Chapter 1

The Failure
Analysis and
Troubleshooting System
Troubleshooting as an Extension of Failure Analysis

For years, the term "failure analysis" has had a specific meaning in connection
with fracture mechanics and corrosion failure analysis activities carried out by static
process equipment inspection groups. Figure 1-1 shows a basic outline of materials
failure analysis steps. 1 The methods applied in our context of process machinery
failure analysis are basically the same; however, they are not limited to metallurgic
investigations. Here, failure analysis is the determination of failure modes of
machinery components and their most probable causes. Figure 1-2 illustrates the
general significance of machinery component failure mode analysis as it relates to
quality, reliability, and safety efforts in the product development of a major turbine
manufacturer. 2
Very often, machinery failures reveal a reaction chain of cause and effect. The end
of the chain is usually a performance deficiency commonly referred to as the
symptom, trouble, or simply "the problem." Troubleshooting works backward to
define the elements of the reaction chain and then proceeds to link the most probable
failure cause based on failure (appearance) analysis with a root cause of an existing or
potential problem. For all practical purposes, failure analysis and troubleshooting
activities will quite often mesh with one another without any clear-cut transition.
However, as we will see later, there are numerous cases where troubleshooting
alone will have to suffice to get to the root cause of the problem. These are the cases
that present themselves as performance deficiencies with no apparent failure modes.
Intermittent malfunctions and faults are typical examples and will tax even the most
experienced troubleshooter. In these cases, troubleshooting will be successful only if
the investigator knows the system he is dealing with. Unless he is thoroughly familiar
with component interaction, operating or failure modes, and functional characteris-
tics, his efforts may be unsuccessful.
2 Machinery Failure Analysis & Troubleshooting

~-/~~~
~. ,-~, ~47

Figure 1-1. Failure analysis steps--materials technology (modified from Ref. 1).

There are certain objectives of machinery failure analysis and troubleshooting:

1. Prevention of future failure events.

2. Assurance of safety, reliability, and maintainability of machinery as it passes
through its life cycles of:
a. Process design and specification.
b. Original equipment design, manufacture, and testing.
c. Shipping and storage.
d. Installation and commissioning.
e. Operation and maintenance.
f. Replacement.

From this it becomes very obvious that failure analysis and troubleshooting are
highly co-operative processes. Because many different parties will be involved and
their objectives will sometimes differ, a systematic and uniform description and
understanding of process machinery failure events is important.

Causes of Machinery Failures

In its simplest form, failure can be defined as any change in a machinery part or
component which causes it to be unable to perform its intended function
 The Failure Analysis & Troubleshooting System 3

CUSTOMER
CONSULTANT

J
Maintenance,
Operating INSTALLATION
SALES Plans SERVICE
CONTRACTORS

I Specs Customer
Requirements QUALITY Products
RELIABILITY
Performance I SAFETY
i Corrective
Standards Action
%

ENGINEERING ANUFACTURING
lI ] j
Specs Manufacture
Test
Failure Mode
Analysis

Figure 1-2. Failure analysis and the "wheel of quality. ''2

satisfactorily. Familiar stages preceding final failure are "incipient failure,"

"incipient damage," "distress," "deterioration," and "damage," all of which
eventually make the part or component unreliable or unsafe for continued use.
Meaningful classifications of failure causes are:

1. Faulty design.
2. Material defects.
3. Processing and manufacturing deficiencies.
4. Assembly or installation defects.
5. Off-design or unintended service conditions.
6. Maintenance deficiencies (neglect, procedures).
7. Improper operation.

