
In most instances, weight corrections can be made with the rotor mounted in its normal
installation, operating as it normally does. This process of balancing a part without
taking it out of the machine is called IN-PLACE BALANCING. Balancing in-place
eliminates costly disassembly and eliminates the possibility of damage during the
transportation of the rotor. Rotors that are totally enclosed such as some motors, pumps
and compressors, can be removed and transported to a balancing machine. The
principles of balancing are similar either way. Before we discuss balancing, we should
first understand unbalance, where it comes from and what must be done to correct it.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 2-1
TYPES OF UNBALANCE
Figure 1 helps illustrate unbalance. Here, assuming a perfectly balanced rotor, a 5 ounce
(141.75 gram) weight is placed on the rotor at a 10-inch (254 mm) radius. This produces an
unbalance of 50 oz-in (36000 gram-mm). Note that the same 5 ounce (141.75 gram) weight
placed at 5 inches (127 mm) from the center would produce only a 25 oz-in (18000 gram-mm)
unbalance. Figure 2 illustrates the same rotor as that in Figure 1 but with a correction made
such that the rotor is balanced by countering the 50 oz-in (36000 gram-mm) original
unbalance by placing a correction of 50 oz-in (36000 gram-mm) (by placing an identical
weight at a 10 inch (254 mm) radius directly opposite the original weight).
FIGURE 1. ILLUSTRATING UNBALACING
FIGURE 2. BALANCE CORRECTION
CHAPTER
2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
2-2
Static unbalance is sometimes known as force unbalance or kinetic unbalance. Static
unbalance is a condition where the mass centerline is displaced from and parallel to the shaft
centerline as shown in Figure 4. This is the simplest type of unbalance, which has classically
been corrected for many years by placing a fan rotor on knife-edges and allowing it to roll to the
bottom. That is, when the fan wheel is released, if the heavy spot is angularly displaced from the
bottom (or 6:00 position), it will tend to roll to the bottom hopefully ending up in the 6:00 position if
the rotor was sufficiently free to rotate. So-called correction of this unbalance was then
accomplished by placing a weight opposite this location (or at about 12:00).
Referring to Figure 3, in a perfectly balanced rotor, both the shaft and mass centerlines would
coincide with one another with equal mass distribution throughout the rotor.
As mentioned before, unbalance occurs when the mass centerline does not coincide with the
shaft centerline as shown in Figure 3. The mass centerline can be thought of as an axis about
which the weight of the rotor is equally distributed. The mass centerline is also the axis about
which the part would like to rotate if free to do so. However, if the rotor itself is restricted in its
bearing, vibration will occur if the shaft and mass centerlines do not coincide. Following below
will be a discussion on each of the four major types of unbalance, which include - STATIC,
COUPLE, QUASI-STATIC and DYNAMIC UNBALANCE. Each of these types of unbalance
will be defined by the relationship between the shaft and mass centerlines of the rotor.
A. Static Unbalance
FIGURE 3. STATIC UNBALANCE
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 2-3
FIGURE 4. STATIC UNBALANCE
Actually, there are two types of static unbalance as shown in Figure 4A and Figure 4B. In
Figure 4A, the unbalance is centered directly above the rotor center of gravity (CG). Figure 4B
likewise shows static unbalance, but with equal masses placed at identical distances from the
mass centerline and rotor CG on each end. Whether the static unbalance occurs as in either
Figures 4A or 4B, each can be corrected by placement of a correction weight in only one plane
at the CG, or by attaching two weights with one half the total weight at either end assuming the
CG is equidistant from each bearing.
NOTE: This text is being written with the assumption that we would be able to attach a weight
of suitable size at the appropriate radius. If this cannot be accomplished, then the appropriate
amount of weight can be removed from the heavy spot.
B. Couple Unbalance
Couple unbalance is a condition where the mass centerline intersects the shaft centerline at
the rotor center of gravity as shown in Figure 5. Here, a couple is created by placement of
equal weights 180 opposite each other and equidistant from the CG in opposite directions. A
couple is simply two equal and parallel forces acting opposite one another, but not in the
same plane. Instead, they are offset from one another, which would tend to rotate the rotor.
Significant couple unbalance can introduce severe instability to the rotor causing it to wobble
back and forth (like a seesaw) with the fulcrum at the rotor CG.
Unlike static unbalance, couple unbalance only becomes apparent when the shaft rotates. In
other words, if the rotor is placed on knife-edges, it would not tend to rotate no matter what
position it is placed since it would be statically balanced. Like static unbalance, couple
unbalance likewise causes high vibration at 1X RPM. Unlike static unbalance, couple
unbalance will bring about a very different phase behavior, which will be discussed in Section
IV. Unlike static unbalance, couple unbalance must be corrected in two planes with corrections
180 opposite each other.
FIGURE 5. COUPLE UNBALANCE
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
2-4
The axial location of the correction couple will not matter as long as it is equal in magnitude,
but opposite in direction to the unbalance couple. For example, looking at Figure 6, placement
of the two 5 oz (141.75 gram) weights at an axial distance with 10 inches (254 mm) to the left
and right of the CG as shown will create a clockwise couple unbalance. This can be
counteracted either by placing identical 5 oz (141.75 gram) weights at a 10-inch (254 mm)
distance directly opposite the original weights or by placing 10 oz (283.5 gram) weights at an
axial distance only 5 inches (127 mm) from the CG.
Only a very few cases will a rotor have true static or true couple unbalance. Normally, an
unbalanced rotor will have some of each type. Combination of static and couple unbalance is
further classified as quasi-static and dynamic unbalance.
FIGURE 6. CORRECTION OF COUPLE UNBALANCE
C. Quasi-Static Unbalance
Quasi-static unbalance represents a specific combination of static and couple unbalance
where the static unbalance is directly in line with one of the couple moments as shown in
Figure 7. Quasi-static unbalance is that condition where the mass centerline intersects with the
shaft, but at a point other than the rotor center of gravity (CG). In Figure 7, the Figures 7A and
7B illustrate quasi-static unbalance. In Figure 7A, the unbalance mass is placed at a location
other than the CG which introduces both static and couple unbalance. In reality, Figure 7B
represents the same unbalance as that in Figure 7A. The two unbalance masses acting
opposite one another close to the CG counteract one another statically, but do not compensate
for the unbalance introduced by the unbalance mass on the top left-hand side of the rotor.
FIGURE 7A FIGURE 7B
FIGURE 7. QUASI-STATIC BALANCE
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 2-5
Figure 8 illustrates another type of quasi-static unbalance often not even considered by
analysts. In this case, assuming you had a perfectly balanced rotor, when this is connected to
an unbalanced coupling, a quasi-static unbalance is created. This is a very common type of
unbalance since most couplings are not balanced unless they are of great size or speed.
Similarly, quasi-static unbalance can be introduced by inserting the wrong size key into the
shaft or pump impeller, which again will create both a static and a couple unbalance. In each
case, the required correction in addition to a static correction at the same location as the
couple component nearest the coupling, key, etc.
FIGURE 8. UNBALANCED COUPLING CAUSING QUASI-STATIC UNBALANCE
D. Dynamic Unbalance
Dynamic unbalance is the most common type of unbalance and can only be corrected by
mass correction in at least two planes. Figure 9 illustrates dynamic unbalance which again is a
combination of both static and couple unbalance, but with unbalance masses at different
angular positions from one another as shown in Figure 9. Because the unbalance masses are
at different angular positions, dynamic unbalance is that condition where the shaft centerline
and mass centerline do not intersect with one another, nor are they parallel with one another.
As will be pointed out in Section III, dynamic unbalance causes phase differences between
the horizontal on one bearing versus the horizontal on the other bearing to be far different from
either 0 or 180. That is, the horizontal phase difference may be 60 or 180, or most anything.
However, if the horizontal phase difference is 60, the vertical phase difference should be the
same as the horizontal within one clock position (+/- 30).
FIGURE 9. DYNAMIC UNBALANCE
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 3-1
TYPES OF BALANCE PROBLEMS
CHAPTER
3
In order to achieve a satisfactory balance with the minimum number of start-stop operations it
not only is important that we recognize the type of balance problem we have (static, couple,
quasi-static or dynamic), it should now be obvious that not all balancing can be achieved by
balancing in a single correction plane. A guide to determining whether single plane, two
plane or multi-plane will be required will be determined by the ratio of the length to diameter
of the rotor along with the speed of the rotor. It is also very important to recognize whether the
rotor is flexible (one that bows)or rigid (one that maintains it geometric shape.) The L/D ratio is
calculated using the dimensions of the rotor exclusive of the supporting shaft. See Table 1.
The selection of single plane versus two-plane balancing based on the L/D ratio and rotor
speed is offered only as a guide and may not hold true in all cases. Experience reveals that
single plane balancing is normally acceptable for rotors such as grinding wheels, single-
sheave pulleys, and similar parts even through their operating speed may be greater than
1000 RPM.
TABLE 1
A. Rigid Vs Flexible Rotors
Only a few rotors are made of a single disc, but instead they are made of several discs
on a common shaft, often times in complex shapes and sizes. This makes it practically
impossible to know which disc(s) the heavy spot is located. The unbalance could be in any
plane or planes located along the length of a rotor, and it would be most difficult and time
consuming to determine where. In addition, it is not always possible to make weight
corrections in just any plane. Therefore, the usual practice is to compromise by making weight
corrections in the two most convenient planes available. This is possible because any
condition of unbalance can be compensated for by weight corrections in any two balancing
planes. This is true only if the rotor and shaft are rigid and do not bend or deflect due to the
forces caused by unbalance.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
3-2
Whether or not a rotor is classified as rigid or flexible depends on the relationship between the
rotating speed (RPM) of the rotor and its natural frequency. You will recall that every object
including the rotor and shaft of a machine has a natural frequency, or a frequency at which it
likes to vibrate. When the natural frequency of some part of a machine is also equal to the
rotating or some other exciting frequency of vibration, there is a condition of resonance.
A flexible rotor balanced at one operating speed may not be balanced when operating at
another speed. If a rotor were first balanced below 70% of the its first critical speed with the
correction weights added in the two end planes, the two correction weights added would
compensate for all sources of unbalance distributed throughout the rotor. If the rotor were
increased to above 70% of the critical speed, the rotor would deflect due to the centrifugal
force of the unbalance located at the center of the rotor as shown in Figure 10. As the rotor
bends or deflects, the weight of the rotor is moved out away from the rotating centerline
creating a new unbalance condition. It would then be necessary to rebalance the two end
planes at this new operating speed, and then the rotor would be out of balance at the slower
operating speed. The only solution to insure smooth operation at all speeds is to make the
balance correction in the actual planes of unbalance, thus a multi-plane balance.
This subject will be discussed in greater detail later in the course.
FIGURE 10. ROTOR DEFLECTION DUE TO UNBALANCE ABOVE CRITICAL SPEED
FIGURE 11. ROTOR FLEXURAL MODES
Fig. 10A
Rotor with dynamic unbalance,
balanced in two planes below
critical speed
Fig. 10B
Operating above critical
speed the rotor deflects due
to unbalance in the center
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 3-3
The rotor in Figure 10 represents the more simple type of flexible rotor. A Rotor can deflect in
several ways depending on its operating speed and the distribution of unbalance through out
the rotor. Figure 11 illustrates the first, second and third flexural modes a rotor could take while
going through the first, second and third criticals. These rotors may require that balance
corrections be made in several planes to insure smooth operation through all speed ranges.
Whether a flexible rotor requires multi-plane balancing depends on the normal operating
speeds of the rotor and the significance of rotor deflection on the functional requirements of the
machine. Flexible rotors generally fall into one of the following categories:
1. If the rotor operates at only one speed and a slight amount of deflection will
not accelerate wear or hamper the productivity of the machine, then
balancing in any two correction planes to minimize bearing vibration may
be all that is required.
2. If a flexible rotor operates at only one speed, but it is essential that rotor
deflection be minimized, then multi-plane balancing may be required.
3. If it is essential that a rotor operate smoothly over a broad range of speeds
where the rotor is rigid at lower speeds and flexible at higher speed, then
multi-plane balancing is required.
B. Critical Speeds
The rotating speed at which the rotor itself goes into bending resonance is called a critical
speed. Depending on how many bending modes the rotor goes through is dependant upon
the number of operating speeds coincide with the rotors natural frequency. In general, rotors
operating below 70% of their natural frequency are considered to be rigid rotors and above
70% of their natural frequency are considered to be flexible rotors.
When a rotor bends or deflects due to operating through its critical speed, the weight of the
rotor is moved out away from the rotating centerline creating a new unbalanced condition. This
rotor could be corrected by rebalancing in the two end planes; however, the rotor would then
be out of balance at slower speeds where there is no deflection. The only solution to insure
smooth operation at each speed is to make the corrections in the planes of unbalance, thus
multi-plane balancing. Remember, any unbalance can be corrected by making corrections in
any two balance planes, but only if the rotor is a non-flexible rotor.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 4-1
HOW TO ENSURE THE DOMINANT PROBLEM IS
UNBALANCE
CHAPTER
4
Before analysts begin balancing a machine, they should always ensure that the dominant problem
is in fact unbalance before they begin. Vibration consultants commonly report that on over one-half
the jobs on which they are requested to balance machinery, they do not in fact perform any
balancing, but find other problems requiring different corrective measures instead. An analyst
should always employ both spectral and phase behaviors for some of the more common
machinery problems, each of which can cause high vibration at 1X RPM, including eccentric rotor,
bent shaft, misalignment, resonance and even certain types of mechanical looseness/weakness.
While there are still other problems that generate 1X RPM vibration, a review of Table II will help
the analyst distinguish which problem is at hand.
It should be pointed out that the column entitled TYPICAL SPECTRUM in Table II means just that
- that is, these spectra are not intended to be all-inclusive. For example, it is quite possible for
misalignment to generate only high 1X RPM vibration in certain cases, however, they most often
generate a noticeable 2X RPM peak. Therefore, such a spectrum is shown under the TYPICAL
SPECTRUM column. Following below will be a quick review of Table II pointing out the more
common spectral and phase behaviors of the problems shown. Later, a more detailed look will be
taken specifically on unbalance symptoms.
A. REVIEW OF TYPICAL SPECTRA AND PHASE BEHAVIORS FOR COMMON
MACHINERY PROBLEMS
1. Mass Unbalance: Table II shows that mass unbalance always generates high
vibration at 1X RPM. The centrifugal forces caused by unbalance always act in the radial
direction, but can sometimes generate high axial vibration in the case of overhung rotors
like in Unbalance Case C of Table II. Pure force, or static unbalance, is evidenced by identical
phase in the radial direction on both the outboard and inboard bearings supporting the rotor.
On the other hand, pure couple unbalance is evidenced by a 180 phase difference in the
radial direction between the outboard and inboard bearings (the horizontals will be 180 out of
phase with one another as well as the outboard and inboard verticals with one another in pure
couple unbalance). Overhung rotors represent a special case of unbalance on which high
axial vibration can be generated which is in phase between the inboard and outboard
bearings supporting the overhung rotor as shown in Table II.
2. Eccentric Rotor: Like unbalance, an eccentric rotor will generate high vibration at
1X RPM of the eccentric rotor itself with the highest vibration normally being in a direction
through the centers of the two rotors as shown in Table II under ECCENTRIC ROTOR.
However, the main difference between an eccentric rotor and an unbalanced one is with
respect to phase behavior-pure unbalance will normally cause the phase difference in the
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
4-2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 4-3
horizontal and vertical directions to be about 90 while in the case of an eccentric rotor, the
horizontal and vertical phase difference will normally be either approximately 0 or 180 (each
of which indicate straight-line motion). One of the problems with an eccentric rotor occurs if
one attempts to balance the eccentric rotor. What will often result is that the balance exercise
may in fact reduce vibration in one radial direction, but increase it in the other, depending on
the amount of eccentricity.
3. Bent Shaft: A bent shaft will most always generate high axial vibration with the
greatest component being 1X RPM if bent near the shaft center, but can create a high 2X RPM
component if bent near the coupling. One of the things that sets apart bent shaft symptoms
from those of unbalance is with respect to phase behavior - a bent shaft will cause axial
vibration on the outboard bearing of a rotor to approximately 180 out of phase with respect to
that of the inboard rotor bearing, while unbalance will normally cause axial outboard and
inboard phase to be about the same.
4. Misalignment: Although misalignment normally generates a 2X RPM component
greater than or equal to 30% of the amplitude at 1X RPM, it can sometimes cause only a high
1X RPM component, particularly in the axial direction. However, one of the things that again
differentiate it from unbalance is its phase behavior - misaligned shafts will cause phase
across the coupling to be approximately 180 different, whereas unbalance will normally
cause almost equal phase on either side of the coupling. As Table II shows, angular
misalignment is evidenced by a 180 phase change across the coupling in the axial direction
whereas parallel, (or offset misalignment), causes a 180 difference in the radial direction
across the coupling. Finally, a misaligned bearing cocked on the shaft generates spectra very
similar to that of shaft misalignment. However, it can be detected by measuring at each of 4
points in the axial direction on each bearing as shown in Table II. This measurement should
show that the phase is almost the same at each of the 4 points around the clock if the bearing
is properly oriented. If there is a 180 phase difference across either points 1 and 3, or
between 2 and 4 as shown in Table II, a cocked bearing is indicated.
5. Resonance: Resonance occurs when a forcing frequency coincides with a
system natural frequency and can cause excessive vibration amplitudes. Even a small amount
of unbalance, for example, can be greatly amplified if the rotor is operating at or near a natural
frequency. Such a resonant problem is evidenced if the phase changes dramatically for only a
small change in speed (Figure 12 shows that a rotor will experience almost a full 180 phase
change when its speed passes completely through a natural frequency). At the same time, the
amplitude first increases dramatically and then decreases as the rotor passes through the
natural frequency (as shown in Table II).
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
4-4
6. Mechanical Looseness/Weakness: Table II shows three different types of
mechanical looseness, one of which is lesser known, but causes high radial vibration
predominantly at 1X RPM which again causes a spectrum almost identical to an unbalance
vibration spectrum as shown under mechanical looseness Type A. Type A looseness is
caused by a looseness or weakness of machine feet, base plate, foundation, loose hold-down
bolts at the base, distortion at the frame or base, etc. Again, the thing which sets it apart from
unbalance is its phase behavior. Referring to MECHANICAL LOOSNESS Type A in Table I
note that a problem is evidenced between the base plate and its base by a 180 phase
change between these two sections. In other words, when a phase measurement is taken, if
everything is moving together as it should, the phase should be almost identical as one
moves his probe in the vertical direction from the foot to the baseplate, and then down to the
base. One of the most important points about this type of looseness/weakness problem is that
even if one is able to temporarily correct the problem by balancing and alignment procedures,
the vibration will likely reoccur when even the least bit of unbalance or misalignment
symptoms return. They must first correct the looseness/weakness problem, then balance or
align if any further correction measures are still required.
B. SUMMARY OF PHASE RELATIONSHIPS FOR VARIOUS MACHINERY
Section A above summarized the typical spectral and phase relationships for some of the
more common machinery problems. One of the most important points that this section
hopefully made was that the key parameter that helped differentiate one problem from another
was phase. Therefore, because of the importance of phase, following below will be a
summary showing how phase generally behave for each particular problem scenario (see
Table II):
1. Force (or static) unbalance is evidenced by nearly identical phase in the
radial direction on each bearing of a machine.
2. Couple unbalance shows approximately a 180 out-of-phase relationship when
comparing the outboard and inboard horizontal, or the outboard and inboard
vertical direction on the same machine.
3. Dynamic Unbalance is indicated when the phase difference is well removed
from either 0 or 180 but importantly is nearly the same in the horizontal and
vertical directions. That is, the horizontal phase difference could be almost
anything ranging from 0 to 180 between the outboard and inboard bearings;
but the key point is that the vertical phase difference should be almost identical
to the horizontal phase difference (+/- 30). For example, if the horizontal phase
difference between the outboard and inboard bearings is 60, and dominant
problem is dynamic unbalance, then the vertical phase difference between these
two bearings should also be about 60 (+/- 30). If the horizontal phase
difference varies greatly from the vertical phase difference when high 1X RPM
vibration is present, this strongly suggests the dominant problem is not
unbalance.
4. Angular misalignment is indicated by approximately a 180 phase difference
across the coupling, with measurements in the axial direction.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 4-5
F
I
G
U
R
E

1
2
C
H
A
N
G
E

O
F

V
I
B
R
A
T
I
O
N

D
I
S
P
L
A
C
E
M
N
T

A
N
D

P
H
A
S
E

L
A
G

W
I
T
H

R
P
M

A
B
O
V
E
,

B
E
L
O
W
A
N
D

A
T

R
O
T
O
R

R
E
S
O
N
A
N
C
E
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
4-6
5. Parallel misalignment causes radial phase differences across the coupling to
be approximately 180 out of phase with respect to one another.
6. Bent shaft causes axial phase on the same shaft of a machine to approach a
180 difference when comparing axial measurements on the outboard with
those on the inboard bearing of the same rotor.
7. Resonance is shown by exactly a 90 phase change at the point when the
forcing frequency coincides with a natural frequency, and approaches a full 180
phase change when the machine passes through the natural frequency
(depending on the amount of damping present).
8. Rotor rub causes significant, instantaneous changes in phase.
9. Mechanical looseness/weakness due to base/frame problems or loose
hold-down bolts is indicated by nearly a 180 phase change when one moves
the probe from the machine foot down to its baseplate and support base.
10. Mechanical looseness due to a cracked frame, loose bearing or loose
rotor causes phase to be unsteady with widely differing phase measurements
from one measurement to the next. The phase measurement may noticeably
differ every time you speed up the machine.
C. Summary of Normal Unbalance Symptoms
Sections A and B above summarized how the analyst can ensure that the dominant problem is
unbalance. Following below will be a more detailed look at the symptoms normally present
when some type of unbalance is the major problem:
1. Special Characteristics - unbalance is always indicated by high vibration at 1X RPM
of the unbalanced part. Normally, this 1X RPM vibration will dominate the spectrum. In
fact, the amplitude at 1X RPM will normally be greater than or equal to 80% of the overall
amplitude when the problem is limited to unbalance (may be only 50% to 80% of the
overall if other problems exist in addition to unbalance).
2. Centrifugal Force Due to Unbalance - Mass unbalance produces centrifugal
forces proportional to the following equation:
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 4-7
For example, assuming a sample rotor with a 1 oz (28.35 grams) unbalance at an 18 inch
(457.2 mm) radius (U= 18 oz-in) (12,962 gram-mm) turning 6000 RPM.
F
C
= (.000001775)(18 oz-in)(6000 RPM)
2
F
C
= 1150 lbs (from centrifugal force due to unbalance alone)
That is, only a 1 oz (28.35 gram) unbalance on a 3 foot (914.4 mm) diameter wheel turning
6000 RPM would introduce a centrifugal force of 1150 lbs (521.6 kg) that would have to be
supported by the bearings in addition to the static rotor weight they must support. Importantly,
note that the centrifugal force varies with the square of RPM (that is, tripling the speed will
result in an increase in unbalance vibration by a factor of 9 times).
3. Unbalance Force Directivity - Mass unbalance generates a uniform rotating force
that is continually changing direction, but is evenly applied in all radial directions. As a
result, the shaft and supporting bearings tend to move in somewhat a circular orbit.
However, due to the fact that vertical bearing stiffness is normally higher than that in the
horizontal direction, the normal response is a slightly elliptical orbit. Subsequently,
horizontal vibration is normally somewhat higher than that in the vertical, commonly
ranging between 1.5 and 2 times higher. When the ratio of horizontal to vertical is higher
than about 5 to 1, it normally indicates problems other than unbalance, particularly
resonance.
