FATIGUE LOADING

BY:
Dr. S. SUBRAHMANYA SWAMY
DEAN- R&D,
GLOBAL ACADEMY OF TECHNOLOGY,
RAJARJESHWARI NAGAR,
BENGALURU-560098
E - MAIL: [email protected]
[email protected]
Introduction to Fatigue failure:

Some machine elements are subjected to static loads and for such elements failure theories are
used to predict failure (Yielding or Fracture). However, most metallic structures like aircrafts,
ships, bridges, automobiles, and machine elements like- gears, axles, shafts, bearings, cams
and followers, are subjected to varying or fluctuating stresses. Fluctuating stresses (repeated
over long periods of time) will cause a part to fail (fracture) at a stress level much smaller than
the ultimate strength or even yield strength. Unlike static loading where failure can be detected
before it happens, fatigue failures are sudden and therefore catastrophic.
Fatigue failures are similar to brittle fracture and the fracture surfaces are perpendicular to the
load axis. More than 85% of the failures are due to fatigue loading. Fatigue failure is due
to crack formation and propagation.

Mechanical failures were observed to take place in metals and materials subject to repetitive
stresses well below their yield strength. The theory came to be that the metal became "tired" or
"fatigued", hence the term "fatigue" or "metal fatigue". A typical Railway axle failure, which
was the origin of a different school of thought in terms of material failure mechanism, is shown
in Fig.1. The axle, although made of steel, behaved like a brittle material in the way it failed.
The failure was sudden and catastrophic and the material did not show any evidence of
yielding; the two halves of the axle had absolutely no deformation and did not have any
microstructural changes.

Fig.1 Railway Axle failure- origin of fatigue loading concepts

Between 1852 and 1870, the first systematic fatigue tests were carried out on specifically
designed laboratory specimens by August Wohler, a German railway engineer, simulating the
loading conditions of a railway axle. These tests enabled Wohler to relate his experimental
results to the stresses in locomotive axles. In 1870, Wohler compiled a report of his
experimental work which contained several conclusions known as Wohler’s laws.

Rotating beam fatigue testers are one of the oldest methods used to determine a material’s
fatigue behaviour. A sample is placed in the machine and a force is applied via a bending
moment using weights hung off the sample. The force induces a surface stress that will be
tensile on one side of the sample (generally the top) and compressive on the opposite side.
When the test is started, the sample will rotate at the desired rate and this rotation will cause
the surfaces to interchange so that each surface experiences alternating tensile and compressive
stresses. This is illustrated in Fig.2.
 Fig.2 Rotating beam Bending test carried out by Wohler

Stages in Fatigue failure:

Stage-1: Crack Nucleation
Fatigue cracks usually start at locations of high stress concentrations (notches) such as sudden
changes in cross section, sharp corners, cracks, blowholes, inclusions, welding defects,
scratches, inspection stamps, holes etc. Localized yielding and slip along grain
boundaries lead to micro-cracks.
Stage-2: Crack propagation
-Macro crack development, orderly crack growth
Stage 3: Unstable crack leading to failure
- Remaining material cannot support stress, which will lead to rapid fracture.
The fatigue failure stages is shown in the Fig.3

Fig.3 Mechanism of fatigue failure

Fundamentals of Fatigue Loading:
A typical fatigue load is shown in Fig. 4.

Fig.4 A typical Fatigue load

The stress varies between a maximum stress, , and a minimum stress, , during a
load cycle. In the field of fatigue, the variation in stress is often defined using the stress
amplitude, , and the mean stress, . Further, variables defining the stress range, ,
and the R-value are frequently used to describe a stress cycle.
Maximum Stress= σmax
Minimum Stress= σmin
Mean Stress, Average stress σm = (σmax + σmin) / 2
Variable stress, Alternating stress σa = (σmax – σmin) /2

R= Stress Ratio = σ /σ
min max

Stress Range= ∆σ = σmax - σmin

A= Amplitude Ratio = σa / σ
m

MEAN STRESS
A stress component always there on the member.

VARIABLE COMPONENT
Superimposed on the Mean stress component to obtain Cyclic/ Fluctuating stress.

σmax = σm + σa
σmin = σm - σa
Different types of cyclic loads:

Different types of cyclic/ fluctuating stresses are shown in Fig. 5.

Completely reversed stress Released Tension Fluctuating Stresses

σm = 0, R= -1 σm = 0.5 σmax, R= 0 , σmin = 0 Non Zero mean stress

Zero mean and changing Amplitude stress Changing mean and changing
Amplitude stress

A quasi-random stress–time pattern that might be typical of an operational aircraft

during any given mission.

Fig.4 Types of Cyclic loads

FATIGUE TESTING:

In stress based fatigue tests, multiple samples of identical size, shape and composition are
subjected to different levels of stress amplitude, σa, and the number of cycles to failure, N, is
measured for each sample.

Various types of instruments and machines are used to apply cyclic loading and include rotating
bend and cantilever bend machines, servo-hydraulic or servo-electric axial push-pull testing
systems, and electric motor driven torsion fatigue testers.
The resulting S-N data for each identical specimen is plotted on either a log-log or semi log
graph. Regression is used to fit a curve through the points resulting in an S-N diagram.
A typical rotating bending test machine and the test specimen is shown in Fig. 5.
Most fatigue experiments are performed with m = 0 (e.g. rotating beam tests).
Rotating Beam Fatigue Testing Machine:

Fig.5. Rotating Bending test machine and Test specimen for

Engineering fatigue data is usually plotted as S-N curve. Here S is the stress and N the number
of cycles to failure (usually fracture). The x-axis is plotted as log (N). It should be noted that
the stress values plotted are nominal values and does not take into account local stress
concentrations.

