Implementing Reliability Improvements…a team approach

By Timothy A. Dunton and Michael L. Snider

Universal Technologies. Inc.

Abstract

The progress of machinery and component condition monitoring during the past twenty years
due to technology advancement and information gained from statistical analysis of equipment
failures is remarkable. Many companies have evolved to the point where breakdown
maintenance and preventive maintenance are the predominant maintenance approaches only
when reliability-centered maintenance (RCM), Streamlined Reliability Centered Maintenance
(SRCM), and other maintenance review processes indicate that these more rudimentary
approaches are justified.

Despite these improvements in technology, and a seemingly endless stream of new

“maintenance management strategies”, plant / maintenance performance is in many cases not
where it should be… Why?

This paper explores the common pitfalls and barriers to implementing real improvements in
machinery reliability and plant / maintenance performance and presents practical real world
procedures and processes for improving machinery reliability. We will examine the problem
from several perspectives including:

• Costs and benefits

• The decision making process
• Who should be involved?
• Standards, tolerances and specifications
• Required procedures, processes and documentation
• Training and task qualification requirements
• Implementing quality assurance checks
• Dealing with motivational/cultural issues
• A 12 step road map to reliability improvement

Page 1 of 1
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 Table of Contents
Implementing Reliability Improvements…a team approach .................................................. 1
Abstract..................................................................................................................................... 1
Introduction ............................................................................................................................... 3
Costs and Benefits.................................................................................................................... 6
The Maintenance Decision Making Process............................................................................. 8
Who should be involved?........................................................................................................ 10
Standards, Tolerances and Specifications.............................................................................. 11
Required Procedures, Processes and Documentation ........................................................... 12
Training/Task Qualification Requirements .............................................................................. 14
Implementing Quality Assurance Checks ............................................................................... 16
Dealing with Motivational and Cultural Issues ........................................................................ 18
Reliability Scenario ................................................................................................................. 20
12 Step Roadmap ................................................................................................................... 21
Appendix ................................................................................................................................... 22
New/Rebuilt Motor Vibration Test Form (0- 20 HP) ................................................................ 23
New/Rebuilt Motor Vibration Test Form (20 - 400 HP) ........................................................... 24
New/Rebuilt Motor Vibration Test Form (400 HP +) ............................................................... 25
Sample Balance Record Form................................................................................................ 26
Sample Centrifugal Pump Repair Form .................................................................................. 28

Page 2 of 2
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Introduction

The progress of machinery and component condition monitoring during the past twenty years is
remarkable. This is largely due to technology advancement and information gained from
statistical analysis of equipment failures. Most companies have now evolved to the point where
breakdown maintenance and preventive maintenance are the pre-dominant maintenance
approaches only when detailed studies such as reliability-centered maintenance (RCM),
Streamlined Reliability Centered Maintenance (SRCM), and other maintenance review
processes indicate that these more rudimentary approaches are justified. On critical equipment,
repetitive failures, and machinery where repair costs are high, a multitude of technologies are
available for diagnosing the failures.

Using such technologies as vibration data collection and

analysis, oil analysis, ultrasonics, infrared thermography,
performance monitoring, and motor current signature
analysis, we have the ability to accurately and effectively
monitor machinery condition and schedule repairs only when
machine symptoms indicate deterioration. This ability alone
will result in improvements in effectiveness of the
maintenance organization. However, most companies have
begun to realize the need to go further…not just to accurately
detect impending machinery and component failures, but to
pinpoint and correct the root causes of failure. After all, most
failures are preventable…if we find and correct the root
cause, rather than just treat the symptoms.

Even when we succeed in selecting and implementing the correct maintenance approaches,
implementing appropriate condition monitoring technologies, and performing maintenance tasks
in a precision manner, premature equipment failures continue to dominate. Reliability goals are
not achieved, mean times between failures are too low, maintenance costs are too high, and
production goals are not met. For even the most successful programs, opportunities for
improvement are tremendous.

As specialists in the development and delivery of technical training related to maximizing the life
of rotating machinery, we at Universal Technologies have experience with a variety of personnel
and condition-based reliability programs from a wide diversity of industries, ranging from power
generation, petrochemical, automotive, mining, paper, steel, and other manufacturing. We
conduct over a hundred technical training seminars per year. Yet our primary goal IS NOT to
conduct training, but to ensure the subjects of our seminars are effectively put into practice.

Much too often, having attended a seminar or training course that shared sound technical
information, the attendees return to the plant with new information and the motivation and
inspiration to make a change. They then, for a variety of reasons, hit a brick wall and the
concepts learned are never implemented. Why?

Page 3 of 3
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Consider a few scenarios that may sound much too familiar:

Case 1: A repetitive rolling element bearing failure in a critical centrifugal pump.

The bearing is a double row angular contact bearing,

serving as the fixed bearing in the pump. It is a catastrophic
failure. Examination of the most recent bearing shows build
up of cage material in the raceways, one of the two steel
cages destroyed. High heat multiple-color discoloration is
present. Vibration data collection has been performed, but
no fundamental bearing defect frequencies were noted in
the vibration spectrum. During the last repair, technicians
replaced the bearing with one from a different manufacturer,
but the vibration analysis department was not informed of
the change. The software was setup to look for defect
frequencies of the old bearing type. Further investigation
reveals that the application requires a bearing with no
shields or seals, yet all that are stocked are shielded
bearings. Further inspection shows where a technician
pried out the shields with a screwdriver, most likely bending the cage in the process.