4. Radial/Axial Vibration Comparison - When unbalance is dominant, radial vibration
(horizontal and vertical) will normally be quite higher than that in the axial direction
(except for overhung rotors).
5. Overhung Rotor Unbalance Directivity - Generally causes high 1X RPM vibration in
both the axial and radial directions. Overhung rotors most often have both static and couple
imbalance, which will normally require correction in at least two planes.
F
C
= mr
2
g
C
F
C
= .000001775 Un
2
= .00002841 Wrn
2
where:
F
c
= Centrifugal Force (lb)
u = Unbalance of Rotating Part (oz-in)
w = Weight of Rotating Part (lb)
r = eccentricity of the rotor (in)
n = Rotating Speed (RPM)
(EQUATION 1)
= Wr
(386)(16)
2n
60
2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
4-8
6. Steadiness & Repeatability of Phase Due to Unbalance - Unbalanced rotors
normally exhibit steady and repeatable phase in radial directions. When the rotor is trim
balanced, phase can begin to dwell back and forth under a strobe light as you
achieve a better and better balance, particularly if problems other than unbalance are
present.
7. Resonant Amplitude Magnification - The effects of unbalance may sometimes be
amplified by resonance. Only a slight unbalance vibration can increase by a factor of 10
up to as much as 50 times if the rotor is operated at or near resonance with a system
natural frequency.
8. Phase Behavior for Dominant Static, Couple and Dynamic Unbalance - Figure
13 illustrates typical phase measurements for a machine which has either static (Table
A), couple (Table B) or dynamic (Table C) unbalance.
Static Unbalance Phase - Table A shows a machine having dominant static unbalance. Note
that the horizontal phase difference between the #1 and #2 bearings is about 5 (30 minus
25) compared to a vertical phase difference of about 10 (120 - 110). Similarly, over on the
pump, the horizontal phase difference at positions 3 and 4 is about 10 and the vertical phase
difference is about 15.
Couple Unbalance Phase - Table B illustrates typical couple unbalance phase readings. Note
the 180 phase difference between positions 1 and 2 horizontal (210 - 30), and the 175
phase difference between positions 1 and 2 vertical (295 - 120).
Dynamic Unbalance - Table C illustrates typical behavior for dynamic unbalance. Note that the
horizontal phase difference between outboard and inboard bearings can be anything from 0
to 180. However, whatever the phase difference in horizontal, the phase difference in the
vertical should then be almost identical (within one clock position or +/- 30). In the Figure 13
example in Table C, note the 60 phase difference between positions 1 and 2 in both the
horizontal and vertical directions; while over on the pump at positions 3 and 4, the 10
difference in the pump horizontal readings compared to the 5 difference in the vertical (170 -
165).
Key Point About Unbalance Phase Behavior - Whatever the phase difference between the
outboard and inboard horizontal phase measurements on a rotor, the vertical phase difference
between outboard and inboard bearings must be about the same (within +/- 30), or the
dominant problem is not unbalance. If, for example, the horizontal phase difference on a
motor between its outboard and inboard bearings were 30, while the outboard and inboard
vertical phase difference was approximately 150, an analyst would likely waste much time
and effort if he attempted to balance the rotor.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 4-9
FIGURE 13
TYPICAL PHASE MEASUREMENTS WHICH WOULD INDICATE EITHER
STATIC, COUPLE OR DYNAMIC UNBALANCE
TABLE A
TABLE B
TABLE C
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 5-1
CAUSES OF UNBALANCE
CHAPTER
5
There are a variety of causes of unbalance. These can be summarized as follows:
A. Assembly Errors - Sometimes occur after assembly when the mass center of rotation of
one part does not line up with the mass center of rotation of the part to which it was
assembled. For example, even if both a pump impeller and the pump shaft were separately
precision balanced and then assembled, this can happen if the pump impeller had been
balanced on a balancing shaft that fit its bore within 1 mil, but then was mounted on the shaft
which itself allows a clearance of over 3 mils. This would shift the mass of the impeller/shaft
rotor away from the shaft center which would throw the assembly out of balance, or at least
cause it to have noticeably more unbalance than that when each part was separately
balanced.
B. Casting Blow Holes - Cast parts occasionally will be left with blow holes within them that
might not be detectable by visual inspection means. Depending on the diameter of the rotor
as well as its speed, this can throw it considerably out of balance.
C. Fabrication Tolerance Problems - A common problem with parts such as a sheave deals
with stack up of clearance tolerance. In this case, since the bore of the sheave is necessarily
larger than that for the shaft diameter, when a key or setscrews is employed, the take-up in
clearance shifts the rotating centerline of the sheave away from that of the shaft on which it is
mounted.
D. Key Length Problems - Use of no key or the wrong size of key can cause noticeable
unbalance problems. Mr. Ralph Buscarello of Update International points out the great
importance of employing a half-key (full key length, but half-key depth) when balancing
couplings, impellers, sheaves, etc. Figure 14 helps explain why this is important. Mr.
Buscarello recommends that a tag like the one shown in the figure should be attached to the
finish balanced rotor any time a machine part is to be balanced and then mounted on a shaft.
For example, if the coupling shown in Figure 14 had a B dimension of about 4 inches (101.6
mm) and an A dimension of 8 inches (203.2 mm), Mr. Buscarello recommends a final key
length of about 6 inches [1/2 X (8 + 4) inches] or 152.4 mm [1/2 X (203.2 + 101.6) mm].
To further illustrate, assume a machine is to be outfitted with a 1/4" X 1/4" X 6"
(6.4 mm X 6.4 mm X 152.4 mm) key, then both the coupling and the shaft should be outfitted
with1/4" X 1/8" X 6" (6.4 mm X 6.4 mm X 152.4 mm) keys when balancing. Also assume this
coupling is perfectly balanced, weighs 5 lbs (2.3 kg), will operate at 1800 RPM and will be
mounted on a 4" (101.6 mm) shaft diameter. The following will illustrate the effect of not using
the proper half-key:
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
5-2
(b) Weight of 4" long key = (1.698)(4) = 1.132 oz (4" key)
6
(c) Unused key weight if used only 4" half-key
= 1.698 - 1.132 = .283 oz (unused half-key weight)
2
(d) Distance of Key CG from shaft center = 2" radius + (1/2 x 1/8") = 2.0625"
(e) So, if 6" half-key rather than a 4" half-key used, unbalance introduced when you
insert 6" full key will be: 2.0625" x .283 lb. = .584 oz-in
(unbalance introduced by wrong half-key length used to balance the coupling).
Then, referring to the ISO balance tolerance table shown in Figure 54 (on page 11-16), let us
see how this would affect an otherwise perfectly balanced coupling installed on a rotor turning
1800 RPM. Assuming the coupling weight of 5 lbs and the unbalance of .584 oz-in introduced
by the key, this corresponds to a residual unbalance of .1168 oz-in/lb which equals .0073 lb-
in/lb. Referring to Figure 54 at 1800 RPM, this would degrade the perfectly balanced coupling
down to an ISO Balance Quality G 40, or one with a poor balance quality grade.
Of course, if no half-key were used at all when balancing the coupling, this would introduce
even more unbalance to the system. And, one of the real problems with this being a coupling
is that the weight would be overhung from the motor bearing meaning that it could introduce
considerable couple unbalance. This fact is often overlooked, particularly when dealing with
couplings, most of which are not even factory balanced unless specifically requested by the
end user.
Figure 14 SUGGESTED TAG THAT SHOULD ACCOMPANY FINISH
BALANCED KEYED ROTOR
(Ref. Practical Solutions to Machinery and Maintenance Vibration Problems, Update International).
Coupling outfitted with a 1/4" x 1/8" x 4" key
(a) Final key weight = (1/4" x 1/4" x 6" ) x .238 lb/in
3
x 16 oz/lb
= 1.698 oz (6 key)
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 5-3
E. Rotational Distortion - Sometimes, a part might be well balanced, but might distort when
rotating due to stress relieving or thermal distortion. Parts fabricated by welding process or
shaped by pressing, drawing, bending and so forth will sometimes have high internal
residual stresses. If not relieved during the fabrication, they may begin doing so over a period
of time when operating, distorting slightly and taking on a new shape. This can throw the rotor
out of balance. In addition, some machines have problems with thermal distortion caused by
such problems as uneven thermal expansion of parts when brought up to operating
temperatures. This sometimes mandates that the rotor be balanced at its normal elevated
operating temperature.
F. Deposit Buildup or Erosion - Fan or impeller wheels are often thrown well out of balance
due to buildup of deposits of dirt or other foreign matter brought into them by the pumping
fluid or air. When small pieces of these deposits break away, it can sometimes introduce
serious unbalance. On the other hand, some high-speed centrifugal compressor rotors are
susceptible to erosion from small droplets of water traveling at very high speeds which impact
the impeller rotors. This can cause uneven erosion of impeller surfaces and eventually can
introduce considerable
unbalance.
G. Unsymmetrical Design - Unbalance can be introduced if good symmetry is not
maintained in all parts. For example, rotor windings on electric motors are sometimes difficult
to keep symmetrical; the thickness in sheaves sometimes vary from on side to the other; the
density of coating finishes sometimes varies around the rotor periphery. Other problems can
affect rotor symmetry, each of which can detrimentally affect rotor balance.
In summary, all of the above causes of unbalance can exist to some degree in a rotor.
However, the vector summation of all unbalance can be considered as a concentration at a
point termed the heavy spot. Balancing, then, is the technique for determining the amount
and location of this heavy spot so that an equal amount of weight can be removed at this
location or an equal amount of weight added directly opposite.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 6-1
CHAPTER
6
WHY DYNAMIC BALANCING IS IMPORTANT
The forces created by unbalance can be among the most destructive forces in rotating
machinery if left uncorrected. Not only will these forces damage the bearing but they have
been know to crack foundations, break welds, etc. In addition, the vibration displacement due
to unbalance can be detrimental to product quality in many applications. The amount of force
created by unbalance depends on the speed of rotation and the weight of the heavy spot.
Figure 15 represents a rotor with a heavy spot (W) located at some radius (R) from the rotating
centerline.
If the unbalance weight, radius and machine RPM are known, the force (F) generated can be
found using the following formula:
F = 1.77 x (RPM/1000)
2
x ounce-inches (EQUATION 2)
In this formula the unbalance is expressed in oz-inches and (F) is the force in pounds. The
constant 1.77 is required to make the formula dimensionally correct. When the unbalance is
expressed in terms of gram-inches, the force (F) in pounds can be found by using the
following formula:
F = 1/16 x (RPM/1000)
2
x gram-inches (EQUATION 3)
For unbalance expressed in gram-mm, the force (F) in kg can be calculated using the
following formula:
F = 0.001 x (RPM/1000)
2
x gram-mm (EQUATION 4)
From these formulas it can be seen that the centrifugal force due to unbalance actually
increases by the square of the rotor RPM. For example, from Figure 16 we see that the force
created by a 3 ounce weight attached at a radius of 30 inches (90 oz-in unbalance) and
rotating at 3600 RPM is over 2000 lbs (907 kg). By doubling the speed to 7200 RPM, the
unbalance force is increased to over 8000 pounds (3630 kg). From this we can see, especially
on high-speed machines, a small amount of unbalance can create a tremendous amount of
force.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 7-1
CHAPTER
7
UNITS OF EXPRESSING UNBALANCE
Units of unbalance in a rotating work piece is normally expressed as the product of the
unbalance weight (lbs., oz., grams, etc.) and its distance from the rotating centerline (inches,
mm, etc.), see Figure 15. The units for expressing unbalance are generally oz-inches, gram-
inches, gram-mm, etc. For example, a 1 oz (28.35 gram) weight located at 10" (254 mm) from
the rotating centerline would be 10 oz-inches, (7200 gram-mm) and a 2 oz (56.7 gram) weight
located 6" (152.4 mm) from the rotating centerline would be 12 oz-in (8641 gram-mm). Figure
16 represents other examples of unbalance expressed as the product of weight and distance.
FIGURE 15. THE FORCE DUE TO UNBALANCE CAN BE FOUND IF THE
UNBALANCE WEIGHT (W), RADIUS (R) AND ROTATING SPEED ARE KNOWN
FIGURE 16. CENTRIFUGAL FORCE EXERTED BY
UNBALANCE (OZ-IN) AT VARIOUS SPEEDS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
7-2
FIGURE 17. UNITS OF UNBALANCE ARE EXPRESSED AS THE PRODUCT OF THE
UNBALANCE WEIGHT AND ITS DISTANCE FROM THE ROTATIONAL CENTER
UNBALANCE UNITS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 8-1
CHAPTER
8
Scalar quantities such as mass, time, volume, or force may be represented by a length of a
single line in any arbitrarily chosen direction. A quantity, which has both magnitude and
direction, is called a vector quantity. Describing a vector is giving it magnitude (length) and
direction.
Unbalance forces generate a magnitude equivalent to a certain number of ounces of weight or
ounce-inches and an angular direction with respect to a reference point on the rotor, can be
represented by a vector. It should be apparent that unbalance forces that tend to move the
rotor away from its axis of rotation cause a certain magnitude. These forces and their exact
location on the rotor cannot be measured directly. However, their effects on the rotor and/or
bearing supports can be measured.
An unbalance vector, then, can be described as a straight line whose length is proportional to
the amount of unbalance and the angular direction measured from a reference point.
The combined effect of several unbalances or balance weights can be determined by vector
calculations. Examples of several vectors are shown in Figures 18. In Figure 18A, vectors are
drawn to represent the radial location of weights. The length of the vector represents the radius
in inches. Figure 18B vectors are shown to represent the weight in ounces. In Figure 18C, the
vectors represent the amount of unbalance in ounce-inches.
Balancing vectors are used to represent the amount and angular location of unbalance, as
well as to measure the effect of trial weight when solving balancing problems.
VECTORS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
8-2
FIGURE 18B. VECTOR WEIGHT FIGURE 18A. VECTOR RADIUS (LENGTH)
FIGURE 18C. VECTOR UNBALANCE (RADIUS X WEIGHT)
FIGURE 18
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-1
CHAPTER
9
DYNAMIC FIELD BALANCING TECHNIQUES
Generally, it is best to balance a good majority of rotating machines in place since it can be
done under the actual operating conditions and speed which exist during operation, in its own
bearings and on its own foundation. In addition, balancing in-place eliminates the possible
damage to the rotor during disassembly and transportation to a balancing machine. Following
will be information on what techniques should be mastered in order to best accomplish field
balancing using portable balancing equipment.
Information on recommended techniques on performing field balancing including single-
plane, two-plane, multi-plane and over-hung rotors will be discussed. In addition, instructions
will be provided on directly related topics such as how to properly size trial weights, how to
split balance correction weights when is not possible to place a single weight at the angular
location specified by the solution, and how to vectorially combine the effects of several
weights into one correction weight of just the right size and at just the right location.
To begin with we will be discussing balancing using the vector method of balancing.
Although there are many instruments with balancing programs in them on the market today,
the mastering of the vector solution will give us a very good understanding of the effects that
we should get and how to read the vector to determine if we made an error in our weight
selection and location. We will later discuss the use of instruments with built in balancing
programs.
Although it is possible to balance any object with amplitude alone, we will begin our
discussion of balancing using conventional vectors, both amplitude and phase. At the end of
this chapter you will find the instructions for balancing using just the amplitude, called the
Four Point Method of Balancing.
A. Recommended Trial Weight Size
It is important that the size of the trial weight be carefully chosen as well as the location at
which the trial weight will be placed. If the trial weight is too large, damage may be done to
the machine if the trial weight happens to be installed at or close to the rotor heavy spot
producing even more vibration, particularly if the rotor is operating above critical speed. On
the other hand, if the trial weight is too small, it may bring about no significant change in
amplitude or phase that can cause significant error when calculations for the proper correction
weight and location are made.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-2
As a general rule, a trial weight should produce either/or both a 30% change in amplitude or a
30 phase change. In order to provide a sufficiently large trial weight effect, but without risking
damage to the rotor, it is recommended that a trial weight which will produce an equivalent
unbalance force at each bearing of about 10% of the rotor weight supported by each bearing
should be installed. Therefore, referring to Equation (1), a similar equation can be derived to
help the analyst choose a proper trial weight:
F
C
=.000001775 Un
2
= .00002841 Wrn
2
(EQUATION 1 Repeated)
Solving for U:
U = 563,380 F
C
(EQUATION 5)
Now, assuming the trial weight should cause a 10% effect (.10 X U),
TW = .10 X U = .10 (563,380) W = 56,338 W (with W=Bearing Load at this point)
n
2
n
2
In order to make the equation easier for the analyst to use, double the constant (56,338) so that
W can be considered the full rotor weight.
Therefore,
TW= 112,676W (EQUATION 6)
Where:
TW =Recommended Trial Weight Effect (oz-in)
W =Weight of Rotating Part (lb)
n =Rotating Speed (RPM)
n
2
n
2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-3
Now, if the radius at which the trial weight will be placed is known, the trial weight size that
should be employed can be calculated as per the following:
TW = U = mr
Therefore:
m = TW (EQUATION 7)
Where: m = Trial Weight Size (oz or grams)
r = Radius at which Trial Weight will be placed (in)
TW = Trial Weight Effect (oz-in or gram-in)
r
An example will serve to illustrate the use of these equations:
Example - The rotor shown in Figure 19 is to be balanced. It has a weight of 453.6 kg.,
operates at 1800 RPM and has a 24" (609.6 mm) diameter wheel. To determine
the recommended trial weight size (oz),
TW = 112,676 W = (112,676)(1000) = 34.78 oz-in
n
2
(1800)
2
(at 12" radius)
Then,
m = TW = 34.78 = 2.90 oz (Record trial weight size)
r 12
The centrifugal force that would be developed by this 2.90 oz (82.2 gram) trial weight is:
Centrifugal Force = (.000001775)(34.78 oz-in)(1800)
2
= 200.0 lb.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-4
is operated. In addition, the machine casings and inspection doors should be closed before
operation in case the trial weights do accidentally come off the rotor. If it is not possible to
close the casing or inspection door, a shield should be placed between the machine and
the analyst for protection. The analyst and all others should place themselves to the side of
the machine away from direction of rotation when the machine is operated. When attaching
temporary clips or set-screwed trial weights, attempt to fasten these so that the centrifugal
force is working for you to hold the weights (for example, if an analyst desires to attach a
balance clip to fan blades, fasten them on the inside of the blades so that the throat rests
against the blades inside surface).
Finally, it is a good idea to identify the location of the trial weights by marking them in case
they do happen to come off.
B. How A Strobe-Lit Mark on a Rotor Moves When a Trial Weight Is Moved
Figure 20 shows an important concept about how a phase reference mark moves relative to
the movement of a trial weight. This often confuses analysts, but really is a simple concept if
one takes a close look. In Figure 20A, a rotor is shown with the key weight at the top (or 0).
Most analysts will put their phase reference mark in line with the key weight or some other
convenient reference point, but it really does not matter exactly where the reference mark is
applied. If the pickup is located at point A, the 90 position, and has a zero response time (no
electronic lag), the strobe light will flash when the heavy spot is at the 90 position, and the
phase mark will be seen at the top or 0 position. Now please refer to Figure 20B where the
weight has been moved 90 clockwise to point B at the 180 position. Again, note that the
phase mark is still at the 0 location, or 180 away from point B where the trial weight is now
located. If the strobe light now flashes when B is at the pickup (90 position), this means that
point A written on the rotor is at the 0 position while the phase mark is over at the 270
position (180 away from the heavy spot). Note what happened. The weight was moved 90
clockwise, but the phase mark moved 90 counterclockwise. The direction of rotation does not
matter. You get the same results. The point is this: If you want to move a phase
reference mark clockwise, move the trail weight counterclockwise and vise versa.
The reference mark always will shift in a direction opposite a shift of the heavy spot;
and the angle that the reference mark shifts is equal to the angle that the heavy spot
has shifted.
FIGURE 19. EXAMPLE ROTOR TO BE BALANCED
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-5
Instruments used to measure phase may or may not have an electronic lag, however, the effects
will still be the same as discussed above. The fact that the phase shift is predictable can be used
and the lag figured once the trial weight effect has been calculated. This will be discussed in more
detail under Balancing in One Run later in this text.
FIGURE 20.
HOW A STROBE-LIT REFERENCE MARK MOVES WHEN A TRIAL WEIGHT IS MOVED
FIGURE 20A.
TRIAL WEIGHT AT LOCATION A
(90 CLOCKWISE FROM PHASE
REFERENCE MARK)
FIGURE 20A.
TRIAL WEIGHT MOVED TO LOCATION B
(180 CLOCKWISE FROM PHASE
REFERENCE MARK)
C. Single-Plane Balancing Using A Strobe Light And A Swept-Filter Analyzer
At the start of a balancing problem we have no idea how large the heavy spot is, nor do
we know where on the part it is located. The unbalance in the part at the start of our problem is
called the ORIGINAL UNBALANCE, and the vibration amplitude and phase readings that
represent the unbalance are called our ORIGINAL READINGS.
In the beginning we must tune our analyzer to a frequency of 1X RPM at which time our
strobe light will flash at a rate equal to 1X RPM. When in the filtered mode on the analyzer, this
flashing strobe will appear to freeze the rotor and our reference mark will appear to be
stopped.
For example, the part in Figure 21 has an original unbalance of 5.0 mils (127 microns)
at 120. Once the original unbalance has been noted and recorded, the next step is to change
the original unbalance by adding a TRIAL WEIGHT to the part. The resultant unbalance in the
part will be represented by a new amplitude and phase of vibration. The change caused by
the trial weight can be used to learn the size and location of the original unbalance, or where
the trial weight must be placed to be opposite the original unbalance heavy spot, and how
large the trial weight must be to be equal to the original heavy spot.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-6
FIGURE 21.
THIS ROTOR HAS AN ORIGINAL UNBALANCE
OF 5.0 MILS (127 microns) AND 120 PHASE
By adding a trial weight to the unbalanced part, one of three things might happen:
1) First, if we are lucky, we might add the trial weight right on the heavy spot. If we do, the
vibration will increase, but the reference mark will appear in the same position it did on
the original run. To balance the part, all we have to do is move the trail weight directly
opposite its first position, and adjust the amount of the weight until we achieve a
satisfactory balance.
2) The second thing that could happen is that we could add the trial weight in exactly the right
location opposite the heavy spot. If the trial weight were smaller than the unbalance, we
would see a decrease in vibration, and the reference mark would appear in the same
position as seen on the original run. To balance the part, all we would have to do is
increase the weight until we reached a satisfactory vibration level.
If the trial weight were larger than the unbalance, then its position would now be the heavy
spot, and the reference mark would shift 180, or directly opposite where it was originally.
In this case, all we would have to do to balance the part is reduce the amount of the trial
weight until we achieved a satisfactory level.
3) The third thing that can happen by adding a trial weight is the usual one where the trial
weight is added neither at the heavy spot, nor opposite it. When this happens, the
reference mark shifts to a new position, and the vibration amplitude may change to a new
amount. In this case, the angle and direction the trial weight must be moved, and how
much the weight must be increased or decreased to be equal and opposite the original
unbalance heavy spot, is determined by making a VECTOR DIAGRAM.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-7
D. Single-Plane Vector Method Of Balancing
A vector is simply a line whose length represents the amount of unbalance and whose
direction represents the angle of the unbalance. For example, if the vibration amplitude is
5.0 mils (127 microns) and the phase reference mark position is 120, the unbalance can be
represented by a line with an arrowhead (a vector) 5.0 divisions long pointed at 120 as
illustrated in Figure 22. To simplify drawing vectors, polar coordinate graph paper like that
shown is normally used. The radial lines, which radiate from the center, or origin, represent the
angular position of the vector and are scaled in degrees increasing in the clockwise direction.
