T echnical

&
R esearch Bulletin 3-51
Practices and
Procedures for the
Alignment of Marine
Main Propulsion
Shafting Systems

The Society of Naval Architects and Marine Engineers

601 Pavonia Avenue, Jersey City, New Jersey 07306

Technical and Research Bulletin 3-51

PRACTICES AND PROCEDURES
FOR THE
ALIGNMENT OF MARINE
MAIN PROPULSION SHAFTING SYSTEMS

Prepared by
PANEL M-16
PROPULSION SHAFT SYSTEMS
of the
SHIPS' MACHINERY COMMITTEE

Published by
THE SOCIETY OF NAVAL ARCHITECTS AND MARINE ENGINEERS
601 Pavonia Avenue, Jersey City, New Jersey 07306

T&R Bulletin 3-51 has been prepared by Panel M-16

for

THE SOCIETY OF NAVAL ARCHITECTS AND MARINE ENGINEERS

TECHNICAL AND RESEARCH PROGRAM

Reviewed and Approved by:

PANEL M-16 (Propulsion Shafting)

Kevin D. Prince, Chairman
Jerry V. Havel
Richard S. Kaminsky
Daniel T. Norton
Derek C. Paschal
Dave Rickman
Steven C. Shepstone
Richard T. Steinhilber
Andrew Szypula

David A. Carlson
John E. Ancarrow, Jr.
Gordon L. Blatt
Kevin F. Danahy
Stephen M. Donley
Thomas W. Frenzinger II
Sujit K. Ghosh
Bendt H. Hansen Sr.

Approved by the:

SHIPS' MACHINERY COMMITTEE

David R. Rodger, Chairman
Thomas P. Mackey
William L. McCarthy
Charles A. Narwicz
Mark F. Nittel
Charles H. Piersall
Kevin D. Prince
Alan L. Rowen
E. Gregory Sanford
Peter George Schaedel
William J. Sembler
Kenneth Siegman
Matthew F. Winkler
Richard T. Woytowich
Ivan Zgalji

Robert S. Behr
Karl E. Briers
Roger K. Butturini
Allen Chin
Joseph H. Comer III
Thomas F. Conroy
James J. Corbett
W. Mark Cummings
Richard D. Delpizzo
Earl W. Fenstermacher
Joseph P. Fischer
Richard W. Harkins
John F. Hennings
Bahadir Inozu

It is understood and agreed that nothing expressed herein is intended or shall be construed to give any person, firm,
or corporation any right, remedy, or claim against SNAME or any of its officers or members.

ii

UNIT CONVERSIONS
The original document contained only English units. Metric units are now provided as well.
Where the values in English units are general the converted values are rounded and approximate,
but where precise numbers were given precision was maintained.

ACKNOWLEDGEMENTS
The panel acknowledges the cooperation and assistance of the following personnel and
organizations that co-authored portions of this document:
D. Sverko
P. Diehl
R. Greene
J. Bordeaux
W. Daube
B. Cowper
L. Douthwaite
R. Kaminsky
J. Reed
D. Norton
K. Danahy

American Bureau of Shipping

Diehl Engineering Co.
Electric Boat Corporation
John J. McMullen Associates, Inc.
John J. McMullen Associates, Inc.
LamaLo Technology
Lloyds Register
Naval Sea Systems Command
Naval Surface Warfare Center
Seaworthy Systems, Inc.
US Coast Guard

The final manuscript was prepared for release by Beda Angelo I. Pormentilla, Intern at SNAME
Headquarters, 2003 to 2007.

DISCLAIMER
It is understood that nothing expressed herein is intended or shall be construed to give any
person, firm, or corporation any right, remedy, or claim against SNAME or any of its officers or
members.

iii

TABLE OF CONTENTS
PAGE
SECTION 1.0

INTRODUCTION

1.1 INTRODUCTION AND DEFINITION OF PROPULSION SHAFTING

SYSTEM ALIGNMENT
1.2 HISTORICAL PERSPECTIVE
1.3 PURPOSE OF THE ALIGNMENT GUIDE

1
1
1
2

SECTION 2.0

DESCRIPTION OF MARINE SHAFTING SYSTEMS

SECTION 3.0

THE IMPORTANCE OF PROPER PROPULSION SHAFTING

SYSTEM ALIGNMENT

SECTION 4.0

CRITERIA FOR THE ACCEPTABLE ALIGNMENT OF A

MARINE PROPULSION SHAFTING SYSTEM

SECTION 5.0

FACTORS THAT AFFECT THE ALIGNMENT OF MARINE

PROPULSION SHAFTING SYSTEMS

10

5.1

5.2

FACTORS THAT AFFECT THE ANALYSIS OF THE SHAFT ALIGNMENT

A. FLEXIBILITY OF THE SHAFTING SYSTEM
B. THERMAL GROWTH
C. WEARDOWN OF WATERBORNE BEARINGS
D. DRYDOCK VERSUS WATERBORNE ALIGNMENT
E. BALLAST, CARGO AND FUEL LOADS
F. VESSEL DESIGN
G. PROPULSOR HYDRODYNAMIC EFFECTS AND THE BORE
SLOPE AT THE AFTERMOST BEARING
H. SHIP SPEED AND POWER
I. DYNAMIC CONDITIONS
J. DEFLECTION OF THE MAIN THRUST BEARING AND
FOUNDATION
K. EFFECT OF PRESSURE AND TEMPERATURE AT DEPTH FOR
SUBMERSIBLE VESSELS
FACTORS THAT AFFECT ALIGNMENT MEASUREMENTS
A. HULL TEMPERATURE
B. STRUCTURAL WORK
C. TRANSIENT SHIP OPERATING CONDITIONS AND SEA
CONDITIONS
D. JOURNAL ALIGNMENT ISSUES

SECTION 6.0

DEVELOPING THE DESIGN ALIGNMENT

6.1 THE DESIGN ALIGNMENT PROCESS

6.2 CREATING A MATHEMATICAL MODEL
iv

10
10
12
13
14
15
15
16
16
16
17
17
17
17
18
18
18
19
19
20

6.3
6.4
6.5
6.6
6.7
6.8

INFLUENCE COEFFICIENTS
ASSUMED INITIAL ALIGNMENT
CALCULATING BEARING AND SHAFT REACTIONS
DETERMINING THE DESIGN ALIGNMENT
CALCULATING THE BUILDERS SETTINGS
INFORMATION THAT SHOULD BE DETERMINED FROM AN
ALIGNMENT ANALYSIS
6.9 VIBRATION CONSIDERATIONS

SECTION 7.0

METHODS FOR MEASURING THE ALIGNMENT OF

MARINE PROPULSION SHAFTING SYSTEMS

7.1 INTRODUCTION
7.2 HYDRAULIC JACK METHOD
7.3 STRAIN GAGE METHOD
7.4 GAP AND SAG METHOD
7.5 OPTICAL AND LASER METHODS
7.6 WIRE METHOD
SECTION 8.0

ATHWARTSHIP ALIGNMENT OF PROPULSION SHAFTING

SYSTEMS

8.1 INTRODUCTION
8.2 MEASUREMENT OF ATHWARTSHIPS ALIGNMENT
8.3 ATHWARTSHIPS ADJUSTMENT OF LINE SHAFT BEARINGS
8.4 ATHWARTSHIPS ADJUSTMENT OF STERN TUBE BEARINGS
8.5 ATHWARTSHIPS ALIGNMENT OF REDUCTION GEARS
SECTION 9.0

ADJUSTING THE SHAFT ALIGNMENT

9.1 NEW CONSTRUCTION VESSELS

9.2 THE ALIGNMENT PLAN
9.3 WHEN TO MEASURE THE SHAFT ALIGNMENT FOR
IN-SERVICE VESSELS

21
23
24
25
26
26
26
28
28
28
31
33
34
35
37
37
37
38
39
39
40
40
41
41

FIGURES
1a.
1b
2.
3.
4.
5.
6.
7.

TYPICAL SHAFTING SYSTEM

TYPICAL SHAFTING SYSTEM FOR A SHIP WITH A LOW-SPEED
DIESEL ENGINE
EXAMPLE OF DEFLECTION, MOMENT, AND SHEAR DIAGRAMS
HYDRAULIC JACK INSTALLATION
EXAMPLE OF BEARING REACTION DIAGRAM PLOT OF LOAD
VERSUS LIFT FOR HYDRAULIC JACK METHOD
STRAIN GAGE METHOD FREE BODY DIAGRAMS
TWO AND FOUR STRAIN GAGE WHEATSTONE BRIDGE
CONFIGURATIONS
GAP AND SAG MEASUREMENTS

44
45
46
47
48
49
50
51

REFERENCES

52

BIBLIOGRAPHY

53

vi

Practices and Procedures for the Alignment of Marine

Main Propulsion Shafting Systems

SECTION 1.0 INTRODUCTION

1.1 INTRODUCTION AND DEFINITION OF PROPULSION SHAFTING SYSTEM
ALIGNMENT
Common sense suggests that the bearings supporting marine propulsion shafting be aligned
so they hold the shaft in a straight line, since offsetting one bearing with respect to another
would bend the shaft. Moreover, industrial machinery is typically aligned so that the shafts
of the driving and driven machines are collinear [1]. However, this approach is not always
appropriate for marine main propulsion shafting because: a) its bearings move as conditions
change, and b) the shafting is never really straight because it deflects under its own weight
and the weight of the propeller and other attached components. Therefore, the vertical and
athwartships position of the bearings must be aligned so that the limits established to ensure
proper system operation are not exceeded for all normal ship operating conditions. It follows
that shaft alignment could also be called bearing alignment, since the alignment is a direct
result of the position of bearings that support the shafts.
1.2 HISTORICAL PERSPECTIVE
In the early 1960s, it was discovered that serious misalignment of propulsion shafting caused
gear skew on large, steam turbine bull gears, which in turn initiated pitting on the second
reduction gear teeth. At that time shafting systems usually had too many bearings, and a
straight-line alignment was considered to be a good alignment. However, the straight-line
alignment tended to place a large amount of the attached line shaft weight on the aft bull gear
bearing in the cold condition. Then, as the gear heated up, even more shafting weight shifted
onto this bearing, while the forward gear bearing became more unloaded. This unequal
loading of the gear bearings caused gear skew. Computerized shafting analysis was used to
misalign the bull gear in the cold condition relative to the propulsion shafting, so that in the
hot condition the vertical loads on the two bull gear bearings would be equal within an
acceptable limit. The SNAME paper Coordinated Alignment of Line Shaft, Propulsion
Gear and Turbines, by Andersen and Zrodowski [2] firmly established the relationship
between shaft misalignment and gear problems and offered practical solutions.
As a result of information accumulated since then, todays shipbuilders produce vessels with
practically no built-in propulsion shafting alignment problems. Nonetheless, alignment
problems still occur and propulsion shaft alignment remains a challenging task for both
shipyards and designers. Some of these alignment problems are caused by: a) worn or failed
bearings, b) extensive hull damage which may have disturbed the bearing settings, and c)
improperly bored stern tubes.

1.3 PURPOSE OF THE ALIGNMENT GUIDE

Ship operators surveyed by SNAME, as well as the US Coast Guard and the US Navy,
expressed a need for a technical reference guide on the alignment of main propulsion shafting
[3]. This document is intended to provide these guidelines for personnel involved in the
design, installation, and operation of marine propulsion shaftlines. This guide provides a
general overview of shaft alignment for the novice, and more detailed information for the
person with some knowledge on one facet of shaft alignment that needs more information on
another alignment topic.

SECTION 2.0 DESCRIPTION OF MARINE SHAFTING SYSTEMS

A shaft is a rotating or stationary member, usually of circular cross section, having mounted
upon it such elements as gears, pulleys, flywheels, cranks, sprockets, and other powertransmission elements [4]. While this definition was taken from a mechanical engineering
design text it is nonetheless an accurate description of marine propulsion shafts. This same
text also provides a good description of what a shaft must endure throughout its intended life
span: Shafts may be subjected to bending, tension, compression, or torsional loads, acting
singly or in combination with one another. When these loads are combined, one may expect
to find both static and fatigue strength to be important design considerations, since a single
shaft may be subjected to static stresses, completely reversed stresses, and repeated stresses,
all acting at the same time [4]. From these excerpts, it is apparent that marine propulsion
shafting has similar design considerations to other power transmission shafting.
A marine main propulsion shafting system must accomplish the following objectives:

transmit the power from the main engines or reduction gear to the propulsor.
transmit the thrust developed by the propulsor to the ships hull.
support the propeller or propulsor rotor.

In the present context, a marine propulsion shaft connects a driving device, such as a
reduction gear, engine, or electric motor to a propeller or other propulsor. For example, a
reduction gear power take-off shaft that drives a generator or pump is not considered a
propulsion shaft, but the shaft connecting the final gear to the propeller is a propulsion shaft.
Marine propulsion shafts can be found in many sizes and configurations, serving installations
from the smallest boat to the largest container ship, over a range from less than 20 hp (15
kW) to more than 87,000 hp (65,000 kW). The shafts are driven by many different types of
prime movers, including diesel engines, gas turbines, and steam turbines. In many instances
a speed reduction is required which may not only dictate different alignment criteria, but also
may result in complex shafting arrangements such as V- and Z-drives.
Figures 1a and 1b illustrate typical shafting systems. Propulsion shafting is comprised of
inboard and waterborne shafting. The inboard shafting is located inside the hull forward the
shaft seal and consists of the line shafting and thrust shaft. The corresponding bearings are
called line shaft or steady bearings. The waterborne shafting is located aft of the forward end
of the shaft seal and consists of the propeller shaft or tail shaft section that supports the
propeller and in some cases the stern-tube shaft, which passes through the stern tube.
The basic shaft system can be designed after the fundamentals of the vessel propulsion
system have been determined, i.e. prime mover, the maximum propulsor speed, and shaft
power. The classification societies have formulas for calculating the minimum required shaft
diameter for the different shaft sections. After the shaft diameters have been estimated, the
bearings can be located along the shaftline. The bearing loads, shafting flexibility, and the
shafting system lateral natural frequencies should be considered when determining the
number and location of the bearings.
Reference 5 includes a formula for estimating the number of line shaft bearings. This
reference also indicates that as a general rule the ratio of bearing center distance to shaft
diameter may range from 12 to 22, although the final ratio must be based on the requirements
3

for vibration, shaft strength, and flexibility requirements to maintain alignment during
service conditions.
A typical shaft system may contain several different types of bearings, each with different
load and lubrication requirements. There are two principle types of bearings used with
marine shaft systems: anti-friction (ball or roller type) bearings and hydrodynamic (journal or
stave type) bearings. Within each type there are numerous variations involving different
bearing materials. Each different configuration will have distinct criteria with regard to the
methods of lubrication, maximum and minimum allowable load, cooling requirements, and
wear characteristics.