To prevent recurrence of this failure, where do we begin? Do we train all technicians at

the facility how to properly remove shields without damaging the cage? How do we
ensure that when bearing types or manufacturers are changed that the reliability
department is informed so that vibration software can be updated? Why do we only
stock shielded and sealed bearings of this type when all the pump applications in this
particular facility are oil-lubricated and require no seals or shields? Is only one type of
bearing stocked by purchasing to reduce inventory?

Case 2: Four repetitive mechanical seal failures in one week.

These mechanical seals lasted less than a day.

After the first two failures, the seal manufacturer
is contacted to verify the seal is correct for this
application. Two highly trained mechanical
technicians are selected to install the new seals
while being supervised by an engineer from the
seal manufacturer. Root cause analysis reveals
the reason for the seal failure to be a lack of seal
flush. Further investigation shows that this pump
is being started from the control room with no
operator at the pump. The correct pump
operating procedure requires an operator to start
the pump locally after opening the seal flush line.

Page 4 of 4
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Case 3:Successive motor failures on two new chiller pumps shortly after being installed
by a contractor.

Symptoms include:

1. Horizontal misalignment at the rear

feet of the machine of 0.233”.
2. Motor is bolt bound.
3. Mounting pads have been welded,
warping surfaces and inducing soft
foot of 0.022.
4. Pipe strain moves pump 0.042 at
non-drive end bearing. The
problem is resolved by on-site
maintenance personnel after the
second failure on one of the motors.

In this case, the real root cause is a lack of acceptance standards between
engineering and the contractor regarding machinery installation and alignment
procedures.

Cases like these occur daily. Many more examples could be cited. The bottom line is that there
are many challenges that must be dealt with in order to effectively implement reliability-centered
programs. Having studied this subject carefully we have found several common areas where the
process breaks down. Several key aspects must be addressed, such as:

• Costs and Benefits

• Maintenance Decision Making Process
• Who Should Be Involved
• Standards, Tolerances and Specifications
• Required Procedures, Processes and Documentation
• Implementing Quality Assurance Processes
• Training and Task Qualification Requirements
• Dealing with Motivational and Cultural Issues

Page 5 of 5
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Costs and Benefits

Essentially, we must make maintenance decisions

based on business sense, not emotion or past
practices. There are many different ways to
calculate costs and determine the pro’s and con’s
associated with different forms of maintenance.
Some of the factors to be considered are described
below.

• It is not cost effective to apply all precision

techniques across the board. Though certain
elements of precision maintenance do not result
in higher maintenance costs and can be applied for all maintenance tasks, care must be
taken to ensure “full blown” precision maintenance tasks are carefully selected. For
example, consider key length. When installing any keyed component, a key must be cut.
To cut it to the correct length involves no additional cost. On the other hand, consider a
repetitive failure of a boiler feed pump due to severe misalignment and excessive pipe
strain. To correct this problem will require the pump to be out-of-service for an extended
period of time and involve costly piping modifications. The latter task must be very carefully
planned and a cost benefit analysis performed.

• Effective key performance indicators (KPI’s) must be established and maintained to track
progress of reliability efforts. Without use of KPI’s many successful programs are eventually
terminated. As reliability and production goals increase, often the perceived need of
precision maintenance subsides.

• There is no “default” set of KPI’s. They must be selected for each facility and even for
certain systems and components within the facility. Be careful of establishing maintenance
costs as a KPI. It is easy to cut maintenance costs by stopping maintenance, but the long
term consequences are obviously adverse.

• Use risk management principles to determine the most appropriate maintenance strategy for
each system and component. Risk can be defined as:

Risk = Probability x Consequence

A variety of risk management processes are available for making risk based decisions.
Essentially, the task involves weighing the consequences of failure as well as the probability
of failure. A thorough look at all of the consequences of failure is important, Typical
consequences in a plant environment include:

ƒ Production losses
ƒ Production opportunity losses
ƒ Costs of repair due to catastrophic failure vs. a scheduled repair
ƒ Environmental costs
ƒ Safety costs

Page 6 of 6
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
• It should be noted that the consequences of failure are frequently in transition due to factors
that are often beyond the control of any plant level personnel. Such factors include market
forces, regulatory requirements, etc. As these variables change, failure consequences
change and therefore risk changes.

• One model for communicating risk is to track “at risk” maintenance dollars.

1. This can be done by creating a “watch list” of all machines that are in a failure mode and
being monitored by the Condition Monitoring group.

2. Assign a cost figure to failures that are currently on the watch list. Determine the cost of
each failure assuming that an “unscheduled” failure will occur, and include estimated
production losses, maintenance costs, and safety/environmental costs.

3. Add all the failure costs for all “at risk” machines to get a total figure. This is the amount
of money that is currently “at risk” due to machinery failure.

4. This total can then be compared to production and or profit figures to establish what
percentage of plant production or profit is currently at risk.

5. Work with management to determine a “risk” percentage figure that they feel comfortable
carrying in the current operating context. Use the above calculations to manage the risk.