The concentric circles with a common center at the origin are spaced equally for plotting the
length of vectors.
When a trail weight is added to a part, we actually add to the original unbalance. The resultant
unbalance will be at some new position between the trail weight and original unbalance. We
see this resultant unbalance as a new vibration amplitude and phase reading. In Figure 22,
our ORIGINAL unbalance was represented by 5.0 mils (127 microns) and a phase of 120.
After adding a trial weight, Figure 23A, the unbalance due to both the ORIGINAL PLUS THE
TRIAL WEIGHT is represented by 8 mils (203 microns) and a phase of 30. These two
readings can be represented by vectors. Using polar graph paper, the ORIGINAL unbalance
vector is plotted by drawing a line from the origin at the same angle as the reference mark, or
120, as shown in Figure 22. A convenient scale is selected for the length of the vector. In this
example, each major division equals 1.0 mil (25.4 microns). Thus, the ORIGINAL unbalance
vector is drawn 5 major divisions in length to represent 5 mils (127 microns). The vector for the
ORIGINAL unbalance is labeled O.
Next, the vector representing the ORIGINAL PLUS THE TRIAL WEIGHT unbalance is drawn
to the same scale at the new phase angle observed. For our example, this vector will be
drawn 8 major divisions in length to represent 8.0 mils (203 microns) at an angular position of
30 that was the new phase angle. The ORIGINIAL PLUS THE TRIAL WEIGHT vector is
labeled O + T in Figure 23A. These two vectors, together with the known amount of trial
weight, are all thats needed to determine the required balance correction - both weight
amount and location.
FIGURE 22
AN UNBALANCE OF 5 MILS (127 Microns) AT 120 CAN BE REPRESENTED BY A
VECTOR DRAWN 5 DIVISIONS LONG AND POINTING AT 120
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-8
To solve the balancing problem, the next step is to draw a vector connecting the end of the
O vector to the end of the O + T vector as illustrated in Figure 23 B. This connecting
vector is labeled T and represents the difference between vectors O and O + T
(O + T) - (O) = T. Thus, vector T represents the effect of the trial weight alone. By measuring
the length of the T vector using the same scale used for O and O + T, the effect of the
trial weight in terms of vibration amplitude is determined. For example, vector T in Figure 3B
is 9.4 mils (239 microns) in length. This means that the trial weight added to the rotor
produced an effect equal to 9.4 mils (239 microns) of vibration. This relationship can now be
used to determine how much weight is required to be equivalent to the original unbalance,
O. The correct balance weight is found following the formula:
Correction weight = Trial weight x O (EQUATION 8)
For our example, assume that the amount of trial weight added to the rotor in Figure 21 is 10
grams. From the vector diagram, Figure 23B, we know that O = 5.0 mils (127 microns) and
T = 9.4 mils (239 microns). Therefore:
Correction weight = 10 grams x 5 mils = 5.3 grams
9.4 mils
or
Correction weight = 10 grams x 127 microns = 5.3 grams
239 microns
To balance a part, our objective is to adjust vector T to make it equal in length and pointing
directly opposite the original unbalance vector O. In this way, the effect of the correction
weight will serve to cancel out the original unbalance, resulting in a balanced rotor. Adjusting
the amount of weight according to the correct formula will make vector T equal in length to
the O vector. The next step is to determine the correct angular position of the weight.
The direction in which the trial weight acts with respect to the original unbalance is
represented by the direction of vector T. See Figure 23B. Vector T can always be thought
of as pointing away from the end of the O vector. Therefore, vector T must be shifted by the
included angle (O) between vector O and vector T in order to be opposite vector O. Of
course, in order to shift vector T the required angle, it will be necessary to move the trial
weight by the same angle. From the vector diagram, Figure 23B, the measured angle (O)
between O and T is 58. Therefore, it will be necessary to move the weight 58.
A
B
FIGURE 23. THE SINGLE-PLANE VECTOR SOLUTION
T
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-9
Remember, the trial weight is moved from its position on the part through the angle
determined by the vector diagram. This is not an angle from the reference mark, but is the
angle from the initial position of the trial weight to the required position.
To determine which direction we must move the weight, i.e., clockwise or counterclockwise,
you will recall from our experiment in Figure 20 that the reference mark shifts in a direction
opposite a shift of the heavy spot. Therefore, the following rule should be used to determine
which direction the weight must be shifted:
Always shift the trial weight in the direction opposite the observed shift of the reference
mark from O to O + T. Thus if the reference mark shifts counterclockwise from O to
O + T, the trial weight must be moved in a clockwise direction. Or, if the observed phase
shift is clockwise, then the weight must be moved counterclockwise. This rule applies
regardless of the direction of rotation of the rotor.
By following these instructions carefully, the part should now be balanced. However, very
small errors in measuring the phase angle, in shifting the weight, and adjusting the weight to
the proper amount can result in some remaining vibration still due to unbalance.
If further correction is required, simply observe and record the new amplitude and phase of
vibration. For example, assume that the balance correction applied according to the vector
diagram in Figure 24 resulted in a new amplitude reading of 1.0 mil (25.4 microns) and a new
phase reading of 270. Plot this new reading as a new O+T vector on the polar graph paper
along with the original unbalance vector O as shown in Figure 24. Next, draw a line
connecting the end of the original O vector to the end of the new O+T vector to find the
vector T. Measure the length of the new T vector. In the example, Figure 24, T = 5.9 mils
(150 microns). Using the new value for vector T proceed to find the new balance correction
weight using the familiar formula:
CORRECT WEIGHT = TRIAL WEIGHT X O
T
FIGURE 24
UNBALANCE CAN BE FURTHER REDUCED BY MAKING A VECTOR DIAGRAM USING
THE NEW O+T VECTOR ALONG WITH THE ORIGINAL O VECTOR
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-10
Remember that the value for the trial weight applied to this formula is the amount of weight
presently on the rotor and not the value of the trial weight applied on the first trial run. In this
example, the original trial weight was 10 grams; however, this was adjusted to 5.3 grams as a
result of our first vector solution, Figure 23. Therefore, to solve for the new correct weight the
formula becomes:
CORRECT WEIGHT = 5.3 grams X 5.0 mils = 4.5 grams
5.9 mils
or
CORRECT WEIGHT = 5.3 grams X 127 microns = 4.5 grams
150 microns
To determine the new location for the correction weight, measure the included angle
between the original vector O and the new T vector. In the example, Figure 24, this
measured angle is approximately 5, and since the phase shift from O to the new O+T
is clockwise, the weight must be shifted 5 counterclockwise.
Applying this new balance correction should further reduce the unbalance vibration. This
procedure may be repeated as many times as necessary using the new O+T and trial
weight value, but always using the original O vector.
E. Balancing in One Run
At the start of a balancing problem, we have no way of knowing exactly how much weight is
required or where the weight must be added to balance the part. However, once a part has
been balanced, it is possible to determine how much and where weight must be added
(or removed) to balance the unit or similar units in the future - in only one run.
We have learned that the amplitude of vibration is directly proportional to the unbalance
weight. Further, we also know how much vibration will result from a given amount of
unbalance. We have also learned that the phase of the reference mark moves in a direction
opposite the shift of the heavy spot.
Should it be necessary to rebalance this rotor or a like rotor in the future, it will be a simple
matter to determine the amount of correction weight needed. We may find that we have many
like rotors in our plant and this One Run Balancing will save extensive hours attempting to
balance rotors.
An unbalance constant can be worked out for any rotor. After you have successfully
balanced the part the first time using the 4-step or vector method, simply divide the final
balance weight by the original amplitude of vibration. For example, if the original amplitude of
vibration was say, 12 mils (305 microns), and after balancing you note that a correction weight
of 18 grams has been added, then this rotor has an unbalance constant or rotor sensitivity of:
18 grams = 1.5 grams/mil
12 mils
or
18 grams = 0.059 grams/micron
305 microns
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-11
If this rotor requires rebalancing in the future, the amount of balance weight needed can easily
be determined by simply multiplying the new original amplitude times the constant of
1.5 grams/mil. In addition to the UNBALANCE WEIGHT/VIBRATION AMPLITUDE constant,
there is another constant relationship, which can be determined for finding the location of the
unbalance. The position of the heavy spot in a rotor relative to the vibration pickup is defined
as the FLASH ANGLE of the system. The flash angle of a rotor is the angle, measured in the
direction of shaft rotation, between the point where the vibration pickup is applied and the
position of the heavy spot when the strobe light flashes. See Figure 25.
FIGURE 25. FLASH ANGLE
The reference mark has nothing to do with this relationship since it can be placed
anywhere on the rotor. The reference mark simply allows us to see the position of the rotor
when the strobe light flashes.
To find the flash angle for a part, proceed as follows:
1. Note the original unbalance readings and proceed to balance the part
using the vector or 4-step method.
2. After the rotor has been balanced successfully, stop the work piece and turn it
until the reference mark is in the same position observed under the strobe
light on the original run.
3. With the rotor in this position, note the location of your applied balance
weight. This represents the location of the original light spot of the rotor. Of
course, 180 away or directly opposite the original light spot is the original
heavy spot.
4. Following the direction of shaft rotation, note the angle between the point
where the vibration pickup is applied and the position of the heavy spot.
This measured angle is the flash angle.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-12
After the weight constant and flash angle for a part have been learned, it is a simple matter to
rebalance the part in the future. The information learned by balancing one rotor can be used
for other rotors. The RPM, pickup location and machine configuration (i.e. mass, stiffness, etc.)
must be the same each time. To balance a part in one run proceed as follows:
1. Operate the machine and record the unbalance data - amplitude and phase.
2. Stop the machine and turn the rotor until the reference mark is in the same
position observed under the strobe light.
3. With the rotor in this position, measure off the flash angle from the pickup in
the direction of shaft rotation to find the heavy spot of the rotor.
4. Next, multiply the unbalance constant times the amplitude of unbalance
vibration to find the amount of weight, which must be either removed from the
heavy spot or added on the light spot directly opposite.
F. Two-Plane Balancing Techniques
Two-plane balancing is done in much the same manner as single-plane balancing. There
are, however, a number of balancing techniques commonly in use, which will yield good
results depending upon the type of unbalance problem encountered. The choice of
balancing techniques will depend on several factors, such as unbalance configuration,
length-to-diameter ratio, balance speed compared to operating speed of the rotor, rotor
flexibility and amount of cross-effect.
Two-plane balancing techniques are:
1. Separate single-plane approach - used when the rotor length-to-diameter
ratio is large.
2. Simultaneous single-plane approach - used when the rotor length-to-
diameter ratio is large and the original unbalance vector indicates a
predominantly static or dynamic unbalance configuration.
3. Force/Couple Derivation - used on overhung rotor configuration and some
standard rotors.
4. Two-Plane Vector Calculations
a) Graphical method
b) Using an automatic balancing instrument or programmable hand-
calculator.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-13
G. Cross-Effects
Two-plane balancing requires special attention because of cross-effects. Cross-effect,
sometimes called correction plane interference, can be defined as the effect on the
unbalance indication at one end of a rotor caused by unbalance at the opposite end.
Cross-effect can best be explained by assuming the rotor in Figure 26A is perfectly
balanced. Adding an unbalance in the right correction plane, Figure 26B, results in a vibration
reading at the right bearing of 5.0 mils (127 microns) at 90. At the left bearing a vibration of
.66 mils (17 microns) is also noted with a phase of 300. This vibration is due to cross-effect.
That is, the vibration at the left bearing is caused by the unbalance in the right correction
plane.
To see what this does to two-plane balancing, note that an unbalance added in the left correction
plane, Figure 26C, changes the amount and phase of vibration at the right bearing to 6.4 mils
(163 microns) at 120.
FIGURE 26. CROSS-EFFECTS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-14
H. Single-Plane Method for Two-Plane Balancing
Cross-effect must be taken into consideration when balancing in two planes. There are many
ways to do this. The most popular way is to treat each correction plane problem using the nearest
bearing for the vibration readings. With this procedure, each plane is balanced
individually, one at a time.
1. Observe the amplitude and phase of vibration at both bearings and select the
bearing with the most vibration to balance first.
2. Using the single-plane vector method described earlier, proceed to balance the end
having the highest vibration by making weight corrections in the nearest correction
plane.
3. After the first plane has been successfully balanced, observe and record the new
amplitude and phase data for the second end. These amplitudes and phase readings
are the original readings for starting the second plane balancing operation.
Balancing the first end will usually result in a new set of readings at the second end
because the unbalance in the first correction plane creating cross-effect has been
removed.
4. Using the new data, proceed to balance the second end using the standard single
plane vector technique.
5. After the second plane has been balanced, you will likely find that the first plane has
changed. This is due to the fact that the cross-effect of unbalance in the second
plane to the first plane (which was originally compensated for in the first plane) has
now been eliminated. In any case, if the change is an increase to an unacceptable
level, the first correction plane must be rebalanced. Therefore, observe and record
the new unbalance data for the first plane and using this data as your original
reading, proceed to rebalance. Do not disturb the previously applied balance
corrections. Start with a new trial weight and rebalance as a new problem.
6. If the cross-effect is especially severe, this procedure may have to be repeated
several times, alternately balancing first one end and then the other end until both
ends are balanced to an acceptable level. Each time correction planes are
changed, a new problem is started using the new original readings. Do not
disturb the previous corrections.
Because of cross-effect, the unbalance indications observed at each end of a rotor do not truly
represent the unbalance in their respective planes. Instead, each indication will be the
resultant of unbalance in the associated correction plane plus cross-effect from the opposite
end. At the start of a balancing problem, there is no way of knowing the amount and phase of
cross-effect. In addition, the amount and phase of cross-effect will be different for different
machines.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-15
The single-plane vector method for two-plane balancing is a good example of where knowing
the flash angle and weight constant of the rotor would be most helpful in reducing the
number of balancing runs. After balancing the first end, this information can be learned and
used for all subsequent balancing operations needed to reduce the unbalance at each end of
the rotor.
In some cases, extremely severe cross-effect may be encountered to make two-plane
balancing very difficult using the single-plane vector method. Some systems may reveal
cross-effect where unbalance in one correction plane has great effect on the indicated
vibration at the bearing furthest away instead of the closest bearing. When this happens, the
cross-effect is said to be greater than 100%. The rotor configurations in Figure 27 will often
have cross-effect greater than 100%. When this is encountered, one solution might be to
simply switch correction planes. For example, referring to the rotor in Figure 27A, balance in
correction plane X using the vibration readings at bearing B, and balance in correction
plane Y using the vibration readings at bearing A. A special procedure is outlined later for
balancing overhung rotors such as that illustrated in Figure 27B.
FIGURE 27
THE ROTOR CONFIGURATION SHOWN WILL OFTEN HAVE VERY
LARGE CROSS-EFFECTS
A B
I. Vector Calculations For Two-Plane Balancing
If it were not for cross-effect, two-plane balancing could be accomplished in only three
balancing runs or start-stop operations by making trial weight additions in both balancing
planes at the same time, and constructing vector diagrams to get the proper solution.
Unfortunately, cross-effect is always present to some degree. Therefore, you can expect to use
many balancing runs to get a good balancing using the single plane vector technique.
However, some machines may require from one-half hour to a full day for only one start-stop
operation. On such machines, it would be most helpful to be able to minimize the number of
balancing runs. When a considerable amount of time is required to start and stop a machine,
or where severe cross-effect is encountered, the balancing problem can be greatly simplified
by using the TWO-PLANE VECTOR METHOD.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-16
In brief, the two-plane vector solution makes it possible to balance in two planes with only
three start-stop operations. First, the original unbalance readings are recorded at the two
bearings of the machine. Next, a trial weight is added to the first correction plane and the
resultant readings at both bearings are again recorded. Finally, the trial weight is removed
from the first correction plane and a trial weight added to the second correction plane. With this
weight in the second plane, the resultant readings at both bearings are again noted and
recorded. Using the data recorded from the original and two trial runs, together with the known
amount and location of the trial weights, a series of vector diagrams and calculations make it
possible to eliminate the cross-effect of the system, and find both the amount and location of
balance weight needed in each correction plane.
The two-plane vector solution requires from 15-30 minutes to complete. Therefore, it is
essential that the data used be as accurate as possible. The most important readings taken
are phase measurements. It is suggested that a phase reference card be used. This card can
be made from a piece of vector paper and attached to cardboard of other stiff material. The
center is cut out so that it may be held up to the center of shaft and the phase mark read
directly. Other means such as a plastic phase reference card may be used. The calculation
data sheet in Figure 28 has been developed for the two-plane vector calculation to serve as a
guide and simplify recording of data. The Roman Numerals in the far left column correspond
to the steps outlined in the detailed procedure below. The NEAR END (N) refers to the bearing
observed; and the FAR END (F) refers to the opposite bearing and correction plane. Phase
measurements for both the near end and far end must be taken using the same reference
mark and phase reference card.
The procedure is as follows:
1. With the machine operating at the balancing speed and your analyzer filter properly
tuned to a frequency that is equal to 1X RPM, observe and record the original phase
for the near end (Item #1); the original amplitude for the near end (Item #2); the
original phase for the far end (item #3), and the original amplitude for the far end
(Item #4).
2. Stop the machine and add a trial weight in the NEAR END correction plane. Record
in (Item #5) the angular position of the trial weight in degrees clockwise from the
reference mark. (For example, with the trial weight in the position shown in Figure 29,
we would record 240.) Enter the amount of the trial weight as (Item # 6).
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-17
FIGURE 28. TWO-PLANE VECTORS CALCULATION DATA SHEET
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-18
FIGURE 29. RECORD THE POSITION OF THE TRIAL WEIGHT IN DEGREES
FROM THE REFERENCE MARK. HERE THE WEIGHT IS AT 240
3. With the trial weight in the near end correction plane, operate the rotor at
balancing speed. Observe and record the new phase for the near end
(Item #7); the new amplitude for the near end (Item #8); the new phase for the
far end (Item #9); and the new amplitude for the far end (Item #10).
4. Stop the machine and REMOVE the near end trial weight. Using the same
weight or different weight if you prefer, add a trial weight at the far end
correction plane. Record as (Item #11) the position of the weight in degrees
clockwise from the reference mark (as viewed from the near end). Record the
amount of this trial weight as (Item #12).
5. With the trial weight in the far end, again operate the rotor at the balancing
speed. Observe and record the new phase reading for the near end (Item #13);
and new amplitude for the end (Item #14); the new phase reading for the far
end (Item #15); and the new amplitude for the far end (Item #16).
6. Using polar graph paper, construct vectors N, F, N2, F2, N3 and F3 by
drawing each at the observed phase angle, and to a length corresponding to
the measured amplitude of vibration. For example, the vectors in Figure 30
have been drawn from the sample data in Figure 28.
NOTE: For accuracy, use the largest scale possible for
constructing your vectors.
FIGURE 30
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-19
7. Construct vector A by drawing a line connecting the end of vector N to the end
of vector N2. See Figure 31. You will note on the data form in Figure 28 that
vector A is designated A = (N2). This notation is given to indicate the direction
of vector A and means that vector A is pointing from the end of vector N towards
the end of vector N2. This direction is very important for finding the angle of
vector A, (Item #17). The angle of vector A is found by transposing vector A
back to the origin of the polar graph as illustrated in Figure 31. A parallel ruler or
set of triangles can be used to accurately transpose vector A parallel back
to the origin. For our example, the angle of vector A is 201 and is entered as
(Item #17). The amplitude of vector A, (Item #18) is found by simply measuring
its length using the same scale selected for vectors N, F, N2, etc. From our
example, Figure 31, vector A = 7.6 mils (193 microns).
Following the same procedure used to find the angle and amplitude of vector A,
proceed to find the values for vector B = (F3); A = (F2); and B = (N3).
Enter these values on the data form as (Items #19 through Items #24).
FIGURE 31
(Figure repeated for convenience of reader)
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-20
8. Do the calculations as indicated to find the values for (Items #25 through Items
#32). Note that the numbers indicated in the Calculation Procedures column of
the data sheet are all referring to Item Numbers. Thus 25 = 21-17 means that the
value of Item #25 is found by subtracting the value of (Item #17) from the value of
(Item #21).
NOTE: During the calculations, you may find that some of your answers will be
negative (-) angles or angles larger than 360. A negative angle, say -35, may be
converted to an equivalent positive angle by subtracting the angle from 360. Thus
360- 35= 325. An angle which is larger than 360 is converted to one less than 360
by subtracting 360 from the angle. For example, 463 - 360= 103.
9. Construct vectors N and BF in the same way and to the same scale used for
vectors N, F, etc. The angle and length of vector N are obtained from your
calculated data, (Items #31 and #32) to construct vector BF.
10. Following the same procedure used to construct vectors A, B, etc., in step 7
above,proceed to construct vector C = (N BF) and vector D = (F N). Find and
enter the values for vectors C and D, (Items 33 through 36).
11. Calculate the values for (Items #37 and #38) following the same procedure
outlined for (Items #25 through #32) in step 8 above.
12. Using a new sheet of polar graph paper, construct the UNITY VECTOR (U), 1.0 unit
long at an angle of 0. Note that the values for the unity vector have already been
entered on the data form as (Items #39 and #40). The unity vector is always 1.0
unit at 0 for all two-plane vector problems. A suggested scale for the unity vector
is 1.0 unit = 2.5 inches. (63.5 mm) See Figure 32.
NOTE: Do not confuse the UNITY VECTOR scale with that used to designate the
amplitude of vibration for vectors N, F, N2, etc. The unity vector can be thought of as a
dimensionless vector. This is why it is suggested that a separate sheet of graph paper
be used, to help avoid confusion.
FIGURE 32. UNITY VECTOR PLOT
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-21
13. On the same graph paper with the unity vector, construct vector B using the
same scale selected for the unity vector. The values for B are obtained from
your calculated data, (Items #37 and #38). Remember, the value of vector B
(Item #38) is expressed in units. Therefore, in the sample, Figure 32, B = 0.22
units long at an angle of 311.
14. Following the same procedure used to construct vectors A, B, etc, in step 7,
construct vector E =( B U).
Find and enter the values for vector E, (Items #41 and #42). Remember to
measure the length of B, (Item #42) using the same unity scale.
15. Calculate the values for (Items 43 through #54), following the same procedure
outlined in step 8 above. (Items #51 and #52) Represent the position and
amount of the balance weight needed for the NEAR END correction plane. The
angles for locating the balance weights are clockwise from the reference mark.
16. Before applying the new balance correction weights as indicated by (Items #51
through #54), it is suggested that a graphic check of your solution be made as
outlined below. This check will reveal whether or not any errors have been made
in the solution.
A. On a new sheet of polar graph paper, construct vector O-A from your
calculated data (Items #43 and #44); and construct vector O/B from
(Items #45 and #46). For the length of vectors, use the same scale
selected for your original vectors N, F, N2, etc.
B. Calculate the amplitude and angle values for vector /BB. Amplitude
= (Item #50 x Item #24); and the angle = (Item #49 + Item #21).
C. Calculate the amplitude and angle values for vector - A. Amplitude
= (Item #40 x Item #22); and the angle = (Item (#47 + Item #21).
D. Using the calculated values, proceed to construct vectors BB and
A to the same scale used for B and A. See Figure 33.
E. Construct vector X by adding vector A and BB. This is done by
completing the parallelogram as shown in Figure 34. The diagonal of
this parallelogram is vector X that should be equal in length but
directly opposite the original N vector.
F. Construct vector Y by adding vectors /B and - A, again by
completing the parallelogram. Vector Y should be equal in length and
directly opposite your original F vector. See Figure 34.