Typically the stress value chosen for the stress is low (< 𝑦 ) and hence S-N curves deal with
fatigue failure at a large number of cycles (>104 cycles). These are the high cycle fatigue tests.
The results of the test are plotted on a log-log sheet and a typical S-N diagram is shown in
Fig.6. As obvious, if the magnitude of alternating stress increases the fatigue life decreases.
 b
Equation of the line SNf A (Nf)
Taking log on either side,
Log SNf = Log A + b Log Nf

Stress below Fatigue

limit gives ‘infinite life’

Fig.6. S-N diagram

 Steel, Ti show fatigue limit or

Endurance Limit (σ )
en
 Al, Mg, Cu show no fatigue limit.

Fig.7. S-N diagram for steel and Aluminum

Broadly two kinds of S-N curves can be differentiated for two classes of materials.
 Those where a stress below a threshold value gives a ‘very long’ life; this stress value
is called the Fatigue Limit / Endurance limit. Steel and Ti come under this category.
 Those where a decrease in stress increases the fatigue life of the component, but no
distinct fatigue life is observed. Al, Mg, Cu come under this category. ( Fig.7)
  For these materials, the fatigue response is specified as fatigue strength/endurance
strength, which is defined as the stress level at which failure will occur for some
8
specified number of cycles (e.g., 10 cycles).
8
 For Aluminum, 100 MPa @ 5X10 cycles.

Low Cycle Fatigue and High Cycle fatigue:

Fig.8.Low and high cycle fatigue

Low-Cycle fatigue: Domain associated with high loads and short service life. Significant
plastic strain occurs during each cycle. Low number of cycles to produce failure. (Fig.8)
3
Number of cycles to produce fatigue failure: 1<N˂10

High-cycle fatigue: For low stress levels wherein deformations are totally elastic, longer
lives result. This is called high-cycle fatigue in as much as relatively large numbers of cycles
are required to produce fatigue failure. Domain associated with low loads and long service
life. Strains are mostly confined to the elastics range.
3
High number of cycles to produce fatigue failure. N>10

Endurance Strength Modification Factors:

The most important deviations that occur in design situation compared to standard test
conditions are:

• Load variations
• Size variations
• Surface finish differences
• Temperature differences
• Reliability
• Other miscellaneous-effects- corrosive environment, fretting, residual
stresses, plating, metal spraying etc.
To account for these conditions a variety of modifying factors, each of which is intended to
account for a single effect, is applied to the endurance limit value of test specimen obtained
under laboratory conditions.

If 𝜎𝑒𝑛 = Un-corrected Endurance limit in reversed bending of the highly polished test
specimen under laboratory testing conditions,
Then, Corrected Endurance limit = (𝑨 × 𝑩 × 𝑪) 𝝈𝒆𝒏
Where ‘A’ is the load correction factor, ‘B’ is the size correction factor,
and ‘C’ is the surface correction factor.

Effect of Loading on Endurance Limit— Load Correction Factor (A)

The endurance limit (𝜎𝑒𝑛 ) of a material as determined by the rotating beam method is for
reversed bending load.
There are many machine members which are subjected to loads other than reversed bending
loads. Thus the endurance limit will also be different for different types of loading.
The endurance limit depends upon the type of loading and may be modified as discussed below:
A = Load correction factor for the type of loading.
Its value is usually taken as 1 for reversed or rotating bending load
= for the reversed axial load, its value may be taken as 0.7.
= for the reversed torsional or shear load, its value may be taken as
0.5 to 0.6 for ductile materials and 0.8 for brittle materials.
Effect of Size on Endurance Limit—Size Correction Factor (B)
The rotating beam specimen is small with 7.5 mm diameter. The larger the machine part, the
greater the probability that a flaw exists somewhere in the component. The chances of fatigue
failure originating at any one of these flaws are more. The endurance limit, therefore, reduces
with increasing the size of the component.

Effect of Surface Roughness on Endurance Limit—Surface Correction Factor (C)

The surface of the rotating beam specimen is polished to mirror finish. The final polishing is
carried out in the axial direction to smooth out any circumferential scratches. This makes the
specimen almost free from surface scratches and imperfections.
Fatigue properties are very sensitive to surface conditions, Fatigue initiation normally starts at
the surface since the maximum stress is at the surface.
The factors which affect the surface of a fatigue specimen can be roughly divided into three
categories:

Different surface finishes produced by different machining processes can appreciably affect
fatigue performance. Polished surface normally known as ‘par bar’ which is used in laboratory,
gives the best fatigue strength. The surface correction factors can be obtained by Fig.9 for
different surface conditions of the components.

Fig.9. Surface Correction Factor based on surface finish

and ultimate strength

(Reference: Data Hand Book-Mahadevan and Balaveera Reddy)

Temperature: This factor accounts for reductions in fatigue life which occur when the
operating temperature of the part differs from room temperature (the testing temperature). Upto
450˚C, the correction factor is taken as 1.
Reliability: This factor accounts for the scatter of test data. Generally for 50% reliability, the
correction factor is 1.