As operating context and failure consequences change the “at risk” percentage will change. By
prioritizing the machines on the list by failure cost it is possible to determine the appropriate
repairs to perform that will have the biggest impact on the amount of risk that the plant is
currently carrying.

Page 7 of 7
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
The Maintenance Decision Making Process

While everyone at a facility has some

impact on reliability, it is essential that all
plant personnel who directly or indirectly
influence the maintenance decision making
process or are involved in maintenance or
operations understand how and why
machines fail and reliability concepts.

An overview of machinery failures is provided by 6 common failure patterns, as shown below:

Use a recognized logical process such as Reliability Centered Maintenance (RCM) to help in
determining the appropriate strategy. There are seven (7) questions for studying
machines/systems from an RCM perspective:

A. What are its’ functions and what do its’ users want it to do?
B. In what ways can it fail? FEMA (failure effects and modes analysis)
C. What causes it to fail? RCFA (root cause failure analysis)
D. What happens when it fails? FEMA
E. Does it matter if it fails? FEMA
F. Can anything be done to detect/predict/prevent the failure?
G. What if we cannot detect/predict/prevent the failure?

Page 8 of 8
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Determine the costs of condition monitoring in your facility. To accomplish this, calculate the
sum of the following:

• Salaries plus benefits of condition monitoring personnel

• Condition Monitoring Department Overhead
• Instrumentation and Software Costs amortized over 2-3 years
• Software Support and Upgrades
• Condition Monitoring Personnel Training Costs

Divide this sum by the number of points monitored per month to derive a per point or per
machine cost.

Calculate or estimate the consequences of failure. These can include lost production, lost
production opportunity, maintenance costs, environmental and safety costs.

This will enable you to make good business decisions regarding whether or not it is appropriate
to perform condition monitoring or use some other maintenance strategy on a specific piece of
equipment or system at your facility.

Once detailed decision making processes are established at your facility, ensure everyone has
a basic understanding of how the process works. Select a small group of machines historically
demonstrating a reliability problem; apply these processes; document and evaluate successes;
and, fine-tune the process.

Page 9 of 9
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Who should be involved?

Everyone in the organization has a role in reliability improvement. A common misconception is

that the Maintenance department is solely responsible for solving machinery reliability problems.
While it is obviously true that the Maintenance department has primary responsibility for actual
performance of maintenance tasks, results of failure analysis prove that the true root causes of
premature component failure often lie outside the maintenance department.

To illustrate the need for all parties to be involved, consider the following questions:

1. Who determines the general time frame in which the task will be performed and how much
time the equipment can be out of service?
2. Who decides if the amount of time allotted will be sufficient for the type of work that needs to
be performed in order to truly correct the root cause?
3. Who determines the quality of replacement parts, e.g., that all rolling element bearings meet
or exceed ABEC and RBEC grades of precision for the application?
4. Who ensures that before contracts are issued on new equipment installations the contractor
will meet or exceed the quality standards and specifications used by maintenance for follow-
up tasks.
5. Who ensures that before the sign-off occurs on a new installation that these standards have
indeed been satisfied?
6. Who ensures that start-up and operating procedures do not contribute to premature
component failure?
7. Are maintenance personnel informed prior to performing a repair the nature of a pump’s
poor performance from a production perspective?
8. Are consultants who determine lubrication requirements provided the details needed to
select the right type of lubricant for specific applications, for example the bearing operating
temperatures?
9. If contractors are used to provide services with various PdM technologies, how are they
informed when a pump impeller with five (5) vanes is changes to one with six (6) vanes or
that an SKF 7203A bearing is replaced with a Fafnir 7203?
10. Who ensures that when bearings arrive at the door of the shipping and receiving
department, they are not dropped or mishandled in a manner that destroys the bearing due
to true brinelling of the bearing material?

Even without answering all these questions, it is hopefully evident that a team approach is not
only desirable; it is a mandatory element of effective machinery reliability programs.

Page 10 of 10
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Standards, Tolerances and Specifications

An unfortunate reality is that in a majority of cases,

when we purchase a new piece of equipment, send an
existing machine out for repair, or even perform a
particular task in-house, the quality of work performed
varies significantly based on who’s performing the task,
what parts are available, and the methods of work
performance which are commonly accepted. Indeed
many of the sources of premature machinery and
component failure are traced back to poor parts, poor
tolerances, poor practices, and poor standards.

Consider as an example a new or rebuilt electric motor.

Are the bearings that are installed the correct bearings?
Were the bearings damaged before they were installed?
How were they installed? Heated with a torch? Drove on with a hammer? If an induction
heater was used, were the bearings demagnetized after they were heated? Was the rotor
balanced? If so to what quality balance grade? And was the balance a single plane or two
plane balance? Was a precision soft foot check performed on the motor feet? How was the
motor transported? Did false brinelling or true brinelling occur during transport? And if any of
the above are in question, did we detect the problem by running the motor solo and checking
the vibration before installing the motor?