G. If vectors N and X or vectors F and Y are not equal and opposite, this
indicates that an error has been made in the solution.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-22
17. If the graphic check indicates that your solution has been done correctly,
proceed to make the balance corrections as indicated in step 15. Be sure the
trial weight added in step 9 has been removed.
18. With the balance corrections applied, operate the rotor and check to be sure the
vibration has been reduced to an acceptable level.
FIGURE 33. GRAPHIC CHECK SOLUTION
FIGURE 34.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-23
19. If the applied corrections significantly reduced the unbalance, yet further
correction is required, observe and record the new unbalance data - amplitude
and phase - for the near and far ends. Enter this data as items 1 through 4 on a
new two-plane vector data sheet. Also enter on the new form those items marked
by an asterisk (*) from the original data (i.e. items 5, 6, 11, 12, 17, 18, 19, 21, 25,
26, 27, 28, 41 and 42). Now, simply recalculate items 29 through 36 and 43
through 54 to find the additional balance corrections required. Do not disturb
your previous corrections. Here also, the GRAPHIC CHECK can be performed to
verify that your solution is correct before applying the additional corrections.
The procedure for applying further balance corrections can be of great value if this rotor should
require rebalancing any time in the future.
Simply attach the vibration pickups in the same position used during the original balancing,
and take phase readings using the same reference mark. Enter the new unbalance data on
the data sheet as items 1 through 4. From the original balance data, enter those items marked
by an asterisk (*) and simply recalculate to find the new required balance corrections. In
summary, once the two-plane vector calculation has been worked successfully for a particular
rotor, this rotor can be balanced in two-planes in the future in only one run.
J . Rotor Balancing By Static Couple Derivation
The static-couple method is a multi-plane (3 plane) balancing technique frequently used for
balancing large turbo-rotors. This procedure, referred to here as STATIC COUPLE
DERIVATION, is based on the premises that:
Any condition of rotor unbalance can be identified as either static unbalance, couple
unbalance or a combination of static and couple unbalance.
For a condition of combined static and couple unbalance, the static and couple unbalance
components can be vectorially derived and corrected separately.
By correcting for a portion of the derived static unbalance in a reference at or very near the
plane which includes the rotor center of gravity, the static unbalance responsible for the first
rigid (structural) resonance as well as the first rotor critical will be minimized. Further, by
correcting the couple unbalance in reference planes near the ends of the rotor, the second
rigid mode and second rotor critical will be minimized.
Although the static-couple derivation technique is clearly a compromise approach and may
not prove effective for all flexible rotor balancing, experience reveals that this technique does
produce satisfactory results when applied to slow-speed balancing of large turbo-generator
rotors. However, the application is by no means limited to multi-plane balancing of flexible
rotors, such as large motor armatures and industrial fans.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-24
One argument for this procedure, even when applied to two-plane balancing problems, is that
deriving the static and couple unbalance components allows balancing to be performed in
two reference planes simultaneously - without fighting annoying cross-effect. Others prefer this
method for in-place balancing of high-speed equipment which must pass through resonant
points during startup and coast down. By locating static and couple trial weights with the aid
of phase-lag information, the result will nearly always be a reduction in vibration.
The procedure of balancing by STATIC-COUPLE DERIVATION can be illustrated using a
typical dynamic unbalance problem with original readings of:
Original Right (OR) = 6 mils (152 microns) at 30
Original Left (OL) = 8 mils (203 microns) at 130
1) Using polar-coordinate graph paper, construct vectors OR and OL to the
same scale. See Figure 35.
2) Connect the end of vector OR to the end of vector OL and find the
mid-point of the intersecting line.
3) Draw a line from the origin to the midpoint of the interconnecting line.
This vector represents the original static unbalance So . For the example,
Figure 35, SO = 4.6 mils (117 microns) at 90. The divided
interconnecting line further represents the couple unbalance, with CR
representing the couple component acting on the right side and CL
representing the couple component on the left. In Figure 35, the couple
unbalance vectors have been transposed parallel through the origin
revealing CR = 5.4 mils (137 microns) at 343 and CL = 5.4 mils
(137 microns) at 163.
With the static and couple unbalance vectors derived, either the static or the couple unbalance
can be corrected first, whichever is preferred. Or, since true static corrections will not influence
the couple and vice versa, with a little care both the static and couple corrections can be
carried out simultaneously. However, for simplification, the following example illustrates the
correction of the static unbalance first followed by correction of the couple unbalance.
4) With the original static unbalance vector (SO) derived as illustrated in
Figure 35, proceed to apply a static trial weight. A static trial weight can be
applied as a single weight in the reference plane that includes the rotor
center-of-gravity (C.G.) as illustrated by the examples in Figure 36.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-25
If it is not possible to add or remove balance weight at the center plane of the rotor, the static
correction can be divided and added in-line at the end planes as shown in Figure 36B.
Figure 36 also illustrates that the static-couple technique can be applied to non-symmetrical
rotors as long as the moments created by the trial and correction weights are equal about the
rotor center of gravity. This may require that balance weights be adjusted to compensate in
each plane.
FIGURE 36. STATIC TRIAL/CORRECTION WEIGHT LOCATIONS
FIGURE 35. STATIC COUPLE DERIVATION UNBALANCE
VECTORS DETERMINED GRAPHICALLY
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-26
For example, assume that a static trial weight of 50 grams (T.W. = 50 grams) has been added
at the center reference plane, resulting in new unbalance reading of:
(O + T)R = 3.6 mils (91 microns) at 313
(O + T)L = 7.8 mils (198 microns) at 176
5) On polar graph paper, construct vectors (O + TR) and (O + TL) to the
same scale used for OR and OL. See Figure 37.
6) Connect the end of vector (O +TR) to the end of vector (O + TL) and find
the midpoint of this interconnecting line.
7) Draw a line from the origin to the midpoint of the interconnecting line. This
vector represents the original-plus-trial static unbalance (SO + T). From
Figure 37, SO + T =2.9 mils (74 microns) at 202.
NOTE: It is significant to note in Figure 37 that the interconnecting line
joining (O + TR) and (O + TL) is equal in length and parallel to the
interconnecting line in Figure 35. This indicates that the static trial weight
has been located in the reference plane which includes the center of
gravity, and the couple unbalance has not been disturbed. Determining
that static and couple trial weights do not disturb one another is important,
particularly when working both static and couple solutions simultaneously.
FIGURE 37. VECTOR DIAGRAM SHOWING RESULTS OF ADDING TRIAL WEIGHT
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-27
8). With vectors SO and SO + T derived from Figures 35 and 37, a standard
single-plane vector solution can be completed to correct for the static
unbalance. This vector is shown in Figure 38.
The resultant vector ST = 6.3 mils. From this the correct weight can be
calculated.
CW =TW X O
T
= 50 grams X 4.6 mils
6.3 mils
= 36.5 grams
or
CW = 50 grams x 117 microns
160 microns
= 36.5 grams
The included angle between SO and ST is 25, and the direction of phase
shift from SO to SO + T is clockwise. Thus, the 50 gram trial weight
should be reduced to 36.5 grams and shifted 25 counterclockwise to
correct for the static unbalance.
9). Repeat steps 5 through 8 above as required to reduce the static
unbalance to acceptable limits. If the static unbalance was completely
corrected, the result would be equal amplitudes by opposite (180) phase
readings, indicating that only the couple unbalance remains. For the
example given in Figure 35, eliminating the static unbalance would result
in a couple unbalance with readings of:
CR = 5.4 grams at 343
CL = 5.4 grams at 163
FIGURE 38. VECTOR DIAGRAM SHOWING STATIC PART OF COMPUTATION
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-28
10) The derived or resultant couple unbalance is corrected by applying trial
weights in the form of a couple as illustrated in Figure 39. The vector
calculations can be carried out using measurements from either the right
or left side. When a trail weight or correction weight is applied at one end,
simply apply an equal weight at the other end - 180 away.
11) Using trial and correction weights in the form of a couple, proceed to
balance for the couple unbalance using standard vector methods.
FIGURE 39. COUPLE UNBALANCE CORRECTIONS
When the static-couple technique is used for slow speed balancing of turbo-generator rotors or
other flexible rotors, a common practice is to divide the final static correction over several
reference planes along the length of the rotor instead of concentrating the entire correction at
the C.G. reference plane. This practice is based on the assumption that the static unbalance of
the rotor will not all be concentrated at the center, but will generally be distributed along the
entire rotor. Thus, if the static unbalance of the entire rotor was corrected by a single weight in
the center, this would very likely over compensate for unbalance at the center, resulting in
excessive deflection at the rotor bending critical speed and excessive unbalance vibration at
normal operating speeds. In some cases, the static balance may be distributed equally
among the number of reference planes available. In other cases, a measure of rotor run-out is
used as the criteria for determining static correction distribution. If there is no run-out, the
weight may be divided evenly. If run-out exceeds a particular value, then a higher percentage
of the static correction might be applied to the center plane with less added at the end planes.
K. Single-Plane Balancing With Remote Phase And A Data Collector
The use of a programmable Data Collector used in a Predictive Maintenance Program and
loaded with a Balancing Program has become a very successful method of balancing. Even
though this technology has made it very easy to collect and store the data, and even perform
the balancing correction data, the setup of this instrument and a good understanding of
balancing effects is still the most important part of obtaining good results.
There are several Data Collectors on the market today with down-loadable balancing
programs, both single plane as well as two-plane. They may include other functions of
balancing such as Splitting Weights, Combining Weights, Safe Trail Weight Calculations,
Static-Couple Solutions and Result Storage for rebalancing of like rotors in One-Run.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-29
It will not be the intent in this text to describe in detail the operation of each of the various Data
Collector balancing programs but instead to give a general overview of them and the
application practice to successfully balance rotors. During the practical work exercise that is
performed during the formal training, each data collector or the data collector balancing
programs of choice will be demonstrated and used to solve balancing programs.
The previous chapters of this manual have gone into some detail on the principles of
balancing. Although the data may be obtained in a different manner, these principles of
balancing still remain in effect except the one that will be noted, especially in respect to which
direction we move our weights.
When using a data collector there will be different methods of collecting or viewing the phase
readings as well as the amplitude readings. Figure 40 shows a typical setup using a Data
Collector when performing balancing operations.
FIGURE 40. FIELD BALANCING SETUP USING A PORTABLE DATA COLLECTOR
All of the Data Collectors have a menu selection for selecting the type of balancing you want to
perform. The menus of each and the way they are accessed may be different but essentially each
data collector will perform the same functions.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-30
As we remember from our earlier lessons, before we attempt to balance a rotor a complete
analysis needs to be performed to insure that the vibration is indeed occurring at a frequency
equal to 1X RPM of the rotor. Using our data collector, we can take readings in all three
directions, i.e. Horizontal, Vertical and Axial at each bearing location. This can be
accomplished either in the Route Mode or Analysis (Off-Route) mode. In addition, we also
need to capture phase readings at these locations to assist us in determining that we have
predominate vibration due to unbalance.
L. Taking Phase Readings With a Data Collector
Taking phase readings with a Data Collector is somewhat different than that which was
discussed in the earlier chapters of this text. With most Data Collectors we have the option of
taking phase readings with a Strobe Light, a Photocell, a Laser-Tach or a Keyphasor (Proximity
Probe). Following will be the method in which the phase is calculated using the Data Collector
with a reference pickup.
Using a tach pulse from a photocell, laser-tach or proximity probe connected to the Data
Collector and a vibration pickup the vibration signal is obtained from the transducer and the
phase obtained from the reference pickup. The phase angle is determined from the difference
in time between the tach pulse and the positive peak amplitude of the vibration signal. This
difference is expressed as a ratio that is multiplied by 360 degrees to give us the phase
difference. This is done internally in the Data Collector.
This method of phase measurement is more accurate than the hand-held strobe light method
since the instrument measures the phase angle within very accurate tolerances, whereas the
strobe methods include human error in attempting to accurately read the angular position of
the reference mark. Figure 40 shows a typical setup where a piece of reflective tape will cause
the light from the photocell to reflect back to provide a 1X RPM speed reference pulse and
phase angle.
M. Single-Plane Balancing Using Data Collector
1. MOUNT TRANSDUCER
Mount the transducer in the direction with the highest amplitude of vibration. This
will usually be the horizontal direction due to less support stiffness provided in
this direction. For our example, Figure 40, the transducer is mounted in the
horizontal direction at bearing A to accomplish a balance in plane 1 (the plane
nearest the C.G.).
2. INSTRUMENT SETUP
From the main menu selection, select instrument configurations, select the
transducer, sensitivity, type of reference pickup and which unit of measure will be
used. Do not change units of measurement once the balancing program has
started as the phase will differ based on the unit of measure. For example, there
is a 90 phase difference between a velocity reading and a displacement
reading.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-31
Select the Single Plane Balancing Program. Depending on the instrument you
will be using you may have to select the direction the weight will be moved, i.e.
with rotation or against rotation. Some programs let you choose while others only
go with rotation and still others go only against rotation. Again, your exact
instrument operation will be covered in the hands on exercises.
3. ESTIMATE TRIAL WEIGHT
This program feature calculates a trail weight size when the rotor weight (weight
of all rotating parts including the shaft, pulley, impeller, etc.), the run speed
(rotational speed of the rotor to be balanced), and the radius (distance from the
center of the shaft that the trial weight is to be placed) are entered. Obtain a trial
weight as per this step and set aside for balancing.
Note: It is important that an accurate set of scales be used for weighing all
weights.
4. COLLECT ORIGINAL MEASUREMENTS
Select the Acquire Data for Single Plane. Enter a rotor description at this point
for identifying if you should select to save this data for future balancing. Take the
original readings. The instrument will record the RPM, the amplitude and the
phase. This reading should be taken two or three times do ensure that the
readings are stable. Once you are satisfied that the readings are stable, store
these readings.
5. ATTACH THE TRIAL WEIGHT
Turn the rotor off and attach you measured trail weight. Make sure the trial weight
is secure on the rotor and rotate the rotor by hand if possible to ensure that the
trial weight does not strike any parts of the machine or housing. Select trial
weight data entry information on the data collector. Enter the amount of the
weight and the location of the weight in angular direction from the reference
mark. The trial weight can be placed at any location but for simplicity reasons we
place it a 0. Again, the weight can be placed anywhere on the rotor but the
location must be entered into the data collector. Depending on the instrument
you are using, you must enter this information as a direction in degrees with
rotation. In addition, with some instruments the degrees are measured from the
leading edge of the reference tape, while others are measured from the trailing
edge. Always look at the rotor in the same direction when placing the trial weight
so as not to get confused when measuring the degrees.
6 MEASURE NEW VIBRATION WITH TRIAL WEIGHT
Start the rotor and measure the resultant reading. Observe the reading before
storing the data to insure that this trial weight has had an effect of at least a 30%
change in phase and/or amplitude. If this change is not noted, stop the rotor and
increase the size of the trial weight.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-32
7. ORIGINAL CORRECTION WEIGHT
Shut the machine down and remove the trail weight. It is always a good idea to
mark the location of this weight for reference if you should forget where it was.
Advance to the next menu selection on the data collector, you will now see the
new weight amount and placement location in degrees. Make the weight
addition/removal in the location indicated on the instrument making sure you
place it in the direction indicated, i.e. with rotation or against rotation.
8. SPLITTING WEIGHTS
If the angular location for a single weight cannot be achieved due to an
obstruction or void (if the program directed the analyst to place the correction
weight between blades or spokes) the Split Vectors portion of the software can
split the single weight vectorially into two weights that can then be attached in
two more convenient locations.
9. MEASURE NEW VIBRATION WITH ORIGINAL CORRECTION WEIGHT
Start the rotor and select and measure new vibration with the original correction
weight added. Compare this vibration level to your balance criteria. If the
resulting vibration amplitude is within quality tolerance, stop the rotor and
permanently attach the original correction weight. If not within tolerance, select
Trim Balance for new weight addition or removal and location.
10 TRIM BALANCE
After taking the new readings and obtaining your trim weight readings, weigh the
weight to the desired size. Stop the rotor and add the weight to the proper
location as indicated on the instrument, again making sure that the weight is
placed at the proper location and direction. Do not remove the original correction
weight. Start the rotor and compare the new vibration amplitude with your
balance quality standards. If the new readings are with in tolerance, balancing is
now complete. If the readings still are not within balance quality standards, the
trim program can be repeated as often as necessary. Remember, always leave
the balance weights on the rotor that were placed there in previous runs.
11 COMBINE CORRECTION WEIGHTS
If more than one weight is used to achieve a satisfactory balance tolerance the
weights can be combined. Select the Add Vectors selection in your balance
program. Enter the weight amount and angular location for any two weights. The
combined vector sum and angle is noted as the Sum. If only two correction
weights were used, this single final correction weight will replace them and can
be attached as one weight, instead of two, except where you have had to do a
vector split. If more than two correction weights were used, the weights can be
combined in steps, i.e., start by combining two weights vectorially, then
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-33
combining this vector sum with the next weight, etc. until all weights are
combined into a single weight. This process must be done very carefully since
the final balance results can be greatly affected if inaccurate weight amount and
angles are entered. After combining weights, it is recommended that the rotor be
re-tested for balance with the single weight and additional corrections made if
required.
12 SECURLEY FASTEN FINAL CORRECTION WEIGHTS
If measurements are now in compliance throughout the machine with balance
and vibration specs, the analyst should take great care to securely fasten
permanent correction weights; or if they desire, remove weight at locations 180
away from the final correction weights. If possible, secure these weights in such
a manner that if they do happen to be thrown off, they will not be thrown into
areas where personnel might be stationed.
13 REPEAT VIBRATION MEASUREMENTS AT ALL MACHINE LOCATIONS
(BOTH DRIVER AND DRIVEN)
Complete vibration measurements should next be captured at each location on
both the driver and driven machine components. Ensure not only that the
unbalance problem has been resolved, but also that other problems have not
been introduced (or have now surfaced since the dominating unbalance
problem has been resolved). Also, ensure that balancing in one direction on
Plane 1 has not now caused an increase in vibration at other locations and
directions. For example, eccentric rotors will often cause vibration to go up in the
vertical direction when balancing in the horizontal or vice versa. This can also
sometimes be caused by looseness and/or resonant problems.
N. Two-Plane Balancing Using A Data Collector
When a second plane is balanced, it involves more than just two single plane balancing
exercises. When balancing is performed in two or more planes, one must consider
something known as cross-effect which is also sometimes known as corrective plane
interference. This can be defined as the change of unbalance indication on one orrection
plane caused by unbalance in the other correction plane. For example, if a rotor were perfectly
balanced on both the left and right planes and a trial weight was placed on the left plane, the
trial weight would cause vibration not only in the left plane, but also in the right. For example, it
may cause 5 mils (127 microns) of vibration on the left plane and increase the vibration on the
right from almost nothing up to say 1 or 2 mils (25 or 50 microns). Then, if the same trial weight
is moved over to the right plane, it might cause 1 or 2 mils (25 or 50 microns) additional
vibration back to the left plane. This is known as cross-effect. Because of such cross-effect,
unbalance indications observed on any one plane do not truly represent the unbalance in just
that plane. Instead, each measurement will be a combination of the unbalance in that
particular plane plus the cross-effect transmitting into this plane from other planes. When one
begins a balancing exercise, he does not know the amount and the phase of cross-effect, but
he must take this into account if he is to successfully balance the machine. This will be
covered in the following sections.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-34
1 DETERMINE IF THE DOMINANT PROBLEM IS UNBALANCE
First, and foremost, determine if the real problem is unbalance using the analysis
procedures outlined early in this text. It is not unusual to find that more than 50% of the
times an analyst will be asked to balance a machine he will find the dominant problem
to be something else. This procedure should include a complete set of measurements
that can be used as before balance measurements and later compared to after
measurements.
2 MOUNT TRANSDUCERS
Mount a transducer securely at each bearing in the radial direction of the highest
1X RPM vibration. This will normally be in the horizontal direction due to less support
stiffness than that provided by the vertical direction. In any case, both transducers must
face the same direction on both the outboard and inboard bearing housings. Two
transducers are recommended to avoid having to move one transducer between the
two bearings. For identification purposes they will be referred to as left and right
transducers.
3 MOUNT PHOTO-TACH AND REFLECTIVE TAPE
Mount photo-tach and place reflective tape on the machine using the recommended
directions given below. The three quantities needed to balance are the frequency,
amplitude and phase at 1X RPM. The accelerometer will provide the amplitude and
frequency information to the data collector. A photo-tach is connected to the data
collector to provide phase information by sensing the reflective tape each time it passes
by the photo-tach during each shaft revolution. The photo-tach can be targeted to any
portion of the rotating shaft supporting the rotor to be balanced. The reflective tape can
be mounted on any exposed, clean surface which rotates with the shaft including a
coupling, sheave, flywheel or the shaft itself. If there is an option, it is usually best to
mount it on a part with a larger diameter with a continuous, unbroken surface (i.e., not
on a broken surface such as a gear). Large diameters will give better accuracy for
angular measurements.
The photo-tach itself can be targeted on the tape at right angles to the shaft, or it can
even be targeted in a direction along the shaft axis, (for example, on the fan wheel itself
in a direction parallel to the shaft). However, the best positioning choice for the photo-
tach will normally be with it facing upwards looking at the target surface from below.
Then, should the photo-tach fall off during operation, it will not likely strike the rotating
member. In addition, if this convention is adopted, the analyst will be able to remember
how the photocell was positioned (the photo-tach position must be the same in order to
make meaningful comparative phase measurements). If the photo-tach is moved, it will
change the resultant phase readings for all subsequent measurements by the angular
amount it was altered when moved. Check for triggering of the photo-tach by slowly
rotating the shaft by hand and noting if the red LED photo-tach indicator goes out as the
reflective tape passes the photo-tach indicating triggering is taking place.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-35
4 INSTRUMENT SETUP
From the main setup menu, select instrument configuration. Choose the units that you
will use for balancing, i.e., gs, in/sec or mils. Also make sure that the sensitivity of the
transducer you are using in properly set. It is recommended that the same type of
transducer with the same sensitivity is used on both locations. If you are not using a
two-channel instrument, it is suggested that the cable from these transducers be
connected to a selector switch and then to the instrument. This will allow switching from
the left and right transducer without switching cables.
5 ESTIMATE TRIAL WEIGHT
From the menu, select estimate trial weight. This selection will calculate a trial weight
size when the rotor weight (weight of all rotating parts including the shaft, pulley,
impeller, etc.), the run speed (rotation speed of the rotor to be balanced), and the radius
(distance from the center of the shaft to the point where the trial weight is to be placed)
are entered, Obtain a trial weight as per this step and set aside for use during the
balancing.
6 COLLECT ORIGINAL READINGS
From the main balancing menu select Begin New Balance. Enter a rotor description
and notes if so desired. Start the rotor and select to begin original reading for the left
plane. It is suggested that this reading be taken two or three times to make certain that
you are getting repeatable data. Upon the completion of this data collection, switch to
the right plane and take the data from the right plane.
7 ATTACH TRIAL WEIGHT
Since this is a two-plane balance, a trial weight must be put in both the leftand right
correction planes, and vibration readings taken from both transducers while the trial
weights are at each location. To accomplish this, first stop the rotor and attach the trial
weight to the left rotor plane. Record the amount of the trial weight and its angular
location, (with or against rotation depending on instrument), from the reflective tape.