SECTION 3.0 THE IMPORTANCE OF PROPER PROPULSION SHAFTING SYSTEM

ALIGNMENT
The shafting system can be considered as a beam supported by multiple, unevenly spaced
bearings. The bearing positions relative to the shaftline, or offset, must be set so that the
loads on the bearings and the shaft stresses are within acceptable limits. Performing regular
alignment checks to maintain proper shafting system alignment is essential for reliable ship
operation.
a. Proper shaft alignment is necessary for correct internal alignment of main
reduction gear slow speed elements and direct-connected diesel engine bearings.
One of the primary goals of shafting system alignment is to ensure that the main
reduction slow speed gear (bull gear) or direct-connected main engine is properly aligned
and that the bearing loads are within the acceptable limits set by the gear or engine
manufacturer. In fact, in geared installations, much of the emphasis on achieving an
acceptable shaft alignment is based on the need to prevent damage to the main reduction
gears. In addition to the usual maximum and minimum bull gear bearing load limits,
reduction gear vendors also prescribe a maximum load differential between the forward
and aft bull bear bearings in the static condition. This bull gear bearing static load
differential relates to bull gear-to-pinion tooth stresses in the dynamic (running)
condition. An excessive differential can produce unacceptable gear tooth stresses, which
can cause destructive gear tooth wear, pitting, and failure. Overstressed gear teeth can
fail due to fatigue after many years of operation. However, the propulsion shaft-toreduction gear alignment is not as critical for a system with a flexible coupling between
the reduction gear or direct-connected engine and shafting system that isolates the gear or
engine from the affects of shafting system alignment. Nonetheless, the bearings and
propulsion train components forward of the flexible coupling should still be included in
the alignment analysis.
b. Proper shaft alignment prevents overloading of the shafting system bearings. All
bearings have a maximum allowable pressure load stipulated by the bearing
manufacturer. Exceeding these limits can lead to premature bearing wear or failure.
c. Proper shaft alignment prevents excessive shaft vibration from a lightly or unloaded
bearing. When a shaft is misaligned, a support bearing in the system can become lightly
loaded. Then the ship can experience pounding of the bearing, as the load on the
bearing varies from a loaded to unloaded condition during each shaft revolution. When
the bearing actually unloads, the bearing span of the system dramatically increases and
this increased span - from the bearings forward and aft of the unloaded bearing - can
cause excessive lateral (whirling) vibration in the shafting system.
d. Proper shaft alignment prevents shaft failure from excessive bending stresses.
Shafting systems are designed to achieve the allowable shaft bending stress levels defined
by the classification societies or other good engineering standards. The shafting
alignment has a direct effect on these shaft stresses and a misaligned shafting system can
have overstresses that lead to eventual shaft fatigue failure. Misaligned shafting systems
can produce very significant bending loads [5].

SECTION 4.0 CRITERIA FOR THE ACCEPTABLE ALIGNMENT OF A MARINE

PROPULSION SHAFTING SYSTEM
The benefits of proper shafting system alignment, as discussed in Section 3, logically lead to
criteria that will be used to achieve the proper alignment on a new construction or
refurbished shafting system. Obviously, the criteria have engineering, economic, and
scheduling ramifications. The shafting system should be designed and aligned to meet the
following criteria for all normal ship operating conditions, including the changes caused by
bearing weardown, thermal growth, ship loading, and the other factors which affect
alignment as discussed in Section 5.
a. The maximum allowable bull gear bearing load differential shall not be exceeded.
The shafting system alignment must be designed so it meets the criteria for the bull gear
bearing differential load. This differential applies to the athwartship, as well as the
vertical loads. The reduction gear manufacturer will usually provide the allowable
differential. However, if this value cannot be obtained, then a general rule is that the
static load differential should be less than 25% of the combined total load on the two bull
gear bearings in all operating conditions to ensure that the gear tooth meshing contact is
not misaligned by a skewed attitude of the gearwheel in its bearings. Sometimes it may
be cost effective to perform a skew analysis to evaluate the acceptability of the measured
gear loads if the gear manufacturers load differential appears overly restrictive.
To meet this load-differential criterion, the shafting system designer must provide for a
sufficiently flexible system at the shaft/gear interface using an appropriate bearing
spacing. A flexibility factor (or allowable setting error) is commonly used to determine
whether the shafting system has reasonable flexibility. The flexibility factor is
determined from the values for maximum allowable bull gear differential and the
influence coefficients of the aft and forward bull gear bearings on themselves. A
minimum absolute value of 0.010 inch (0.25 mm) for the flexibility factor is considered
acceptable. Sections 5.1.a and 6.3 contain more information on flexibility factors and
bearing influence coefficients, respectively.
Some gear manufacturers use the gear/line shaft gap and sag to specify the alignment
limits (See Section 7.4 for information on the gap and sag method). This is an older
method of aligning the gears to the propulsion shafting, although it is still used during
new construction for making an approximate alignment until the shaft sections are
connected. At least one gear manufacturer specifies that the gear be aligned with zero
gap and sag to the free-hanging shaft in order to prevent any moment or shear from being
imposed on the gear shaft. However, when there is a long span from the coupling to the
first line shaft bearing, this alignment can overload the line shaft bearing and/or produce
high bending stresses in the line shaft. Some gear manufacturers and almost all engine
manufacturers specify limits on the moment and shear applied to their couplings (which
can be measured by the strain gage method discussed in Section 7.3)
b. No support bearing in the system shall be loaded above its maximum allowable
pressure. Generally, the support bearings are sized to accommodate the shaft vertical
design loads for all anticipated alignments and ship conditions. The maximum allowable
bearing loads are usually defined in terms of pressure and are specified by the bearing
vendors or based on historical results. The maximum allowable load for a particular
bearing is usually determined by using the projected area of the bearing (the product of
6

the effective bearing length and the shaft outer diameter), although not all classification
societies compute the pressure load on the bearing in this manner.
(1) Line shaft bearings will run well with pressure limits of 75 psi (c. 500 kPa) for disc
lubricated bearings, and 50 psi (c. 350 kPa) for ring lubricated bearings. It is very
important that the bearings have a uniform load distribution in both the vertical and
horizontal planes since very high-localized pressures can occur on a bearing that is
tipped down or skewed athwartships relative to the shaft. The load distribution on
the line shaft bearings can be checked by applying a strip of bluing axially along the
journal and rotating the shaft through the bearing. All the bluing should be wiped
off; if not, the area with the bluing is not in contact with the bearing. The end of the
bearing should be shimmed until the bluing is evenly removed when the shaft is
rotated. The bearings athwartships load distribution can also be estimated by
removing the bearing cap, and measuring the side skew at the four corners with a
feeler gage. A hot-running line shaft bearing usually indicates an overloaded
bearing if the lubrication system is operating satisfactorily [6].
(2) In general, an anti-friction bearing large enough to fit the shaft diameter is capable of
supporting a much higher load than can be applied from a reasonable alignment.
The B10* life for this type of bearing typically exceeds 100,000 hours. Therefore,
almost any positive, downward load will be acceptable. Although anti-friction
bearings are not nearly as sensitive to angular alignment as sleeve bearings, some
care needs to be taken during installation so that even self-aligning bearings start out
with the proper angular alignment.
(3) Waterborne-shaft bearings may be either oil- or water-lubricated. Oil-lubricated
bearings usually have a length-to-diameter (L/D) ratio of 2. Bearings using synthetic
material are usually longer, with an L/D between 2 and 4. The American Bureau of
Shipping allows oil-lubricated, white-metal lined outboard bearings to be shorter
(but not less than L/D = 1.5) as long as the bearing nominal pressure is not more than
116 psi (800 kPa). Likewise, oil-lubricated synthetic bearings may be less than 2D
in length (limit is L/D = 1.5) if the bearing nominal pressure is not more than 87 psi
(600 kPa).
The following pressures are the maximum allowable limits for each bearing type:
Oil-lubricated synthetic liner:
87 psi (600 kPa)
Oil-lubricated babbitt liner:
116 psi (800 kPa)
Water-lubricated synthetic liner:
70 psi (c. 500 kPa)
Water-lubricated rubber or wood liner:
40 psi (c. 300 kPa)
(4) The propeller bearing is located just forward of the propeller and its primary
function is to support the propeller and shaft weight, and counteract the propeller
induced hydrodynamic forces. The slope (or cant) of the propeller bearing is critical
and the bearing must be aligned with the tail shaft so that the load is distributed over
most of the available bearing area. Otherwise, the overhung weight of the propeller
will deflect the tail shaft and the total propeller weight will be concentrated on a
small area at the aft end of the bearing. A change in the bearing offsets rarely causes
a significant change in the load on the propeller bearing and many more propeller
bearings have failed from a poor load distribution than from an excessive overall
*

B10 is the life in hours that a group of apparently identical bearings will exceed or complete.

load. This is particularly true for oil-lubricated metal-lined bearings, which have less
tolerance for slope misalignment because of their stiffness.
One way to align the propulsor bearing considering its slope in relation to the
theoretical shaft slope through the bearing is to run the shafting alignment computer
program with the propulsor bearing located at its fore-and-aft center. The propulsor
bearing is then set to a vertical slope corresponding to 60%-100% of the calculated
slope. A rule-of-thumb for the slope limit on oil-lubricated bearings is 0.003
radians, and a rule-of-thumb for water-lubricated strut or stern-tube bearings is that
the bearing may be out of parallel to the shaft by no more than the design minimum
bearing clearance. A particularly poor alignment is one where the shaft bears
downward at one end of the propulsor bearing, and upward at the other end of the
bearing, passing through the bearing clearances.
Propulsor bearings with a misaligned slope also need a higher shaft rotational speed
to develop an adequate hydrodynamic lubrication film, and therefore are more apt to
generate stick-slip noise at slow shaft speeds. The shaft speed required to prevent
stick-slip noise increases as the slope misalignment worsens.
(5) In general, the load on a forward stern tube bearing varies considerably with the
vertical or athwartship offset. It is just as important to achieve a good load
distribution across a forward stern tube bearing as it is for a propulsor bearing. This
bearing should also be sloped if the output from the alignment software shows a
difference in slope of more than 0.003 radians between the shaft and a non-canted
forward stern tube bearing. Sometimes sloping the forward stern tube bearing can be
avoided by lowering the propulsor bearing, which achieves the same net result.
c. No support bearing in the system shall be loaded less than its minimum allowable
load. A minimum allowable load is required for each bearing to stabilize the shaft in the
bearing and prevent shaft vibratory problems. Determining the minimum load is a
somewhat arbitrary decision, although zero or negative (upward) loads on the bearings
are not acceptable. The minimum load can be chosen so that the bearing will not become
unloaded because of the normal, expected shaft eccentricities (runout). One guideline is
that the minimum load shall be at least 50% of the bearings initial design load.
d. The maximum allowable stresses in the shafting shall not be exceeded. Obtaining
acceptable shaft bending stresses is a high-priority alignment criterion since the shaft can
fail from excessive bending stresses. Since the alignment directly affects the bending
moments and bending stresses in the shaftline, the alignment affects the overall shaft
stress and shaft diameter calculations. However, the classification societies, such as the
American Bureau of Shipping (ABS), generally calculate the minimum shaft diameters
based on a torsional stress limit without regard to the bending moments in the shaft (or
other loads, such as thrust, which produces relatively low stress). Therefore, the
classification society equations use other means to compensate for the loads not included
in their formulas. For example, the ABS accounts for the higher bending moments
experienced by tail shafts and stern tube shafts by: a) using an appreciably larger shaft
design factor than for line shafting, and b) limiting the material strength used in the tail
shaft calculations.
When the bending stresses are high, it is important to evaluate the fatigue risk. For
instance, the US Navys shaft stress formulas [7] require a more detailed analysis than
8

those used by the classification societies, and include the bending moments determined
from the alignment analysis and the resulting fatigue stresses. The US Navy design rules
specify the minimum factors of safety at various points along the shaftline, and limit the
maximum bending stress in steel waterborne shafting to 6000 psi (41,400 kPa). The
factor of safety calculation accounts for stresses caused by: a) steady state torque, b)
steady state thrust, c) alternating (cyclic) bending moments determined from the
computerized shaft alignment, and d) alternating torques. The minimum allowable factor
of safety depends on whether the vessel is a surface ship, icebreaker, or submarine, and
whether the shaft section is waterborne or line shafting.
e. The design alignment criteria for directly connected engines, special couplings, or
other equipment, shall not be exceeded. The manufacturers of large main engines
directly connected to propulsion shafting usually specify shaft alignment limits to prevent
excessive crankshaft deflections. A limit on the bending moment and shear at the engine
coupling is a relatively common alignment requirement. The strain gage method is an
ideal method (although not the only way) to measure the coupling moment and shear.
Existing shafting installations that put excessive moment or shear loads on the engine
coupling can usually be corrected by a vertical or athwartships adjustment of the line
shaft bearings. The engine manufacturer may also specify limits on the coupling gap and
sag (See Section 7.4 for information on this method). This is a valid requirement if strain
gages are not used, although it is necessary to uncouple the shaft and measure the gap and
sag. If strain gages are used, it is worthwhile to ask the engine manufacturer to provide
the equivalent moment and shear limits. Otherwise, the alignment software will produce
the equivalent moment and shear corresponding to the specified gap and sag. If the
shafting and bull gear contain special couplings, such as flexible or spline couplings, then
the coupling alignment limits should be obtained from the coupling vendor.

SECTION 5.0 FACTORS THAT AFFECT THE ALIGNMENT OF MARINE

PROPULSION SHAFTING SYSTEMS
The actual alignment of a shafting system changes as its bearings move relative to one
another because of changes in the condition of the ship and its environment. The goal is to
obtain a satisfactory alignment so that the bearing reactions, shaft stresses, and clearances
between the shaft and adjacent components and structures remain within acceptable limits for
all normal operating conditions. The alignment analysis and the alignment measurements
must quantify and account for the various factors that affect the alignment. Any effect that
might move one or more bearings relative to the others must be considered. After the
individual effects are quantified, they can be combined using superposition. The practical
aspects of shipyard construction must also be considered when specifying the alignment. If
the alignment criteria are too stringent, then excessive costs will be incurred during
construction. If some criteria are too impractical, then they should be re-evaluated.
5.1 FACTORS THAT AFFECT THE ANALYSIS OF THE SHAFT ALIGNMENT
a. Flexibility of the shafting system. The alignment of shaft systems in land-based
installations relies on the foundations of the equipment to be fixed relative to each other.
However, ships are inherently flexible and so there is an almost continuous movement of
the shafting systems supports. The required flexibility of the shafting system depends on
the flexibility of the ship along the shaft length, and as a general rule, the shaft system
should be more flexible than the ship. The flexibility of a marine shaft system has a
distinct effect on its performance and a shaft whose flexibility is well suited to its
installation will have little problem, regardless of the vessels load or other conditions.
Some ships are much more sensitive to shaft alignment than others, and the flexibility of
the shafting system or conversely its stiffness is one of the most important factors. A
short, rigid shaft system is very sensitive to the small changes in alignment that occur
normally, such as thermal growth or vessel loading. A stiff shaft is generally less
forgiving of excursions outside its design envelope, and may be damaged by a permanent
or transient foundation change caused by sea conditions, grounding, or loading. For
example, a large diameter shafting system of short overall length tends to be more
sensitive to changes since it is less flexible. In contrast, a soft shaft with high lateral
flexibility relative to the hull structure is less sensitive to thermal growth, ship loads, and
sea conditions. Accordingly, it is preferable to install as few bearings as possible in order
to produce a lower bending stiffness. A shaft system that is designed to be flexible and
insensitive to small changes will generally have fewer problems caused by shaft
misalignment.
A large number of vessels with direct-coupled propulsion systems (propeller-shaftingdiesel engine) have very rigid shafting, and are very sensitive to changes in the bearing
offsets that may result from hull deflection, bearing wear, or thermal growth.
Fortunately, this shafting is generally installed in a very rigid bottom structure, so there is
less hull deflection, which somewhat reduces this effect. Nonetheless it is strongly
recommended that the shaft alignment analysis for certain categories of vessels (in
particular those with rigid direct-coupled propulsion systems) consider the influence of
the surrounding structure on the bearing offset.