A number of other factors can act to reduce the fatigue resistance of a part.
These include tensile residual stresses, corrosion, plating, metal spraying, cyclic frequency and
other factors. Factors that reduce fatigue resistance must be accounted for when designing
parts. Surface treatments such as shot peening can induce compressive residual stresses
and increase the fatigue resistance of a part (cracks don’t open and grow well in
compressive stress fields).
Electroplating, especially chromium plating, while improves corrosion resistance and/or the
looking of surface finish, generally decreases the fatigue limit of steel.
Grinding is a necessary process to improve surface finish, abrade hard materials, and tighten
the tolerance. However, it often introduces surface tension and the heat generated in the
grinding process might temper the previously quench hardened components.
Forging refines the grain structure and improves physical properties of the metal.
Nevertheless, forging can cause decarburization (loss of surface carbon atoms) which is
harmful to fatigue life.
Hot rolling can also cause decarburization (loss of surface carbon atoms), a damaging loss
regarding the fatigue life.
Carburizing and nitriding produce higher strength and hardness at the surface and thus
improves fatigue life.

Commercial methods introducing favourable compressive stresses:

 Surface rolling - Compressive stress is introduced in between the rollers during sheet
rolling
 Shot peening - Projecting fine steel or cast-iron shot against the surface at high
velocity
 Polishing - Reducing surface scratches
 Thermal stress - Quenching or surface treatments introduce volume change giving
compressive stress.
Surface rolling and shot peening operations are shown in Fig. 10.

Surface rolling
Fig. 10 surface rolling and Shot peening.
Relation between Endurance limit in reversed bending and ultimate tensile
strength:

For Steels: 𝝈𝒆𝒏 = 0.5 𝝈𝒖 for 𝝈𝒖 < 1400 MPa

= 700 MPa for 𝝈𝒖 > 1400 MPa
For Cast iron: 𝝈𝒆𝒏 = 0.4 𝝈𝒖
For Aluminium: 𝝈𝒆𝒏 = 0.4 𝝈𝒖
For Copper alloys: 𝝈𝒆𝒏 = 0.4 𝝈𝒖

NOTCH SENSITIVITY: The term “NOTCH” is referred to any discontinuity in shape

or non-uniformity of material. A notch is a stress raiser, and at the notch, the local stresses will
be much higher than the nominal stresses. The notch serves as a starting point for fatigue crack.

The notch sensitivity of a material is a measure of how sensitive a material is to notches or

geometric discontinuities. Notch sensitivity is defined as the degree to which the theoretical
effect of stress concentration is actually reached in a notch.

• Different types of Notches:

• Metallurgical Notches- inclusions, blowholes, quenching cracks etc.
• Mechanical Notch- grooves, holes, threads, keyways, fillets, serrations, surface
indentations etc.
• Service notch- chemical or corrosion pits, scuffing, fretting, impact indentations,
etc.
It is observed that actual reduction in endurance limit of a material due to stress concentration is less
than the amount indicated by the theoretical Stress concentration factor is Kt.
To take into this reduction in stress concentration, Fatigue stress concentration factor Kft is
introduced in Fatigue analysis.

𝐸𝑛𝑑𝑢𝑟𝑎𝑛𝑐𝑒 𝑙𝑖𝑚𝑖𝑡 𝑜𝑓 𝑎 𝑛𝑜𝑡𝑐ℎ 𝑓𝑟𝑒𝑒 𝑠𝑝𝑒𝑐𝑖𝑚𝑒𝑛

Kft =
𝐸𝑛𝑑𝑢𝑟𝑎𝑛𝑐𝑒 𝑙𝑖𝑚𝑖𝑡 𝑜𝑓 𝑎 𝑛𝑜𝑡𝑐ℎ𝑒𝑑 𝑠𝑝𝑒𝑐𝑖𝑚𝑒𝑛

Notch sensitivity is defined as the susceptibility of the material to succumb to the damaging
effects of stress raising notches in fatigue loading. The notch sensitivity factor ‘q’ is defined as
𝐼𝑛𝑐𝑟𝑒𝑎𝑠𝑒 𝑜𝑓 𝑎𝑐𝑡𝑢𝑎𝑙 𝑠𝑡𝑟𝑒𝑠𝑠 𝑜𝑣𝑒𝑟 𝑛𝑜𝑚𝑖𝑛𝑎𝑙 𝑠𝑡𝑟𝑒𝑒
q=
𝐼𝑛𝑐𝑟𝑒𝑎𝑠𝑒 𝑜𝑓 𝑡ℎ𝑒𝑜𝑟𝑒𝑡𝑖𝑐𝑎𝑙 𝑠𝑡𝑟𝑒𝑠𝑠 𝑜𝑣𝑒𝑟 𝑛𝑜𝑚𝑖𝑛𝑎𝑙 𝑠𝑡𝑟𝑒𝑠𝑠

q = (Kft - 1) / (Kt - 1)
where ‘q’ is the notch sensitivity index, Kft is the fatigue notch factor and Kt is the stress-concentration
factor.

A material is said to be fully notch sensitive if q approaches a value of 1.0; it is not notch sensitive
if the ratio approaches 0. Notch sensitivity index q can be obtained from charts shown in Fig.11.
 Fig. 11. Notch sensitivity index ‘q’ based on notch radius

Effect of Mean Stresses on Fatigue Life:

When a component is subjected to fluctuating stresses, there is a mean stress component as

well as alternating stress component as illustrated in the Fig.12.

Fig.12. Fluctuating stresses

Mean stress has effect on fatigue life when present in combination with alternating component.
In general, compressive mean stresses are beneficial and tensile mean stresses are detrimental
to fatigue life as shown in the Fig.13.

Fig.13. Effect of mean stress on fatigue life

When the component is subjected to both mean stress and alternating stress, the actual failure
occurs at different scattered points as shown in Fig.14. There exists a border, which divides
the safe region from unsafe region for various combinations of 𝜎𝑚 and 𝜎𝑎 .

This border can be: (i) Soderberg’s line

(ii) Goodman’s line
(iii) Gerber’s parabola

Fig.15 illustrates the way failure prediction is made using the above criteria.