Note that even though the weight can be placed at any location on the rotor it is
recommended that the trial weight be placed at 0 making sure that you are reading the
leading edge or trailing edge of the reflective tape depending on your instrument. If the
trial weight cannot be placed at 0, measure the angle between the reflective tape and
the trial weight and enter that value. See Figure 41 for angular measurement
conventions. Use the trial weight from Step 5. Always look at the rotor from the same
direction when measuring this trial weight angle and any correction weight angles. Start
the rotor and measure the new right vibration using the right transducer. Insure
measurement repeatability. Now, switch to the left transducer and measure the new
left vibration and phase. Store this information. Stop the rotor and remove the left
plane trial weight. Attach a trial weight to the right rotor correction plane.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-36
Enter the amount and location of this weight. It is recommended this weight be the same
size as the left trial weight and be placed at the same angle that the left trial weight had
been located. If this cannot be done, enter the amount and angular location of the actual
position. Remember to always look at the rotor from the same direction when measuring
angles. Making sure that you are switched to the left transducer, measure and store
the new left vibration and phase. Then, switch to the right transducer and measure
and store the new right vibration and phase. Note during this procedure a significant
change of at least 30% amplitude and/or phase is needed for accurate correction
weight calculations. If this amount of change was not achieved, increase the trial weight
size and repeat the procedure.
8 ORIGINAL CORRECTION WEIGHT
Shut the machine down and remove the right trial weight. Select correction weight
readout on the instrument. Both the left and the right correction weight should be
displayed. Obtain the proper correction weights and attach at their proper locations in
the left and right correction planes. Remember to always view the angle from the
same direction.
9 SPLITTING WEIGHTS
If the angular location for a single correction weight cannot be achieved due to an
obstruction or void (as if between blades or spokes), the Split Vector portion of the
software can split the single weight vectorially into two weights that can then be
attached in two more convenient locations.
FIGURE 41. ANGULAR MEASUREMENT CONVENTIONS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-37
Enter the number of equally spaced positions where split weights could be attached, the
amount and angular position of the weight to be split. The result is two weights (Amt 1
and Amt 2) with the weight and angular location of each noted. Note the reflective tape or
reference position is considered Position 1, and both weights must be added to the rotor
to replace the original un-split single weight. Repeat this procedure for both correction
weights if needed.
10 MEASURE NEW VIBRATION WITH ORIGINAL CORRECTION WEIGHTS
Start the machine with correction weights from Step 8 in both the left and right
correction planes. Select the menu selection to measure new vibration making sure that
the proper transducer has been selected. Measure both the new vibration and phase
for both the left and right planes. Insure that these reading are stable. Stop the rotor.
Compare this vibration to your balance criteria. If the vibration is within tolerance,
permanently attach the correction weights. If the vibration is not in tolerance, stop the rotor
and select Trim Balance to access the new weight and location for further correction.
11 TRIM BALANCE
The trim balance screen list both the left and right corrections weight size and
angular location. Obtain trim weights equal to these and attach to the rotor in their correct
angular location in their respective left and right correction planes at their correct
angular location. Do not remove the original correction weights from Step 8 and use the
weight splitting technique of Step 9 if needed. Start the rotor and measure the new left
transducer and right transducer vibration with both correction and trim weights in place.
Check for measurement repeatability. Stop the rotor. Compare the results to your balance
criteria. If within tolerance, the balance is complete. If not within tolerance, repeat the trim
procedures list above. This procedure can be repeated as often as necessary, however,
always leave on the previous attached correction weights.
12 COMBINE CORRECTION WEIGHTS
If more than one weight is used in each (left and right) correction planes, the weights can
be combined by using the Add Vector function. Note only the weights in each plane
may be combined. That is, only the weights in the left plane can be combined to make
one left correction weight and only the weights in the right plane can be combined to
make a single right correction weight. The left and right correction weights cannot
be combined.
Enter the weight amount and angular location (in degrees from the reference tape) for any
two weights in the same plane. The combined vector sum and combined vector angle
are noted as the sum on the screen. If any two correction weights were used, this single
final correction weight will replace them and can be attached as one weight instead of
two. If more than one correction weight was used, the weights can be combined in steps
starting by combining two weights vectorially, then combining that combined vector sum
with the next weight, etc., until all weights in the same plane are combined to a single
weight. This process must be done very carefully as the final balance results can be
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-38
greatly affected if inaccurate weight amounts and angles are entered. After combining
weights it is recommended that the rotor be re-tested for balance with the left and
right combined weights in place. Additional balance corrections may be required.
13 SECURELY FASTEN CORRECTION WEIGHTS
If measurements are now in complince throughout the machine with balance and
vibration specs, the analyst should take care to securely fasten permanent correction
weights or, if he desires, remove weight at locations 180 away from the final
correction weights. If possible, secure these weights in such a manner that if they do
happen to be thrown off, they will not be thrown into areas where personnel might be
stationed.
14 REPEAT VIBRATION MEASUREMENTS AT ALL LOCATIONS (BOTH THE
DRIVER AND DRIVEN).
Complete vibration measurements should next be captured at each location on
both the driver and driven machine components. Ensure not only that the unbalance
problem has been resolved, but also that other problems are not now indicated. Also,
ensure that balancing in one direction on Plane 1 has not now caused an increase in
vibration at other locations and directions.
O. Overhung Rotors
Overhung rotors are machine configurations like that shown in Figure 42 where the fan wheel
to be balanced is outboard of its two supporting bearings. This configuration is very often
found with machines such as blowers, pumps, etc. The planes where balance corrections are
made do not necessarily respond to standard single and two-plane balancing techniques. In
addition, the unbalance planes alone will create a couple unbalance proportional to the
distance of the unbalance plane from the rotor C.G. Therefore, when attempting to balance
overhung rotors, the analyst needs to take into account both static and couple unbalance
forces, and treat the accordingly.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-39
FIGURE 42. EXAMPLE SETUP FOR BALANCING AN OVERHUNG ROTOR
When balancing an overhung rotor, one of the two following procedures should be taken:
1 BALANCING OVERHUNG ROTORS BY CLASSIC SINGLE-PLANE STATIC-
COUPLE METHOD:
Figure 42 helps explain methods of balancing overhung rotors. Classically,
Bearing A is most sensitive to static unbalance whereas the bearing farthest from
the fan wheel to be balanced (Bearing B) is most sensitive to couple unbalance.
Since Plane 1 is closest to the rotor center of gravity (C.G.), static corrections
should be made in this plane while measuring the response on Bearing A. On
the other hand, measurements should be made on Bearing B when making
couple corrections in Plane 2. However, placing a trial weight in Plane 2 will
destroy the static balance achieved at Bearing A. Therefore, in order to maintain
the static balance at Bearing A, a trial weight placement which will generate a
couple must be used. Thus, a trial weight of identical size should be placed in
Plane 1 at an angle 180 opposite the trial weight in Plane 2.
Therefore, either the data collector can be used using single-plane balance
software or the single-plane graphic technique previously explained can be
successfully employed on many overhung rotors, particularly if the ratio of rotor
length-to-diameter (L/D) is less than approximately .50 (where L is length
of the rotating component on which correction weights will be placed and D is
the diameter of this component. See Figure 42.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-40
Following below will be a description of this classic single-plane balancing technique
for overhung rotors.
a. SET UP DATA COLLECTOR OR SPECTRUM ANALYZER INSTRUMENT
The data collector, photo-tach and transducer should be set up as previously described
under Sections K & L and Figure 40 showing the two-plane balancing procedure.
Alternatively, the analyst may wish to employ either a swept-filter analyzer that drives a
strobe light, or a spectrum analyzer which will fire a photo-tach for phase
measurements.
b. TAKE INITIAL MEASUREMENTS
Take initial measurements of 1X RPM amplitude, frequency and phase before adding
any trial weights. Measurements should be taken on both the outboard and inboard
bearings in both vertical and horizontal directions. The radial direction measurement
having the highest amplitude will normally be employed for initial balancing (however,
after correcting unbalance in the radial direction, measurements will have to taken in
the other directions to ensure amplitudes in all directions likewise acceptable).
c. DETERMINE IF THE DOMINANT UNBALANCE PROBLEM IS STATIC OR
COUPLE UNBALANCE
Looking at the amplitude and phase measurements taken on both bearings in the
radial and horizontal directions, determine if the problem is dominated by either static
or couple unbalance. If phase differences between the outboard and inboard bearings
are between 90 and 180 in both the vertical and horizontal directions, the dominant
problem will be couple unbalance. On the other hand, if these differences are both
anywhere from 0 to approximately 40, a static unbalance is dominant. Of course,
phase differences ranging from approximately 40 to 140 are truly dynamic balance
once again with a combination of static and couple. If the problem appears to be mostly
couple unbalance, use couple unbalance procedures outlined below. However, if the
problem appears to be predominantly static or dynamic unbalance, employ static
balance procedures. For now, we will assume that the problem is mostly static.
d. MAKE A SINGLE-PLANE STATIC BALANCE
Referring to Figure 22, use single-plane techniques taking measurements on bearing A
and placing trial and correction weights in Plane1.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-41
e. DETERMINE IF RESULTANT VIBRATION AMPLITUDES MEET REQUIRED
CRITERIA
After completing the single-plane static balance using Plane 1, repeat vibration
measurements on both the outboard and inboard bearings in each direction (including
axial) and ensure that amplitudes now meet allowable criteria.
f. IF CONSIDERABLE COUPLE UNBALANCE NOW REMAINS, CONTINUE WITH
SINGLE-PLANE BALANCE FROM BEARING B
Overhung rotors often have large cross-effects which means that single-plane
balancing from Plane 1 will often cause high vibration on bearing B. Therefore, the
analysts will perform another single-plane balance, this time making their
measurements from bearing B farthest from the component to be balanced. When they
arrive at the single-plane correction weight solution, they should place this weight in
Plane 2; and then place an identical size correction weight in Plane 1 180 away from
the weight location in Plane 2.
g. DETERMINE IF AMPLITUDES NOW MEET ALL CRITERIA
After completing the single-plane couple correction, the analyst must again make
measurements in horizontal, vertical and axial directions on each bearing and
determine that all amplitudes now meet allowable criteria. Often, further balancing must
be done at this point beginning with another single-plane balance using Bearing A and
Plane 1 which might possibly be followed by another couple balance correction.
h. IF ALLOWABLE CRITERIA CANNOT BE MET IN ALL THREE DIRECTIONS OF
EACH BEARING, PROCEED TO TWO-PLANE BALANCE PROCEDURE
OUTLINED BELOW
Sometimes, this single-plane approach will not successfully reduce amplitudes below
allowable criteria in all three direction on each bearing, particularly if the L/D ratio is
greater than .5 or if the component to be balanced is located far away from the closest
bearing. If this happens, two-plane techniques outlined below will have to be taken.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-42
2. BALANCING OVERHUNG ROTORS BY CLASSIC TWO-PLANE STATIC-COUPLE
METHOD:
Due to the significant cross-effects that are often present in overhung rotors, two-plane
balance correction techniques often are more successful than those employing single-
plane methods. However, one of the problems with two-plane methods is that it can
sometimes be a little confusing on deciding which bearing is the left and which is the
right bearing; similarly, which plane is the left and which is the right plane? (Some data
collectors refer to these as the near and far planes as opposed to left and right;
terminology does not matter - only that the analysts remain consistent in their
convention.) Referring to Figure 42, when using two-plane techniques, Bearing A will
be considered the bearing closest to the overhung rotor while Bearing B will be closest
to the pulley. Similarly, Plane 1 will be on the inboard side of the wheel closest to the
bearings whereas Plane 2 will be outboard.
Here again, a static/couple solution will be employed when the two-plane
correction weight calculations are completed. Since most overhung rotors are so
sensitive to static unbalance, only the static correction will be placed when this static/
couple solution is obtained. Then, after trim balancing, if considerable couple
unbalance remains, the analyst will proceed to correct this as well. They should follow
the procedure outlined below:
a. SET UP INSTRUMENTS AS OUTLINED IN TWO-PLANE BALANCING
METHOD AND FIGURE 42
Here again, this same procedure can be used with either data collectors, swept-
filter analyzers or real-time analyzers. However, if using either a swept-filter or real-
time analyzer, the analyst should have a two-plane calculator program that is
capable of providing static/couple solutions.
b. TAKE INITIAL MEASUREMENTS ON BOTH BEARINGS
Here again, 1X RPM amplitude, frequency and phase should be measured in
horizontal, vertical and axial directions on both the outboard and inboard bearings.
c. COMPLETE A TWO-PLANE BALANCING PROCEDURE, BUT DO NOT
YET PLACE BALANCE CORRECTION WEIGHTS
A two-plane balance procedure like that outlined in Section L should be employed,
but final correction weights not put in place. Instead, when the trial weights, sizes
and locations are calculated for each plane, the analyst should ask for a static/
couple solution and should initially only make the static correction. For example, it
the static solution called for 1 oz (28.35 grams) in Plane 1 whereas the couple
solution called for a 2 oz (56.70) grams correction in Planes 1 & 2 180 opposite
one another, make only the static correction at this point.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-43
d. DETERMINE IF AMPLITUDES NOW MEET ALLOWABLE CRITERIA
After making the static correction in Plane 1, see if amplitudes in all three directions
on each bearing are now within compliance with allowable criteria. If not, trim as
required. Again, when the two-plane corrections are determined, ask for the static/
couple solution and once again, make only the static correction. Most of the time, the
problems are resolved at this point. However, if considerable couple unbalance still
remains, complete another two-plane procedure asking for the static/couple solution
- this time making the couple correction called for, and not the static correction.
e. DETERMINE IF AMPLITUDES NOW MEET ALLOWABLE CRITERIA
After each of the two trials making these static corrections and the single trial making
the couple correction, compare amplitudes in horizontal,vertical and axial directions
on both the outboard bearings with allowable criteria. A small percentage of the
time, the couple correction will throw the static balance back off. If this is the case, it
may require one more static correction before the rotor is successfully balanced.
P. MULTI-PLANE BALANCING
When a rotor is rigid and running well below its first critical speed, it can be balanced
successfully by employing two-plane methods. In this case, the correctionweights can be
placed in any two planes as long as they satisfy both the static and the couple unbalance
problems. However, when this rotor is operated at a speed roughly 70% of critical, it will often
experience noticeable bending, particularly as it gets closer and closer to its critical. In this
case, it becomes a flexible rotor. In addition, multi-plane balancing may be required on long
rotors (high L/D ratios) which again are subject to bending due to the effect of unbalance
forces, Figure 44 helps illustrate why a rotor operating near critical speed will likely require
multi-plane balancing. Note the different mode shapes the rotor will take when passing
through the first, second and third critical speeds. When such a rotor approaches critical, the
internal bending moments producing great deflection will have to be counterbalanced by
placement of correction weights in three or more planes. In these cases, it may be required
that one perform a two-plane balance at speeds significantly below critical and when it is
operating as a rigid rotor. Then, the rotor can be brought up to the desired operating speed
and balanced once again, but this time employing multi-plane techniques. In these cases, it
will be most important to know whether the rotor is operating near its first, second or third
critical speed since the mode shapes will dictate how many planes require correction weights.
Most often, if the rotor speed approaches 70% of critical, it will require multi-plane balancing.
Software has been developed for many of todays data collectors that will allow multi-plane
balancing.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-44
FIGURE 43. BALANCING USING A PHOTOCELL
Make sure to check the instrument set up to know if you are measuring the angle with or
against rotation. Also, always view the rotor from the same direction.
FIGURE 44
MODE SHAPE DURING FIRST, SECOND, AND THIRD CRITICAL
SPEEDS FOR A SIMPLY SUPPORTED ROTOR
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-45
Q. SPLITTING BALANCE CORRECTION WEIGHTS
1. Splitting Weights Using Data Collector Software:
Often, when balancing pumps, blowers, centrifuges and other machines having rotating
blades or vanes, balance calculations will call for correction weights to be placed between
vanes or blades. In these cases, many of todays data collectors and balance machines can
help the analysts determine how they can place smaller correction weights on two adjacent
vanes which will give the same effect as would the one correction weight which was specified
to be installed between the two vanes. In these cases, the analyst will be asked by the
software to enter the number of vanes, the amount of the single correction weight for which the
software calls, and the angle at which the software specifies the correction weight to be
placed. The software, will then calculate the size of the two correction weights, assuming that
the correction weights will be installed at the same radius as the single weight would
have been. If it is not possible to install them at the same radius, the following formula should
be invoked to determine the ultimate correction weight size based on the new radius:
Where: W
c
r
c
= W
n
r
n
(Equation 9)
W
c
=Correction weight size specified by Software (oz or gram)
r
c
= Radius at which W was to be installed (in)
W
n
= New correction weight size to be installed at new radius r (oz or gram)
r
n
= New radius at which W is to be installed (in)
2. Manual Formula for Splitting Weights (see Figure 45):
Figure 45 is provided for those situations in which an analyst does not have a software to
calculate split weight. Figure 45 shows two equations solving for CW1 and CW2 that are
calculated correction weight sizes that would be installed on each of two vanes rather than
CWR that was specified to be installed between vanes. An example is provided which will
help explain the use of the two equations given in Figure 45 (Equations 10 and 11). In this
example, a fan wheel with six blades is shown in Figure 45A with a required correction weight
of 1 oz (28.35 grams) to be installed at an angle of 75 (this places it 15 clockwise of Blade 1
and 45 counter-clockwise of Blade 2 as shown in Figure 46). When the example is solved in
Figure 47, it shows that the singular correction weight of 1 oz (28.35 grams) at 75 is
equivalent to one correction weight of .816 oz (23.13 grams) on Blade 1 and another weight
of .299 oz (8.48 grams) on Blade 2.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-46
3. Splitting weights using the graphic vector method:
As an example, we are going to be using a fan with 6 blades evenly spaced at 60 as
shown in Figure 47. After adding a trial weight on blade #1, the vector diagram directs
us to move the weight 75 clockwise and adjust the weight to 20 grams. As you can see,
there is no blade 75 clockwise to which we can add the required balance weight.
Therefore, we must add weights on the adjacent blades which will provide the required
result. The problem now is to find how much weight must be added on each blade.
FIGURE 45.
HOW TO SPLIT ONE CORRECTION WEIGHT INTO TWO
EQUIVALENT CORRECTION WEIGHTS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-47
FIGURE 46.
WEIGHT CORRECTIONS ARE MADE TO BLADES #2 AND #3 TO
PRODUCE THE NEEDED RESULTANT CORRECTIONS
To find the amount of weight required on blades #2 and #3, we will construct a vector
diagram. On a sheet of polar graph paper, mark off the angular position of blades #2 and #3
as shown in Figure 47A. Next draw a vector representing therequired correction weight,
Figure 47B. The angular position of this vector is 75 clockwise from blade #1 and 20 grams in
length as dictated by our vector calculations.
Now, complete the parallelogram as shown in Figure 47C by drawing a line fromthe end of
vector CW, parallel to blade #3 until is intersects blade #2; and draw a line from the end of
vector CW parallel to blade #2 until it intersects blade #3. To find the amount of weight
required on blade #3, simply measure the length of vector OA using the same scale used for
vector CW. Similarly, measure vector OB to find the amount of weight needed on blade #3. In
Figure 47D, vectors OA and OB show that 16.3 grams are needed on blade #2 and 6.0 grams
are needed on blade #3. Of course, these are required weights added at the same radius as
the original trial weight on blade #1. Note that the two weights total more than 20 grams. This
is normal as the two will always total more than the resultant vector CW.
R. Combining Balance Correction Weights Using Vectors
After balancing a rotor in two planes using the single plane vector method, you may find that 2,
3 or more balance weights have been added in a correction plane as a result of repeated
balancing to eliminate cross-effect. Instead of permanently correcting for each of these smaller
weights, it is often more convenient to combine these weights so only one permanent
correction weight must be added. Any number of balance correction weights in a given plane
can be combined into a single weight by constructing a vector diagram.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-48
FIGURE 47. VECTOR SOLUTION FOR SPLITTING WEIGHTS
For example, consider the three balance weights on the rotor in Figure 48A. To combine these
weights, their amounts and angular positions must be known.
First, draw a vector representing balance weight #1. See Figure 48B. For convenience, we
have selected the largest weight for #1 and have constructed its vector at 0 . The length of this
vector corresponds to the amount of weight, 25 grams. Next, from the end of the #1 balance
weight vector, construct a vector 10 grams in length representing balance weight #2 as shown
in Figure 48C. Note that the vector for balance weight #2 is drawn at an angle of 30 clockwise
from balance weight #1. Now, from the end of the #2 balance weight vector, construct the
vector for #3 that is 5 grams in length. See Figure 48D. This vector is constructed at an angle
45 clockwise from balance weight vector #1.
After vectors have been constructed for each balance weight as shown, construct vector R by
drawing a line from the origin (O) to the end of the last balance vector as shown in Figure 48D.
This vector R is the resultant and represents the amount and position of a single weight
whichwill be equivalent to the three balance weights. From vector R, we see that a weight of
38 grams located at a position 13 clockwise from balance weight #1 is required. will be
equivalent to the three balance weights. From vector R, we see that a weight of 38 grams
located at a position 13 clockwise from balance weight #1 is required.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-49
S. Effect of Angular Measurement Errors on Potential Unbalance Reduction
Two types of angular measurement errors can occur, each of which can have detrimental
effects on the amount of unbalance reduction achievable. The first type includes phase
measurement errors made by the instruments themselves and the second type includes errors
by the analyst in placing the balance correction weights at the specified angular locations on
the rotor. Each of these will be separately discussed.
1. EFFECT OF PHASE ANGLE MEASUREMENT ERRORS BY INSTRUMENTS
THEMSELVES.
Table III shows the best possible ratio of unbalanced reduction for various
degrees of error when measuring the phase angle of unbalance:
Referring to Table III, note that if the instrument makes a phase measurement of
30 that the maximum possible unbalance reduction during this trial or trim run is
by a ratio of only 2:1 (i.e., if the initial was 10 oz (283.5 grams), it will not allow
FIGURE 48
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
9-50
one to go below 5 oz (141.7 grams) during this trial). And it should be pointed out
that this assumes just the right balance weight was employed. Similarly, a phase
measurement error of only 7.5 using the correct weight will not allow more reduction
than by a factor of 8 times. If such errors can be kept below +/- 2, considerably higher
reduction ratios are possible.
Therefore, this points out the importance of using instruments that have very good
phase resolution capability down to +/- 2 if possible. Strobe light instruments with
the analyst attempting to interpret phase angles do allow this phase measurement
precision (usually fortunate to be within 10 to 15). Therefore, instruments with the
capability of measuring remote phase with a reference pickup are recommended
since many of them do have this measurement resolution capability. Use of such
remote phase instruments will often allow the machine to be balanced in much
fewer runs.
TABLE III.
BEST POSSIBLE RATIO OF UNBALANCE REDUCTION FOR VARIOUS
DEGREES OF PHASE ANGLE MEASUREMENT ERROR
2. EFFECT OF ANGULAR MEASUREMENT ERRORS WHEN ATTACHING BALANCE
CORRECTION WEIGHTS
The analyst should take great pains to correctly locate the balance correction weights
as closely as possible to the angular locations called for by the software or graphical
solution. Of course, when attempting to balance some fan wheels or pump impellers
and wanting to place such weights on the blades, the analyst may have to invoke the
split weight capability if the solution calls for weight placement between blades. In
any case, they should not just attempt to ball park either the weight size or angular
location. Figure 49 illustrates the relationship of how errors in attaching correction
weights will cause unnecessary residual unbalance to remain that should have been
removed. For example, looking at Figure 49, a vector that would be directly opposite
the initial unbalance of 100 grams was required. However, the analyst made a 10 error
when attaching the correction mass of 100 grams. Therefore, the figure shows that a
residual unbalance of 17.4 grams would be left in place. This residual unbalance will
be indicated in the next run as 17.4 grams at 85, which is nearly at right angles to the
initial unbalance. Table IV shows the amount of residual unbalance error for
varying degrees of correction weight angular error.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C. 9-51
FIGURE 49
RESIDUAL UNBALACNE DUE TO AN ERROR WHEN
ATTACHING A BALANCE CORRECTION WEIGHT
TABLE IV
RESIDUAL UNBALANCE AMOUNT ERROR FOR VARIOUS DEGREES
OF ERROR WHEN PLACING CORRECTION WEIGHTS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
10-1
BALANCING MACHINES - SOF BALANCING MACHINES - SOFT T-BEARING V -BEARING Vs s
HARD-BEARING MACHINES HARD-BEARING MACHINES
Figure 50 shows the distinctly different approaches taken by soft and hard bearing
balancing machines in the attempt of each design to provide a relatively constant
amplitude and phase lag throughout their operating speed range. These differences
and some of the attributes of each will be discussed below.