10

The issue of shaft stiffness will become increasingly important as ship powers continue to
increase. Until the 1980s the general limit for power transmitted on one shaft was 60,000
hp (45,000 kW). Formally naval vessels had the highest-powered shafts, which usually
operated with relatively high shaft speeds. Since the shaft torque decreases with
increasing shaft speed (rpm) and the shaft diameter is determined in part by torque (and
not power), naval shafts do not have inordinately large diameters. Naval shafts are also
relatively long, incorporating several line shaft, stern-tube, and strut bearings. Therefore,
naval shafts are fairly soft since they have a relatively small diameter and long length.
The engine powers in recent commercial ships have increased to over 90,000 hp (67,000
kW) at less than 95 rpm. Furthermore, modern long-stroke diesel engines can fully
exploit the efficiency of slower turning propellers. Consequently, large diameter, stiff
shafts are required to transmit the higher torque from heavy propellers. Also, the trend to
place the main engines further aft on commercial vessels reduces the shaft lengths, and
the shafting becomes stiffer. In some cases, the minimum shaft diameter is larger than
normally required by the design rules to prevent torsional vibrations and fatigue failure.
The propulsion shafting is therefore stiffer, and large shear force and bending moment
changes that can jeopardize the machinery alignment are induced by variations in the hull
deflections. This does not imply that only high-powered, low-speed shafts are stiff,
because the stiffness of a shafting system depends on the shaft diameters and lengths, as
well as on the bearing location and support stiffness.
In addition to the cases already cited, the shaft alignment needs to be evaluated more
carefully for the following alignment- sensitive installations:

propulsion shafting with a reduction gear where the bull gear is driven by two or
more ahead pinions
propulsion shafting with a power take-off or with booster power arrangements
propulsion shafting for which the tail shaft bearings are to be slope bored.

As discussed earlier, the flexibility factor can be used to evaluate the flexibility of the
shafting in way of the reduction gear, and therefore, the sensitivity of the gear to
improper alignment [3]. The flexibility factor represents the total of the permissible error
in estimating the thermal rise of the slow speed gear bearings relative to the line shaft
bearings and the permissible error in setting the gear to the line shafting without
exceeding the maximum allowable difference in the static slow speed gear bearing
reactions. An absolute minimum value of 0.010 inch (0.25 mm) for the flexibility factor
is considered acceptable [5]. See reference 8 for more information on the flexibility
factor. The flexibility factor (FF) is mathematically defined as:
FF = R/|I11 - I22|, where

(5-1)

R = maximum allowable difference between the two reduction gear bearing

static reactions, lbf (N)
I11 = influence coefficient of forward reduction gear bearing on itself, lbf/mil
(N/m)
I22 = influence coefficient of aft reduction gear bearing on itself, lbf/mil (N/m).
The influence coefficient matrix can be used as a rough guide for determining whether a
shaft is stiff or soft, but there are no established criteria. However, gear bearing influence
11

coefficients over 1000 lbf/mil (175 N/m) or line shaft bearing influence coefficients over
500 lbf/mil (88 N/m) may indicate a rigid system.
b. Thermal growth. Thermal expansion of equipment causes relative movement of the
bearings, which changes the shafting system alignment. The thermal growth occurs
primarily at the bull gear, thrust bearings, and in direct-connected diesel engines, where
hot lubricating oil sumps located below the machinery lift it when going from the cold to
the hot operating condition. Thermal changes in the line shaft and tail shaft bearings can
be ignored since they are generally small compared to the engine or gear thermal growth.
Propulsion shafting is generally aligned when cold (and it may be very cold in northern
shipyards). The cold alignment must be set so that the desired bearing loads are achieved
when the ship is in the hot, running condition. Since the thermal growth of the bull gear
changes the bearing loads, the thermal growth must be calculated and integrated into the
alignment analysis. This leads to the establishment of design cold and hot alignments.
The bull gear has two bearings relatively close together, so the influence coefficients are
high, but not equal, and sensitive to errors in the assumed thermal growth. The gear
manufacturer may estimate the thermal growth of the gear bearings. However, in some
practitioners experience, the gear manufacturers load tolerance is often unnecessarily
tight, and rarely do the two bull gear bearings undergo the same thermal growth. This is
particularly true on large, double reduction, steam turbine gears. The first reduction gear
often runs 10F (6C) or so hotter than the second reduction gear because of the higher
pitch-line velocities, which will cause the bull gear bearing located underneath the first
reduction gear to expand more that the bearing on the other side of the bull gear. The
gear bearing thermal growth can be calculated using the equation below, with different
operating temperatures for the two bearings as appropriate.
The following general rules are useful:
(1) The gear manufacturers estimate of the thermal growth can usually be accepted.
When this is not available, the thermal growth, Hv, for steel can be calculated as:
Hv = k Dv T, where
(5-2)
-6
-6
k = coefficient for thermal expansion of steel, 6.610 /F (11.710 /C)
Dv = vertical distance from shaft centerline to the bearing foundation, inch
(mm)
T = difference between the hot operating and cold temperatures, F (C)
When the hot operating temperature is unknown, then usually a value of 120F
(50C) may be assumed [9].
(2) To satisfy the reduction gear load differential criteria it is necessary to make the best
thermal growth corrections possible, but then check the hot alignment. This can be
done on a sea trial using the strain gage method described in Section 7.3. Ideally,
telemetry could be used to measure the shaft strains underway, but it is possible to
get a fairly good representation of the hot loads without telemetry by shutting down
the propulsion system to measure the strains after running several hours so that the
gear casing and foundations are at their operating temperatures.

12

(3) Although the thermal growth of the line shaft can be ignored, there may be an
exception to this where the line shaft bearing foundation is located near equipment
that has some measurable thermal impact.
It is generally recommended that the criteria in Section 4 for the maximum bearing load,
minimum bearing load, and bull gear bearing load differential be met for both the hot and
cold aligned conditions, although there is some debate over the need to meet these criteria
for the cold condition. Some practitioners believe that the cold condition can be
disregarded since it is not an actual operating condition. Others believe that since the
ship actually initiates its operations in the cold condition, that this condition be
considered a short-term operating condition. A reasonable compromise is to design for
acceptable alignment in the cold condition as much as is practical.
c. Weardown of waterborne bearings. Outboard bearings are often water-lubricated and
have staves or liners that experience significant weardown over long periods of time,
which can have a significant influence on the alignment. Extreme bearing weardown can
cause the tail shaft and line shaft to fail. Inboard bearings are often oil-lubricated, metallined bearings that wear very little over their lifetimes. The weardown of inboard or
outboard oil-lubricated bearings can usually be ignored, although occasionally an oillubricated bearing will wipe and cause a small change in alignment.
The design shafting alignment must account for the change in loads from bearing
weardown; an acceptable alignment must be maintained throughout the allowable
weardown range. Generally, weardown tends to decrease the waterborne bearing loads
while increasing the inboard bearing loads. A bearing weardown analysis is usually
performed as part of the computerized shaft alignment analysis. The weardown analysis
examines the effects on bearing loads at wear intervals typically 0.005 to 0.010 inches (c.
0.13 to 0.25 mm) throughout the allowable weardown range. It is usually assumed that
the water-lubricated bearings wear at a rate proportional to their relative pressure loading,
although many other factors can accelerate the weardown. For example, if the pressure
load on the main propulsor bearing is twice that of the aft stern tube bearing, then the
propulsor bearing will wear at twice the rate of the stern tube bearing.
Since bearings can wear unevenly in a shaftline with multiple bearings, several
weardown scenarios are possible. The most conservative approach is that each bearing
may reach its wear limit while the others are nearly new. This approach is probably not
realistic, except if one bearing is replaced while the others remain worn. Another
approach, as discussed above, is to perform a collective weardown analysis, which
assumes that all the bearing staves or liners are new or replaced before the weardown
begins. Starting with this assumption, the unit load on each bearing subject to weardown
is calculated. One of the bearings is then assumed to wear a fixed incremental amount.
The incremental weardown on the other bearings is then calculated assuming that they
wear in proportion to their unit load. The unit load on each bearing is then recalculated
from the incremental wear on each bearing. This iterative calculation is repeated until
one or more bearings reaches its weardown limit or the bearing load goes outside of its
acceptable range.
It is relatively easy to predict the alignment changes that will occur over the weardown
range when there is only one wear-prone waterborne bearing. The load and bending
stresses will increase at the next bearing forward as bearing wear progresses. This next
bearing will usually be a line shaft bearing, which can be lowered to keep its load or
13

bending stress within limits. However, if there is only one line shaft bearing (or none)
between the tail shaft and the gear or engine, then the resulting misalignment can have
serious consequences on that equipment. In that case, periodic alignment checks should
be made to prevent adverse effects on the bearings, shafting, gear, or engine.
Alignment changes in a shaft system with more than one wear-prone waterborne bearing
are much more problematical because it is more difficult to predict the wear progression
in each wear-prone bearing. Initially, the aft bearing is prone to wear faster than the
forward bearing because of heavier unit pressures, possibly poor load distribution, and
the ingestion of sand or silt in the aft bearing. As the aft bearing wears, the load increases
at the forward bearing, which may or may not cause it to wear. There will be a
significant change in the tail shaft to line shaft alignment that is not easy to plan for
without experience in this area. Therefore, as in the single wear-prone bearing case,
periodic alignment checks are needed to monitor the weardown.
d. Drydock versus waterborne alignment. The drydock alignment condition can be an
important consideration during the alignment design process since it may be necessary to
align the shafting in drydock. The shaft alignment varies between the drydock and
waterborne conditions, mostly because of buoyancy effects. The alignment will also
change for different drydock/keel blocking arrangements. The shafting should be aligned
in drydock so it achieves the proper alignment when the ship is afloat. In this way, the
shipbuilder avoids the more cumbersome alignment adjustments in the waterborne
condition.
The shafting alignment software calculates the change in bearing loads caused by the
buoyancy of the waterborne shafting and propulsor. However, the vessels hull
deflection along the length of the shaft is also important because it determines the relative
location of each bearing. Advances in computer technology and structural analysis give
the designer a much better understanding of how changes in the hull deflection affect the
shaft alignment. The hull deflection along the shaftline should be evaluated for all
normal waterborne ballasted and loaded conditions and for all building ways and
drydocked conditions where alignment work might be performed. The hull deflection of
a ship in drydock or under construction is affected by the number, position, and type of
blocks in the docking or construction support plan.
Since the hull deflection calculation is subject to relatively large tolerances compared to
the bearing offsets, it is preferable to obtain the necessary hull deflection data from
similar ship designs. If these data are not available, then a naval architects analysis of
the hull deflections can be used to estimate the position of the bearing foundations. ABS
requires that alignment calculations consider the effect of the deformation of the aft end
of the hull, where known. A common case concerns ships with waterborne shafting,
where hull deflection in way of the strut may be of the order of several inches
(centimeters). However the strut bearing has a relatively low influence coefficient and the
shafting system is not extremely sensitive to errors in the hull deflection calculation.
When work that will affect the shafting alignment will be performed in drydock, the
alignment should be checked while the ship is waterborne before going into drydock and
then again immediately after drydocking. Then the correlation between the waterborne
and drydock bearing loads can be used to predict the waterborne bearing loads when the
ship comes out of drydock. It is important to perform a final waterborne alignment check
and adjustment after undocking and under the normal draft and load conditions. If this
14

cannot be done, then it is necessary to make the alignment adjustments in drydock based
on the correlation between the drydock and waterborne loads. When coming out of
drydock, the ships hull should be given some time to stabilize (up to two days, if
practical) before performing the final alignment check.
e. Ballast, cargo, and fuel loads. Although the ships ballast, cargo, and fuel loads affect
the hull shape they are often neglected when considering factors that affect the shaft
alignment. Their effect on shaft alignment varies with the type and size of the vessel and
the shafting arrangement. Cargo loading can have a significant impact on the shaft
alignment, especially for ships that have a wide range of loading options that affect the
hull deflections and the relative bearing positions. Unfortunately, many ships so affected
tend to have relatively stiff, short, and stubby shafting systems that are sensitive to
misalignment. There have been cases where ship classes whose designs did not account
for hull loading had to be assigned restricted loading patterns to maintain acceptable
propulsion performance.
The analysis of ship loading effects on shaft alignment can be difficult, which may
explain why this factor is often ignored. As a minimum, the alignment analysis for cargo
ships should account for the no-load and full-load conditions. However, for some
ships, the intermediate cargo loading condition may create a shaft alignment that falls
outside of the no-load and full-load conditions, and so the intermediate ship loading
should also be analyzed. If there is a large variation in results, then the shafting should
be aligned to the mean draft condition to minimize the bearing load changes on both sides
of the mean condition, which will hopefully produce an acceptable alignment.
Since the calculation of these ship-loading effects may be inaccurate, it is preferable to
obtain the necessary data from similar ship designs. If the engineer is fortunate, there
will be measurements of the ship load versus shaft bearing loads, or hull deflection data
that can be related to the relative bearing foundation positions. If data from similar ship
designs is not available, then a naval architects analysis of the hull deflections for the
various load conditions can be used to estimate the positions of the bearing foundations.
Here are other guidelines:
(1) Short, stiff vessels can probably be safely aligned in the light-draft condition, which
is apt to be the condition in a shipyard after undocking.
(2) The alignment of ships with aft engine rooms short shafting systems will vary more
with aft draft than with cargo load. Unless experience says that the load variation is
not significant, it is wise to measure the bearing loads at both extremes of normal
operating draft aft.
(3) Shaft bearing loads on ships with amidships engine rooms can be expected to vary
significantly with cargo load.
The storage of fuel affects the hull deflections and the shaft alignment in a manner
similar to cargo loading and ballast. The shaft alignment measurements should be made
under similar fuel load conditions when possible. The US Navy rule-of-thumb is to
measure the alignment with about 80% of the fuel and/or cargo load.
f. Vessel design. Different types of vessels and propulsion trains have different influences
and concerns. For example, vessels with water-lubricated propulsor and stern tube
15

bearings experience weardown, while oil-lubricated bearings have practically no

weardown and can accept higher load limits. Consequently a shaft to be installed in a
water-lubricated bearing should be designed to be more flexible than one intended for an
oil-lubricated bearing, since the former shaft will have a larger range of deflection as the
bearing wears down.
Vessels that incorporate a gear in the drive train have a combination of design criteria
that must be satisfied. Gear manufacturers place limits on the allowable bearing loads
and allowable bull gear load differential. After a shaft is hard coupled to the final gear,
the gear shaft becomes part of the shaft system and its bearing loads are influenced by the
entire shaft system. The shaft must be relatively soft so that the allowable loads and load
differential are not exceeded when the gear bearing offsets change because of by thermal
growth, oil film thickness, and torque reaction.
Direct-coupled diesel engines have their own design criteria. The engine structure is
relatively stiff and the shaft section coupled to it must be relatively soft. Movement of
the shaft will influence the engine bearing loads and the engine movement will influence
the shaft bearing loads. Unlike gear elements, the engine crankshafts usually have many
bearings. While slight misalignment of a gear system may cause tooth contact problems,
slight misalignment of a direct-coupled diesel system may cause lightly loaded or
unloaded bearings at idle or stopped conditions. These same unloaded or lightly loaded
bearings, however, may be forced into their normal configuration once the engine load
increases causing excessive crank web deflections and stresses. This phenomenon is
becoming more prominent as diesel engine shaft systems become stiffer.
g. Propulsor hydrodynamic effects and the bore slope at the aftermost bearing. A
significant consideration is whether to slope the aftermost bearing supporting the
overhung propulsor. The hydrodynamic force on an operating propulsor is off-center
relative to its centerline. These hydrodynamic forces are imposed on the shaft, which
may significantly alter its relative position within the aft bearings and create horizontal
reactions. This eccentricity creates a moment that varies in magnitude and direction,
depending primarily on the stern configuration and vessel draft. Since this moment, in
deep water when underway, acts opposite to the weight moment, the slope of the shaftline
will decrease relative to the static conditions, and may even reverse. It is important to
ensure that the shaft journal remains in contact with the aft end of the bearing, because if
the effective support point moves forward when underway, then the longer overhang of
the propulsor can cause dynamic pounding. Thus, if the bore is to be sloped, the slope
should usually not be the full angle predicted by the alignment analysis, and a bias in
favor of the aft end support is recommended.
h. Ship speed and power. The range of shaft speeds and powers affect the running position
of the shaft within the bearings. By knowing the range of operating positions within each
bearing, the designer can place each bearing in the proper location. The shaft speeds and
powers also determine the range of thrust loads and the off-thrust loads, which can have a
significant effect on the running position of the aftermost bearing journal. In some
applications the off-center thrust can affect the load on the propulsor bearing. The
propulsor designer should be able to provide information on the off-center thrust. The
shaft speed and power affect the shaft response to various forms of vibration.
i. Dynamic conditions. The shafting system alignment changes during ship operations
because of ship flexing and reduction-gear torque roll. In most cases, the shaft alignment
16