All statistics and references dealing with machinery failures, their sources and causes,
generally use these classifications. And, as will be shown in Chapter 4, remembering
4 Machinery Failure Analysis & Troubleshooting

these seven classifications may be extremely helpful in failure analysis and troubleshoot-
ing of equipment.
For practical failure analysis, an expansion of this list seems necessary. Table 1-1
shows a representative collection of process machinery failure causes. The table
makes it clear that failure causes should be allocated to areas of responsibilities. If
this allocation is not made, the previously listed objectives of most failure analyses
will probably not be met.
Failure causes are usually determined by relating them to one or more specific
failure modes. This becomes the central idea of any failure analysis activity. Failure
mode (FM) in our context is the appearance, manner, or form in which a machinery
component or unit failure manifests itself. Table 1-2 lists the basic failure modes
encountered in 99 percent of all petrochemical process plant machinery failures.
In the following sections, this list will be expanded so that it can be used for
day-to-day failure analysis. Failure mode should not be confused with failure cause,
as the former is the effect and the latter is the cause of a failure event. Failure mode
can also be the result of a long chain of causes and effects, ultimately leading to a
functional failure, i.e a symptom, trouble, or operational complaint pertaining to a
piece of machinery equipment as an entity.
Other terms frequently used in the preceding context are "kind of defect,"
"defect," or "failure mechanism." The term "failure mechanism" is often described
as the metallurgical, chemical, and tribological process leading to a particular failure
mode. For instance, failure mechanisms have been developed to describe the chain of
cause and effect for fretting wear (FM) in roller bearing assemblies, cavitation (FM)
in pump impellers, and initial pitting (FM) on the surface of a gear tooth, to name a
few. The basic agents of machinery component and part failure mechanisms are al-
ways force, time, temperature, and a reactive environment. Each of these can be sub-
divided as indicated in Table 1-3.
For our purpose, failure mechanisms thus defined will have to stay part of the
failure mode definition: They will tell how and why a failure mode might have
occurred in chemical or metallurgical terms, but in so doing, the root cause of the
failure will remain undefined.

Root Causes of Machinery Failure

The preceding pages have shown us that there will always be a number of causes
and effects in any given failure event. We need to arrive at a practical pointmif not all
the way to the beginning----of the cause and effect chain where removal or
modification of contributing factors will solve the problem.
A good example would be scuffing (FM) as one of the major failure modes of
gears. It is a severe form of adhesive wear (FM) with its own well-defined failure
mechanism. Adhesive wear cannot occur if a sufficiently thick oil film separates the
gear tooth surfaces. This last sentence--even though there is a long chain of cause
and effect hidden in the adhesive wear failure mechanismmwill give us the clue as to
the root cause. What then is the root cause? We know that scuffing usually occurs
quite suddenly, in contrast to the time-dependent failure mode of pitting. Thus, we
cannot look for the root cause in the design of the lube oil system or in the lube oil
 The Failure Analysis & Troubleshooting System 5