FIGURE 50
COMPARSION OF PHASE LAG ANGLE AND DISPLACEMENT
AMPLITUDES VERSUS ROTATIONAL SPEED IN SOFT-BEARING
AND HARD-BEARING BALANCING MACHINES
CHAPTER
10
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
10-2
A Soft-Bearing Balancing Machines
Figure 51 illustrates the motions of an unbalanced rotor suspended in a soft-
bearing machine having flexible supported bearings. The soft-bearing
balancing machine gets its name from the fact that it supports the rotor to be
balanced on bearings that are free to move in at least one direction, usually
horizontal, perpendicular to the rotor axis as shown in Figure 51. Referring to
Figure 50, the natural frequency of the rotor and bearing system occurs at
one-half or less of the lowest balancing speed. Therefore, when the rotor is
brought up to balancing speed, the phase lag angle and vibration displacement
amplitude have stabilized and can be measured with accuracy.
Until recently, a direct indication of unbalance was only possible on soft-bearing
machines after calibrating the indicating system for any particular rotor by
making several calibration runs with calibration weights of known size attached
to the rotor in specific correction planes (calibration settings will vary from one
rotor to the next due to different mass and distribution of the mass since
displacement of the principle axis of inertia in the balancing machine bearings
depends on rotor mass bearing and suspension mass, rotor moments of inertia
and the distance between bearings). However, in recent years, new technical
developments have overcome many of the soft-bearing machine calibration
problems making it possible to obtain accurate vibration before the first spin-up
of the rotor, and also making it possible to provide updated calibration of the
system just minutes before actually correcting and checking a rotor.
FIGURE 51
MOTION OF AN UNBALANCED ROTOR AND BEARING IN A SOFT-BEARING
BALANCING MACHINE WITH FLEXIBLE SUPPORTED BEARINGS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
10-3
In addition, this updated calibration also provides a built-in means of verifying
and validating the fact that the machine is responding properly. These newer
techniques involve several steps. First, the balance machine operator will dial in
several geometric parameters that are specific to each rotor. Then, the soft-
bearing machine will apply several predetermined calibrated forces against the
bearing supports with the rotor at rest. Then, transducers mounted on the
balance machine will sense the response of the rotor to the applied forces and
will determine the rotor sensitivity (oz-in/mil or gr-in/mil of vibration). Then, the
calibration can be checked by cross checks to verify the apparent level. The
balance machine is switched into the mils mode, and a direct measurement of
the actual vibration is obtained. One-half of this value will be multiplied by the
rotor weight to provide a cross check with the actual balance level.
B Hard-Bearing Balancing Machines
Figure 52 shows a typical hard-bearing machine along with some of the
dimensions which must be dialed into its analog (or digital) computer. Hard-
bearing machines have similar construction to soft-bearing machines, except
that their bearing supports are significantly stiffer in the horizontal direction. This
results in a horizontal resonance which is several orders of magnitude higher
than that for a comparable soft-bearing balancing. The natural frequency of the
rotor in a hard-bearing machine is normally designed to occur a 3X greater than
the maximum balancing machine. Referring to Figure 50, the hard-bearing
machine is designed to operate at speeds well below the natural frequency in an
area where the phase lag angle is constant and practically zero, and where the
vibration displacement amplitude, though small, is directly proportional to the
centrifugal forces produced by unbalance.
FIGURE 52
TYPICAL HARD-BEARING MACHINE WITH PERMANENTLY CALIBRATED
INSTRUMENTATION. ROTOR DIMENSIONS a, b, c AND r1, r2 ARE DIALED
DIRECTLY INTO THE ANALOG (OR DIGITAL) COMPUTER.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
10-4
Recall that centrifugal force is proportional to the following formula:
F
C
= .000001775 Un
2
= .00002841 Wrn
2
(EQUATION 1 Repeated)
where:
F
c
= Centrifugal Force (lb)
U = Unbalance of Rotating Part (oz-in)
W = Weight of Rotating Part (lb)
r = eccentricity of the rotor (in)
n = Rotating Speed (RPM)
Since the centrifugal force (F
c
) for a given amount of unbalance (Wr) at a given
speed is always the same whether this unbalance occurs in a small or large
rotor, the output from the sensing elements attached to the balancing machine
bearing supports remains proportional to the centrifugal force resulting from
unbalance in the rotor (remember that W in this equation is the unbalance
weight, not the rotor weight; r is the radius at which this unbalance weight is
acting). This output will depend only on the speed and amount of unbalance
(Wr). Therefore, the use of calibration masses is not required to calibrate the
machine for a given rotor.
(EQUATION 12)
F
C
= mr
2
g
C
= Wr
(386)(16)
2n
60
2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-1
RECOMMENDED VIBRA RECOMMENDED VIBRATION AND BALANCE TION AND BALANCE
TOLERANCES TOLERANCES
A. Vibration Tolerances
Todays condition monitoring programs are recognizing that two sets of vibration
tolerance specifications need to be developed for their rotating equipment-
(1) Overall Vibration Specifications; and (2) Spectral Band Alarm Specifications.
That is, a number of standards have been written through the years by a number of
professional organizations and individual corporations concerning allowable overall
vibration. However, spectral analysis particularly within the last few years has proven
time after time that serious problems can occur with certain components such as
bearings and gears which can have little or no real effect on a change in the overall
vibration itself, but will noticeably affect certain areas of the spectrum (in fact, in the final
stages of deterioration, rolling element bearings can actually cause amplitude to drop).
Therefore, procedures have been developed which will help the analysts break up
the spectrum into a number of bands, thereby allowing them to set much higher alarm
levels for problems that cause high vibration such as unbalance affecting 1X RPM, but
much lower amplitudes for those things which can withstand significantly less vibration
such as rolling element bearings. Unfortunately, although there are hundreds of plants
and installations that have condition monitoring software that will allow the user to
specify such spectral alarm bands, very few sites are effectively using them since there
is very little information available on how they should properly do so. While setting such
spectral alarm bands is not the purpose of this paper on field balancing, and since
space considerations do not allow complete publication of a paper by the author on this
subject, the following sections under vibration specifications will at least introduce this
topic, hoping that the reader will gain an appreciation of how important properly
specifying these bands is to his entire condition monitoring program.
CHAPTER CHAPTER
11 11
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-2
1. Recommended Overall Vibration Specifications
Much work continues today on establishing standards for allowable overall
vibration. Some of the better known standards now available include:
a) ISO 2372 - Mechanical Vibration of Machines with Operating Speeds
from 10 to 200 Revolutions per Second - Basis for specifying evaluation
standards (measurements made on structure).
b) ISO 3945 - Mechanical Vibration of Large Rotating Machines with
Speeds Ranging from 10 to 200 Revolutions per Second Measurement
and evaluation of vibration severity in situ (measurements made on
structure at various elevations).
c) ISO 7919 - Mechanical Vibration of Non-Reciprocating Machines -
Measurements on rotating components and evaluation (measurements
made on shafting).
d) ANSI S2.44, Part 1 - 1986, Measurements and Evaluation of Mechanical
Vibration of Non-Reciprocating Machines as measured on rotating shafts,
American National Standards Institute, NY (1986).
e) AGMA Standard, Specifications for Measurement of Lateral Vibration on
High Speed Helical and Herringbone Gear Units, Standard 426.01.
f) API 610 Centrifugal Pumps for General Refinery Services, 1971,
American Petroleum Institute, Washington, DC.
g) API 617, Centrifugal Compressors for Refinery Services, 4th Edition
1979, American Petroleum Institute, Washington, DC.
h) API 670, Non-Contacting Vibration and Axial PositionMonitoring System,
1976, American Petroleum Institute, Washington, DC.
i) MIL STD-167-1 (Ships) 1974, Mechanical Vibrations of Shipboard
Equipment, US Government Printing Office, Washington, DC.
j) MIL-M-17060B (SHIPS) 1959, Military Specifications - Motors, Alternating
Current, Integral Horsepower (Shipboard Use), US Government Printing
Office, Washington, DC.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-3
Some attempts have been made to provide overall vibration criteria based on the type
of machine and its drive configuration (centrifugal pump, direct coupled fan, belt driven
fan, turbine/generator, etc.), and on its mounting (isolated versus non-isolated). It is
recognized that there is often a significant difference in the amount of vibration for one
machine type versus another. For example, a reciprocating air compressor obviously
has significantly frequency, forcing frequencies of the machine itself, machine center of
gravity relative to the placement of isolators, etc. Thus, it is important that users of
todays predictive maintenance hardware and software take into account the type of
machine and its mounting when they begin to specify alarm levels of overall vibration
for each machine that they will input into the computer database.
In addition, it is important for users to know how their particular predictive
maintenance data collector and software system measures overall vibration. Some
systems have a fixed frequency range which is completely independent of any
frequency range chosen on any particular spectrum. In fact, this overall measurement
is completely independent from the spectra measurement parameters specified for the
data collector (F
max
, # lines, # averages, etc.). In these cases, the analog time waveform
is used as the basis for computing the overall vibration. In other data collector systems,
the overall is computed directly from the spectra themselves using the following
formula:
(EQUATION 13)
The danger with the latter technique of calculating overall vibration is that significant
vibration can possibly be occurring outside the maximum frequency (F
max
) that was
specified by the analyst. There would be no way one could be aware of this if the
overall were not computed from the time waveform. For example, if a user specified a
maximum frequency of 30,000 CPM and, unknown to him, there was a very large peak
out at 60,000 CPM with an amplitude of .50 in/sec (12.7 mm/sec), the data collector may
only display an overall of about .20 in/sec (5.1 mm/sec) whereas the true overall may
be up on the order of .60 in/sec (15.2 mm/sec). Thus, it is most important for the analyst
to know how the overall is computed and, if given the option, one should choose the
analog time waveform technique.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-4
If they still wish to know the total RMS energy within the spectrum as calculated by the
above equation, they could possibly specify one of the spectral alarm bands that would
go all the way from the low end up to the high end of the spectrum since the RMS
energy in each one of the bands is calculated using the same formula within the
software of most condition monitoring vendors.
Table V is offered as a specification which takes into account the various machine types,
how they are mounted and where measurements are taken when specifying peak
velocity overall alarm levels. This table assumes that the data collector has used the
analog time waveform technique to compute the overall peak velocity. In addition, this
overall peak velocity assumes an actual measurement of RMS vibration multiplied by
2 since most all data collectors and analyzers today actually take RMS measurements
and multiply them by this mathematical constant when displaying so-called peak
velocity. Note that each of three ratings is provided in Table V including GOOD,
FAIR and ALARM. After reviewing of all spectra captured on a machine, if no
problems are found, the first two columns (GOOD and FAIR) are offered to give the
analyst a general feel for the overall condition of each machine based on the highest
overall level measured on their machine. However, even if the highest overall in the
machine is still within the GOOD range, it is still possible for the machine to be in
alarm, depending on what frequencies were generated and the amplitudes of those
frequencies. That is where the spectral alarm bands come into play to ferret out the
apparently good condition machines from those that truly have problems.
2. Synopsis of Spectral Alarm Band Specifications
Written, tabulated procedures have been developed to help the analyst specify
spectral alarm bands for a series of machine types and configuration using those
types of predictive maintenance software systems which allow the spectrum to be
broken up into 6 individual bands. Each of these bands can be set at any span of
frequencies and at any alarm level for each individual band as chosen by the
analyst. Again, please note that a complete paper has been prepared by the author
on this subject. However, due to space considerations and due to the fact that the
purpose of the paper is instruction and specifications for balancing, only a portion of
the procedure can be presented within this paper.
Therefore, Table VI is a portion of a complete set of tables that has been developed
to specify these spectral alarm bands. Importantly, like the Table V overall alarm
table, this Table VI spectral alarm band table assumes casing measurements of
peak velocity using instruments which measure RMS and convert them to peak
levels by electronic multiplication of amplitudes by 1.414. Also of importance is the
fact that Table VI specifies spectral bands whose alarm levels are compared to the
total power within the band (so-called power bands). That is, the amplitude
for each band is computed by summing the amplitudes of each of the FFT lines as
per the earlier equation. Therefore, it is not necessary for any individual peak to
equal or exceed the level of any of these alarm bands for the band alarm itself to be
violated. Instead, it is only necessary that the total RMS content of the band as
computed equal or exceed this band alarm specification.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-5
Table VI shows how spectral alarm bands are set up for several different types of
machines including general machines with rolling element bearings and without
vanes (Type A), general machines with sleeve bearings and without vanes (Type
B), special gearbox high frequency points with known number of teeth (Type C),
and special gearbox high frequency points with unknown number of teeth (Type
D). A number of other machine types are covered in the remainder of the complete
spectral alarm band tables that are in the referenced paper.
When using Table VI, the analyst first should identify the particular machine type
looking beneath the overall alarm specification in Table V, and find the overall alarm
for this particular machine type. This will be used as direct input into the spectral
alarm band specs of Table VI. If this particular machine type is not included in Table
V, the user should either refer to the manufacturer of the machine, or other similar
vibration severity charts, or use alarm levels for another machine type that most
closely resembles the particular machine.
Please refer to the entries under the first column of Table VI. BAND LOWER
FREQUENCY specifies at what frequency each band should begin, whereas
BAND HIGHER FREQUENCY shows where each band should end (for example,
from 0 to 1000 CPM). In general, no gaps should be left between bands, nor
should bands overlap one another (although some analysts using power bands
sometimes extend one band from the beginning to the end of a complete spectrum
in order to have the system calculate the Spectral Overall Level, and then compare
this to the overall level provided separately by their instrument). Next, the column
entry entitled BAND ALARM specifies how high to set the alarm level of each
band.
Notice that many of the cases described in Table VI have the BAND LOWER
FREQUENCY set at 1% of F
max
rather than at 0 CPM. The reason for this is that data
collectors and spectrum analyzers most always have built-in noise within the first 1
to 3 FFT lines, particularly when data from an accelerometer are electronically
integrated to velocity. In fact, some instruments have been know to display peaks
with so called amplitudes over 2.0 in/sec (50.8 mm/sec) within these first 3 FFT
lines. If F
max
is properly specified, this most always is garbage data which can be
neglected. Therefore, Band 1 will never begin within these first 3 lines in Table VI.
Each of the cases specify the maximum frequency (F
max
which is always given along
with the case title). Therefore, each case will tell where to set both the frequency
range and alarm level of each band, and will describe what each band covers (i.e.,
bearing defect frequencies, gear mesh frequencies, etc.) Case A will be discussed
in detail to illustrate the alarm band specification technique, whereas only highlights
of each remaining case will be given. Then, several examples will follow the
discussion to further illustrate how these techniques should be applied. A detailed
explanation will be given on how to use Case A on general rolling element bearing
machines without rotating vanes in order to give a detailed example on how to use
the remainder of this portion of the alarm band tables. Also, an example showing
how spectral alarm bands are specified for a pump is shown in Figure 53.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-6
Note the special points called 1.0 and 2.0 which evaluate electrical problems
and note the slightly different band setup for the pump spectra at positions 3 and 4
versus that for the motor at positions 1 and 2. The remaining three cases on this
page of Table VI will be handled in like manner. However, it should be pointed out
that Cases C and D specify particular gearbox high frequency points which are
taken in addition to standard measurements on the gearbox.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-7
CRITERIA FOR OVERALL CONDITION RATING (PEAK OVERALL VELOCITY, IN/SEC)*
1. Assuming Machine Speed = 600 to 60,000 RPM.
2. Assuming Measurements by Accelerometer or Velocity Pickup securely mounted as Close as Possible
to Bearing Housing.
3. Assuming Machine Is Not Mounted on Vibration Isolators (for Isolated Machinery - Set Alarm 30% - 50%
Higher).
4. Set Motor Alarms the Same as that for the Particular Machine Type unless Otherwise Noted.
5. Consider Setting Alarms on Individual External Gearbox Positions about 25% Higher than that for a
particular Machine Type.
MACHINE TYPE GOOD FAIR ALARM 1 ALARM 2
COOLING TOWER DRIVES
Long, Hollow Drive Shaft 0 - .375 .375 - .600 .600 .900
Close Coupled Belt Drive 0 - .275 .275 - .425 .425 .650
Close Coupled Direct Drive 0 - .200 .200 - .300 .300 .450
COMPRESSORS
Reciprocating 0 - .325 .325 - .500 .500 .750
Rotary Screw 0 - .300 .300 - .450 .450 .650
Centrifugal With or W/O External Gearbox 0 - .200 .200 - .300 .300 .450
Centrifugal - Integral Gear (Axial Meas.) 0 - .200 .200 - .300 .300 .450
Centrifugal - Integral Gear (Radial Meas.) 0 - .150 .150 - .250 .250 .375
BLOWERS (FANS)
Lobe-Type Rotary 0 - .300 .300 - .450 .450 .675
Belt-Driven Blowers 0 - .275 .275 - .425 .425 .650
General Direct Drive Fans (with Coupling) 0 - .250 .250 - .375 .375 .550
Primary Air Fans 0 - .250 .250 - .375 .375 .550
Vacuum Blowers 0 - .200 .200 - .300 .300 .450
Large Forced Draft Fans 0 - .200 .200 - .300 .300 .450
Large Induced Draft Fans 0 - .175 .175 - .275 .275 .400
Shaft-Mounted Integral Fan (Extended Motor Shaft) 0 - .175 .175 - .275 .275 .400
Vane-Axial Fans 0 - .150 .150 - .250 .250 .375
MOTOR/GENERATOR SETS
Belt-Driven 0 - .275 .275 - .425 .425 .675
Direct Coupled 0 - .200 .200 - .300 .300 .450
CHILLERS
Reciprocating 0 - .250 .250 - .400 .400 .600
Centrifugal (Open-Air) - Motor & Compressor Separate 0 - .200 .200 - .300 .300 .450
Centrifugal (Hermetic) - Motor & Impellers Inside 0 - .150 .150 - .225 .225 .350
LARGE TURBINE/GENERATORS
3600 RPM Turbine/Generators 0 - .175 .175 - .275 .275 .400
1800 RPM Turbine/Generators 0 - .150 .150 - .225 .225 .350
CENTRIFUGAL PUMPS
Vertical Pumps (12' - 20' Height) 0 - .325 .325 - .500 .500 .750
Vertical Pumps ( 8' - 12' Height) 0 - .275 .275 - .425 .425 .650
Vertical Pumps ( 5' - 8' Height) 0 - .225 .225 - .350 .350 .525
Vertical Pumps ( 0' - 5' Height) 0 - .200 .200 - .300 .300 .450
General Purpose Horizontal Pump - Direct Coupled 0 - .200 .200 - .300 .300 .450
Boiler Feed Pumps - Horizontal Orientation 0 - .200 .200 - .300 .300 .450
Piston Type Hydraulic Pumps - Horizontal Orientation (under load) 0 - .150 .150 - .250 .250 .375
MACHINE TOOLS
Motor 0 - .100 .100 - .175 .175 .250
Gearbox Input 0 - .150 .150 - .225 .225 .350
Gearbox Output 0 - .090 .090 - .150 .150 .225
Spindles:
a. Roughing Operations 0 - .065 .065 - .100 .100 .150
b. Machine Finishing 0 - .040 .040 - .060 .060 .090
c. Critical Finishing 0 - .025 .025 - .040 .040 .060
*NOTE: The ALARM 1 and ALARM 2 overall levels given above apply only to in-service machinery which has been
operating for some time after initial installation and/or overhaul. They do not apply (and are not meant to serve
as) Acceptance Criteria for either new or rebuilt machinery.
R-200002-PK
TABLE V
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-8
TABLE VI
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-9
FIGURE 53
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-10
That is, one set of measurements should be taken on the gearbox using either Case A
or B, depending on whether the gearbox is outfitted with rolling element (Case A) or
sleeve (Case B) bearings. Then, a second set of measurements should be taken at
various gearbox points close to each mesh, with a much higher F
max
on this second
measurement up to 3.25X gear mesh frequency as shown in Table VI. Therefore, the
spectra having lower F
max
which are given in Tables A or B will detect problems such as
misalignment, unbalance, looseness, etc., which are concentrated within lower
frequencies; and the special high frequency point specified by Cases C and D will look
much higher for the presence of gear problems which commonly show up at gear mesh
frequency, 2X gear mesh frequency and even one-half harmonics of gear mesh
frequency up to 5.5X gear mesh frequency. When taking measurements up to these
high F
max
frequencies, the analyst is cautioned to read the notes under Cases C and D
concerning mounting of the accelerometer. When getting up to frequencies which
exceed 360,000 CPM or more, the analyst may have to stud mount the accelerometer
and may even have to employ special high frequency accelerometers.
Case A - General Rolling Element Bearing Machine without Rotating Vanes
Case A on Table VI applies to both the driver and driven components of a wide range of
general rotating process and utility machines which are outfitted with rolling element
bearings (ball, roller or needle bearings). Before entering Table VI, refer to Table V to
obtain the alarm level of overall peak velocity for your machine type. Then, determine
the type of rolling element bearings. For common rolling element bearings Case A
specifies a spectrum with a maximum frequency (F
max
) of approximately 40X RPM (for
example, for a nominal speed of 1800 RPM, set F
max
at approximately 72,000 CPM as in
the example of Figure 53). However, for tapered roller bearings (Timken cup and cone
arrangement, or equivalent) or for special roller bearings, Case A specifies a maximum
frequency of approximately 50X RPM. The reason for the higher F
max
for these bearing
types is the fact that, with their particular geometries, they inherently have higher
calculated rolling element bearing defect frequencies. Also note in Case A that if the
speed is from 500 to 999 RPM, F
max
should be set at 60X RPM. The reason for this is to
ensure that the spectra designed for this machine will detect a rolling element
bearing in only the second of four failure stages through which it will normally pass
rather than waiting late in the life of the bearing before problems are detected. Referring
to Table IV for Rolling Element Bearings, note that the natural frequencies of bearing
components will be excited during this second stage. Since these natural frequencies
normally range from 30,000 to 120,000 CPM for most bearings, it is important to keep
F
max
sufficiently high to detect these when excited.
Please note that it is not necessary to specify F
max
at exactly 40X or 50X RPM, but should
be somewhere in this vicinity (certainly not less than 30X RPM). If one sets F
MAX
too low it
can cause a spectrum to completely miss potentially serious developing bearing wear,
particularly during earlier stages. On the other hand, if F
max
is set too high, this can result
in poor frequency resolution that can cause the user to misdiagnose problems, since
they do not have sufficiently precise frequency resolution to properly identify such
frequency components as true running speed harmonics versus bearing defect
frequencies and vibration from adjacent machines. Also, if one sets F
max
too high,
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-11
this can cause potentially valuable information on subsychronous vibration to be
buried at the left-hand side of the spectrum. In general, the rule of thumb is to keep
F
max
as low as you can without missing anything important.