Fig.14 Failure locus with different ratios of Alternating stress and Mean stress

Fig.15 Failure prediction using different criteria

SODERBERG’S RELATION:

Soderberg’s relation is based on yield strength of the material whereas all other failure relations
for dynamic loading are based on ultimate strength of the material.

This theory proposes that designs for fluctuating normal stress states should be based on a
limiting condition defined by a straight line drawn from the endurance limit on the vertical axis
to the yield stress on the horizontal axis in the first quadrant.

σa
A Failure line

C
σen
Working stress line
R σm P
(σen /n)

σa
O Q D B
σm +
(σy /n)
σy

All mean stresses are plotted on X axis and alternating stresses are plotted on Y axis.
When 𝜎𝑎 =0, the component fails when the stress value reaches yield strength 𝜎𝑦 .
When 𝜎𝑚 =0, the component fails when the stress value reaches endurance strength 𝜎𝑒𝑛 .
The line joining 𝜎𝑦 and 𝜎𝑒𝑛 indicates the failure line for different combinations of 𝜎𝑎
and 𝜎𝑚 . This theory proposes that designs for fluctuating stress should be based on limiting
condition defined by a straight line drawn from the endurance limit on vertical axis to the yield
point on the horizontal axis in the first quadrant. The working stress line is drawn considering
the Factor of safety ‘n’ as shown in the diagram.

Comparing the similar triangles DQP and DOC

 𝜎𝑎 𝜎𝑚
+𝜎 =1
𝜎𝑒𝑛 /𝑛 𝑦 /𝑛

𝜎𝑎 𝜎𝑚 1
+ =
𝜎𝑒𝑛 𝜎𝑦 𝑛
Considering the fatigue stress concentration factor, correction factors for type of loading, size
and surface, the Goodman’s equation for ductile materials can be modified and written as:

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑦 𝑛
The Soderberg’s equation for brittle materials can be modified and written as:

𝐾𝑡𝑓 𝜎𝑎 𝑘𝑡 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑦 𝑛

𝑘𝑡 is considered in brittle materials for the fact that stress concentration is serious in brittle
materials even in static loading. Mean stress being a stress component whose magnitude does
no vary with time,

Goodman’s Relation:
σa
Failure line
A

C
Working stress line
σen
R σm P
(σen /n)

σa
O Q D B
σm +
(σu /n)
σu
This theory proposes that designs for fluctuating normal stress states should be based on a
limiting condition defined by a straight line drawn from the endurance limit on the vertical axis
to the Ultimate tensile strength on the horizontal axis in the first quadrant.

The derivation can be made on similar lines to that of Soderberg’s Relation.

Considering the fatigue stress concentration factor, correction factors for type of loading, size
and surface, the Goodman’s equation for ductile materials can be modified and written as:

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑢 𝑛
The Goodman’s equation for brittle materials can be modified and written as:

𝐾𝑡𝑓 𝜎𝑎 𝑘𝑡 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑢 𝑛
Problem 1:

Determine the magnitude of the load ‘P’ for a simply supported beam of 400 mm length
if the load at the mid span varies cyclically from 2P to 4P. Size of the beam is 50 mm
diameter. The endurance limit for reversed bending is 350 MPa and yield point stress in
tension is 520 MPa. Take size factor as 0.85, and surface correction factor as 0.9. Use a
design factor of safety of 1.9. There is no keyway present in the critical section on the
shaft.

Solution:

Since there is no keyway present in the critical section, the effect of stress concentration can
be neglected. 𝑘𝑡 = 𝐾𝑡𝑓 =1.
A= Correction factor for the type of loading= 1 (as the member is subjected to Bending stress)
B= Size factor= 0.85 (given)
C= Surface correction factor = 0.9 (given)
n= Factor of safety= 1.9 (given)
2P 200

B C D
400

P 4P P

2P
2P
Bending moment at B=0, at D=0. The bending moment at C varies from minimum to
maximum as the bending load varies from minimum to maximum.

𝑀𝑚𝑎𝑥 = Maximum Bending moment at C= 200 X 2P= 400P N-mm

𝑀𝑚𝑖𝑛 = Minimum Bending moment at C= 200 X P= 200P N-mm

𝑀𝑚𝑎𝑥 +𝑀𝑚𝑖𝑛 400𝑃 +200𝑃

𝑀𝑚 = 𝑀𝑚 = = 300P
2 2

𝑀𝑚𝑎𝑥 −𝑀𝑚𝑖𝑛 400𝑃 −200𝑃

𝑀𝑎 = 𝑀𝑎 = = 100P
2 2

32𝑀𝑚 32×300𝑃
𝜎𝑚 = 𝜎𝑚 = = 0.0244P
𝜋𝑑 3 𝜋𝑑 3
32𝑀𝑎 32×100𝑃
𝜎𝑎 = 𝜎𝑎 = = 0.00815P MPa
𝜋𝑑 3 𝜋𝑑 3

Substituting the values in the Soderberg’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑦 𝑛

1 0.00815P 0.0244𝑃 1
+ =
1 × 0.85 × 0.9 350 520 1.9

P = 6806 N
Problem 2:

A component machined from a steel plate made of 45C8 is shown in the Fig.
It is subjected to a completely reversed axial loading of 50 kN.
The factor of safety is 2. Size factor is 0.85 Determine the plate thickness for an infinite
life. Take the notch sensitivity factor as 0.8. Yield strength in tension is 315 MPa. The
Ultimate Tensile strength is 610 MPa.
Solution:

Consider the smaller section of the plate for design as the stress is maximum in that section.
The plate is subjected to completely reversed axial loading.