is measured in the static condition and dynamic effects are ignored because it is much
more difficult to analyze and measure them. However, the US Navy has evaluated the
dynamic alignment when the reduction gear is supported on resilient mounts for structure
borne noise attenuation. Analytic studies and dynamic alignment measurements
performed on ships with resiliently mounted reduction gears indicate that dynamic torque
roll can be significant and should be included in the alignment analysis. The reduction
gear manufacturer should provide an analysis of dynamic torque roll, which can be used
to offset the reduction gear in the cold static condition to compensate for this dynamic
motion. This compensation will most likely be primarily in the horizontal direction.
j. Deflection of the main thrust bearing and foundation. In addition to experiencing
axial deflections, the thrust bearings may rise or be compressed as the applied thrust load
increases, depending upon the their foundation design. This effect should be small or
negligible for a properly designed foundation.
k. Effect of pressure and temperature at depth for submersible vessels. The combined
effect of increasing pressure on the hull and decreasing water temperature at depth is a
significant factor affecting the shaft alignment of submersible vessels. Both the high
water pressure and the cold water cause the pressure hull of a submersible vessel to
contract. As the depth increases, the bearings mounted on top of foundations fixed to the
lower portion of the pressure hull will rise relative to bearings supported either uniformly
from above and below or by unaffected non-pressure hull structure. Therefore, the
deflection of the affected inboard bearing foundations caused by pressure and
temperature at the maximum design depth must be calculated.
5.2 FACTORS THAT AFFECT ALIGNMENT MEASUREMENTS
The following factors should be considered when making alignment measurements, although
they are not normally included in the shaft alignment process.
a. Hull temperature. The temperature of the hull can affect its shape. The water
temperature for a waterborne ship or the air temperature for a drydocked ship will
influence the hull temperature. The significance of the hull temperature on shaft
alignment depends in part on the type of ship; for some designs, it may be negligible.
The effect of uneven heating of a ships hull structure caused by solar radiation, whether
the ship is in drydock or waterborne, is more important. Tests and experience have
demonstrated that thermal growth caused by solar heating can deflect the hull enough to
significantly affect the shaft alignment. Daylight heating tends to expand the topside of
the hull relative to the bottom. These hull deflections affect the relative positions of the
shaft bearings and the shafting alignment. Therefore, the shaft alignment should be
measured several hours after sunset and before sunrise, or with the ship inside a covered
facility.
The shafting system should be designed so that the solar radiation and hull temperature
do not affect the shaft alignment when the vessel is waterborne. If necessary, the
alignment should be checked under the two extreme temperature conditions: a) under a
hot sun, and b) in the late evening or early morning. Changes to the design alignment
may be required if the alignment is unsatisfactory in some climates.

17

b. Structural work. Structural work, such as major welding and equipment change-outs
near the shaft bearing foundations, can affect the shaft alignment. Thus, all structural
work near bearing foundations should be completed before the final alignment
measurements.
c. Transient ship operating conditions and sea conditions. Although transient conditions
are not always included in the shafting system alignment and stress analyses, the
designers should consider how they affect the shaft alignment. Transient conditions are
usually extreme situations that exist for short intervals but may still accumulate a
significant number of load cycles. Transient ship operating conditions include loads from
hydrodynamic control surfaces and high sea states. In general, rough seas do not affect
the alignment because they cause random transient events. The magnitude of such
transient effects can be quite large, depending upon the particular ships operating profile.
d. Journal alignment issues. In the past, the shaft alignment was normally only concerned
with maintaining the correct bearing loads within the range of static journal positions.
While the weardown of water-lubricated stave bearings was included in this range, the
hydrodynamic or off-center thrust influence on the journal position was seldom included.
This latter influence causes the journals to shift within the confines of the bearing much
more than previously suspected, and in some cases the journal positions are considerably
different than accounted for in a static alignment. This issue is particularly important for
ice-class ships where there are ice impact forces on the propeller.
Traditionally, the terms shaft alignment and bearing alignment have been almost
interchangeable because shaft alignment is almost always considered from the viewpoint
of positioning the bearings. However, shaft alignment should not be understood solely as
bearing alignment, rather shaft alignment should be viewed as journal alignment. Only in
the static condition does the bearing position determine the shaft position. When both
static and dynamic influences are included, the shafting must be aligned using the
journals and not just the bearings.

18

SECTION 6.0 DEVELOPING THE DESIGN ALIGNMENT

6.1 THE DESIGN ALIGNMENT PROCESS
Because the position of the bearings is not constant, the design alignment must meet the
design criteria discussed in Section 4 for all anticipated ship operating conditions. The
design alignment must consider the condition of the ship during the initial alignment, and
how the alignment will change over time because of various ship operating conditions and
long-term effects such as bearing wear. Consider the following two examples, which address
the effects of hull deflections and bearing weardown.

Under a certain ship ballast condition, the hull deflection causes an outboard intermediate
bearing to rise relative to the adjacent bearings. As the bearing rises, the load on it
increases, and the bending moment in the shaft at the bearing also increases. At the same
time, the loads on adjacent bearings decrease. Initially lowering the intermediate
outboard bearing by one-half of the predicted motion can compensate for this effect,
although the impact on the propulsor bearing must be considered before this is attempted.
The weardown of a water-lubricated propeller bearing of 0.250 inches (6.5 mm) is
allowed before its staves or liners must be replaced. In this case, initially raising the
bearing by one-half of the allowable weardown will minimize the allowable variation
from the design alignment.

Propulsion shaft alignment has advanced from considering the individual sections or
components as separate pieces of equipment, which are simply aligned to each other on a
straight-line, to considering the shafting system as a continuous beam. The continuous beam
analysis allows designers to better understand the interaction of the complete shafting
system, how the movement of one bearing support affects the others, and how minimum
stresses are achieved. The purpose of the design alignment analyses is to determine the
acceptable bearing reactions, shaft-bending moments, shaft deflections, and slopes for all
known alignment conditions. The following general outline summarizes a process for
developing an acceptable design alignment. Simpler approaches may be used for simpler
systems. These items noted below are discussed in more detail in the subsequent paragraphs.

Create a mathematical model of the shafting system and perform a static analysis.
Identify all the factors affecting the shaft alignment and determine the effect each one has
on the shaft offsets at the bearings.
Identify the ship operating conditions for which the design criteria must be met.
Assume an initial alignment of the bearings for a known baseline condition. For
example, assume that all bearings are in a straight-line when the ship is waterborne at
dockside.
Calculate and tabulate the shaft offsets at the bearings for each ship operating condition.
Calculate and tabulate the bearing reactions and shaft bending moments, deflections,
slopes, and bending stresses for each ship operating condition for the assumed initial
alignment. Compare the maximum and minimum bearing reactions, and the maximum
shaft bending stresses for the assumed initial alignment with the design criteria.
If the design criteria are not met, calculate a revised alignment for one or more bearings
to meet the criteria. Even if the criteria are satisfied, it may be possible to calculate an
optimal alignment that will minimize the life-cycle costs by reducing the bearing loads,
shaft stresses, or number of bearings.
After the design alignment is determined, calculate the builders settings such that the
shipbuilder can achieve a proper alignment of the bearings.
19

6.2 CREATING A MATHEMATICAL MODEL

The mathematical model should provide a reasonable representation of the shafting systems
geometry, stiffness, weight, boundary conditions, and applied loads. The results of the static
analysis will only be as good as the mathematical model. The following documents are
needed to create the mathematical model for the shafting alignment calculation that will be
analyzed with either an alignment computer program based on continuous beam theory or a
general finite element analysis program:
(1)
(2)
(3)
(4)
(5)
(6)
(7)

Shafting arrangement drawing

Shafting detail drawing
Propeller drawing, including the propeller centerline, weight, and center of gravity
Reduction gear drawing (if applicable), with bull gear weight and dimensions
Diesel engine manufacturers project guide (if applicable)
Bearing design data for the line shaft and waterborne bearings
Shaft seal detail drawing

The mathematical model includes the shafting outside and inside diameters, the shaft sleeves,
the concentrated weights (such as the propeller and bull gear), the bearings, and the modulus
of elasticity and density of the shaft material. The shaft bore plugs, raised lands in way of the
bearings and sleeves, and non-metallic shaft coverings are usually ignored. Tapered bores
are modeled by assuming an abrupt change in the bore diameter at the midpoint of the taper.
The bearings are typically modeled as rigid knife-edge supports, although in some cases a
bearing may be modeled as a distributed elastic support for a more accurate computation of
the shaft bending moment at the bearing. The bearing support point is placed at the bearing
centerline, except for the bearing just forward of the propulsor. The relatively long length of
this bearing makes it more difficult to accurately define the net reaction point. The tail shaft
has a significant slope at this bearing because of the propulsor weight and so the reaction
point tends to be in the aft part of the bearing. For waterborne bearings with L/D ratios of
about 4, the bearing reaction point is assumed to be one shaft diameter forward of the aft
bearing face [5]. The US Navy specifies that the propulsor bearing support point be located
forward from the aft end of the bearing the larger of one-fourth of the bearing length or one
shaft outer diameter [7]. For oil-lubricated bearings, with an L/D ranging from 1 to 2, the
bearing reaction point is assumed to be one-half shaft diameter forward of the aft bearing
face [5].
The weights of propellers, collars, and other components attached to the shafting that have
little effect on the overall system flexibility are modeled as concentrated loads. The effect of
buoyancy on the propulsor and waterborne shafting is included in the analysis. The model
should include a station at each change in shaft geometry, bearing support points, flange
faces, concentrated weights, hydraulic jack locations (for subsequent bearing load
measurements), shaft seals, and other locations as appropriate. Additional stations may be
required between long lengths of shafting to determine the bending moments along the shaft.
Static analyses of the mathematical model are required to determine the bearing reactions and
shaft bending moments, stresses, deflections, and slopes caused by the systems own weight,
applied loads, and deflections. Analyses for both the buoyant and airborne (non-buoyant)
conditions are usually performed. The US Navy requires the analysis of the straight-line,
non-buoyant (in air) condition [7].
20

6.3 INFLUENCE COEFFICIENTS

Since the bearings relative position effects the bearing reactions and shaft stresses, it is
useful to determine the influence that a unit deflection of the shaft at each bearing has on the
reaction at each bearing and on the bending moments, deflections, and slopes throughout the
shaftline. These calculated influence coefficients (or influence numbers) are solely a
function of the shafting systems geometry and stiffness, and the effect of gravity is not
relevant to this analysis. After the influence coefficients have been determined, the bearing
and shaft reactions can be calculated for any ship operating condition or proposed alignment.
A similar approach can be used to determine the influence of a unit load applied to the shaft.
For example, since the system is linear, the influence of a 1000 lbf (4.5 kN) load applied at
the propeller can be used to determine the effect of a heavier propeller or a hydrodynamic
load applied to the propeller during a turn.
The bearing reaction influence coefficient matrix is an N x N symmetric matrix, where N is
the number of bearings and each element in the matrix is an influence coefficient of the form
Iij. Mathematically, the influence coefficient Iij is defined as Iij = Ri/j, where Ri is the
change in the bearing reaction on the ith bearing for a unit displacement j of the jth bearing.
Tables 1a and 1b are examples of bearing reaction influence coefficient matrices. For this
example, raising line shaft bearing no. 1 by 1 mil (0.025 mm) increases the load on itself by
30.89 lbf (137 N), decreases the load on the aft bull gear bearing by 112.35 lbf (500 N),
increases the load on the forward bull gear bearing by 92.58 lbf (412 N), and so forth. Since
the influence coefficient matrix is symmetric, the affect of a given movement of a bearing A
upon a bearing B is the same as the affect of the same movement of bearing B on bearing A.
For example, in Tables 1a and 1b, raising line shaft bearing no. 1 by 1 mil (0.025 mm)
decreases the load on line shaft bearing no. 2 by 18.50 lbf (82 N), and raising line shaft
bearing no. 2 by 1 mil (0.025 mm) decreases the load on line shaft bearing no. 2 by 18.50 lbf
(82 N).

21

Table 1a. Example of an Influence Coefficient Matrix (English Units)

Bearing Reaction Influence Coefficients,
lbf per mil (lbf/mil) of bearing displacement
Bearing

Forward
bull gear

Aft bull
gear

Line shaft Line shaft Line shaft Stern tube

no. 1
no. 2
no. 3
bearing

Propeller
bearing

Forward
bull gear

535.34

-612.53

92.58

-21.98

8.01

-1.64

0.22

Aft bull
gear

-612.53

704.03

-112.35

29.78

-10.84

2.22

-0.30

Line shaft
no. 1

92.58

-112.35

30.89

-18.50

8.97

-1.84

0.25

Line shaft
no. 2

-21.98

29.78

-18.50

27.54

-23.88

8.15

-1.10

Line shaft
no. 3

8.01

-10.84

8.97

-23.88

33.04

-20.78

5.50

Stern tube
bearing

-1.64

2.22

-1.84

8.15

-20.78

21.59

-7.71

Propeller
bearing

0.22

-0.30

0.25

-1.10

5.50

-7.71

3.15

Table 1b. Example of an Influence Coefficient Matrix (Metric Units)

Bearing Reaction Influence Coefficients,
Newtons per millimeter (N/mm) of bearing displacement (multiply by 110-3)
Bearing

Forward
bull gear

Aft bull
gear

Line shaft Line shaft Line shaft Stern tube

no. 1
no. 2
no. 3
bearing

Forward
bull gear

93.75

-107.27

16.21

-3.85

1.40

-0.29

0.039

Aft bull
gear

-107.27

123.29

-19.68

5.22

-1.89

0.39

-0.053

Line shaft
no. 1

16.21

-19.68

5.41

-3.24

1.57

-0.32

0.044

Line shaft
no. 2

-3.85

5.22

-3.24

4.82

-4.18

1.43

-0.19

Line shaft
no. 3

1.40

-1.89

1.57

-4.18

5.79

-3.64

0.96

Stern tube
bearing

-0.29

0.39

-0.32

1.43

-3.64

3.78

-1.35

Propeller
bearing

0.039

-0.053

0.044

-0.19

0.96

-1.35

0.55

22

Propeller
bearing

Every bearing in a shafting system is affected when the offset of one bearing is altered, although
the effect becomes smaller as the distance from the altered bearing increases. Observe that Iii is
always positive since lifting a bearing always increases the load on it. Also, note that the sum of
the elements of each row or column of the influence coefficient matrix equals zero. In general,
the signs of the adjacent terms in any row or column should alternate, e.g., from positive to
negative to positive, although there are certain shaft arrangements for which this may not occur.
Influence coefficient matrices can also be determined for the bending moments. In this case, the
coefficients represent the change in moment (e.g., N-m) for a 1 mil (0.025 mm) change in the
bearing position. The bending moment influence coefficient matrices are not symmetrical.
As a minimum, static alignment analyses should be performed to determine:

The influence coefficient matrix for a unit deflection of the shaft at each bearing
The bearing reactions and shaft bending moments, deflections, and slopes for the
following conditions:
a. straight-line, dry-docked
b. cold aligned, dry-docked
c. straight-line, waterborne
d. cold aligned, waterborne
e. hot aligned, waterborne
f. cold aligned with collective weardown, waterborne
g. hot aligned with collective weardown, waterborne

Once the static analyses are complete and the results have been tabulated, the following checks
are recommended to confirm the validity of the results. If the static analyses results fail any of
these tests, then the mathematical model input file should be checked for errors.