Referring to Case A in Table VI, note that each one of the bands has a specific purpose
and zone of coverage. For example, Band 1 ranges from subsynchronous vibration
(below 1X RPM) up through operating speed. Bands 2 and 3 cover 2X and 3X RPM,
respectively. Band 4 will include fundamental bearing defect frequencies for most
rolling element bearings. Similarly, Bands 5 and 6 will include bearing defect
frequency harmonics as well as natural frequencies of bearing components for most
common rolling element bearings.
Now, referring back to Table VI, note that Band 1 extends from 1% of F
max
to a frequency
at 1.2X RPM. In the case of the example 1800 RPM machine having F
max
at 72,000 CPM
as previously discussed, Band 1 would extend from 720 to 2160 CPM. The Band 1
alarm spec calls for 90% of the overall level, thus, if the overall alarm were .300 in/sec
(7.6 mm/sec) (from table V), then the Band 1 alarm would be set at .270 in/sec (6.9 mm/
sec) for this machine. Similarly, Table VI specifies the frequency range of Band 2 to
extend from 1.2 to 2.2X RPM (in the 1800 RPM case, this would extend from
approximately 2160 to 3960 CPM). The Band 2 alarm spec calls for 50% of the overall
alarm (thus from the example .300 in/sec (7.6 mm/sec) overall, Band 2 would be set at
.150 in/sec) (3.8 mm/sec). Finally, Bands 3 through 6 are specified similarly (note that
fixed alarm levels are specified for Bands 4 through 6, independent of the overall alarm
for that machine type since bearing frequencies exceeding these fixed amplitudes
should be considered serious, no matter what the machine, particularly when more than
one harmonic defect frequency is present). The fixed alarm levels specified for Bands 4
thru 6 should work out satisfactorily for the great majority of machines. However, some
users might have to drop these considerably, particularly if they have high vibration
path impendence, slow moving machinery such as paper machines or calendar rolls.
Here again, this is a case where the alarm band and overall level refinement procedure
discussed and illustrated in a later section of the paper will to be put into play. Finally, if
readers find that they need additional information on how to properly specify spectral
alarm bands, they can get in touch with Technical Associates and acquire the complete
paper. This paper is also given along with complete instruction on how it should be
used along with examples in vibration seminars offered by Technical Associates.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-12
B. Balance Tolerances On Allowable Residual Unbalance
Balance specification are provided by a number of sources. These include each of the
following:
1) ISO 1940 Balance Quality of Rotating Rigid Bodies (same as ANSI
S2.19-1975); The International Standards Organization (ISO); Geneva 20,
Switzerland.
2) NEMA MG1-12.06 Balance of Motors 1971; National Electrical
Manufactures Association.
3) NAVY MIL STD - 167-1 (SHIPS) 1974, Mechanical Vibrations of
Shipboard Equipment; U S Government Printing Office, Washington, DC
4) NAVY MIL - M - 17060B (SHIPS) 1959, Military Specification - Motors,
Alternating Current, Integral Horsepower (Shipboard Use).
5) SAE ARP 1136, Balance Classification of Turbine Rotor Blades; Society
of Automotive Engineers, Inc. SAE; Warrendale, PA.
6) API 610, Centrifugal Pumps for General Refinery Service, 1971; American
Petroleum Institute, Washington, DC
(API offers vibration and balance standards for many types of rotating
equipment).
A discussion will first be provided on balance quality as per ISO 1940. Then, a
comparison of ISO 1940 will be made with standards provided by the Navy and API.
1. ISO 1940 Balance Quality Grades
Dynamic balancing is the process of attempting to align the mass and shaft
centerlines so that the rotor will rotate with an absolute minimum of unbalanced
centrifugal force. Since there is no such thing as a perfectly concentric rotor with
perfectly fitting concentric parts that are supported with perfectly fitting concentric
bearings, there will always be some degree of residual unbalance in each rotor.
Residual unbalance, sometimes called final unbalance, is the unbalance of any
kind that remains after balancing.
Calculations can be performed when balancing a rotor to determine the degree
of residual unbalance (this will be covered in Section 2 which will follow). During
the 1950s, a small group of experts active within the balancing field got together
and began to formulate discussions on how a set of balance standards
could be developed so that one could numerically determine to what quality his
rotor is balanced in comparison with other similar rotors.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-13
Several years of intensive study ensued, culminating with this group joining
Technical Committee 108 on Shock and Vibration of the International
Standards Organization (ISO). Later, the group name was changed to
Sub-Committee 1 on Balancing and Balancing Machies (ISO TC-108/SC1).
Request were made by the committee to groups throughout the world to submit
more and more data on balance qualities achieved. Finally, ISO Standard No.
1940 was issued entitled Balance Quality Requirements of Rigid Rotors which,
after a period of time, was also adopted as S2.19 by The American National
Standards Institute (ANSI). Some of the more important points of this standard
are summarized below.
Table VII provides each of the balance quality grades as per ISO1940
delineated by Balance Quality Grade. Beside each quality grade G number is
a listing of various rotor types, grouped according to these quality grades. As
Table VII states, the list of rotor types is meant to include general examples and
is not meant to be all-inclusive.
Please note that beside each quality grade number in Table VII is a column
entitled e
per
x . It shows that each quality grade number represents the
maximum permissible orbital velocity (e
per
x ) of the center of gravity in mm/sec
around the shaft axis () expressed in radians/sec. For example, referring to
quality grade G 6.3, this grade allows an orbital velocity of:
6.3 mm/sec, RMS = .248 in/sec, RMS = .351 in/sec peak.
Figure 54 provides the ISO Permissible Residual Unbalance bands for each
rotor group as a function of the maximum service speed (RPM). This graph is
used to establish the actual residual unbalance tolerance (called U
per
) in terms
of lb-in/lb or gram-in/gram. Therefore, this tolerance specifies the permissible
residual unbalance per unit of rotor weight (rotor weight includes all portions of a
machine which are in fact rotating, not the total weight of the entire machine
assembly). For example, a small motor may have a total static weight of 20 lbs.
(9 kg), whereas the weight of its rotor alone is only 10 lbs. (4.5 kg). It is the
10 lb. (4.5 kg) rotor weight which should be employed when using these
residual unbalance guidelines.
Referring again to Figure 54, note that each of the balance quality grades
incorporates four bands except for those in the very upper or lower extremes of
the graph. These unofficial bands might be considered (from top to bottom in
each grade) as sub-standard, fair, good and precision. Therefore, the
graph permits some adjustment to individual circumstances within each quality
grade.
Importantly, the difference in permissible residual unbalance between the
bottom and top edge of each grade is a factor of 2.5. For critical applications,
one should employ the lower band within each grade.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-14
It is important to point out that the tolerances recommended in Figure 54 apply only to
rigid rotors. Recommendations for flexible rotor tolerances are contained in ISO 5343
entitled Criteria for Evaluating Flexible Rotor Unbalance which must be read in
conjunction with ISO 1940 and ISO 5406 entitled The Mechanical Balancing of Flexible
Rotors. Looking again at Figure 54, note that in general, the larger the rotor mass, the
greater the permissible unbalance; while the higher the RPM, the lower the allowable
residual unbalance. Also, note that when obtaining the residual unbalance from the
left-hand vertical axis, it will be expressed in units of (for example) lb-in/lb. As note (2)
below Figure 54 states, to obtain this in units of oz-in/lb, multiply by a value of 16.
For example, referring to Figure 54, if you had a 10 lb rotor turning at 1000 RPM and
wanted to single-plane balance it to the precision level of G 2.5 (lowest of the four
G 2.5 bands), this would correspond to a permissible residual unbalance of
.00065 lb-in/lb which equals .0104 oz-lb/lb of rotor weight. Therefore, if this were a
single-plane problem, the total allowable residual unbalance would be
.0104 X 10 lb = .104 oz/in total in this one plane (had it been a two-plane problem,
the permissible residual unbalance would have been .0052 oz-in/plane if this were a
symmetric rotor centered between bearings as per note (1) below Figure 54).
Following below will be a discussion on how ISO applies to both single-plane and two-
plane problems with standard rotors of symetric design centered between bearings, as
well as how it applies to rotors of various other geometries (including overhung rotors).
a. Application of Tolerances to Single-Plane Problems:
A single-plane rotor is normally considered to be disc shaped requiring
only one correction plane which may be sufficient if the distance between
the bearings is large and the disc has small axial run-out. The entire
tolerance provided in Figure 54 will be allowed for the single plane. If the
couple unbalance (referred to the bearing planes) exceeds one-half the
total rotor tolerance, the rotor may require two-plane balance.
b. Application of Tolerances to Two-Plane Problems:
In general, one-half of the permissible residual unbalance is applied to each
of the two correction planes provided each of the following three conditions
is met:
1) The rotor CG is located within the mid-third of the bearing span;
2) The distance between correction planes is greater than 1/3 of the
bearing span, and the correction planes are between bearings;
3) The correction planes are approximately the same distance from the
CG, having a ratio no greater than 1.5.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-15
TABLE VII
BALANCE QUALITY GRADES FOR VARIOUS GROUPS OF REPRESENTATIVE
RIGID ROTORS IN ACCORDANCE WITH ISO 1940 AND ANSI S2.19-1975
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-16
FIGURE 54
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-17
FIGURE 55 (1OF 2)
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-18
FIGURE 55 (2 OF 2)
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-19
c. Application of Tolerances to Special Rotor Geometries
Figure 55 is provided to show how ISO 1940 tolerances can be applied to
rotors of special geometries that do not meet each of the three conditions
stipulated above in Section b. For example, rotor A at the top of the figure
shows a rotor with widely spaced inboard (between bearings) correction
planes; rotor B has each of two overhung rotors outboard of each of the
bearings; rotor C shows a classic overhung rotor with each of two
corrections planes on the single wheel which is outboard of each of its
bearings; whereas rotor D is situated between bearings, but away from
the bearing span mid-point. Note that permissible tolerances (U
per
) are
stipulated on Figure 55 for each of these rotor geometries.
Following below are several examples which help illustrate how ISO 1940
tolerances are applied to various rotor types and geometries:
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-20
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-21
C How to Determine Residual Unbalance Remaining in a Rotor After
Balancing
When field balancing, you must know when to determine that the job is
complete. This will be not only when low vibration levels are achieved but also
when the rotor is balanced within allowable specifications. To know this, you
must determine the residual unbalance remaining in the rotor. This can be
accomplished by following the procedure below;
a. Make original measurements of amplitude and phase and graph
this to scale on polar coordinate paper. Call this vector the O
vector.
b. Attach a trial weight and document the trial weight size (oz) and
radius (in) to which it is attached. (mr = trial weight size X trial
weight radius.
c. After attaching the trial weight, spin the rotor and measure
amplitude and phase. Graph this on the polar graph paper as the
O + T vector.
d. Draw a vector called T from the end of vector O to the end of
vector O + T. Vector T represents the effect of the trial weight
alone. Measure the length of vector T to the same scale as that
used for vectors O and O + T. Using this scale, determine the
equivalent vibration level (mils).
e. Calculate Rotor Sensitivity as per the following equation:
Rotor Sensitivity = (Trial Wt. Size) (Trial Wt. Radius)
oz-in/mil Trial Weight Effect
f. Calculate Residual Unbalance using the equation shown below. If
Residual unbalance is not brought within tolerances, trim balance
using current correction weight as the trial weight for the trim run.
Continue trim balancing until Residual Unbalance is reduced
within required balance tolerances.
Residual Unbalance = Rotor Sensitivity X Vibration amp. after Bal.
(oz-in) (oz-in/mil) (mils)
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-22
Example (see Figure 54)
Given:
Required ISO Balance Quality = G 2.5
Rotor Weight = 100
Rotor speed = 800 RPM
Amplitude after balancing = 2.0 mils
Therefore, Required U
per
= 1.76 oz-in total (single-plane)
a. Original reading = 10 mils @ 240 = O vector.
b. Trial weight of 3 oz is attached in the balance plane at a 6 inch
radius (mr = 3 oz X 6 in =18 oz-in)
c. Trial run reading = 8 mils @ 120 = O + T vector
d. Effect of trial weight alone = T = 15.5 mils (From Figure 54)
e. Rotor Sensitivity = 18 oz-in = 1.16 oz-in
15.5 mils mil
f. Residual Unbalance = (1.16 oz-in) (2.0 mils) =2.32 oz-in
mil (not with specs)
Continued balancing and reduced vibration to 1.0 mil
Residual Unbalance = (1.16 oz-in)(1.0 mil) = 1.16 oz-in
mil (in compliance)
D Comparison of ISO 1940 with API and Navy Balance Specifications:
In general, the author has not found ISO tolerances to be sufficient tight. Recall that they were
initially developed in the 1950s when much less was know about the detrimental effect of
vibration on machine life. Therefore, when using the ISO 1940 tolerances, the author has
found much greater success by going one quality grade better than that called for by ISO
1940 guidelines that is given in Table VIII.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-23
FIGURE 56 STANDARD SINGLE-PLANE VECTOR SOLUTION
Prior to the development of ISO 1940, the United States Navy had developed a balancing
standard for its noise reduction program on submarines. This standard was known as
NAVSEA STANDARD ITEM 009-15 and the Navy MIL STD-167-1 (SHIPS) 1 May 1974
prescribed the following balance tolerances:
For Rotor Service Speeds Below 150 RPM,
U
per
= 0.177 W
For Rotor Service Speeds Up to 1000 RPM,
U
per
= 4000 W
N
2
For Rotor Service Speeds Above 1000 RPM,
U
per
= 4W
N
Where:
U
per
= Permissible residual unbalance in oz-in
W = Rotor weight (lbs)
N = Maximum service speed (RPM)
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
11-24
The American Petroleum Institute (API) finally adopted the third of the above
standards that are written on a variety of machines. Therefore, the API standards are
equal to those of the Navy MIL-STD-167-1 (SHIPS) 1 May 1974.
Examples Comparing Navy and ISO Standards:
Example 4. Blower Requiring Two-Plane Balancing
Machine Type = Blower
Machine Class. = ISO G 6.3 from Table VII
Blower Speed = 2000 RPM
Rotor Weight = 1000 lbs (including fan wheel, shaft & sheave)
a) Navy
U
per
= 4W = (4)(1000) =2 oz-in/plane
N 2000
b) ISO Grade G 6.3 (From Figure 52)
U
per
= (.0012 lb-in)(16 oz)(1000 lb) =19.2 oz-in total = 9.6 oz-in/plane
lb lb 2 Planes
Example 5. Fractional HP Electric Motor Requiring Two-Plane Balancing
Machine Type = Fractional HP Electric Motor
Machine Class.= ISO G 2.5 from Table VII
Machine Speed = 1780 RPM
Rotor Weight = 5 lbs (Total Motor Static Weight = 8 lb)
a) Navy
U
per
= 4W =(4)(5) =.0112 oz-in/plane
N 1780
b) ISO Grade G 2.5 (from Figure 52)
U
per
= (.00062 lb-in)(16 oz)(5 lb) = .0496 oz-in total =.0248 oz-in/plane
lb lb 2 planes
Based on Examples 4 and 5 along with a number of other test examples comparing
Navy and ISO standards, one will see that the ISO tolerances are noticeably more
lenient than those of the Navy. In some cases, the Navy Specs, written for submarine
noise protection, may sometimes be difficult for the average balance shop to achieve.
On the other hand, those of ISO are almost always too lenient.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-1
APPENDIX
A
BALANCING TERMINOL BALANCING TERMINOLOG OGY ISO Y ISO
A general vocabulary on vibration and shock is given in ISO 2041.
NOTE: Terms in italics in the definitions are themselves defined elsewhere
in this vocabulary. Italicized terms that are defined separately are
separated by an asterisk.
ISO 1940-1, Mechanical vibration Balance quality requirements of rigid rotors
Part 1: Determination of permissible residual unbalance.
ISO 2041, Vibration and shock Vocabulary.
ISO 2953, Balancing machines Description and evaluation.
Acceptability limit: That value of an unbalance parameter which is specified as the
maximum below which the state of unbalance of a rotor is
considered to be acceptable.
Amount indicator: On a balancing machine, the dial, gauge or meter with which a
measured amount of unbalance or the effect of this unbalance is
indicated.
Amount of unbalance: Quantitative measure of unbalance of a rotor (referred to a
plane), without referring to its angular position. It is obtained
by taking the product of the unbalance mass and the
distance of its center of gravity from the shaft axis.
NOTE:
1. Units of unbalance are, for example, gram-millimeters and ounce-inches.
2. In certain countries, the terms weight and mass are used
interchangeably.
Angle indicator: Device used to indicate the angle of unbalance.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-2
Angle of unbalance: Given a polar coordinate system fixed in a plane
perpendicular to the shaft axis and rotating with the rotor, the
polar angle at which an unbalance mass is located with
reference to the given coordinate system.
Angle reference generator: In balancing, a device used to generate a signal
which defines the angular position of the rotor.
Angle reference marks: Marks placed on a rotor to denote an angle reference system
fixed in the rotor; they may be optical, magnetic, mechanical
or radioactive.
Axis of rotation: Instantaneous line about which a body rotates
NOTES:
1. If the bearings are anisotropic, there is no stationary axis of rotation.
2. In the case of rigid bearings, the axis of rotation is the shaft axis, but
if the bearings are not rigid, the axis of rotation is not necessarily
the shaft axis.
Balance quality grade: For rigid rotors, the product of the specific unbalance and
the maximum service angular velocity of the rotor,
(See ISO 1940-1).
Balance tolerance; maximum permissible residual unbalance U
per
: In case of rigid
rotors, that amount of unbalance with respect to a plane
(measuring plane or correction plane) which is specified as
the maximum below which the state of unbalance is
considered to be acceptable.
Balancing: Procedure by which the mass distribution of the rotor is checked and, if
necessary, adjusted to ensure that the residual unbalance or the vibration
of the journals and/or forces on the bearings at a frequency corresponding
to service speed are within specified limits.
Balancing machine: Machine that provides a measure of the unbalance in a rotor
and which can be used for adjusting the mass distribution of
that rotor mounted on it so that the once-per-revolution
vibratory motion of the journals or the force on the bearings
can be reduced if necessary.
Balancing machine accuracy: Limits within which the amount and angle of
unbalance can be measured under specified
conditions.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-3
Balancing machine minimum response: Measure of the machines ability to sense
and indicate a minimum amount of
unbalance under specified conditions.
Balancing machine sensitivity: Of a balancing machine under specified conditions,
the increment in unbalance indication expressed as
indicator movement or a digital reading per unit
increment in the amount of unbalance.
Balancing run (on a balancing machine): Run consisting of one measuring run and
the associated correction process.
Balancing speed: Rotational speed at which a rotor is balanced.
Bearing support: Part, or series of parts, that transmits the load from the bearing to the
main body of the structure.
Bias mass: The mass added to a mandrel (balancing arbor) to create a desired
unbalance bias.
Calibration: Process of adjusting a machine so that the unbalance indicator(s) read(s)
in terms of selected correction units in specified correction planes for a
given rotor and other essentially identical rotors: it may include
adjustment for angular location if required.
Calibration mass: A known mass used
a) in conjunction with a proving rotor, to calibrate a balancing ma-
chine, and
b) on the first rotor of a kind, to calibrate a soft-bearing balancing
machine for that particular rotor and subsequent identical rotors.
Calibration rotor: Rotor (usually the first of a series) used for the calibration of a
balancing machine.
Center of gravity: The point in a body through which the resultant of the weights of its
component particles passes, for all orientations of the body with
respect to a gravitational field.
NOTE: - If the field is uniform, the center of gravity coincides with the center of
mass.
Center of mass: That point associated with a body which has the property that an
imaginary particle placed at this point, with a mass equal to the
mass of a given material system, has a first moment with respect to
any plane equal to the corresponding first moment of the system.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-4
Centrifugal (rotational) balancing machine: Balancing Machine that provides for the
support and rotation of a rotor and for the measurement of
once-per-revolution vibratory forces or motions due to unbalance in
the rotor.
Claimed minimum achievable residual unbalance: The value of minimum
achievable residual unbalance stated by the manufacture for his
machine, and measured in accordance with the procedure
specified in ISO 2953.
Compensating (null-force) balancing machine: Balancing machine with a built-in
calibrated force system which counteracts the unbalance forces in
the rotor.
Compensator: Facility into a balancing machine which enables the initial
unbalance of the rotor to be nulled out, usually electrically, so
speeding up the process of plane setting and calibration.
Component correction: Correction of unbalance in a correction plane by mass
addition or subtraction at two or more of a predetermined
number of angular locations.
Component measuring device: Device for measuring and displaying the amount and
angle of unbalance in terms of selected components of the
unbalance vector.
Controlled initial unbalance: Initial unbalance which has been minimized by
individual balancing of components and/or careful attention to
design, manufacture and assembly of rotor.
Correction (balancing) plane: Plane perpendicular to the shaft axis of a rotor in which
correction for unbalance is made.
Correction mass: A mass attached to a rotor in a given correction plane for the
purpose of reducing the unbalance to the desired level.
NOTE: The same correction can be effected by removing mass from the
opposite side of the rotor.
Correction plane interference (cross-effect): Change in balancing machine
indication for one correction plane of a given rotor, which is
observed for a certain change in unbalance in the other plane.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-5
Correction plane interference ratios: Interference ratios I
AB
and I
BA
of two correction
planes A and B of a given rotor are defined by the following
relationships.
Where U
AB
and U
BB
are the unbalances referring to planes A and B respectively,
caused by the addition of a specified amount of unbalance in plane B:
Where U
BA
and U
AA
are the unbalances referring to planes B and A respectively,
caused by the addition of a specified amount of unbalance in plane A.
Counterweight: Weight added to a body to reduce a calculated unbalance at a
desired place.
NOTE: Such weights may be used to bring an asymmetric body to a state
of balance or to reduce bending moments within a body, e.g.,
crankshafts.
Couple unbalance: That condition of unbalance for which the central principal
axis intersects the shaft axis at the center of gravity.
NOTES:
1. The quantitative measure of couple unbalance can be given by the
vector sum of the moments of the two dynamic unbalance vectors
about a certain reference point in the plane containing the center of
gravity and the shaft axis.
2. If static unbalance in a rotor is corrected in any plane other than
that containing the reference point, the couple unbalance will be
changed.
Couple unbalance interference ratio: The interference ratio I
SC
= U
S
IU
C
.
Where U
S
is the change in static unbalance indication of a balancing machine
when a given amount of couple unbalance U
C
is introduced into the rotor.
NOTE: This ratio is generally used in the testing of single-plane
balancing machines and may be expressed as a
percentage by multiplying it by the maximum distance
between the test plane on a proving rotor.
I
AB
=
U
AB
U
BB
I
BA
=
U
BA
U
AA
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-6
Critical speed: Characteristic speed at which resonance of a system is excited.
NOTE: Depending on the relative magnitudes of the bearing stiffness
and mass and the rotor stiffness and mass, the significant
effect at a critical speed may be the motion of the journals or
the flexure of the rotor (see flexural critical speed, and
rigid-rotor-mode critical speed).
Cycle rate: The number of starts and stops that a balancing machine, for a given rotor
having a specified moment of inertia and for a given balancing speed, can
perform per hour (without damage to the machine) when balancing the
rotor.
Design axis: Axis about which parts and assemblies are designed and about
which it is intended that the body be balanced.
Differential test masses: Two masses, representing different amounts of unbalance,
added to a rotor in the same transverse plane at
diametrically opposed positions.
NOTES:
1. Differential test masses are used, for example, in cases where a
single test mass is impractical.
2. In practice, the threaded portion and the height of the head of the
test mass are kept constant. The diameter of the head is varied to
achieve the difference in test mass.
3. The smaller of the two differential test masses is sometimes called
the tare mass, the larger the tare-delta mass.