Among the standards, tolerances and specifications that should be established and
implemented are the following:

1. Approved Balancing Methods

2. Precision Balancing Tolerances
3. Precision Balancing Standards
4. Motor Solo Vibration Standards
5. Installed Component Vibration Standards
6. Condition Monitoring Tolerances
7. Foundation, Base and Machine Frame Standards
8. Approved Alignment Methods
9. Precision Alignment Tolerances
10. Precision Pipe Strain and Soft Foot Method and Standards
11. Approved Methods for Measuring Dynamic Movement
12. Component Part Specifications an Substitution Policies
13. New and Rebuilt Machinery Specification Requirements

With such detailed standards, tolerances, and specifications in place, our suppliers, vendors
and contractors, and internal personnel will know our expectations…that we expect reliability to
be built in from the start. Beyond establishing these standards, tolerances, and specifications,
appropriate processes must be implemented to train all involved parties on the rationale for the
standards, as well as how to interpret and apply them. Finally, mechanisms must be
established to monitor compliance with these standards and hold appropriate personnel
accountable.

Page 11 of 11
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Required Procedures, Processes and Documentation

In addition to establishing and implementing precision standards, tolerances and specifications,

essential to successful implementation of reliability programs is the development and
implementation of required procedures and
processes for certain tasks, including
adequate documentation requirements.
Use of these guidelines creates
expectations that things will be done a
certain way.

In addition, craftsman should view these

procedures as valuable tools. While
certain tasks, such as precision alignment
using a certain laser system, may be
performed on a regular basis by certain
personnel, there are many times when an
individual is assigned to perform a task not
recently performed. In such cases, a
written procedure provides guidance on
key elements of task performance and
helps to ensure efficient and high quality
task performance.

While very detailed step-by-step procedures are needed for certain tasks, shorter working
procedures providing an overview of major steps are sufficient in many cases. The level of
detail needed should be determined according to the nature of specific tasks and the frequency
of task performance.

A critical part of this process is inclusion of appropriate documentation. For example, for a
precision alignment task, documentation typically required includes a completed pre-alignment
checklist, as-found shaft alignment conditions, initial and final precision soft foot records, and
final alignment conditions. These documents create vital formal history sorely lacking in most
facilities. Wherever possible, we should utilize modern technology to create effective
documentation processes, e.g., electronic log sheets.

The specific list of written procedures, processes and documentation requirements needed
varies from facility to facility, largely based on how the organization is structured, the total
number of people who may be assigned a specific task, and the history of machinery failures
due to poor quality task performance. In general, a procedure or other formal process is
recommended for any task where inconsistency in work quality is predominant.

Procedures and processes are not limited to maintenance activities. Effective procurement
processes should be established to ensure that the correct parts and components of the
required quality are purchased. Examples include components balanced to the required quality
grade, bearings of the correct type and precision grade, belts that are matched, etc. Processes
should be in place to ensure machine start up procedures for specific equipment do not lead to
catastrophic failure of components such as mechanical seals, and that certain preventive
maintenance tasks performed by operations personnel are proper for the equipment.

Page 12 of 12
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Again, the goal here is not to create a mountain of unneeded paperwork or additional
Administrative burdens. The goal is to first identify specific tasks where inconsistent human
performance is a key factor in the mode of failure. Then, in addition to training of personnel,
development of effective procedural guidelines is often an effective alternative.

Page 13 of 13
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Training/Task Qualification Requirements

Appropriate processes must be in place to ensure all personnel that effect machinery reliability
are properly trained, and more importantly, are qualified to perform assigned tasks. Effective
training is outcome based and is developed, delivered, and evaluated using a Systematic
Approach.

1. Identify Key Tasks

First, key tasks must be identified for various personnel. For each work group, there are
common tasks that are performed on a regular basis. For example, common tasks for
maintenance personnel include repair of pumps, installation and alignment of pumps, repair of
valves, installation or modification of piping, etc.
There are also infrequently performed tasks that often fall under a high risk with respect to
reliability.
Key tasks, both common and infrequently performed that affect reliability should be identified for
each work group.

2. Analyze Key Tasks

After the tasks for each work group are identified, a detailed analysis of each task is performed
to identify key elements of task performance.

3. Perform Needs Analysis

Once each task is analyzed in such a manner, individual needs analysis is conducted to assess
the individuals’ abilities to perform assigned tasks.

4. Develop Action Plan

Results of this needs analysis and individual assessments are then used to develop an action
plan, including how areas where improvement is desired will be addressed.

This process can be conducted to various

levels of detail based on numerous variables
beyond the scope of this article. At first
glance, the process briefly described here
may appear a bit complex. However, even if
this process is shortened to a “table top”
format, it is highly recommended. And the
main reason is that this Systematic
Approach serves as the foundation for
development of a training plan that will be
cost effective and truly contribute to
increased machinery reliability goals.

Page 14 of 14
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
In addition, from such an analysis, clear learning objectives are identified for each subject.
These objectives serve as the foundation for training. Whether the training is developed and
conducted by in-house personnel or whether vendor courses are selected, the objectives
provide a clear means by which the training content can be evaluated based on the desired
learning outcomes.

Additional variables that should be considered relative to training include the nature of the
training itself. Training providers should be flexible in delivering training that is:

• Activity-Based,
• Customized according to your specific needs,
• Applies sound conceptual information to your specific instrumentation and internal
procedures.
• In addition, plans must be in-place for effectively evaluating the success of training that
is conducted, and identifying and addressing continuing training needs in each subject
area.