Perform an independent calculation of the total weight and longitudinal center of gravity
of the shafting system, both in air and waterborne, using the geometry, material
properties, and applied loads specified in the mathematical model. The calculated system
weight should equal the sum of the bearing reactions determined by the static alignment
analysis. The sum of the moments of the bearing reactions about any point should equal
zero.
Confirm that the bearing reaction influence coefficients are in an N x N symmetric
matrix. The sum the elements of each row or column of the influence coefficient matrix
should equal zero. For most systems, the signs of the adjacent terms in any row or
column should alternate.
Plot the bending moment and deflection diagrams. The resultant curves should be
continuous and make sense for the system configuration.

6.4 ASSUMED INITIAL ALIGNMENT

The bearings supporting the shafting move relative to one another as the condition of the ship
and its environment changes. Section 5 discusses the common factors that affect shaft
alignment. All factors that may significantly affect the shaft alignment should be identified,
although the list will differ for different ship designs. The effect of each factor on the shaft
offsets at the bearings should be quantified and tabulated.
The calculation of the alignment settings is somewhat a trial and error process. Therefore, an
initial guess at the design alignment is needed and any reasonable assumption of bearing
23

offsets for a particular ship condition will generally be satisfactory. Regardless of the
assumption, the bearing offsets must be assumed for a specific ship condition and relative to
a defined reference axis. A convenient first assumption might be that all bearings are in a
straight-line either when the ship is waterborne at dockside or on the building ways.
The relative positions of the shaft bearings for a particular ship operating condition can be
determined by evaluating all the factors that affect the alignment and then superimposing
their effects on the initial alignment. The changes in shaft offsets at the bearings caused by
various factors can be summed and added to the initial bearing offsets to determine the net
shaft offsets for the specified condition. This is best illustrated by an example. Consider a
ship operating in calm seas, with its machinery at normal operating temperature, its hull
ballasted to a specific trim, with one outboard bearing worn to very near its limit, and the
others worn in proportion to their unit load. The bearings supporting the shafting system
were installed on the building ways to a specified design alignment and tolerance with
respect to one another. The shaft alignment was verified to be within the specified tolerance.
For this steady state condition, the shaft offset at each bearing is the sum of its design
alignment offset on the building ways plus the following effects:

the relative change in the hull deflection in way of the bearings that occurred in going
from the building ways to the specified ballast condition
the thermal growth of the bearing foundation that occurred going from the cold to
normal operating temperature
the weardown on the bearings, assuming that they were initially new
the as-built deviation in the bearing offset from the design alignment allowed by the
specified installation tolerance.

Note that the offsets discussed here may be relative to any straight-line reference axis, such
as an optical line of sight. The axis used to define the offsets should be one that is
convenient to the designer or shipbuilder. It can be the construction baseline for the ship. It
can be any shaftline that can be referenced to the ships structure at two points, such as
machined lands in way of two different bearings. Rigid-body translations and rotations will
have no effect on bearing and shaft reactions.
Note that there is a distinction between bearing offsets and shaft offsets at bearings. Hull
deflections and thermal growth change the bearing offsets, which in turn affect the shaft
offsets. Bearing weardown affects the shaft offsets, but has no effect on the offsets of the
bearings themselves. The shaft offsets at the bearings, and not just the bearing offsets, affect
the shaft alignment.
6.5 CALCULATING BEARING AND SHAFT REACTIONS
The bearing and shaft reactions for any ship operating condition can be calculated from the
influence equations. These equations are formulated from the results of the static analyses
(i.e. the deadweight bearing reactions, shaft bending moments, and the bearing influence
coefficients) and the shaft offsets at the bearings determined for the different ship operating
conditions. The influence equations can be used to calculate the bearing reactions or the
shaft moments at any station for any set of bearing offsets. For example, for a system with N
bearings, the reaction on bearing 3 (R3) is given by an equation of the form
R3 = RI,3 + 1I1,R3 + 2I2,R3 + 3I3,R3 + ... + NIN,R3
24

(6-1)

or more concisely
R3 = RI,3 + jIj,R3 (j = 1 ... N).

(6-2)

Likewise the moment in the shaft at station 24 is calculated using the equation
M24 = MI,24 + jIj,M24 (j = 1 ... N)

(6-3)

Using a general notation, the reaction at any bearing or can be determined using an influence
equation of the form
Ri = RI,i + jIj,Ri (j = 1 ... N)

(6-4)

where i is the bearing or shaft station at which the reaction Ri is calculated, j is the bearing
number (j = 1 ... N), and
RI,i = the straight-line or initial reaction at station i (from the static analysis results)
j
= the shaft offset at bearing j
Ij,Ri = the influence of a unit offset of the shaft at bearing j on the reaction at station i
(from the static alignment analysis).
The number of influence equations can become quite large for a system with several bearings
and shaft stations. However, the matrix formulation of the influence equations is readily
adaptable to a spreadsheet format, which makes practicable an iterative approach for
determining an optimum bearing alignment.
6.6 DETERMINING THE DESIGN ALIGNMENT
After the analysis is completed, identify the maximum and minimum bearing reactions and
the maximum bending moments for the assumed alignment, and determine if any of the
design criteria are violated. An acceptable alignment satisfies all the criteria, although this is
not necessarily an optimal alignment. There is no one best alignment; theoretically there are
an infinite number of alignment configurations that would produce the same bearing load and
component stress values. Calculating the final design alignment settings can require trial and
error, particularly if the system has several bearings. The ability to achieve a satisfactory
alignment with a variety of configurations is the primary reason that shafting systems work
on ships.
The influence coefficients are useful for determining how much a bearing must be offset to
correct a problem. For example, consider a case where the load on a stern tube bearing
exceeds its allowable limit by 5000 lbf (22 kN) under the worst-case ship operating
condition. Repositioning the bearing lower with respect to the other bearings would reduce
its load. Assume that the static analysis shows that the influence coefficient of the stern tube
bearing on itself is 50 lbf/mil (910-3 N/mm). Therefore, setting this bearing 0.100 inches
(2.5 mm) lower will reduce its load by 5000 lbf (22 kN), thus eliminating the overload
condition on this bearing. The same result could have been achieved by raising one or both
adjacent bearings. Unfortunately, the solution is rarely so simple since repositioning one or
more bearings may solve one problem and create another. Realigning a bearing changes the
load on that bearing and every other bearing in the system, (as seen from the influence
coefficient matrix), and it also changes the bending moments, stresses, deflections, and
slopes at every station along the shaft.
25

6.7 CALCULATING THE BUILDERS SETTINGS

Once an appropriate design alignment is determined, it must be expressed in terms that the
shipbuilder or maintenance facility can use so that the ship will have an acceptable alignment
once it is waterborne. The engineer must consider how the ship will be supported when the
bearings are installed and aligned, and how this differs from the baseline condition assumed
in the alignment analysis. This includes an understanding of the expected design bearing
clearances and the possible compression of the waterborne bearing staves. The engineer
must also consider the method that will be used to align the bearings; e.g., optical alignment,
gap and sag, etc.
6.8 INFORMATION THAT SHOULD BE DETERMINED FROM AN ALIGNMENT
ANALYSIS
An alignment analysis report should be prepared for immediate use and for future alignment
checks. An alignment drawing should be prepared with all the data necessary to generate the
alignment. As a minimum, the alignment analysis report should include the following:
A simplified sketch of the shafting system, showing the location of the bearings and
the other data necessary to compute the influence coefficients.
The influence coefficients.
The bearing reactions for the conditions listed in Section 6.3.
The bending moment and shear diagrams (at a minimum, the maximum values and
their location). See Figure 2 for examples.
The weardown limits for the waterborne bearings (if applicable) and the effect of
bearing weardown on the bearing reactions, with all bearings at the maximum
allowable wear and preferably the individual bearings at maximum allowable wear.
The effect of thermal growth on the bearing reactions.
The jack correction factors.
The required offset and slope (cant) of bearings in the aligned condition and the
instructions for boring the slope of the bearings, if required.
The acceptable tolerances for bearing position.
The flexibility factor, which should not be less than 0.010 inch (0.25 mm).
The gaps and sags at the inboard shaft flange couplings.
An alignment procedure is typically prepared for a new construction project. This procedure
should include data entry tables for each step in the process, beginning with the offsets at the
ends and midpoint of the bore actually obtained for the aft bearings, the subsequent sags and
gaps achieved for the uncoupled shafting, and finally, the reactions measured at each
accessible bearing after all the couplings have been connected and the propeller installed, and
with the vessel both in drydock and afloat.
6.9 VIBRATION CONSIDERATIONS
The propeller is a source of potentially serious vibratory excitation because of the
nonuniform wake field in which it operates. Since the shafting system is inherently flexible
it is extremely vulnerable to these vibratory excitations. Unless the overall design is already
proven in service, it is necessary to perform axial, lateral, and torsional vibration analyses of
the shafting system to ensure that the vibratory shaft stresses are below the allowable limits.
This is best achieved by designing the shafting system so its natural frequencies do not
26

coincide with the excitation frequencies. As with all dynamic systems, one type of vibration
can influence the other types. For this reason, various types of coupled-vibration studies can
also be performed, with a coupled axial-torsional analysis commonly performed for motor
ships.
The excitation frequencies of concern include the shaft rotational frequency, the propeller
blade frequency, and the engine firing frequency. While any of these frequencies can
theoretically be changed, they are usually fixed by this point in the design, and the only
practical option is to change the natural frequency of the shafting. This can be done by
adjusting the shaft diameters, the bearing location and loading, and possibly the system
damping. The best method for changing the shafting system natural frequencies depends on
the individual ship and the type of vibration. For example, adding a torsional damper may be
considered before any other change when there is a problem with torsional vibration on
motor ships. If the system cannot be adjusted to avoid a resonance within the normal
operating range, then a barred speed range may be mandated by the regulatory body.

27

SECTION 7.0 METHODS FOR MEASURING THE ALIGNMENT OF MARINE

PROPULSION SHAFTING SYSTEMS
7.1 INTRODUCTION
There are several proven methods for assessing the alignment of main propulsion shafting.
The most common procedures include the hydraulic jack, strain gage, gap and sag, and
optical methods. The hydraulic jack method is a common technique used to measure the
reactions of the line shaft bearings. The strain gage method is a more analytical technique
developed to measure the inboard or outboard shaft bearing reactions using strain gages
mounted at predetermined locations along the shaft. The gap and sag method is still used to
determine the initial alignment settings, but this method generally does not provide the
accuracy necessary for most systems. Optical alignment (or boresighting) is primarily used
to analyze and adjust the position of the waterborne and/or inboard bearings in drydock, and
to determine the initial position of the propulsion machinery. It is typically used prior to the
initial installation of the propulsion shafting since it can only be used when some part of the
shafting is missing. The following paragraphs discuss these methods in more detail.
7.2 HYDRAULIC JACK METHOD
The hydraulic jack method is the most widely used procedure for measuring the alignment of
main propulsion shafting. This method can measure the reactions of the line shaft and bull
gear bearings, detect bent shafts, and calculate the shaft runout. The hydraulic jack method is
more straightforward and requires less training than other methods. It has sufficient
accuracy, although a series of measurements must be performed on most inboard bearings to
obtain a comprehensive view of the system alignment. Although the hydraulic jack method
has shortcomings, it is valuable for measuring bearing reactions and shaft runout, and is often
used to validate the results of the strain gage method.
The hydraulic jack method requires a hydraulic jack (or ram), a dial indicator with
graduations of 0.001 inches (0.025 mm) or less, and a load cell with an electronic output
display accurate to within 100 lbf (450 N). A calibrated jack can be used in place of a load
cell, but this introduces additional errors of 10% to 20% because the jacks are sensitive to
structural resistance and friction within the cylinder. Since a hydraulic jack measures
pressure and not the force on the piston, the friction within the cylinder gives an inaccurate
measurement of the piston load. A load cell installed between the shaft and hydraulic jack is
more accurate than a calibrated hydraulic jack since the load cell measures the force applied
to the shaft. Load cells use very accurate strain gages (0.25% of the full load range), have
very low hysteresis, and their calibration can be easily checked with a universal test machine.
As shown in Figure 3, the hydraulic jack and load cell are positioned under the shaft, either
forward or aft of the bearing, but as close as possible to the bearing. The jack must be
supported by a jacking pad or a very stiff foundation member. Each line shaft bearing must
have its own jack location. A dial indicator for measuring the shafts vertical movement is
installed on the top of the shaft directly above the jack. The dial indicator must be anchored
to a structure that is not affected by the jacking process. Anchoring it to the bearing housing
or foundation may produce errors. Prior to starting the measurements, the test operator
should rotate the shaft several times on the turning gear. The operator should also lift the
shaft at least once to: a) reduce hysteresis in the shaft, b) ensure that the shaft can be raised
above the bottom of the bearing shell, c) ensure that no other bearings rise at the same time,
and d) ensure there is sufficient clearance between the shaft and the upper bearing shell. The
28

position of the shaft in the bearing clearances should also be checked to ensure that the shaft
is in the bottom of the bearing and is centered athwartship.
For a vertical reaction measurement, the test operator lifts the shaft in approximately 0.001
inch (0.025 mm) increments, and plots the load and shaft movement. The operator must be
attentive to prevent the shaft from lifting off (detaching) from adjacent bearings. The
operator also records the load and displacement as the shaft is lowered in approximately
0.001 inch (0.025 mm) increments. The loading and unloading curves usually do not follow
exactly the same line because of internal friction in the shafting and hydraulic cylinder,
which causes nonlinear behavior and hysteresis in the system. The bearing reaction is
determined by plotting the shaft displacement (mil, mm) versus the shaft load (lbf, N).
Figure 4 shows typical plots of the jacking data in English units. In these figures, the shaft
displacement is on the x-axis and the load is on the y-axis, although this can be reversed.
Each curve in Figure 4 can be characterized by two regions. The first region is a high
stiffness region, where the shaft is supported by both the bearing and the hydraulic jack. The
second region occurs as the shaft is raised and the load is transferred from the bearing to the
hydraulic jack, which causes a sharp change in the slope of the curve as it starts to follow the
line of lower stiffness. Any sharp change in the slope of the lifting load curve after the low
stiffness region is reached indicates that an adjacent bearing(s) is being detached or that the
shaft has contacted the upper half of the bearing shell.
The jacking load, which corresponds to the reaction at imaginary bearing located at the
hydraulic jack, can be defined by a mean load line positioned between the lifting and
lowering lines. The US Navy technical manual for bearing reaction measurement [9] only
requires that lifting line data be used; however, it is considered good practice to use the
average of lifting and lowering data. If there is a large difference between the lifting and
lowering load lines, then this is an indication that something is amiss and it should be
investigated.
The reaction at the hydraulic jack is determined by the intersection of the mean load line with
the line of zero shaft displacement, i.e., the y-axis. The load at the bearing is then determined
by multiplying the hydraulic jack reaction by the jack correction factor (C), defined as the
ratio between the bearing reaction and the reaction at the jack location. The jack correction
factor is determined from the computerized alignment analysis by including the jack in the
analysis, and is calculated as Iii/Ijacki, where Ijacki is the influence coefficient of the jack on the
ith bearing. The slope of the mean load line is approximately equal to the influence
coefficient of the bearing on itself (Iii) multiplied by C.
The bearing reactions so determined are usually within 10% of the actual value, however, the
accuracy of the influence coefficients is appreciably less. Consequently, the calculated
influence numbers should be used to adjust the bearing offsets and not those determined from
these measurements [5]. This process is repeated, so that it is performed at four locations 90
apart around the shafts circumference for each bearing as shown in Figure 4. The average of
the load lines for the 0and 180 positions and the 90 and 270 positions should be nearly
congruent. The bearing reaction is the average of the four corrected measurements. When it
has been shown that the shaft is not bent, (see the discussion below), then the reactions at that
bearing are measured at only two angular positions, 0and 180 or 90 and 270.
Since the bearing reactions are usually measured with the cold machinery plant, it is
necessary to convert the measured reactions to the hot condition by considering the rise of
29

the bull gear bearings caused by the thermal growth of their foundations and housings and
the influence coefficients of these bearings, as discussed in Section 6.5.
The hydraulic jack method can also detect bent shafts and determine the runout of a shaft by
comparing the bearing reactions at shaft positions 90 apart. A bent shaft should be
suspected if the bearing reaction changes significantly with the shaft circumferential position,
although loose flange-coupling bolts can cause the same symptom. The shaft runout at a
bearing can be calculated from the hydraulic jack measurements at that bearing according to
the following relation.
TIR = C .
where