σu =Ultimate tensile strength of 45C8- 610 MPa.

σen = Endurance strength in reversed bending= 0.5x σu = 0.5 x 610 =305 MPa
A= Correction factor for the type of loading =0.7 (for axial loading)
B= Size factor = 0.85 (size factor)
C= Surface correction factor =0.86 from table 2.2 (for 630 MPa, for machined plate)
n= Factor of safety = 2 (given)
σy = 315 MPa (given)
q= notch sensitivity index= 0.8 (given)

𝐹𝑚𝑎𝑥 −𝐹𝑚𝑖𝑛 50000−(−50000)

𝐹𝑎 = = = 50000 N.
2 2

The mean axial load is zero and hence the mean axial stress is zero.
𝐹𝑎 50000×4 1000
𝜎𝑚 = 0 and 𝜎𝑎 = = =
𝐴 50×𝑡 𝑡

Determination of Fatigue stress concentration factor:

r/d= 5/50= 0.1

D/d= 100/50=2
Determine Kt from Fig.2.6 of Mahadevan;
Kt =2.25.
Ktf = 1+ q (Kt -1);
Ktf = 1+ 0.8 (2.25-1) = 2;
q is notch sensitivity index.

Substituting the values in the Soderberg’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑦 𝑛
 2 1000 1
+ 0=
0.7 × 0.85 × 0.86 𝑡 × 315 2

t = 24.81 mm= 25 mm

Problem 3:

A steel connecting rod is subjected to a completely reversed axial load of 100 kN.
Suggest a suitable size of the rod using a Factor of safety of 2.
The ultimate tensile strength of the material is 1100 MPa. The yield strength is 930MPa.
The correction factor for Loading may be taken as 0.85, and size factor of 0.85. Neglect
the column action and effects of stress concentration.

Solution:

Connecting rods are subjected to completely reversed axial loading.

It is clearly stated that column action should be neglected, hence effect of bending stresses
due to buckling need not be considered.
It has also been given that the effect of stress concentration has to be neglected.

Given: Correction factor for type of loading (Axial) = A=0.85

Correction factor for size=B=0.85
Correction factor for surface = C= 0.76 (from Table 2.2, Mahadevan,
for σu =1100 MPa, for cold drawn steel )
Factor of safety= n = 2
Neglecting Stress concentration, Ktf =1

The mean axial load is zero and hence the mean axial stress is zero.

𝜎𝑚 = 0
Substituting the values in the Soderberg’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑦 𝑛

1 σ𝑎 1
+ 0=
0.7 × 0.85 × 0.76 550 2
 σ𝑎 = 151 MPa
Let ‘d’ be the diameter of the connecting rod.
𝐹𝑎 100000×4
𝜎𝑎 = = = 151
𝐴 𝜋𝑑 2

d = 29.5 or 30 mm
Problem 4:

A cantilever beam made of cold drawn C40 steel is shown in Fig.2. The force P acting at
the free end varies from -50 N to +150 N as shown. The factor of safety should be 2. Notch
sensitivity index at the fillet cross section is 0.9. Determine the diameter ‘d’ at the fillet
cross section using (i) Soderberg’s relation (ii) Goodman’s relation.

Solution:
Note: The stresses will be maximum at the section where there is change of cross section. Point
A at the fillet experiences maximum bending stress.

σu =Ultimate tensile strength of C40- (from Table 1.8, Mahadevan, page 418)
= 600 MPa. (570-667 MPa range)
σen = Endurance strength in reversed bending= 0.5x σu = 0.5x 600 =300 MPa
σy = Yield strength in tension= 324 MPa
n= Factor of safety= 2
Correction factor for type of loading =A= 1 (bending)
Correction factor for size= B=0.85
Correction factor for surface = C= 0.86 (from Table 2.2, Mahadevan,
for σu =600 MPa, for cold drawn steel)
 Determination of Fatigue stress concentration factor:

r/d= 0.2d/d= 0.2,

D/d= 1.5d/d= 1.5
Determine Kt from Fig.2.16 of Mahadevan;
Kt =1.43.
Ktf = 1+q (Kt -1);
q is notch sensitivity index=0.9 (given).
Ktf = 1+0.9(1.43-1)= 1.387;

Mmax = +150 x 100= 15000 N-mm

Mmin = -50 x 100= - 5000 N-mm

𝑀𝑚𝑎𝑥 +(−𝑀min) 15000 −5000

𝑀𝑚 = 𝑀𝑚 = = 5000 N-mm
2 2
𝑀𝑚𝑎𝑥 −(−𝑀min) 15000 −(−5000)
𝑀𝑎 = 𝑀𝑎 = = 10000 N-mm
2 2

32𝑀𝑚 32×5000 50955

𝜎𝑚 = 𝜎𝑚 = =
𝜋𝑑 3 𝜋𝑑 3 𝑑3
32𝑀𝑎 32×10000 101910
𝜎𝑎 = 𝜎𝑎 = =
𝜋𝑑 3 𝜋𝑑 3 𝑑3

Substituting the values in the Soderberg’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑦 𝑛

1.387 101910 50955 1

+ =
1 × 0.85 × 0.86 𝑑 3 × 300 𝑑 3 × 324 2

d = 11.90 mm
Substituting the values in the Goodman’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑢 𝑛

1.387 101910 50955 1

3
+ 3 =
1 × 0.85 × 0.86 𝑑 × 300 𝑑 × 600 2

d = 11.34 mm
Soderberg’s relation is more conservative in its approach and gives a higher value of the
diameter.
Problem 6:

A polished steel bar shown in Fig. is subjected to an axial tensile stress that varies from
zero 𝑭𝒎𝒂𝒙 . The radius of the groove is 3 mm. The outer diameter of the bar is 30 mm.
The notch sensitivity factor at the groove is 0.95. The ultimate tensile strength of the bar
is 1250 MPa. The endurance limit in reversed bending is 600 MPa. Find the maximum
force the bar can carry for an infinite life based on Goodman’s Criterion with a factor of
safety of 2.