Typically, it is the infrequently performed

tasks are often key candidates for continuing
training as well as tasks that may need
procedures or job aids. For the purposes of
this continuing training, a variety of
alternatives and delivery mechanisms
should be considered, including periodic
task assessments, Computer-based training
products, and performance support tools.

Page 15 of 15
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Implementing Quality Assurance Checks

Another essential component of reliability improvement is implementation of some form of

quality assurance checks. The degree to which these checks are implemented is largely
dependent on the industry in which you are involved. For certain industries, such as the
military, the airline industry, and nuclear power, almost every task performed is carefully
monitored by an independent quality assurance organization. On the other extreme, throughout
most industries, such stringent measures are not typically in place. The question is to what
degree are these measures needed at your facility?

Quality assurance checks create an essential

expectation for performance. To achieve this, they
do not have to be conducted by an independent
organization as long as appropriate personnel
conducting these checks are equipped with
whatever mechanisms are needed to conduct the
checks in a consistent and qualified manner.

Assuming that the above elements of a good

reliability program are in place, the real question is
how do we demonstrate compliance with what has
been established. For example:

• How do we know that the maintenance decision making process is consistently applied?
• How do we know that written procedures are being properly used?
• How do we ensure all work by vendors, contractors, and in-house personnel complies with
established standards, tolerances, and specifications?

As important, on a daily basis, how do we prove that we are seeing the maximum benefit from
our reliability efforts? There are several practical processes that can be used to document the
results. Among them are the following:

• Pre-repair and post-repair vibration readings. (See appendix for sample.)

It is recommended that the craftsman who performs the repair perform the initial vibration
check after a repair has been completed. There are many simple inexpensive vibration
measurement devices that can be used for this procedure.
In addition, with such a process in place, maintenance personnel are embraced by
Condition Monitoring, rather than feeling alienated from the program. In addition, this
process can reduce the workload on full-time PdM personnel and at the same time increase
the number of vibration measurements obtained for the facility.

• Motor solo vibration checks. (See appendix for sample.)

Many organizations are reaping tremendous benefits by checking new and rebuilt motors
before putting them into service. These checks again create an expectation for
performance. They can be performed at the facility or, if a good relationship exists with the
vendor, at the vendor facility. A half key should be installed and the motor set on a rubber
pad during these checks. Remember that these checks will only confirm mechanical

Page 16 of 16
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 integrity. Some companies have invested in dynamometers to load test motors that also
checks electrical integrity.

• Balance reports. (See appendix for sample.)

Many companies now specify precision balancing tolerances, but how do we know that
these tolerances are being met. A simple balance record form containing as a minimum
rotor weight, RPM, and residual unbalance sets an expectation to be achieved. Larger,
more critical machines may warrant witnessing of the balancing process.

• Maintenance repair records. (See appendix for sample.)

Teaching people to perform precision maintenance tasks does not necessarily create an
expectation that the tasks be performed correctly outside of the classroom. While we find
most attendees leave the classroom highly motivated and excited about precision
maintenance, if we do not continue to nurture that motivation by creating an expectation that
real change will occur, the training is often wasted. By creating a group of practical, easy to
use forms to document results of a repair task, the expectation is communicated and
adequate documentation is obtained for machinery history files, two elements missing from
most industrial facilities today.

Page 17 of 17
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Dealing with Motivational and Cultural Issues

In many cases cultural and motivational problems stem from a lack of communication. As
managers and supervisors, we rarely have or make the time to explain the reasoning behind
certain decisions. Failure to do clearly communicate often results in distrust between
management and the workforce. By sharing information about the basic concepts of reliability,
financial issues, and how the maintenance decision making process works, employees will feel
less alienated and more empowered.

Be aware that it is a basic human instinct to resist change. The immediate response to any new
idea is often negative. Expect this response; it is normal. In order to truly effect
change, it is necessary to make the new idea or concept more attractive than the current way of
doing business. Most plant personnel genuinely want to do good work. More importantly, they
want to be seen as doing good work. By establishing measures that clearly demonstrate the
quality of work, these personal goals are often achieved.

Understanding resistance to change explains why cultural changes are slow. Expect changes
to occur slowly. Far too many plants attempt to implement radical changes in a short period of
time, only to find that the culture soon reverts back to its original, more comfortable state.

When trying to influence culture, many companies find more success by making small
incremental changes as opposed to sweeping monumental changes. For example, after
deciding to implement a new procedure for performing a maintenance task, e.g., shaft
alignment, choose a group of ten or so machines. Gather initial vibration data. Assign groups
of individuals to perform the tasks on these machines to new precision standards, using record
forms, etc. Gather post-repair vibration data and share the results.

Cultural change is all about momentum. Once well underway, renewed energy for reliability
improvement tends to be self-sustaining. However, especially in the early stage, it is extremely
important to be aware that even what might appear to be a minor setback can have disastrous
consequences. In every facility, there are certain individuals whose resistance to change is
stronger than average. Usually, these people are waiting for an excuse to criticize what’s new
and encourage the masses to return to the old status quo, where their comfort level is highest.

Conversely, once these particular individuals are converted to the cause, they often become the
program’s strongest champions.