TIR
C
R0
R90
R180
R270
Iii

=
=
=
=
=
=
=

R0

R180

R90

R270

(7-1)
Iii
total indicated runout, mils (mm)
jack correction factor
measured bearing reaction at 0 position, lbf (N)
measured bearing reaction at 90 position, lbf (N)
measured bearing reaction at 180 position, lbf (N)
measured bearing reaction at 270 position, lbf (N)
jack influence coefficient of bearing at the bearing, lbf/mil (N/mm)

The hydraulic jack method has been primarily used to measure the vertical reactions of
inboard bearings. However, there are special circumstances where it may be used to measure
the horizontal reactions. Although not nearly as practical or accurate, this method may be
useful when athwartship misalignment is purposely established in the bearing system. This
condition occurs if significant torque roll and/or thermal growth are expected during the
running state. Using this method in the horizontal plane is very rare and the measurements
can be quite cumbersome since special fixtures and procedures are typically required.
The hydraulic jack method has the following disadvantages.
(1) Many jacking measurements are required to obtain reasonable system alignment data
and account for possible bends in the shaft. The entire process is time consuming and
labor intensive when there are several bearings in the shaft system, and even more so
when there are physical interferences around the jack locations.
(2) The bearing caps or casings on bull gear and motor bearings must often be disassembled
for access to the main shaft. Also, a temporary stub shaft may need to be installed to
provide a lift point for the forward reduction gear bearing.
(3) It is difficult to measure the bearing reactions on bull gears since their influence
coefficients are large and it is easy to overlook the correct slope and unload the second
bull gear bearing. (Therefore, the strain gage method is often used to measure the bull
gear bearing reactions.)
(4) It is difficult to measure the horizontal bearing loads for evaluating athwartship
alignment.
(5) Because of the physics of this technique, the hydraulic jack method does not accurately
measure the bearing reactions on resiliently mounted reduction gears.
30

(6) The hydraulic jack method is not the optimal technique to use with the longer and more
flexible rubber stave bearing; although some outboard bearings may be measured using
this method.
(7) The following potential problems can cause inaccurate measurements and errors.
The jack and load cell are not located directly under the shaft center and the dial
indicator is not located directly over the shaft.
The dial indicator support moves because of motion of the surrounding structure.
There is hysteresis in the shafting system.
There are shifting load centers in adjacent bearings.
7.3 STRAIN GAGE METHOD
The strain gage method is the only available technique for measuring the alignment of the
entire connected propulsion shaftline with the vessel afloat. It provides a more complete
assessment of the shafts alignment condition than the hydraulic jack method and is generally
more accurate. The strain gage method is primarily employed when the hydraulic jack
method cannot be used to measure the bearing loads. The strain gage method has the
following benefits:

The reactions can be measured on bearings that are not accessible for jacking.
Repetitive measurements are quick and easy after the strain gages are installed.
The measurements can be made during ship operations.
The strain gage method is adaptable for ships with resiliently mounted reduction gears,
such as US Navy surface combatants, since there is no vertical shaft movement required
during measurement.
The reactions of bearings in both the vertical and horizontal directions can be calculated
simultaneously. The athwartships shaft alignment can be assessed from the horizontal
bearing reactions.
The reactions at the bull gear and thrust journal bearings can usually be measured without
disassembling these components.

The strain gage method provides shaft alignment information through the fundamental theory
of flexural beam analysis. These principles can be used to derive mathematical equations for
calculating the bending moments along the shaft from the measured strains, which can then
be used to calculate the bearing reactions for the shafting system. There are two proven
strain gage techniques for calculating bearing reactions, the free-body method [10, 11] and
the moment-theorem method [12].
A typical shaft with N bearings is statically indeterminate by N-2 degrees, since there are
only two independent static equilibrium equations useful for determining the vertical bearing
reactions. The two static equilibrium equations are: (1) the sum of forces in the vertical
direction is zero and (2) the sum of moments about any point is zero. Therefore, measuring
the bending moments (or bearing reactions) at N-2 or more points provides the additional
data needed to determine the bearing reactions along the shaft.
The free-body method judiciously divides the shaft into separate artificial free bodies at the
strain gage locations, such as shown in Figure 5. A minimum of N-2 strain gage locations is
required along the shaft. As shown in the figure, where the shaft section is accessible
between bearings, a set of strain gages is located between the bearings, and where the shaft
31

system is not accessible between a pair of bearings, then two sets of strain gages are located
between the next pair of bearings. The moments determined from the strain gages can be
used to calculate the bearing reactions (and the unknown shear forces) from the above two
equations of static equilibrium as shown in the figure for several of the artificial free bodies.
Observe that the shaft segment identified as the free section is statically determinate since
it has no bearing, and therefore, this is the starting point for solving the equilibrium
equations. The free-body method cannot be used when there are three bearings with
inaccessible shafting, such as with a stern tube with a pair of bearings followed by a strut
bearing. In this case, the moment-theorem method must be used.
The bending moments are calculated from the measured strains by M = IE/c, which is
derived from = E and = Mc/I for the single axis stress condition on the shaft surface,
where;
M = bending moment in the shaft caused by the loads on the shaft and bearing offsets
(if any)
I = area moment of inertia of shaft
E = modulus of elasticity
= measured strain (peak)
c = shaft outer radius
= bending stress
The strain gages at each axial position are located in the same plane normal to the shaft
centerline. The gages are installed 180 apart around the shaft circumference for measuring
the axial strains, or 90 apart for determining if the shaft is straight. The gages should be
located in accordance with the following guidelines.

The strain gages should be located where the bending moment is very sensitive to
changes in the bearing load, so that a small change in the bearing load will cause a
significant change in the measured strain. Large absolute values of the moment influence
coefficient obtained from the shafting bending moment diagram (usually determined
from the computerized alignment analysis) identify suitable locations.
The strain gages should be located at least one shaft diameter from shaft discontinuities,
such as flanges, tapers, diameter or bore transitions, and concentrated weights to ensure
that local stress concentrations do not affect the data.
The strain gages should not be installed too close to a bearing since the exact location of
the bearing reaction point is not known, which creates an error in the bending moment
arm from the strain gage location to the bearing reaction point. Mounting a strain gage
further from the bearing reduces the relative magnitude of this error.

The strain gages are typically installed by welding or bonding them to the shaft. Four gages
can be wired into a full Wheatstone bridge or two gages can be wired into a half bridge at
each longitudinal shaft location as shown in Figure 6. The two-gage configuration is usually
adequate to minimize the temperature effects and the shaft longitudinal and torsional loads,
although the four-gage configuration increases the sensitivity of the system to flexural
strains. The bending tensile and compressive strains are recorded at 90 increments by
rotating the shaft on a slow speed jacking gear. The shaft should be rotated at least one or
two revolutions before any strain readings are recorded. The measured strains should
stabilize such that the calculated bearing reactions do not differ significantly between several
successive measurements. The strain gage signals may be transmitted via slip rings and
brushes or by telemetry for measurements during ship operations.
32

Although the strain gage method has significant advantages, its complexity can be a
drawback. For example, it is sometimes difficult for the novice to recognize improper strain
gage readings, and on-site calibration and determination of the results are required to help
eliminate this problem. It is recommended that the results for a sample of the bearings be
determined before leaving the ship so that any errors can be determined and corrected. The
strain gage results can also be compared with bearing reactions measured by the hydraulic
jack method. The results should agree within 15% when the measurements have been taken
under the same conditions. Errors in the strain gage method are caused by improper location
of the strain gages, insufficient gage bonding, improper wiring techniques, damaged or
defective strain gages, poor surface condition underneath the gage, or sign-convention errors.
Any suspect gages should be replaced.
Another disadvantage of the strain gage method is that it cannot easily be used to measure the
shaft runout, as can be done with the hydraulic jack method. An indication of the shaft
runout can be determined from the strain gage measurements as follows, where a strain
runout value of less than four microstrain (strain expressed as parts per million) is usually
acceptable:
Strain runout = (strain90 + strain270) (strain0 + strain180)

(7-2)

The strain gage report should include the following: the strain gage locations and bridge
configuration, strain readings, the calculated bearing reactions, the calculated strain-to-load
conversion factors, the computer model input for the analysis of the strain gage results, the
table of influence coefficients, the evaluation of results, and any realignment
recommendations.
7.4 GAP AND SAG METHOD
The gap and sag method is a traditional technique primarily used to determine the
preliminary alignment of inboard bearings and to establish the alignment of the machinery
relative to the stern tube bearings during new construction prior to the final shaftline
assembly. Figure 7 illustrates the gap and sag. The gap is the difference in opening between
the top and bottom edges of the unconnected flange pair. The sag is the vertical distance
between the centers of the unconnected flanges. The gap and sag are normally measured
with two dial indicators making simultaneous readings on two mutually perpendicular
surfaces of the coupling flanges as the shaft is rotated through 360.
This method, once referred to as drop and gap, is based on the theory that the alignment
can be estimated by the relative position of the mating pairs of shaft flanges. The theoretical
distance between the face and periphery of the unconnected coupling flanges determined
from the alignment software can be compared with the actual measurements. Since the free
end position of a shaft flange depends on the elevation and athwartship position of the
adjacent bearings, this method can be used to establish and adjust the bearings alignment.
The alignment software can be used to determine a family of sag and gap combinations for a
cold condition that will give acceptable bearing reactions in the hot condition.
The alignment must be checked for at least every 90 of shaft rotation through 360. The
tolerance on the flange measurements should be 0.002 0.004 inches (0.05 0.10 mm),
depending on the size of the components and the length of the bearing spans, and the overall
accuracy required for the particular shaftline.
33

This gap and sag method is adequate for establishing the initial shaftline position, especially
for lengthy systems, but it has inherent limitations if it is not used in conjunction with a direct
bearing reaction measurement method. As is the case for all shaft alignment methods, these
measurements should be conducted at night when thermal distortion of the hull is a
minimum. The advantages of the gap and sag method are that: (a) it uses simple measuring
equipment, such as dial indicators and feeler gages, and (b) it is useful for evaluating both the
horizontal and vertical alignment. The disadvantages of this method are that: (a) the load on
the bearings and the deformation of the bearing supports may be different when the shafting
system is assembled, (b) temporary supports may be required for the shafting, and (c) the
bearing reactions are not measured directly, but are obtained from a plot of bearing reactions
for various combinations of gap and sag.
The following requirements must be satisfied before starting the gap and sag measurements:

all major hull work must be completed (i.e., the superstructure is in place, all major
welding is completed)
the prime movers and reduction gears are installed
the propeller is at least 50% submerged for waterborne measurements
the propeller shaft at the forward stern tube bearing is resting on the bearings bottom
shell.
for direct-connected engines proper crankshaft and bedplate alignment has been
confirmed.

The sag and gap method is highly sensitive to reading errors that increase with the number of
aligned shafts. The alignment is checked at the flange surface, causing a reading error
proportional to the ratio of the shaft length to flange diameter. Moreover, the method is
sensitive to the manufacturers practices, and the alignment error may increase if the
alignment does not follow the basic alignment requirements listed above. However, there are
situations where bearing reaction measurements cannot be made and in these cases the sag
and gap method is used. These situations generally involve smaller craft where there may
not be any line shaft bearings.
7.5 OPTICAL AND LASER METHODS
The optical method, also known as boresighting, is mostly used to check and correct the
alignment of waterborne bearings before installing the main propulsion shafting. Although
optical measurements do not provide direct bearing reactions, they are important for
establishing the builders bearing offsets and they dictate how the outboard bearing struts are
bored prior to the bearing installation. The optical method can also be used to determine the
preliminary location of inboard bearings when the shafting is not in place. The method is
typically used to determine two separate alignment parameters: the location of the bearing
supports relative to a datum (i.e., a line representing the straight-line alignment through all
the bearing centers) and the localized alignment of the bearing commonly referred to as cant
and skew. The optical method uses the line-of-slight relationship of the bearing bores to the
shaft axis of rotation to establish an optical reference line and to determine the location of the
shaft bearings relative to that line.
The location of each bearing can be verified using the optical method when the vessel is in
drydock and the shafting is removed. The optical scope or laser may be set up outside the aft
stern tube bearing looking forward, or mounted to a fixture on the driver output coupling
34

looking aft. The offsets of all bearing centers are measured relative to an optical line-of-sight
and compared to the output from the alignment computer model.
Optical methods can use a micro-alignment telescope (a borescope with a cross-hair lens) or
a more elaborate computerized laser sensor and targets, depending on the shaft configuration,
length, and alignment scheme. The laser method is based on the same technique as the
optical scope method, except that the laser beam replaces the visual line-of-sight. The main
advantage of laser alignment is that it can measure large distances with minimal loss of
accuracy.
The scope or laser is attached to the flange of a reference shaft (bull gear, line shaft, directcoupled engine drive flange, or propulsion motor) so that the line-of-sight or laser beam is
perpendicular to the flange face. This should be checked with a target at the after end of the
aftermost bearing. The flange with the scope or laser attached is then rotated through 360.
The line-of-sight or laser will trace a small circle on the target, and the mounting is corrected
until the circle is reduced to a point. Once the correct mounting has been established, targets
should be installed at the forward and aft ends of each bearing. Each bearing is set in place
and adjusted until the light-of-sight or laser beam hits the target. As with all alignment
measurements, the optical work must be performed at night when the ships hull is least
affected by temperature variations.
The optical line should be repeatable with good accuracy, and must be most accurate at the
inboard end. It must be an extension of the shafting continuous beam to permit the
mathematical analysis of the alignment data. The optical line must be a true projection of the
center of rotation of the freely suspended after end of the existing portion of the propulsion
system, i.e. reduction gear flange, direct-connected engine or propulsion motor drive flange,
or line shaft. The deflection and slope must be calculated from the end of the shaft from
which the optical line is projected. These theoretical data are used to compensate for the
differences between the projected optical line and the calculated straight-line.
7.6 WIRE METHOD
The bearing alignment can be checked by anchoring a wire above the shaft at one end and
tensioning it with a weight over a pulley at the other end of the bearing line to be examined.
The location of the bearing in relation to the wire can then be measured. In most cases, the
gear output flange or direct-connected engine drive flange is a good support for the wire.
The wire method is based on the same concepts as the optical method, although there are
some drawbacks because the wire sags and vibrates, and must not be moved during the
measurements. During installation of a bearing, a target can be installed in the bearing with
the wire running through it. The hole in the target is sized to the allowable tolerance in the
alignment procedure, which reduces the need to continually measure the bearing location
relative to the wire [13]. The sag in the wire must be accounted for according to the
following equation:
S = wx2/(2T)
where: S = wire sag
w = weight of wire per unit length
x = one-half the distance between wire supports
T = wire tension, i.e., the weight of the suspended mass
35

(7-3)

The accuracy of the wire method can be within 0.004 inch (0.1 mm) over short distances, but
the accuracy decreases over longer distances.