Solution:

σu =Ultimate tensile strength of the bar = 1250 MPa

σen = Endurance strength in reversed bending= 600MPa
Correction factor for type of loading (Axial)= A=0.7
Correction factor for size=B=0.85
Correction factor for surface = C= 1 as the bar has a polished surface.
Factor of safety= n = 2

Determination of Fatigue stress d=ϕ26mm

concentration factor:
r/d= 3/26= 0.115, D/d= 30/26= 1.15
Determine Kt from Fig.2.9 of Mahadevan;
D
Kt =1.85.
Kft = 1+q(Kt -1); Kft = 1+0.95(1.85-1)=
1.807;
q is notch sensitivity index. r =3

The bar is subjected to released cyclic

 𝐹𝑚𝑖𝑛 = 0
𝐹𝑚𝑎𝑥 − 𝐹𝑚𝑖𝑛 𝐹𝑚𝑎𝑥
𝐹𝑎 = 𝐹𝑚 = = = 0.5 𝐹𝑚𝑎𝑥
2 2
0.5𝐹𝑚𝑎𝑥 0.5𝐹𝑚𝑎𝑥 ×4 0.5𝐹𝑚𝑎𝑥 ×4
𝜎𝑚 = 𝜎𝑎 = = = = 0.00094 𝐹𝑚𝑎𝑥
𝐴 𝜋×𝑑 2 𝜋×262

Substituting the values in the Goodman’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑢 𝑛

1.807 0.00094 𝐹𝑚𝑎𝑥 0.00094 𝐹𝑚𝑎𝑥 1

+ =
0.7 × 0.85 × 1 600 1250 2

𝑭𝒎𝒂𝒙 = 90876 N
The maximum load that the bar can carry is 90876 N.

Problem 6:

A rotating shaft, subjected to a non-rotating force of 5 kN is shown in Fig.4. The shaft is

machined from plain carbon steel with an ultimate tensile strength of 500 MPa. The
equivalent notch radius can be taken as 3 mm. Determine the factor of safety of this
member based on the Goodman’s criterion. Take the notch sensitivity index at the fillet
as 0.9. Comment on the result.

Solution:

Note: Failure possibilities must be investigated at three sections- B, C, and D.

At section C, although the bending moment is maximum, the diameter is more and there is no
stress concentration.
At section D, the diameter is more and bending moment is lower compared to Section B.
Section B is prone to failure and considered for design.
 5000

B D E

300 200
C Re
Ra
400 300

B C E
A D

Bending moment diagram

Taking moments about ‘A’:
5000 x 400 -Re x700=0
Re= 2857 N, Ra= 5000-2857=2143 N
Bending Moment at B=2143 x 300= 642900 N-mm
Bending Moment at C=2143 x 400= 571400 N-mm
Bending Moment at D=2857 x 200= 642900 N-mm

Since the shaft is rotating and the load is stationary, the load on the shaft is of completely
reversed bending type. 𝜎𝑚 = 0

A= 1 (for bending)
B= 0.85 (size factor)
C= 0.89 from table 2.2 (for 500 MPa, for machined shaft)
σu = 500 MPa
σen = Endurance strength in reversed bending= 0.5x σu = 0.5x500 =250 MPa

Determination of Fatigue stress concentration factor:

r/d= 3/30= 0.1, D/d= 45/30= 1.5

Determine Kt from Fig.2.16 of Mahadevan; Kt =1.67.
Ktf = 1+q (Kt -1); Ktf = 1+0.9 (1.67-1)= 1.60;
q is notch sensitivity index.

𝑀𝑚𝑎𝑥 −𝑀𝑚𝑖𝑛 642900 −(−642900)

𝑀𝑎 = 𝑀𝑎 = = 642900 N-mm
2 2

32𝑀𝑎 32×642900
𝜎𝑎 = 𝜎𝑎 = = 242.6 𝑀𝑃𝑎
𝜋𝑑 3 𝜋×303
Substituting the values in the Goodman’s equation,

𝐾𝑡𝑓 𝜎𝑎 𝜎𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜎𝑢 𝑛

1.60 242.6 1
+ 0=
1 × 0.85 × 0.89 250 𝑛

n = 0.467 ˂ 1

The F.S. is less than 1. This component will not have infinite life instead will have a
finite life less than a million cycles.

Problem 7:

A 40 mm diameter steel shaft has 𝝈𝒚 =413 MPa 𝝈𝒆𝒏 = 336 MPa. For a factor of safety of 2,
what (i) repeated (ii) reversed torques can be sustained by the shaft indefinitely? The shaft has a
groove machined on it. The radius of the groove is 2 mm and the diameter at the bottom of the
groove is 36 mm. Take size factor =0.85 and surface factor=1.