Change is a process, not a program… A program, by definition, has a

start and a finish.

Page 18 of 18
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Be sensitive to inter-departmental issues. In many facilities a certain amount of “friction” exists
between the various departments, in particular between operations and maintenance. One
traditional perspective is that maintenance is somewhat subservient to production. This
relationship must become a true partnership if reliability goals are to be met.

To facilitate this, when planning and making important decisions regarding maintenance
strategy a group such as the one illustrated below should be utilized to ensure that all interested
parties are represented. This will help validate the group’s findings.

Maintenance
Maintenance Craftsperson(s)
Supervisor

Management Operator
Representative

PdM Operations
Representation Supervisor

Page 19 of 19
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
It is essential that personnel understand how the above factors relate to the reliability efforts at
the plant level.

Reliability Scenario

As an example of how the lack of a

team approach to reliability can cause
confusion amongst the ranks, consider
how market forces affect a reliability
problem on a boiler feed pump in a
coal-fired power station. There are 2
feed pumps both of which are required
to operate the plant at full capacity.
Mean time between failure on these
units is only 12 months. FEMA
analysis reveals repetitive seal and
bearing failures on the pump. The root
cause is determined as misalignment
due to excessive pipe strain.

Deregulation of the utility industry has created an open market for power in the US. Currently
capacity exceeds demand significantly. As a purchaser of wholesale power from a generation
system one of the key indicators I use in making the decision to purchase power is the efficiency
of the plant or heat rate. At the corporate and plant level senior management being aware of this
instigate programs to improve efficiency and heat rate. Because of a reduction in maintenance
personnel these projects are conducted at the expense of reliability issues. The feed pump
reliability is still poor.

6 months later regulatory changes result in the temporary closure of several nuclear power
plants. The resulting reduction in capacity creates conditions where capacity and demand are
evenly balanced. As a purchaser my new need it to purchase from the most reliable source
possible. In this way I can provide my customers with the power they need when they need it. I
would therefore prefer to buy from the most reliable provider and hence at the plant level
reliability now becomes a major issue. I even apply penalty clauses of $10,000 per MW per day
to the supplier contact. At the plant level it now becomes necessary and appropriate to spend
upwards of several hundred thousand dollars to redesign the piping system for the boiler feed
pump to eliminate a pipe strain induced alignment problem that has reduced the reliability of that
pump.

It is now 12 months after completion of the repairs to the boiler feed pump #1. All condition
based indications are that the reliability of this machine is now excellent. A new change has
occurred in the market place. Country wide generation capacity has increased to the point that
reliability is no longer the critical Issue, having been replaced by environmental issues. The
plant plans, prepares, then cancels repairs to the second feed pump in favor of some boiler
work that improves emissions. Now, consider the perspective of the craftsman:

ƒ I recently have been subjected to a significant changes in maintenance philosophy

(Predictive / Precision Maintenance)
ƒ I am naturally skeptical of the changes in maintenance environment, after all I have been in
maintenance for 20 years and have seen a lot of “programs” come and go

Page 20 of 20
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
ƒ I am also already suspicious of management decisions regarding maintenance as recent re-
organizations have placed personnel from a non-maintenance environment in control of
maintenance activities.
ƒ We have spent thousands of dollars improving insulation in the plant yet one of the most
expensive machines in the place as been “allowed to fail regularly.
ƒ We are finally “allowed” to solve the feed pump problem and I derive a lot of satisfaction
from the improvements made.
ƒ I was prepared and keen to make the same improvements to the sister machine.
ƒ Just before the improvements were to be made to the sister machine the project gets
cancelled.
ƒ The company sends me to “precision” maintenance training.

Is there any wonder frustration sets in ? In order to prevent these issues it is very important to
explain the reasoning behind certain plans processes and strategies not just the strategies
themselves.

12 Step Roadmap

1. Understand the concepts of machinery reliability.

2. Determine the most appropriate maintenance approach for particular machines, systems, or
areas.
3. Establish key performance indicators and a system for tracking progress.
4. Properly plan and schedule maintenance activities.
5. Implement a condition monitoring program as required
6. Establish approved maintenance standards, procedures, and acceptance criteria.
7. Establish an effective maintenance Management information system
8. Establish effective quality assurance processes.
9. Procure necessary tools and equipment.
10. Ensure personnel are properly trained and qualified.
11. Repair machines in accordance with approved “reliability qualification” processes.
12. Verify and evaluate the qualification process and make improvements as required.

Page 21 of 21
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Appendix
Sample Forms and Procedures

Page 22 of 22
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 New/Rebuilt Motor Vibration Test Form (0- 20 HP)

Scope

This form is to used to record the shop test data taken on motors running solo in the shop.

Set up
• Motor to have a ½ key installed.
• Motor to be placed on rubber mat at least ¼” thick.
• Where possible motor to be tested at full service speed.
• Maximum line amplitude is the amplitude value of the highest discrete frequency component
within the stated frequency range in any direction (HVA).
• Scope of this criteria is 2, 4, & 6 pole motors.