36

SECTION 8.0 ATHWARTSHIP ALIGNMENT OF PROPULSION SHAFTING

SYSTEMS
8.1 INTRODUCTION
Ideally, propulsion shafting is aligned with the centers of all bearings in a straight-line in the
horizontal plane so that horizontal loading in the hot operating condition is minimized. On
some shafting and reduction gear designs, this may require a horizontal bearing offset in the
cold, static condition. In some cases the torque roll of the engine requires that a system
utilize a horizontal misalignment when cold in order to achieve a good running condition.
Athwartships misalignment occurs when one or more bearings are positioned off of the
determined ideal line, causing excessive bearing loads in the horizontal plane. The bearing
reaction forces in the athwartship horizontal plane are typically smaller than in the vertical
plane. Nevertheless, the horizontal reaction components of each bearing must be analyzed to
achieve the overall alignment objective, and the position of each bearing within the
horizontal plane must be considered to obtain a comprehensive alignment overview of the
propulsion system.
In concept, the bearings are positioned horizontally so that the side clearances are equal.
Because of these bearing clearances, the horizontal bearing loads are generally much less
sensitive to athwartships misalignment than vertical misalignment, and a fair amount of
athwartships misalignment may exist without causing excessive horizontal loads. The
athwartship force component is added vectorially to the vertical force component to
determine the resultant magnitude and direction of the bearing reaction. Athwartships
bearing loads of up to 20% of the vertical loads are generally not a problem in line shaft or
tail shaft bearings. For example, the total load on a bearing with a 30,000 lbf (133 kN)
vertical load and a 6000 lbf (27 kN) athwartship load (20% of the vertical load), is 30,600 lbf
(136 kN), which is an insignificant increase over the vertical load. However, horizontal
misalignment can have a significant effect on the bull gear alignment, especially with respect
to the allowable bull gear load differential, and on direct-connected engine alignment.
8.2 MEASUREMENT OF ATHWARTSHIPS ALIGNMENT
The following paragraphs summarize information pertinent to measuring the athwartship
alignment. Section 7 provides more details on alignment measurement methods.
a. Strain Gage Method: The strain gage method is the most efficient and direct way of
measuring the horizontal bearing loads.
b. Boresighting: Optical or laser alignment equipment can be used to determine the
athwartships disposition of the gear, line shaft, or stern bearings. This method requires
drydocking the vessel, and removing the shafting up to the gear or direct-connected
engine to achieve a line of sight. Athwartships disposition of the gear shaft or engine
relative to the stern tube bearings can be achieved by using a mirror target on the gear
shaft or engine drive flange, mounting the alignment scope on the gear shaft or engine
drive flange, or in the case of hollow gear shafts, by mounting boresight targets inside the
gear shaft. An estimate of the bearing athwartships loads can be calculated from the
boresight data and the bearing influence coefficients, but this is usually not very accurate
because of the bearing clearances.

37

c. Clearance measurements: If the actual horizontal bearing reactions cannot be measured,

then manual clearance measurements with feeler gages can be used to assess the
transverse position of the bearings relative to the shaft. This method does not provide the
athwartship loads on the bearings, and so it only gives an estimate of the athwartship
alignment. Because of the characteristics of the bearings, it is difficult to make a direct
correlation of the side clearances to transverse forces without direct measurements.
Manual clearance measurements provide good data points for comparison with the
reaction results obtained by direct methods such as the strain gage method.
Usually this procedure requires that the line shaft bearing caps be removed, the gear
bearings opened up, the end seal housings of journal bearings disassembled, and the
fairings on waterborne bearings be removed. The side clearances reveal the localized
alignment of the shaft through the bearing. More specifically, the angle of the shaft
through the bearing, referred to as skew, is determined by comparing the clearances on
opposite sides for each end of the bearing. If there are notable clearances on equivalent
sides of each bearing end, then the transverse offset may need to be adjusted.
For direct-connected engines proper crankshaft alignment should first be confirmed by
taking crankshaft deflection readings, then, with the line shaft uncoupled from the engine
flange, the crankshaft should be rotated with the turning gear to ensure that it is seated at
the bottoms of the bearings. The line shaft is properly aligned athwartships when the
flanges are centered with equal gaps on both sides.
d. Inspections: Inspecting the bearing wear contact surfaces may provide a good perspective
of the overall horizontal alignment. Skewed contact patterns within a fluid film journal
bearing may be traced to deficiencies in the bearing surface, but athwartship
misalignment must also be considered. A skewed contact pattern within the waterborne
stave bearings is more likely to be directly associated with transverse misalignment.
8.3 ATHWARTSHIPS ADJUSTMENT OF LINE SHAFT BEARINGS
The line shaft bearings are the first place to start when considering athwartships realignment, although the horizontal alignment adjustments on line shaft bearings are typically
more involved than the vertical adjustments. The estimate of the horizontal adjustment is
based on the same influence coefficient matrix used in the vertical plane, provided the
shafting foundation is the same for the vertical and horizontal planes (i.e. there are no spring
or resilient mounts in the vertical direction) since the influence coefficients are based on the
shaft stiffness. However, the horizontal adjustments do not necessarily react as expected
because of side clearances within the bearing that are not included in the influence coefficient
matrix. After the alignment adjustments are complete, the bearing slope alignment should be
checked and the bearing loads measured as described in Section 7.
In most cases, the horizontal alignment of line shaft bearings is acceptable if the bearing shell
can be rolled in from either side of the bearing with the line shaft supported on a jack [13].
To correct athwartship misalignment it is usually necessary to release the bearing hold-down
bolts and shift the bearing laterally. The bearing might then have to be re-chocked. The line
shaft bearing housings are positioned to the foundation with fitted bolts. The transverse
adjustments can be made by elongating the foundation holes and installing new fitted
hardware, but this method is limited if the foundation footing is weakened in the process. As
is the case with vertical alignment adjustments, the effects on adjacent bearings and seals
must be considered.
38

8.4 ATHWARTSHIPS ADJUSTMENT OF STERN TUBE BEARINGS

If more adjustment is needed than possible using line shaft bearings, then athwartships
adjustment of the stern tube bearings should be considered. This requires drydocking the
vessel, and either boring the stern tube bores offset from their existing locations, or turning
the outer diameter of new stern bearings eccentric relative to their bores. This method of
athwartships adjustment is a practical solution if the stern bearings will also be adjusted
vertically. Any vertical or athwartships changes in the stern tube bearing position are usually
monitored using optical or laser alignment measurements. Although the US Navy does not
permit the following method, a third alternative is to hold the stern tube bearing loosely in
the stern tube bore using jack screws, align the stern tube bearing using optics, and then fix
the stern tube bearing in place using poured epoxy. When using boresight data to determine
the athwartships alignment, a fair curve alignment using a spline is helpful for determining
the most expedient moves. In this way, it is even possible, with a sufficiently flexible shaft
line, to leave the shafting with a slight bow in the horizontal plane, provided that a fair curve
results. Whenever stern tube bearing adjustments are made, both ends of each bearing must
be changed proportionately to achieve a good alignment of the bearing relative to the shaft.
8.5 ATHWARTSHIPS ALIGNMENT OF REDUCTION GEARS
The greatest concern for horizontal misalignment is the effect it has on the internal alignment
of the bull gear shaft. The horizontal bearing misalignment must be corrected if it
approaches or exceeds the gear manufacturers limits or adversely affects the low speed tooth
contact. Because of the increased risk of gear skew, large athwartships loads should not be
allowed on gear bearings, particularly if they are opposing. These loads should be reduced or
eliminated by athwartships alignment changes, or a gear skew analysis should be performed
to determine if the athwartships misalignment will have a significant effect on gear life. For
a main reduction gear bearing, the misalignment is related to the tooth load distribution at the
gear to pinion mesh. The maximum increase in tooth load occurs at the extreme forward or
aft end of each helix of the mesh.
Fluid film bearings are typically split along the horizontal plane and the shaft resultant
reaction is designed to operate in the lower half of the bearing, so that minimal loading
within the horizontal plane is desirable. Other cases exist where a journal bearing within a
main reduction gear is split along the vertical plane. More athwartship loading is expected in
this design, resulting from the combined forces of multiple mating gear elements. Other
conditions exist where the reduction gear or drive motor bearings are purposely misaligned
within the horizontal plane during the cold static condition to compensate for significant
dynamic adjustments within the horizontal plane expected from the calculated torque roll
and/or thermal growth. These dynamic effects should be verified with actual measurements,
but this is rarely done because they can be cumbersome and expensive.
If there is an excessive athwartships load couple in the gear bearings, the first action should
be to relieve this couple by an athwartships adjustment of the first one or two line shaft
bearings. This may require applying, or increasing, the athwartships load on the line shaft
bearings, but this is preferable to a gear load couple. As a last resort when there is a serious
athwartships misalignment of the gear, the whole reduction gear housing is loosened from its
foundation, and re-aligned in the horizontal plane.

39

SECTION 9.0 ADJUSTING THE SHAFT ALIGNMENT

9.1 NEW CONSTRUCTION VESSELS
Ideally, propulsion system alignment would occur after the vessel is complete and afloat.
Unfortunately, physical and economic realities require that major alignment tasks be
accomplished while the vessel is still on the building ways or blocks. For example, boring
for the stern tube and strut bearings cannot be done with the vessel afloat.
Boring of the bearings should be done after all hot work in the lower part of the stern is
complete and after all superstructure modules and other significant weights in the stern are
placed in position. The boring reference line would be taken from the extended centerline of
the rough-positioned reduction gear shaft or engine crankshaft. These bearing locations then
become fixed references to which the other shaft, gear, and engine bearings must conform.
Any attempt to finalize the alignment of these inboard components, prior to floating the
vessel for the first time, involves uncertainty and risk.
The concern that these realities cause the alignment engineer depends to a large extent on the
type of ship and the shafting arrangement. For example, an aircraft carrier shafting system is
very long and supported by many inboard line shaft bearings. These multiple line shaft
bearings provide the engineer with adjustment tools that can be used to compensate for
unintended relative movements of the strut and stern tube bearings relative to the rest of the
system. This adjustment is accomplished by modifying the vertical and athwartship offsets
of the line shaft bearings to achieve an alternative fair shaftline that produces acceptable
reactions at all bearings. However, the bearing load changes predicted by the influence
coefficient equations are not usually achieved in practice (i.e., the actual changes are less
than the calculated changes) especially when large adjustments are necessary [9]. Often
when a bearing is shimmed upward, the foundation will deflect downward, and so the shaft is
only lifted a fraction of the desired amount. Nonetheless, the influence coefficient equations
allow the initial adjustments to be made so that the desired overall alignment is nearly
achieved and only slight additional adjustments are required.
Many commercial vessels have propulsion machinery well aft, with short shaftlines. The
engine and gear are relatively close to the stern tube with few line shaft bearings. In this
case, the engineer has fewer adjustment tools available to counter the effects of relative
movements in way of the propulsion system supports after the stern tube and strut are bored.
As a result, there is a much higher risk when attempting to enhance the construction schedule
by finalizing the locations of the inboard propulsion system components. Adding line shaft
bearings in an attempt to increase the adjustment tools is counter-productive, since placing
bearings too close to the gear or engine increases the sensitivity of their bearing reactions to
misalignment and thermal movements.
The following example illustrates the constant need for the alignment engineers on newconstruction projects to anticipate downstream effects of alignment changes, to advise the
construction department of areas of risk, and to make every effort to avoid major
construction impacts.
During the new-construction of a large crude carrier, a decision was made to install
the seawater circulating piping to the main condenser before boring of the stern tube
was completed. The piping was large in diameter, about 60 inches or 1.5 m, of
copper nickel. The main condenser was attached to the low-pressure turbine, which
40

was aligned to the reduction gear. The reduction gear had been set on key chocks
based on the intended extended stern tube bore center. After the boring was
completed it was found that the reduction gear was no longer adequately aligned to
the boring center. The misalignment was in the vertical plane and was such that, if
the system were bolted up, the reactions at both the single line shaft bearing and the
aft bull gear bearing would be too high. Adjusting the single line shaft bearing could
not correct the problem. If this bearings vertical offset were increased, then its own
reaction would become even greater, and if its offset were decreased, the aft bull gear
bearing would be more overloaded. In order to solve the problem the gear was tilted
up at the forward end, which shifted the load from the aft to the forward gear bearing,
and the gear was also translated vertically upward. However, as the gear case was
tilted, the turbines had to be moved in order to remain aligned. Since the gear
bearings are closely spaced compared to the distance between those bearings and the
condenser connection for the large seawater pipes, the ratio of the movement in way
of the pipes compared to the change in offset between the two bull gear bearings was
on the order of four to one. Consequently, the seawater circulating piping had to be
removed and reworked.
9.2 THE ALIGNMENT PLAN
The alignment plan, which stays with the vessel, should be updated before delivery of the
ship to indicate the actual alignment. Data included in the as-built summary should include:
(1) The vertical and horizontal offsets measured at the centerline of the line shaft
bearings, the low speed gear or motor bearings, and at the ends and midpoint of the
waterborne bearings.
(2) The measured reactions at all accessible bearings, on-dock in the cold condition and
afloat in the cold and hot condition, if possible.
(3) The sags and gaps for the unbolted couplings, both on-dock and afloat.
(4) The shaft bending moment distribution for the actual alignment.
With the amended alignment plan, the engineer responsible for alignment related work after
the ship goes into service will have the information necessary to assess whether changes have
occurred, and to restore the shafting to an acceptable alignment after the propulsion system
has been disturbed or damaged. The offsets and bearing reactions can be measured directly
and, if they are close to the design alignment values, it is not necessary to re-calculate and
reassess the bending moments. However, if it is necessary to depart from the design
alignment, or if the alignment plan cannot be located, then a new mathematical model of the
system will have to be created and new calculations made. When new calculations are
required for an existing shafting system, the criteria of most interest are the bearing reactions,
the shaft bending moments, and the load distribution in long bearings.
9.3 WHEN TO MEASURE THE SHAFT ALIGNMENT FOR IN-SERVICE VESSELS
After the final alignment check for a new vessel shows compliance with the design
alignment, then subsequent alignment checks for an in-service vessel are only necessary in
the following cases:
(1) after work on the bearings or after the shafting has been removed.
(2) if structural damage may have changed the vertical or horizontal position of a bearing.
(3) when hot work near a bearing may have changed the bearing offsets because of
thermal distortion.
41

(4) when misalignment is suspected from events such as repeated bearing failures,
abnormal gear tooth contact patterns, unexplained shafting system vibrations, or a ship
grounding.
(5) when the bearings, journals, or staves show uneven wear or evidence of overloading or
unloading.
For case (1), the required alignment checks depend on which bearing was repaired or moved.
If a stern tube or strut bearing is to be replaced, and the bearing is mounted in a symmetric
shell (i.e., the bearing shell had not been turned eccentrically to adjust the alignment), then
the bearing can be replaced with a new one whose concentricity has been checked. If the
replaced bearing was eccentrically turned, then the original eccentricity must be used for
setting up the new bearing in a lathe. After turning to the correct outer diameter, the bearing
ends must be marked to indicate the required installation orientation.
If a line shaft bearing is being replaced with an exact duplicate, then the new shell can be
rolled into the housing and no further bearing reaction checks are required, although the
bearing clearances should be measured. However, if a line shaft bearing housing has to be
moved or unbolted for any reason, then its reaction and horizontal shoulder clearances must
be checked after it is bolted down. In addition, the adjacent line shaft bearings should be
checked after the desired reaction is achieved at the disturbed bearing.
For case (2), if there is enough damage to question the alignment, then the hot work
performed during the repairs will cause more changes to the alignment. In these cases, the
shafting system should be completely removed and the bearing offsets should be checked
after the repairs are completed.
Significant structural work in the stern of a vessel can affect the propulsion system
alignment. In the case of major work involving structure in way of the engine and shaftline,
the shafting should be removed and alignment should be monitored during the work. In less
extreme cases, such as local piece-work steel renewals in tanks, changes in the bearing
reactions and offsets of potentially affected bearings should be checked while the work is
underway. If the bearings at risk are line shaft bearings, then periodically checking their
reactions and shoulder clearances can suffice. For repairs affecting a stern tube, indirect
measurements to detect changes in the bearing offsets are recommended.
Bearing re-alignment will also affect the alignment of the shaft seals relative to the shaft,
depending on the location of the bearings relative to the seal assembly and the amount the
bearings are adjusted. Radical changes in shaft position may also require re-positioning the
seal housings as well. The seal vendor should be consulted for additional information.
Note that it is not necessary to change the existing alignment to match the design alignment.
Since a shafting system has many potential satisfactory alignments, an alignment is
acceptable when it does not impose adverse operating conditions on any component, while
allowing the ship to operate without problems until the next scheduled alignment check or
availability. When the measured reactions are different from the design values, then the
system must be re-evaluated to determine what changes are required to achieve a satisfactory
alignment [9].