Differential unbalance: The difference in unbalance between the two differential test
masses.
Direct reading balancing machine: A balancing machine which can be set to indicate
unbalance in terms of angular position and in units of mass such as
grams (or ounces) in any two balancing planes without requiring
individual calibration for the first rotor of a kind.
Dummy rotor: In a balancing procedure, an attachment of adequate stiffness and
of the same dynamic characteristics (center-of mass location, mass
and moments of inertia) as the rotor, or part of a rotor, it replaces.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-7
Dynamic (two-plane balancing machine: Centrifugal balancing machine that
furnishes information for performing two-plane balancing.
NOTE:
Dynamic balancing machines are sometimes used to accomplish
single-plane balancing.
Dynamic unbalance: That condition in which the central principle axis is not
parallel to and does not intersect the shaft axis.
NOTE:
The quantitative measure of dynamic unbalance can be given by two
complementary unbalance vectors in two specified planes (perpendicular
to the shaft axis) which completely represent the total unbalance of the
rotor.
Equivalent nth modal unbalance: The minimum single unbalance U
ne
equivalent to
the nth modal unbalance in its effect on the nth principal mode of the
deflection configuration.
Field balancing: The process of balancing a rotor in its own bearings and
supporting structure rather than in a balancing machine.
NOTE:
Under such conditions, the information required to perform balancing is
derived from measurements of vibratory forces or motions of the
supporting structure and/or measurements of other responses to rotor
unbalance.
Field balancing equipment: Assembly of measuring instruments for providing
information for performing balancing operations on assembled machinery
which is not mounted in a balancing machine.
Flexible rotor: Rotor not satisfying definition 2.2 because of elastic deflection.
Floor-to-floor time:Time including the time for all necessary balancing runs and
measuring runs, together with the times for loading and unloading.
Foundation: Structure that supports the mechanical system.
NOTE:
In the context of the balancing and vibration of rotating machines, the term
foundation is usually applied to the heavy base structure on which the whole
machine is mounted
Gravitational (non-rotating) balancing machine: Balancing Machine that provides for
the support of a rigid rotor under non-rotating conditions and
provides information on the amount and angle of the static
unbalance.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-8
Hard-bearing (force-measuring, below resonance) balancing machine: Machine
having a balancing speed range below the natural frequency of the
suspension-and-rotor system
High speed balancing (relating to flexible rotors): Procedure of balancing at a
speed where the rotor to be balanced cannot be considered to be
rigid.
Inboard rotor: A two-journal rotor which has its center of mass between the
journals, without having significant mass outside the journals.
NOTE:
For a precise description of the rotor, it may be necessary to state the
positions of the center of mass and of the correction planes
indexing: Incremental rotation of a rotor about its journal axis for the purpose of
bringing it to a desired position.
Index balancing (as applied to multipart rotor assemblies): A procedure whereby
each part of a multipart rotor assembly is corrected within itself for the
unbalance errors in it, and caused by it, by indexing one part of the
assembly with respect to the remainder.
NOTES:
1. Index balancing is normally carried out by balancing a multipart
rotor to within desired limits, indexing a specific part through 180
with respect to the remainder and correcting half the indexing
unbalance in each part.
2. If 180 indexing is not possible, other angles can be used: in that
case, however, vector calculations may be required.
Indexing unbalance: The change in unbalance, indicated after indexing two
components of an unbalanced rotor assembly in relation to
each other, which is usually caused by individual
component unbalance, run-out of mounting (locating)
surfaces, and/or loose fits.
NOTE: Given repeatability of the interface fit. The change in unbalance
measured in one component after indexing by 180 is twice the error
in or resulting from the mating component.
Initial unbalance: Unbalance of any kind that exists in the rotor before balancing.
J ournal: That part of the rotor which is in contact with or supported by a bearing in
which it revolves.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-9
J ournal axis: Mean straight line joining the centroids of cross-sectional contours
of a journal.
J ournal center: Intersection of the journal axis and the radial plane of the journal
where the resultant transverse bearing force acts.
Local mass eccentricity (for distributed mass rotors): For small axial elements cut
from a rotor perpendicular to the shaft axis. The distance of the
center of mass of each element from the shaft axis.
Low speed balancing (relating to flexible rotors): Procedure of balancing at a speed
where the rotor to be balanced can be considered to be rigid.
Mandrel; balancing arbor: Machined shaft on which work is mounted for balancing.
Mass centring: The process of determination of the rotors principal axis of inertia
followed by the machining of journals, centers or other reference
surfaces to bring the axis of rotation, determined by these surfaces,
into close proximity with the principal axis.
Mass eccentricity: The distance of the center of mass of a rigid rotor from the shaft
axis.
Master rotor: A calibration rotor with provision for adding calibration masses at a
known location and used for periodically checking the calibration of
a balancing machine.
Measuring plane: Plane perpendicular to the shaft axis in which the unbalance vector
is determined.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-10
Measuring run (on a balancing machine): A run consisting of the following steps.
a) mechanical adjustment of the machine, including the drive, tooling and/or
adaptor.
b) setting of the indicating system.
c) preparation of the rotor for the balancing run.
d) acceleration of the rotor.
e) measurement of the unbalance.
f) deceleration of the rotor.
g) any further operations necessary to relate the readings obtained to the
actual rotor being balanced.
h) any other required operation, e.g., safety measures.
NOTES:
1. In the case of mass production balancing, steps a) and b) are
usually omitted from the initial measuring run. For subsequent
measuring runs, steps a), b) and c) are omitted in all cases.
2. A measuring run is sometimes referred to as a check run.
Mechanical adjustment: of a balancing machine, the operation of preparing the
machine mechanically to balance a rotor.
Method of correction: Procedure by which the mass distribution of a rotor is
adjusted to reduce unbalance, or vibration due to
unbalance, to an acceptable value. Corrections are usually
made by adding material to, or removing it from, the rotor.
Minimum achievable residual unbalance U
mar
: The smallest value of residual
unbalance that a balancing machine is capable of achieving.
Minimum achievable residual specific unbalance U
emar
: the smallest value of residual
specific unbalance that a balancing machine is capable of achieving under
given conditions.
Modal balancing: Procedure for balancing flexible rotors which affects only the nth
principle mode of the deflection configuration of a rotor/bearing system.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-11
Modal balance tolerance: With respect to a mode, that amount of equivalent modal
unbalance that is specified as the maximum below which the state of
unbalance in that mode is considered to be acceptable.
Multiple-frequency vibration: Vibration at a frequency corresponding to an integral
multiple of rotational speed.
NOTE:
This vibration may be caused by anisotropy of the rotor, non-liner
characteristics of the rotor/bearing system, or other causes.
Multi-plane balancing: As applied to the balancing of flexible rotors, any balancing
procedure that requires unbalance correction in more than two
correction planes.
Nodal Bar: A rigid bar coupled through bearings to a flexibly supported rigid
rotor, its motion being essentially parallel to that of the shaft axis.
NOTES:
1. Its function is to provide correction plane separation by locating the
motion transducer at centers of rotation corresponding to centers of
percussion located in connection planes.
2. A motion transducer so located has minimum correction plane
interference ration.
nth modal unbalance: The unbalance which affects only the nth principal mode of the
deflection configuration of a rotor/bearing system.
Overhung (outboard) rotor: A two-journal rotor with significant mass located outside
the journals.
Parasitic mass: Of a balancing machine, any mass, other than that of the rotor being
balanced, that is moved by the unbalance force(s) developed in the rotor.
Perfectly balanced rotor: An ideal rotor which has zero unbalance.
Permanent calibration: That feature of a hard-bearing balancing machine that permits it
to be calibrated once and for all, so that it remains calibrated for any rotor
within the capacity and speed range of the machine.
Plane separation: Of a balancing machine, the operation of reducing the correction plane
interference ratio for a particular rotor.
Plane separation (nodal) network: Electrical circuit, interposed between the motion
transducers and the unbalance indicators, that performs the plane
separation function electrically without requiring particular locations for the
motion transducers
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-12
Polar correction: Correction of amount of unbalance in a correction plane by mass
addition or subtraction at a single angular location.
Practical correction unit: Unit corresponding to a unit value of the amount of
unbalance indicated on a balancing machine. For convenience, it
is associatedwith a specific radius and correction plane and is
commonly expressed as units of an arbitrarily chosen quantity such
as drill depths of given diameter, weight, lengths of wire solder,
plugs and wedges.
Principal axis location: The axis location defined by the offset of the center of mass
from the design axis and the tilt angle of the principal axis
from the design axis.
Principal inertia axes: The coordinate directions corresponding to the principal
moments of inertia l
xixj
(I = j).
For each set of Cartesian coordinate directions at a given point, the values
of the six moments of inertia I
xixj
(I,j =1,2,3) of a body are in general unequal:
for one such coordinate system, the moments I
xixj
(i = j) vanish.
The values of I
xixj
(i = j) for this particular coordinate system are called the
principal moments of inertia and the corresponding coordinate directions
are called the principle axes of inertia.
NOTES:
1. I
xixj
= j
m
I
xixj
= (r
2
x
i
2
) dm, if I = j
m
where
r
2
= x1
2
+x2
2
+ x3
2
xi, xj are Cartesian coordinates.
2. If the point under consideration is the center of mass of the body, the
axes and moments are called central principal axes and central
principal moments of inertia respectively.
3. In balancing, the term principal inertia axis is used to designate the
central principal axis (of the three such axes) most nearly coincident
with the shaft axis of the rotor, and is sometimes referred to as the
balance axis of the mass axis.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-13
Production rate: Reciprocal of floor-to-floor time.
NOTE: The rate is normally expressed in pieces per hour.
Proving (test) rotor: Rigid rotor of suitable mass designed for testing balancing
machines and balanced sufficiently to permit the introduction
of exact unbalance by means of additional masses with high
reproducibility of the magnitude and angular position.
Quasi-rigid rotor: Flexible rotor that can be satisfactorily balanced below where
significant flexure of the rotor occurs.
Quasi-static unbalance: That condition of unbalance for which the central principal axis
intersects the shaft axis at a point other than the center of gravity.
Reference plane: Any plane perpendicular to the shaft axis to which an amount of
unbalance is referred.
Residual (final) unbalance: Unbalance of any kind that remains after balancing.
Resonance balancing machine: Machine having a balancing speed
corresponding to the natural frequency of the
suspension and rotor system.
Resultant unbalance force: Resultant force of the system of centrifugal forces of
all mass elements of a rotor referred to any point on
the shaft axis, provided that the rotor revolves about
the shaft axis.
NOTE:
The resultant unbalance force always lies in the plane
containing the center of gravity of the rotor and the shaft axis.
Resultant unbalance moment: resultant moment of unbalance forces: The re-
sultant moment of the system of centrifugal forces of all mass elements of
the rotor about a certain reference point in the plane containing the center
of gravity of the rotor and the shaft axis.
NOTES:
1. The angle and the magnitude of the resultant moment
depend in general on the position of the reference point.
2. There exists a certain position of the reference point in which
the magnitude of the resultant moment reaches its minimum
(center of unbalance).
3. The resultant moment is independent of the position of the
reference point in the case where the resultant unbalance
force is zero.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-14
Rigid free-body balancing: Procedure by which the mass distribution of a rigid
free-body is checked and, if necessary, adjusted to
ensure that the principal axis location is within
specific limits.
Rigid free-body unbalance: On a balancing machine the condition that exists in
any rotating rigid-free-body, when rotary motion is
imparted about its spin axis as a result of centrifugal
force(s).
NOTE:
1. The rotating motion of the principal axis may be cylindrical or
conical or a combination of both.
Rigid rotor: A rotor is considered to be rigid when its unbalance can be corrected in
any two (arbitrarily selected) planes (see 4.9). After the correction, its
residual unbalance does not change significantly (relative to the shaft
axis) at any speed up to the maximum service speed and running under
conditions which approximate closely to those of the final supporting
system.
NOTE:
A rotor which qualifies as a rigid rotor under one set of conditions, such as
service speed and initial unbalance, may not qualify as a rigid under
other conditions
Rigid-rotor-mode critical speed: Speed of a rotor at which there is maximum motion of
the journals and where that motion is significantly greater than the flexure of
the rotor
Rigid free-body: System of particles with rigid internal connections and no external
constraints.
Rotating rigid free-body: Rigid free-body rotating about an axis.
NOTE: The rotation axis is not stationary if it is not a central principle axis.
Rotor: Body, capable of rotation, generally with journals which are supported by
bearings.
NOTE:
The term rotor is sometimes applied to, for example, a disk-like
mass that has no journals (e.g. a fly wheel). In the sense of the definition
of 2.1, such a disk-like mass becomes a rotor for the purpose of balancing
only when it is placed on a shaft with journals (see 2.4).
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-15
(Rotor) flexural critical speed: Speed of a rotor at which there is maximum flexure of
the rotor and where that flexure is significantly greater than the motion of
the journals.
(Rotor) flexural principle mode: For undamped rotor/bearing systems, that mode
shape which the rotor takes on at one of the (rotor) flexural critical speeds.
Self-balancing device: Equipment which compensates automatically for changes in
unbalance during normal operation.
Sensitivity switch: A control used to change the maximum amount of unbalance that
can be indicated in a range or scale, usually in steps of 10.1 or
smaller.
Service speed: Rotational speed at which a rotor operates in its final installation or
environment.
The definitions in this clause apply to unbalance in rigid rotors.
They may also be applied to flexible rotors, but because
unbalance in such rotors changes with speed, any values of
unbalance given for those rotors must be associated with a
particular speed.
Setting: Of a hard-bearing balancing machine, the operation of entering into the
machine information concerning the location of the correction planes, the
location of the bearings, the radii of correction, and the speed if
applicable.
Shaft axis: The straight line joining the journal centers.
Single-plane (static) balancing:Procedure by which the mass distribution of a
rigid rotor is adjusted to ensure that the
residual static unbalance is within specified limits.
NOTE:
Single-plane balancing can be carried out on a pair of knife edges
without rotation of the rotor but is now more usually carried out on
centrifugal balancing machines.
Single-plane (static) balancing machine: Gravitational or centrifugal balancing
machine that provides information for performing two-plance
balancing.
Soft-bearing (above-resonance) balancing machine: Machine having a balancing
speed above the natural frequency of the suspension-and-rotor
system.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-16
Specific unbalance: The amount of static unbalance U divided by the mass M of the
rotor.
NOTES:
1. The specific unbalance is numerically equivalent to the mass
eccentricity .
2. In the case of a rotor with two correction planes, specific unbalance
sometimes refers to the unbalance in one plane divided by the
rotor mass allocated to that plane
Static unbalance: That condition of unbalance for which the central principal axis is
displaced only parallel to the shaft axis.
NOTE:
The quantitative measure of static unbalance can be given by the
resultant of the two dynamic unbalance vectors.
Swing Diameter: Maximum work-piece diameter that can be accommodated by a
balancing machine.
Test mass: A precisely defined mass used in conjunction with a proving rotor to test a
balancing machine.
NOTES:
1. The use of the term test weight is depreciated: the term test
mass is accepted in international usage.
2. The specification for a test mass should include its mass and its
center-of-mass location, the aggregate effect of the errors in these
values should not have a significant effect on the test results.
Test plane: A plane perpendicular to the shaft axis of a rotor in which test masses may
be attached.
Thermally induced unbalance: That change in condition exhibited by a rotor if its
state of unbalance is significantly altered by its changes in temperature.
Trial mass: a mass selected arbitrarily (or prior experience with similar rotors) and
attached to a rotor to determine the rotor response.
Traverse test: Test by which the residual unbalances of a rotor can be found
(see ISO 1940-1) or which a balancing machine may be tested for
conformance with the claimed minimum achievable residual
unbalance U
mar.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-17
Two-plane (dynamic) balancing: Procedure by which the mass distribution of a rigid
rotor is adjusted to ensure that the residual dynamic
unbalance is within specified limits.
NOTE:
A trial mass is usually used in trial and error balancing or
field balancing where conditions cannot be precisely
controlled and/or precision measuring equipment is not
available.
Unbalance: That condition which exists in a rotor when vibratory force or motion is
imparted to its bearings as a result of centrifugal (See clause 3)
NOTES:
1. The term unbalance is sometimes used as a synonym for
amount of unbalance or unbalance vector.
2. The term imbalance is sometimes used in place of
unbalance, but this is depreciated.
3. Unbalance will in general be distributed throughout the rotor
but can be reduced to:
a) static unbalance and couple unbalance described by
three unbalance vectors in three specified planes, or
b) dynamic unbalance described by two unbalance
vectors in two specified planes.
Unbalance bias of a mandrel (balancing arbor): A known amount of unbalance
added to a balancing arbor.
NOTE: Biasing a balancing arbor generally serves the purpose of either
compensating for the residual unbalance that run-out of the
balancing arbors rotor mounting surface causes when this single
balancing arbor is used in balancing a series of rotors of the same
mass or introducing a specified unbalance at a specific
angular position for the purpose of balancing parts which, after
being removed from the balancing arbor, are to have a specified
unbalance.
Unbalance couple:For the case where the resultant unbalance force is zero, the
resultant couple of the system of centrifugal forces of all mass
elements of the rotor.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
A-18
Unbalance force: In a rotor referred to a correction plane, the centrifugal force at a
given speed (referred to the shaft axis) due to the unbalance in that
plane.
Unbalance mass: That mass which is considered to be located at a particular radius
such that the product of this mass and its centrifugal acceleration is
equal to the unbalance force.
NOTE:
The centrifugal acceleration is the distance between the shaft axis and the
unbalance mass multiplied by the square of the angular velocity of the
rotor, in radians per second.
Unbalance moment: Moment of a centrifugal force of a mass element of a rotor
about a certain reference point in the plane containing the
center of gravity of the rotor and the shaft axis.
Unbalance reduction ratio (URR): The ratio of the reduction in the unbalance by a
single unbalance correction to the initial
unbalance.
Where:
U
1
is the amount of initial unbalance
U
2
is the amount of unbalance remaining after one correction.
NOTES:
1. The unbalance reduction ratio is a measure of the overall efficiency
of the unbalance correction.
2. The ratio is usually given as a percentage.
Unbalance vector: Vector whose magnitude is the amount of unbalance and whose
direction is the angle of unbalance.
Vertical axis freedom: Freedom of a horizontal balancing machine bearing
carriage or housing to rotate by a few degrees about the
vertical axis through the center of the support.
Vector measuring device: Device for measuring and displaying the amount and angle
of unbalance in terms of an unbalance vector, usually by
means of a point or line.
Vibration transducer plane: Plane perpendicular to the shaft axis in which the
vibration transducer in located.
URR = U
1
- U
2
U
2
= 1
U
1
U
1
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-1
APPENDIX
B
WEIGHT REMOVAL CHARTS WEIGHT REMOVAL CHARTS
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-3
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-4
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-5
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-6
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-7
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-8
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-9
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-10
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-11
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-12
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-13
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-14
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-15
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-16
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-17
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-18
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-19
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-20
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-21
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-22
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-23
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-24
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-25
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-26
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-27
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-28
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-29
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-30
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
B-31
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
C-1
APPENDIX
C
CONVERSION CHART FOR CONVERTING CONVERSION CHART FOR CONVERTING
INCHES OF FLA INCHES OF FLAT STOCK #1020 STEEL TO T STOCK #1020 STEEL TO
OUNCES OF WEIGHT OUNCES OF WEIGHT
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
C-2
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
C-3
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
C-4
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
D-1
THREE-POINT METHOD OF BALANCING THREE-POINT METHOD OF BALANCING
The following procedure outlines a three-point method for balancing rotors where the
use of a reference pickup or strobe light to read phase is not practical or possible. This
three-point method is often preferred to the familiar two-point method because it does
determine the location of the balance correction weight. This may be particularly
important where it may be necessary to divide balance weights between adjacent
blades of a fan or similar rotor.
1. With the rotor operating at normal speed, measure and record the original
vibration amplitudes as O.
For our example : O = 6 mils (152 micron)
2. Draw a circle with a radius equal to O, as shown in Figure 1.
For our example this circle will have a radius of O = 6 mils (152 micron)
3. Stop the rotor. Mark on the rotor three (3) points; A, B and C
approximately 120 apart. These three points on the rotor need not be
exactly 120 apart, however, the precise angles of separation, whatever
they may be, must be known. In our example we will not position our
test weights equally spaced, just to show that this can be done.
FIGURE 1
APPENDIX
D
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
D-2
In our example, Figure 2, point A is our starting point and is thus
considered 0. Mark the respective positions of points A, B and
C on the original circle as shown in Figure 2.
FIGURE 2
4. Select a suitable trial weight and attach at position A on the rotor.
Refer to the formulas for calculating a safe trial weight as required.
For our example, trial weight (TW) = 10 ounces (283.5 grams)
5. Start the rotor and bring to normal operating speed. Measure and
record the new vibration amplitude as O + T
1
.
For our example O + T
1
= 4 mils (102 micron)
6. Using point A on our original circle as the center point, draw a circle
with a radius equal to O + T
1
.
For our example, this circle will have a radius of O + T
1
= 4 mils (102
micron), as shown in Figure 3.
7. Stop the rotor and move the trial weight to position B.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
D-3
FIGURE 3
8. Start the rotor and bring to normal operating speed. Measure and
record the new vibration amplitude as O + T
2
.
For our example: O + T
2
= 8 mils (203 micron).
9. Using point B on our original circle as the center point, draw a circle
with a radius equal to O + T
2
.
For our example, this circle will have a radius of O + T
2
= 8 mils (203
micron), as shown in Figure 4.
10. Stop the rotor and move the trial weight to position C.
11. Start the rotor and bring to normal operating speed. Measure and
record the new vibration amplitude as O + T
3
.
For our example, this circle will have a radius of O + T
3
= 11 mils
(279 micron).
12. Using point C on our original circle as the center point, draw a circle
with a radius equal to O + T
3
.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
D-4
FIGURE 4
For our example, this circle will have a radius of O + T
3
= 11 mils (279
micron), as shown in Figure 5.
FIGURE 5
Note from Figure 5 that the three circles drawn from points A, B and C
intersect at a common point D.
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
D-5
13. Draw a line from the center O of the original circle to point D, as
shown in Figure 6. Label this line T.
14. Measure the length of line T, using the same scale used in drawing
the circles.
For our example, this line T is 5.25 (133 mm) as measured in Figure 6.
FIGURE 6
15. Calculate the amount of the balance correction weight, using
the formula:
CW = TW (O/T)
Where:
CW = Correct Weight
TW = Trial Weight
O = The Original Unbalance Reading
T = The measured Resultant Vector
For our example problem, the solution is found as follows:
CW = TW (O/T) CW = TW (O/T)
CW =10 oz. X (6.0/5.25) CW = 283.5 grams X (152 /133)
CW =11.4 oz. CW = 323 grams
Techncial Associates Field Dynamic Balancing
Copyright 2001 Techncial Associates of Charlotte, P.C.
D-6
16. Using the protractor, measure the angle between line T and line
OA, as shown in Figure 7. THIS MEASURED ANGLE IS THE
ANGULAR LOCATION OF THE CORRECT WEIGHT, LOCATED
RELATIVE TO POINT A ON THE ROTOR.
FIGURE 7
For our example problem, this angle is measured to be 41.
17. Stop the rotor and remove the original trial weight from point C.
18. Attach the correct weight calculated in Step 15 above to the rotor at
the angular position determined from Step 16.
For our example problem, the calculated correct weight of 11.4 oz.
(323 grams) is added to the rotor at an angular position 41 clockwise from
position A on the rotor, as shown in Figure 8.
FIGURE 8
With the balance correction weight calculated and located on the rotor in accordance with
the above instructions, the rotor should now be balanced.