Equipment ID:

Acceptance Criteria Vibration Test Data

0-20 HP Motors New/Rebuilt Motors
Max Line
Frequency Range Amplitude MOV MIV MOH MIH MIA
ips pk
Sub-harmonic (< 1x RPM) 0.02

1 x RPM 0.03

Lower Multiples
0.02
(1x RPM – 10x RPM)
Higher Frequencies
0.01
(10x – 50x RPM)
Overall Acceleration g pk
0.5
(0- 20kHz)

Measurement locations
MIV MOV MIH MOH

MIA
MIA

Page 23 of 23
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 New/Rebuilt Motor Vibration Test Form (20 - 400 HP)

Scope

This form is to used to record the shop test data taken on motors running solo in the shop.

Set up

• Motor to have a ½ key installed.

• Motor to be placed on rubber mat at least ¼” thick.
• Where possible motor to be tested at full service speed.
• Maximum line amplitude is the amplitude value of the highest discrete frequency component
within the stated frequency range in any direction (HVA).
• Scope of this criteria is 2, 4, & 6 pole motors.

Equipment ID:

Acceptance Criteria Vibration Test Data

0-20 HP Motors New/Rebuilt Motors
Max Line
Frequency Range Amplitude MOV MIV MOH MIH MIA
ips pk
Sub-harmonic (< 1x RPM) 0.02

1 x RPM 0.05

Lower Multiples
0.03
(1x RPM – 10x RPM)
Higher Frequencies
0.01
(10x – 50x RPM)
Overall Acceleration g pk
0.5
(0- 20kHz)

Measurement locations
MIV MOV MIH MOH

MIA
MIA

Page 24 of 24
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 New/Rebuilt Motor Vibration Test Form (400 HP +)

Scope

This form is to used to record the shop test data taken on motors running solo in the shop.

Set up

• Motor to have a ½ key installed.

• Motor to be placed on rubber mat at least ¼” thick.
• Where possible motor to be tested at full service speed.
• Maximum line amplitude is the amplitude value of the highest discrete frequency component
within the stated frequency range in any direction (HVA).
• Scope of this criteria is 2, 4, & 6 pole motors.

Equipment ID:

Acceptance Criteria Vibration Test Data

400 HP + Motors New/Rebuilt Motors
Max Line
Frequency Range Amplitude MOV MIV MOH MIH MIA
ips pk
Sub-harmonic (< 1x RPM) 0.02

1 x RPM 0.08

Lower Multiples
0.03
(1x RPM – 10x RPM)
Higher Frequencies
0.01
(10x – 50x RPM)
Overall Acceleration g pk
0.5
(0- 20kHz)
Measurement locations

MIV MOV MIH MOH

MIA
MIA

Page 25 of 25
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 Sample Balance Record Form
Rotor Balance Data Sheet (Shop Balancing)
1.0 ROTOR INFORMATION
(1.1) Equipment ID (1.2) Date
(1.3) Serial # (1.4) Manufacturer
(1.5) Rotor Type (ie fan, blower,cplg, (1.6) Service RPM
shaft, impeller etc)
(1.7) Rotor Weight (lbs)
(1.8) Service Configuration (circle one) (1.9) Balancing Configuration (circle one)
(see note i.) Centerhung Overhung (see note i.) Centerhung Overhung
2.0 BALANCE TOLERANCE CALCULATION
(2.1) Desired Tolerance (circle one) (2.2) Stack Balance Required ? (circle one)
4W/N W/N Yes No
Plane 1 Plane 2
(2.3) Journal Load [W] (see note ii.) (2.4) Journal Load [W] (see note ii.)
(50% of rotor weight from (1.7) if (50% of rotor weight from 1.7 if
symmetrical) Lbs symmetrical) Lbs
(2.5) RPM [N] (from item (1.6)) (2.51) RPM [N] (from item (1.6))
(2.6) Max Permissible Unbalance [U] (2.7) Max Permissible Unbalance [U]
(Calculate based on 4W/N or 1W/N as (Calculate based on 4W/N or 1W/N as
reqd. ) UMAX = [Link] reqd.) UMAX = [Link]
(2.8) Convert to [Link] if required (2.9) Convert to [Link]. if required
U x 28.35 = [Link] [Link] U x 28.35 = [Link] [Link]
3.0 BALANCE DATA
(3.1) Balancing ½ key Used ? (3.2) Balancing Machine Type (circle one)
RPM yes no Hard Bearing Soft Bearing
Plane 1 Plane 2
(3.3) Correction radius inches (3.4) Correction Radius inches
(3.51) 1ST Correction g deg (3.61) 1ST Correction g deg
ND ND
(3.52) 2 Correction g deg (3.62) 2 Correction g deg
(3.53) 3RD Correction g deg (3.63) 3RD Correction g deg
(3.54) 4TH Correction g deg (3.64) 4TH Correction g deg
(3.7) Actual Balance Achieved (3.8) Actual Balance Achieved
(next correction wt x radius) (next correction wt x radius)
g. in [Link] g. in [Link]
(3.9) Desired Unbalance (3.91) Desired Unbalance
from item (2.6) From item (2.7)
g. in [Link] g. in [Link]
Notes / Comments

Name Signature

Page 26 of 26
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
The purpose of this form is to convey important information about the balance quality of rotors
that are balanced in a balancing machine.

This form to accompany every rotor to be balanced

The recommended standard for general machinery is that of the American Petroleum Institute
(API) which states that:

UMAX = 4 W/N

Where:

UMAX = the maximum allowable unbalance per plane in oz-in.