A= 0.6 (for Torsion)

B= 0.85 (size factor, given)
C= 1 (given)
σen = Endurance strength in reversed bending= 336 MPa (given)
σy = Yield strength = 413 MPa
𝝉𝒚 = 𝟎. 𝟓 × 𝝈𝒚 = 0.5 x 413= 206.5 MPa
 (i) Let T be the released torque acting on the component

Tm = (Tmax + Tmin) / 2 = T/2= 0.5T

Ta = (Tmax - Tmin) / 2 = T/2= 0.5T

16𝑇𝑚 16×0.5 𝑇
Mean shear stress= 𝜏𝑚 = = = 5.46x 10−5 MPa
𝜋𝑑 3 3,14×363

𝐴𝑙𝑡𝑒𝑟𝑛𝑎𝑡𝑖𝑛𝑔 𝑠ℎ𝑒𝑎𝑟 𝑠𝑡𝑟𝑒𝑠𝑠 = 𝜏𝑎 = 5.46x 10−5 MPa

Determination of Fatigue stress concentration factor:

r/d= 2/36= 0.05, D/d= 40/36= 1.33
Determine Kt from Fig.2.13 of Mahadevan; Kt =1.85.

Ktf = 1+q(Kt -1); Ktf = 1+ 1 (1.85-1)= 1.85;

Assuming ‘q’ notch sensitivity index =1 .

Substituting the values in the Soderberg’s equation,

𝐾𝑡𝑓 𝜏𝑎 𝜏𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜏𝑦 𝑛

1.85 5.46 × 10−5 × 𝑇 5.46 × 10−5 × 𝑇 1

+ =
0.6 × 0.85 × 1 413 206.5 2
T= 684931 N-mm

(ii) Let ‘T’ be the completely released torque acting on the component
Tm = (Tmax + Tmin) / 2 = 0

Ta = (Tmax - Tmin) / 2 = 2T/2= T

16𝑇𝑎 16× 𝑇
𝜏𝑚 = 0 𝜏𝑎 = = = 1.09x10−4 T MPa
𝜋𝑑 3 3,14×363

Substituting the values in the Soderberg’s equation,

𝐾𝑡𝑓 𝜏𝑎 𝜏𝑚 1
+ =
𝐴𝐵𝐶 𝜎𝑒𝑛 𝜏𝑦 𝑛

1.85 1.09 × 10−4 × 𝑇 1

=
0.6 × 0.85 × 1 413 2

T= 522466 N-mm
STRESSES DUE TO COMBINED LOADING:

When a machine part is subjected to both variable normal stress and a variable shear stress,
then it is designed by using the following two theories of combined stresses:

• Maximum shear stress theory.

• Maximum normal stress theory.
In practice, the machine elements are subjected to combined bending and torsional stresses
which are cyclic. In case of two dimensional stresses, each of the stresses may have two
components –mean and variable. In such cases, equivalent normal stress component and
equivalent shear stress component are calculated.

Combined Axial and bending fatigue loading:

𝜎𝑦 𝐾𝑡𝑓 𝜎𝑎
𝜎𝑒𝑞−𝑛(𝑎𝑥𝑖𝑎𝑙) = 𝜎𝑚 + ( )
𝜎𝑒𝑛 𝐴 × 𝐵 × 𝐶
𝜎𝑦 𝐾𝑡𝑓 𝜎𝑎
𝜎𝑒𝑞−𝑛(𝑏𝑒𝑛𝑑𝑖𝑛𝑔) = 𝜎𝑚 + ( )
𝜎𝑒𝑛 𝐴 × 𝐵 × 𝐶

When a component is subjected to combined axial and bending fatigue loading,

then, the maximum equivalent normal stress is given by:

𝜎𝑒𝑞 𝑚𝑎𝑥= 𝜎𝑒𝑞−𝑛(𝑎𝑥𝑖𝑎𝑙) +𝜎𝑒𝑞−𝑛(𝑏𝑒𝑛𝑑𝑖𝑛𝑔)

Design will be based on maximum normal stress theory.

𝜎𝑦
𝜎𝑒𝑞 𝑚𝑎𝑥 =
𝑛
Combined bending and torsion fatigue loading:

The equivalent normal stress component is given by:

𝜎𝑦 𝐾𝑡𝑓 𝜎𝑎
𝜎𝑒𝑞−𝑛 = 𝜎𝑚 + ( )
𝜎𝑒𝑛 𝐴 × 𝐵 × 𝐶

The equivalent shear stress component is given by:

𝜏𝑦 𝐾𝑡𝑓 𝜏𝑎
𝜏𝑒𝑞 = 𝜏𝑚 + ( )
𝜎𝑒𝑛 𝐴 × 𝐵 × 𝐶
Using Maximum shear stress theory, Maximum shear stress is calculated.

2
1
𝜏𝑒𝑞 𝑚𝑎𝑥 = √( 𝜎𝑒𝑞−𝑛 ) + (𝜏𝑒𝑞 2 )
2

Design will be based on maximum shear stress theory.

𝜏𝑦
𝜏𝑒𝑞 𝑚𝑎𝑥=
𝑛

Using Maximum normal stress theory, Maximum normal stress is calculated.

2
1 1
𝜎𝑒𝑞 𝑚𝑎𝑥 = 𝜎𝑒𝑞−𝑛 + √( 𝜎𝑒𝑞−𝑛 ) + (𝜏𝑒𝑞 2 )
2 2

Design will be based on maximum normal stress theory.

𝜎𝑦
𝜎𝑒𝑞 𝑚𝑎𝑥 = 𝑛
 PROBLEMS ON COMBINED VARIABLE NORMAL STRESSES
AND VARIABLE SHEAR STRESSES

Problem 1:

Solution:

Consider bending load on the component: The section at the fillet is critical from the
point of view of design as it experiences Max bending stress due to smaller diameter
and change in cross section (point A).