Table 1-1
Causes of Failures
Design and Specification Fabrication [] Defective material
Responsibility [] Welding error [] Inadequate bolting
Application [] Improper heat treatment [] Connected wrong
[] Undercapacity [] Improper hardness [] Foreign material left in
[] Overcapacity [] Wrong surface finish [] General poor workmanship
[] Incorrect physical conditions [] Imbalance []
(temperature pressure, etc.) [] Lub. passages not open
[] Incorrect physical prop. [] Operations and Maintenance
(mol. wt., etc.) Assembly Responsibility
[] [] Improper fit Shock
[] [] Improper tolerances [] Thermal
Specifications [] Parts omitted [] Mechanical
[] Inadequate lubrication system [] Parts in wrong [] Improper startup
[] Insufficient control instrumentation [] Parts/bolts not tight []
[] Improper coupling [] Poor alignmerrt
[] Improper bearing [] Imbalance Operating
[] Improper seal [] Inadequate bearing contact [] Slugs of liquid
[] Insufficient shutdown devices [] Inadequate testing [] Process surging
[] [] [] Control error
[] Controls deactivated
[] Shipping and Storage Responsibility or not put in service
Material of Construction Preparation for Shipment [] Operating error
[] Corrosion and/or erosion [] Oil system not clean []
[] Rapid wear [] Inadequate drainage Auxiliaries
[] Fatigue [] Protective coating not applied [] Utility failure
[] Strength exceeded [] Wrong coating used [] Insufficient instrumentation
[] Galling [] Equipment not cleaned [] Electronic control failure
[] Wrong hardening method [] [] Pneumatic control failure
[]
[] Protection []
[] Insufficient protection Lubrication
Design [] Corrosion by salt [] Dirt in oil
[] Unsatisfactory piping support [] Corrosion by rain or humidity [] Insufficient oil
[] Improper piping flexibility [] Poor packaging [] Wrong lubricant
[] Undersized piping [] Dessicant omitted [] Water in oil
[] Inadequate foundation [] Contamination with dirt, etc. [] Oil pump failure
[] Unsatisfactory soil data [] [] Low oil pressure
[] Liquid ingestion
[] Inadequate liquid drain Physical Damage [] Plugged lines
[] Design error [] Loading damage [] Improper filtration
[] [] Transport damage [] Contaminated oil
[] [] Insufficient support []
[] Unloading damage Craftsmanship
Vendor Responsibility [] [] Improper tolerances
Material of Construction Installation Responsibility [] Welding error
[] Flaw or defect [] Improper surface finish
[] Improper material Foundations
[] Improper treatment [] Settling [] Improper fit
[] [] Improper or insufficient [] General poor workmanship
grouting []
Design [] Cracking or separating Assembly
[] Improper specification [] [] Mechanical damage
[] Wrong selection Piping [] Parts in wrong
[] Design error [] Parts omitted
[] Inadequate or wrong lubrication [] Misalignment [] Misalignment
[] Inadequate liquid drain [] Inadequate cleaning [] Improper bolting
[] Critical speed [] Inadequate support [] Imbalance
[] Inadequate strength [] [] Piping stress
[] Inadequate controls and protective de- Assembly [] Foreign material left in
vices [] Misalignment [] Wrong material of construction
[] [] Assembly damage (crafts) []

(table continued on next page)

6 Machinery Failure Analysis & Troubleshooting

Table 1-1 (cont.)

Preventive Maintenance [] Seal [] Blades

[] Postponed [] Coupling [] Blade root
[] Schedule too long [] Shaft [] Blade shroud
[] [] Pinion gear [] Labyrinth
[] Bull gear [] Thrust bearing
[] Turning gear [] Pivoted pad bearing
[] Casing [] Roller/ball bearing
[] Rotor [] Cross-head piston
Distress, Damages, or Failed Components [] Impeller [] Cylinder
[] Vibration [] Shroud [] Crankshaft
[] Short circuit [] Piston []
[] Open circuit [] Diaphragm []
[] Sleeve bearing [] Wheel []

Comments:

itself--that is, if scuffing was not observed before on that particular gear set. Sudden
and intermittent loss of lubrication could be the cause. Is it the root cause? No, we
still have to find it because we are looking for the element that, if removed or
modified, will prevent recurrence or continuation of scuffing. Is it because this
particular plant is periodically testing their standby lube oil pumps, causing sudden
and momentary loss of lube oil pressure? Eventually, we will arrive at a point where
a change in design, operation, or maintenance practices will stop the gear tooth
scuffing.
Removal of the root cause of machinery failures should take place in design and
operations-maintenance. Quite often the latter, in its traditional form, is given
too much emphasis in failure analysis and failure prevention. In our opinion,
long-term reductions in failure trends will only be accomplished by specification and
design modifications. We will see again in Chapter 7 that only design changes will
achieve the required results. How then does this work? After ascertaining the failure
mode, we determine whether or not the failed machinery component could be made
more resistant to the failure event. This is done by checking design parameters such
as the ones shown in Table 1-4 for possible modification. Once a positive answer has
been obtained, the root cause has also been determined and we can specify whatever
is required to impart less vulnerability to the material, component, assembly, or
system. As we formulate our action plan, we will test whether the mechanic's axiom
holds true:

When in doubt
Make it stout
Out of something
You know about.
 The Failure Analysis & Troubleshooting System 7

Table 1-2
Machinery Failure Mode Classification
Deformation--i.e. plastic, elastic, etc.
Fracture--i.e. cracks, fatigue fracture, pitting, etc.
Surface changes--i.e, hairline cracks, cavitation, wear, etc.
Material changesmi.e, contamination, corrosion, wear, etc.
Displacement--i.e. loosening, seizure, excessive clearance, etc.
Leakage
Contamination

Table 1-3
Agents of Machinery Component and Part Failure Mechanisms
Steady Low
Force Transient Room
~ Cyclic Elevated
Temperature
Steady
Transient
Cyclic
Very Short
Time Sho rt Reactiv e r - - Chemical
Long Environment L..._Nuclear

Table 1-4
Process Machinery Design Properties
Material-of-Construction Level
1. Material properties, i.e. ductility, creep resistance, heat resistance, etc.
2. Properties derived from processing, i.e. cast, rolled, forged, etc.
3. Properties resulting from heat treatment, i.e. not heat treated, hardened, stress relieved,
etc.
4. Surface properties, i.e. machined, ground, lapped, etc.
5. Properties derived from corrosion and wear protection measures, i.e. overlayed,
enameled, painted, etc.
6. Properties resulting from connecting method, i.e. welded, shrunk, rolled-in, etc.

Part and Component Level

7. Properties derived from shape and form, i.e. cylindrical, spherical, perforated, etc.

Part, Component, and Assembly Level

8. Suitability for service, i.e. prone to plugging, wear, vibration, etc.
9. Properties resulting from assembly type, i.e. riveted, pinned, bolted, etc.
10. Assembly quality, i.e. countersunk, flush, tight, locked, etc.
$ Machinery Failure Analysis & Troubleshooting

Table 1-5
Machinery Failure Modes--Process Plant
Fracture
Deformation Separation. Surface/Material Changes Displacement

Wear
~ ,, Adhesive ~

~5,~~ ~ ~
Design Parameter
Failure Resistance
~ ~o l"~ ~:
~ ~
"= ~
~ ~ _~
~ ~ ~'
-~
<
!
" I m m e d i a t e " Cause >- m ~ m U m

Tensile Strength 9 9 9 9
Yield Strength 9
Compressive
Yield Strength 9
Shear Yield
Strength 9 9
Fatigue
Properties 9
Ductility %RA 9 9 9 9
Charpy Energy 9 9
Modulus of
Elasticity 9 9
Creep Rate 9 9
[~r 9 9
,,,
Erosion Resistance
EMF Series
Hardness 9 9 9 9 9 9 9 9 9 9 9

Coefficient of
Thermal Expansion
Melting Point 9
Thermal Conductivity
Properties
from Manufacturing i
Process
Properties from
Corrosion Pro-
tection
Properties from
Heat Treatment
Properties from
Wear Protection
Properties from
Shape and Form

IIIIIIIIIIIII
Properties from
Type and Quality
of Assembly
Static Force aaaallllnlaal
Dynamic Force
IDDDlUlIIDDDDDI 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9

Coefficient of
Sliding Friction
_.
Time
Temperature
Chemical Environment
 The Failure Analysis & Troubleshooting System 9

We will keep in mind our inability to influence machinery failures by simply

making the part stronger in every conceivable situation. A flexibly designed
component may, in some cases, survive certain severe operating conditions better
than the rigid part.
Table 1-5 concludes this section by summarizing machinery failure modes as they
relate to their immediate causes or design parameter deficiencies.

References

1. VDI Guidelines No. 3822., Der Maschinenschaden, Vol. 54, No. 4, 1981, p. 131.
2. Ludwig, G. A., "Tests Performed by the Builder on New Products to Prevent
Failure", Loss Prevention of Rotating Machinery, The American Society of
Mechanical Engineers, New York, N.Y. 10017, 1972, p. 3.