W = static journal load in lbs. (for symmetrical rotors, ½ rotor weight)
N = the maximum continuous rotating speed in RPM

In some special cases a standard of 1 W/N may be requested

Notes

i. Configuration Items (1.8) ; (1.9)

Centerhung Overhung
configuration configuration

ii. Journal Load Items (2.3) ; (2.4)

For symmetrical rotors the journal load is 50% of rotor weight
For Non-symmetrical rotors the individual journal load must be calculated.

Rotor
Wt Rotor
100lb Wt
100lb
50lb 50lb
80% 20%

Page 27 of 27
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 Sample Centrifugal Pump Repair Form

Page 28 of 28
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
Equipment Work Order Date
Number Number
Pump description Running Hrs
Machine History Clearance # Isolation Points
Checked ? Valves Breakers

Pump Pump Size

Manufacturer
Reason that Work Order was Written for Pump

Inspection Section
Pump Foundation & Piping
Item Condition Comments
Grout/Foundation Good Bad Not Checked
Anchor Bolts Good Bad Not Checked
Base Condition Good Bad Not Checked
Pusher Bolts Yes No Not Checked
Suction Piping Strained Yes No Not Checked
Discharge Piping Strained Yes No Not Checked

Pump Coupling
Coupling Spacer Good Bad Not Checked
Coupling Flanges Good Bad Not Checked
Coupling Hubs Good Bad Not Checked
Key Length Good Bad Not Checked
Lubricant Condition Good Bad Not Checked

Pump Bowl
Impeller Rub Yes No Not Checked
Corroded/Eroded Yes No Not Checked
Gasket Leak Yes No Not Checked
Plugged in Suction/Discharge Yes No Not Checked
Wear Ring Condition Good Bad Not Checked

Pump Impeller
Impeller Rub Yes No Not Checked
Corroded/Eroded Yes No Not Checked
Impeller Gasket Condition Good Bad Not Checked
Cavitation Wear Yes No Not Checked
Impeller Tips Bent/Worn Yes No Not Checked

Page 29 of 29
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 Rear Cover
Backplate Corroded/Eroded Yes No Not Checked
Stuffing Box Corroded Yes No Not Checked
Gland Gasket Leaking Yes No Not Checked

Mechanical Seal
Seal Leaking Yes No Not Checked
O-rings Good Bad Not Checked
Rotating Face Good Bad Not Checked
Stationary Face Good Bad Not Checked
Seal Flush Operational Yes No Not Checked

Shaft
Sleeve Yes No Not Checked
Fretted Yes No Not Checked
Grooved Yes No Not Checked
Corroded/Pitted Yes No Not Checked
Threads Worn Yes No Not Checked
Bent/Broken Yes No Not Checked

Cartridge
Front Bearings Good Bad Bearing Manuf. & #
Rear Bearings Good Bad Bearing Manuf. & #
Oil Seals/Isolators Good Bad
Oil Level Good Low
Oil Contaminated Yes No

Page 30 of 30
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 Measurement Section

Dial Indicator Readings/ Run out Checks

Description Total Indicator As Found Rebuilt
Reading Reading Reading
(NOT TO EXCEED)
Shaft End Play Check and measure the thrust bearing .002”
and its fit in the housing.
Radial Check and measure the condition of .002”
Deflection the bearings and their fit in the housing
& shaft.
Shaft Run Out Check and measure that the shaft is .002”
straight and round.
Stuffing Box Check and measure that the rear cover .002”
Face is square with the shaft
Perpendicularly
Concentricity Check and measure the rear cover is .005”
concentric to the shaft.
Bearing Fits
Bearing Shaft/Housing Fit
Inside Diameter of Outboard Bearing vs Outside Diameter of Shaft
Inside Diameter of Inboard Bearing vs Outside Diameter of Shaft
Outside Diameter of Outboard Bearing vs Inside Diameter of Housing
Outside Diameter of Inboard Bearing vs Inside Diameter of Housing
Internal Clearance (tapered adapter sleeve bearings)
Note: The bearing ‘FIT’ is the difference between the ‘Bearing’ reading and the ‘Shaft/Housing’ reading.
Impeller Setting
For Open Vane Impellers For Reverse Vane Impellers
Distance Set Off Front Distance Set Off Rear
Casing Cover
Impeller Diameter Wear ring Clearance
Mechanical Seal
Seal Number Manufacturer
Seal Distance from Rear
Cover
Final Alignment Readings
Vertical Readings Horizontal Readings
Front foot Reading Front foot Reading
Back foot Reading Back foot Reading
Vertical Offset Vertical Angularity Horiz Offset Horiz Anglularity
Hold Down Bolt Torque
Values

Page 31 of 31
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]
 Operational and Quality Checks

Operation Checkout
Dead Head Pressure Reading
Normal Operation Pressure
Reading
Vibration Readings Overall ips pk Drive End Non Drive End

Recommendations/Comments

Page 32 of 32
© Universal Technologies, Inc. • 25797 Conifer Road • Suite #C210 • Aspen Park, CO 80433 USA
Phone: (303) 838-3447 • Fax: (303) 838-3437 • Online: [Link]