42

FIGURES

43

44

45

DEFLECTION (IN)

0.040
0.020
0
-0.020
-0.040
-0.060
-0.080
-0.100

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

2400

2600

2800

3000

2400

2600

2800

3000

2400

2600

2800

3000

LOCATION ALONG SHAFT AXIS FROM FORWARD END

MOMENT (FT-LB)

200,000
100,000
0
-100,000
-200,000
-300,000
-400,000
-500,000

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

LOCATION ALONG SHAFT AXIS FROM FORWARD END

80,000
70,000
60,000

SHEAR (LB)

50,000
40,000
30,000
20,000
10,000
0
-10,000
-20,000
-30,000
0

200

400

600

800

1000

1200

1400

1600

1800

2000

2200

LOCATION ALONG SHAFT AXIS FROM FORWARD END

Figure 2. Example of Deflection, Moment, and Shear Diagrams.

46

SHIP STRUCTURE

DIAL INDICATOR

HYDRAULIC JACK
OIL PRESSURE LINE

LOAD CELL

BEARING
FOUNDATION

Figure 3. Hydraulic Jack Installation.

47

Load On Jack (Thousand Pounds)

10

180

90
270

360

0
4

10

12

14

Vertical Shaft Movement (Mils)

Load On Jack If
Bearing Was Removed

Load On Jack (Thousand Pounds)

10

Bearing Load
Completely
Transferred To Jack

Design

Slope

10

12

14

Vertical Shaft Movement (Mils)

Figure 4. Example of Bearing Reaction Diagram Plot of Load Versus Lift for Hydraulic Jack
Method.
48

FREE SECTION
G

P
Xe

Lp

L8

L7
R7

Xd

Xc

Xb

Xa

Me

Md

Mc

Mb

Ma

Ve

Vd

Vc

Vb

Va

L6

L5

R6

L4

R5

L1

L3

R4

L2

R3

R2

STRAIN-GAGE LOCATION
(TYPICAL)
Rn = BEARING REACTION
Ln = DISTANCE FROM SHAFT FORWARD END
X i = DISTANCE TO STRAIN-GAGE LOCATION
M i = MOMENT IN SHAFT,

vi

POSITIVE

= SHEAR IN SHAFT, POSITIVE

P = PROPELLER WEIGHT
G = GEAR WEIGHT
U(X) = SHAFT WEIGHT PER UNIT LENGTH

EXAMPLES OF EQUILIBRIUM EQUATIONS

REDUCTION GEAR SECTION

xa

M R1 = 0 = Ma + Va(xa - L1) + R2(L2 - L1) - G(LG - L1) Xa

F = 0 = Va + R2 - G - R1 -

L1

x u(x) dx +
L1

x u(x) dx
0

u(x) dx
0

FREE SECTION

Xe

M d = 0 = M e - M d + Ve(xe - xd) -

u(x) dx
Xd

Xe

F = 0 = Ve - Vd -

u(x) dx
Xd

STERN TUBE SECTION

L6

M 6 = 0 = - P(LP - L6) + R7(L7 - L6) + Ve (L6- xe) - M e +

L8

F = 0 = - P + R7 + R6 - Ve -

L8

x u(x) dx xe

x u(x) dx
L6

u(x) dx
xe

Figure 5. Strain Gage Method Free Body Diagrams.

49

LG

R1

GAGE NO. 1
1

2
E

G
A
G

G
A
G

E OUT

GAGE NO. 2
E IN

Two-Gage Configuration

GAGE NO. 1

GAGE NO. 3
1

2
E

G
A

A
G
E

E OUT

GAGE NO. 2

G
A

A
G
E

GAGE NO. 4
E IN

Four-Gage Configuration

Figure 6. Two and Four Strain Gage Wheatstone Bridge Configurations.

50

SAG
B

GAP = B - A

Figure 7. Gap and Sag Measurement.

51

REFERENCES
1. Xu, M., Shaft Alignment - An Overview, Sound and Vibration, (May 1993), 26-29 1986.
2. Anderson, H. C. and J. J. Zrodowski, Co-ordinated Alignment of Line shaft, Propulsion
Gear, and Turbines, 1959 SNAME Transactions.
3. Results of a Survey on Shaft Alignment Procedures Used by American Shipyards,
SNAME Technical & Research Report R-25, March 1978.
4. Shigley, J and L. Mitchell, Mechanical Engineering Design, fourth edition, McGraw-Hill,
1983.
5. Harrington, R.L., Marine Engineering, SNAME 1992.
6. Modern Marine Engineers Handbook, Vol. 1, pp. 5-17/5-19.
7. Design Methods for Naval Shipboard Systems, Part 1, Propulsion Shafting, MIL-HDBK2189-243-1.
8. Guide to Propulsion Reduction Gear Alignment and Installation, SNAME Technical &
Research Bulletin 3-43, September 1987.
9. Handbook, Bearings, Measurement of Load, NAVSEA S6420-AC-HBK-010, October 1,
1986.
10. Grant, R., Shaft Alignment Methods with Strain Gages and Load Cells, Marine
Technology, January 1980.
11. Forrest Jr., A.W., and R. F. Labasky, Shaft Alignment Using Strain Gages, Marine
Technology, July 1981.
12. Keshava Rao, M. N., M. V. Dharaneepathy, S. Gomathinayagam, K. Ramaraju, P. K.
Chakravorty, and P. K. Mishra, Computer-Aided Alignment of Ship Propulsion Shafts by
Strain-Gage Methods, Marine Technology, Vol. 28, No. 2, (March 1991), 84-90.
13. Jackson, P. C., Shaft Alignment in Ship Repair: Three Case Studies, Marine Technology,
Vol. 27, No. 1, (January 1990).
14. Donaghy, A. A., Detecting and Correcting Propulsion Shafting Misalignment,
Propeller/Shafting 97 Symposium.

52

BIBLIOGRAPHY
A Practical Guide to the Alignment and Installation of a Ship's Main Propulsion Gear and Line
shaft System - Long Line shaft System, General Electric Technical Information Series No.
DF62-MSD-206, April 13, 1962.
Andersen, H. C., D. E. Bethune, and E. H. Sibley, Shafting System Programs MGE-402 and
MGE-405 Programmed for an IBM-704 Computer with 32K Capacity and Two Tape Units,
General Electric Technical Information Series No. DF59MSD-202, Sept. 8, 1959.
Anderson, H. C. and J. J. Zrodowski, Co-ordinated Alignment of Line shaft, Propulsion Gear,
and Turbines, 1959 SNAME Transactions.
Bearing Reaction Measurement Using Strain Gages for DD-963 and DDG-993 Class Ships,
NAVSSES Publication 9243-1, May 1988.
Bradshaw, R. T., The Optimum Alignment of Marine Shafting, Marine Technology, Vol. 11,
(July 1974), 260-269.
Bradshaw, R. T. and L. Vassilopoulos, Optimum Shaft Alignment of Super Ocean Carriers,
Proceedings, Super Ocean Carrier Conference, New York, N.Y., January 1974.
Cowper, B., A. DaCosta, and S. Bobyn, Shaft Alignment Using Strain Gages: Case Studies,
Marine Technology, Vol. 36, No. 2, (April 1999), 77-91.
Cowper B., Kolomojcev, A., Solutions to Failures of Shaft Bearings on Vessels Fitted with ZDrive Thrusters, ICMES 2003 Conference, Helsinki, Finland, May 2003.
Cowper, B., M. Edgecombe, R. Ritch, and P. Semery, Kalvik Propulsion System Full-Scale
Tests: Bearing Instrumentation and Data, presented at the SNAME Propellers '88 Symposium,
Virginia Beach, VA, September 1988.
Cowper, B., Rickman, D., Sahr, D., Application of the strain gauge alignment technique on
slow-speed diesel propulsion shafting installations, World Maritime Technology Conference,
London, England, March 2006.
Cowper, B., et. al., Full-Scale Testing and Data Analysis of the Shaftline and Nozzle Dynamics
in Ice on the MV Ikaluk, Transport Canada Report TP 10402E, FTL #C83610C, May 1990.
DeGeorge, V. A., Combined Effect of Vertical and Horizontal Line shaft Alignment on Main
Reduction Gear, Marine Technology, Vol. 19, No. 2, April 1982.
Design Methods for Naval Shipboard Systems, Part 1, Propulsion Shafting, MIL-HDBK-2189243-1.
Donaghy, A. A., Propulsion Shafting Alignment, SNAME New York Metro Section meeting,
January 9, 1990.
Donaghy, A. A., Detecting and Correcting Propulsion Shafting Misalignment,
Propeller/Shafting 97 Symposium.
53

Edgecombe, M., B. Cowper, et. al., MV Kalvik Propulsion System Full-Scale Tests,,
Transport Canada Report TP 8401E, FTL # C72452C, June 1987.
Forrest Jr., A. W., and R. F. Labasky, Shaft Alignment Using Strain Gages, Marine
Technology, July 1981.
Grant, R., Shaft Alignment Methods with Strain Gages and Load Cells, Marine Technology,
January 1980.
Guide to Propulsion Reduction Gear Alignment and Installation, SNAME Technical &
Research Bulletin 3-43, September 1987.
Handbook, Bearings, Measurement of Load, NAVSEA S6420-AC-HBK-010, October 1,
1986.
Harrington, R. L., Marine Engineering, SNAME 1992.
Jackson, P. C., Shaft Alignment in Ship Repair: Three Case Studies, Marine Technology, Vol.
27, No. 1, (January 1990), 23-29.
Keshava Rao, M. N., M. V. Dharaneepathy, S. Gomathinayagam, K. Ramaraju, P. K.
Chakravorty, and P. K. Mishra, Computer-Aided Alignment of Ship Propulsion Shafts by
Strain-Gage Methods, Marine Technology, Vol. 28, No. 2, (March 1991), 84-90.
Kosiba, Lt. R. E., J. J. Francis, and R. A. Woolacott, The Alignment of Main Propulsion
Shaft Bearings in Ships, presented at the January 1956 meeting of the New England Section,
Society of Naval Architects and Marine Engineers.
Kvamsdal, R., Shaft Alignment, European Shipbuilding, No. 1 and No. 2, 1969.
Lehr, Lt. W. E. and Lt. E. L. Parker, Considerations in the Design of Marine Propulsion
Shaft Systems, Presented at the May 1961 meeting of the New England Section of the Society
of Naval Architects and Marine Engineers.
MacDonald, C. W., Main Shaft Alignment, SNAME Hampton Roads Section, October 14,
1957.
Mann, G., LCDR, USCG, Design of Propulsion Shaft Systems Using Fair Curve Alignment
Theory, Naval Engineers Journal, December 1964.
Mann, G., LCDR USCG, Analysis of Shafting Problems Using Fair Curve Alignment Theory,
Naval Engineers Journal, February 1965.
Mann, G., LCDR USCG, Shipyard Alignment of Propulsion Shafting Using Fair Curve
Alignment Theory, Naval Engineers Journal, August 1965.
McKernan, J. C., Propulsion Shafting Installation and Alignment Procedures, Proceedings,
Ship Technology and Research (STAR) Symposium, SNAME 1986.

54

Measuring Bearing Reactions by Jacking Method, General Electric Technical Information

Series No. DF61-MSD-202, November 30, 1961.
Measuring Bearing Reactions by Jacking Method with Athwartship Misalignment, General
Electric Technical Information Series No. DF62-MSD-207, July 10, 1962.
Modern Marine Engineers Handbook, Vol. 1, pp. 5-17/5-19.
Mourelatos, Z. and P. Papalambros, A Mathematical Model for Optimal Strength and
Alignment of a Marine Shafting System, Journal of Ship Research, Vol. 29, No. 3, September
1985.
Owen, S. E., Optimizing and Stochastic Modeling Applied to Propulsion Shafting Alignment,
Propeller/Shafting 97 Symposium.
Piotrowski, John, Shaft Alignment Handbook, Marcel Dekker, Inc., 1986.
Propulsion Shafting Naval Ships Technical Manual, S9086-HM-STM-010/CH-243, Chapter 243.
Propulsion Shafting, General Specifications for Ships of the U.S. Navy, Section 243.
Reeves, J. D., and N. Vlahopoulos, Optimized Alignment of a USCG Polar Class Icebreaker
Wing Shaft Using a Distributed Bearing Finite-Element Model, Marine Technology, Vol. 36,
No. 4, (Fall-Winter 1999), 238-247.
Results of a Survey on Shaft Alignment Procedures Used by American Shipyards, SNAME
Technical & Research Report R-25, March 1978
Shigley, J and L. Mitchell, Mechanical Engineering Design, Fourth edition, McGraw-Hill,
1983.
Special Report: Line shafting Alignment and Vibration, Bulletin Technique Du Bureau
Veritas, Vol. 15, No. 10-11, November - December 1986.
Sullivan, R., Shaft Alignment Over Long Distances, Plant Engineering, April 10, 1995.
Summary of Studies on Recent Ships Main Propulsion Line shaft Systems and
Recommendations for Future Designs, General Electric Technical Information Series No.
DF61-MSD-200, June 12, 1960.
Taylor Hobson Limited, Optical Alignment, Centaprint Limited, 1998
Vassilopoulos, L., Static and Underway Alignment of Main Propulsion Shaft Systems, Naval
Engineers Journal, May 1988.
Vassilopoulos, L., Constraint Equations for Optimum Shaft Alignment Problems, Proceedings,
Conference on Operational Aspects of Propulsion Shafting Systems, Vol. 91, Institute of Marine
Engineers, London, England, (May 21-22, 1979), 47.

55

Vassilopoulos, L., The Influence of Propeller Mean Loads on Propulsion Shaft Alignment,
Proceedings, Propellers '78 Symposium, S-6, SNAME, Paper No. 17, Virginia Beach, VA, (May
24-25, 1978), 17-1/17-8.
Welke, M., Shaft Alignment Monitoring Using 3D-Coordinate Measurement on the Heavy
Icebreaker CCGS Louis S. St. Laurent, Marine Technology, Vol. 31, No. 3, (July 1994), 238243.
Woodward, J. B., Investigating Some Fundamental Notions of Propulsion Shaft Alignment,
Propeller/Shafting 97 Symposium.
Xu, M., Shaft Alignment - An Overview, Sound and Vibration, (May 1993), 26-29 1986.

56