𝑀𝑚𝑎𝑥 = 120 × 150 = 18000 𝑁 − 𝑚𝑚

𝑀𝑚𝑖𝑛 = − 80 × 150 = −12000 𝑁 − 𝑚𝑚

𝑀𝑚𝑎𝑥 +𝑀𝑚𝑖𝑛 18000 −12000

𝑀𝑚 = 𝑀𝑚 = = 3000 𝑁 − 𝑚𝑚
2 2

𝑀𝑚𝑎𝑥 −𝑀𝑚𝑖𝑛 18000+12000

𝑀𝑎 = 𝑀𝑎 = = 15000 𝑁 − 𝑚𝑚
2 2
 32×𝑀𝑚 32×3000 30573
𝜎𝑚 = 3 𝜎𝑚 = 3 = MPa
𝜋×d 𝜋×d 𝑑3

32𝑀𝑎 32×15000 152866

𝜎𝑎 = 𝜎𝑎 = = MPa
𝜋𝑑 3 𝜋𝑑 3 𝑑3

A=1 for bending, B=0.85, C=0.90, q=0.9 (given)

Determination of Fatigue stress concentration factor:

Ktf = 1+q(Kt -1); Ktf = 1+0.9 (1.44-1)=1.396 (Kt is given as 1.44 in bending)
q is notch sensitivity index.

σy = Yield strength of the component = 330 MPa)

σen= Endurance strength in reversed bending= 300 MPa
Equivalent normal stress in bending:

𝜎𝑦 𝐾𝑡𝑓 𝜎𝑎
𝜎𝑒𝑞−𝑛 (𝑏𝑒𝑛𝑑𝑖𝑛𝑔) = 𝜎𝑚 + ( )
𝜎𝑒𝑛 𝐴 × 𝐵 × 𝐶

30573 330 1.396 152866

𝜎𝑒𝑞−𝑛 bending = 3 +( )
𝑑 300 1𝑥0.85𝑥0.9 𝑑3

337424
𝜎𝑒𝑞−𝑛 (𝑏𝑒𝑛𝑑𝑖𝑛𝑔) = MPa
𝑑3

Equivalent axial stress in Tension:

𝐹𝑚𝑎𝑥 = +450 N (Tensile)

𝐹𝑚𝑖𝑛 = - 150 N (Compressive)

𝐹𝑚𝑎𝑥 +𝐹𝑚𝑖𝑛 450−150

𝐹𝑚 = = = 150 N
2 2

𝐹𝑚𝑎𝑥 −𝐹𝑚𝑖𝑛 450+150

𝐹𝑎 = = = 300 N
2 2
 𝐹𝑚 4 × 𝐹𝑚 4×150 191
𝜎𝑚 = 𝜎𝑚 = 𝜎𝑚 = = MPa
𝐴 𝜋𝑑2 𝜋𝑑 2 𝑑2

4 × 𝐹𝑎 4×300 382
𝐹𝑎 𝜎𝑎 = 𝜎𝑎 = = MPa
𝜎𝑎 = 𝜋𝑑2 𝜋𝑑 2 𝑑2
𝐴

A=0.7 for axial loading, B=0.85, C=0.90, q=0.9 (given)

Determination of Fatigue stress concentration factor:

Ktf = 1+q(Kt -1); Ktf = 1+0.9 (1.64-1)=1.576 (Kt is given as 1.64 in axial loading)
q is notch sensitivity index.

𝜎𝑦 𝐾𝑡𝑓 𝜎𝑎
𝜎𝑒𝑞−𝑛(𝑎𝑥𝑖𝑎𝑙) = 𝜎𝑚 + ( )
𝜎𝑒𝑛 𝐴 × 𝐵 × 𝐶

191 330 1.576 382

𝜎𝑒𝑞−𝑛 axial = 2
+( )
𝑑 300 1𝑥0.85𝑥0.9 𝑑 2
1056.6
𝜎𝑒𝑞−𝑛 𝑎𝑥𝑖𝑎𝑙 = MPa
𝑑2

𝜎𝑦
𝜎𝑒𝑞 𝑚𝑎𝑥 = 𝜎𝑒𝑞−𝑛 𝑏𝑒𝑛𝑑𝑖𝑛𝑔 + 𝜎𝑒𝑞−𝑛 𝑎𝑥𝑖𝑎𝑙 =
𝑛

337424 1056 330

+ =
𝑑3 𝑑2 2

Solving the above equation, we get d = 12.9 mm or 13 mm

Problem 2:

A hot rolled steel shaft is subjected to a torsional moment that varies from 330 N.m
clockwise to 110 N.m counterclockwise and an applied bending moment at a critical
section varies from 440 N.m to -220 N.m. The shaft is of uniform cross section and no key
way is present at the critical section. Determine the required shaft diameter. The material
2 2
has an ultimate strength of 550 MN/m and yield strength of 410 MN/m . Take the
endurance limit as half the ultimate strength, factor of safety of 2, Load factor of 0.55,
size factor of 0.85 and a surface finish factor of 0.62.

Solution:

σu =Ultimate tensile strength of the bar = 550 MPa

σen = Endurance limit in reversed bending= 0.5 x550 = 275 MPa
σy = Yield strength in Tension= 410 MPa.
 Kft = Kt =1 since there is no stress concentration;

Equivalent shear stress in Torsion:

𝑇(max) +𝑇(𝑚𝑖𝑛) 330+(−110)
𝑇𝑚 = = = 110 𝑁. 𝑚 = 110 × 103 N-mm
2 2

𝑇(max) −𝑇(𝑚𝑖𝑛) 330−(−110)

𝑇𝑎 = = = 220 𝑁. 𝑚 = 210 × 103 N-mm
2 2

16𝑇𝑚 16(110×103 ) 560×103

Mean shear stress,𝜏𝑚 = = = MPa
𝜋𝑑3 𝜋𝑑3 𝑑3

16𝑇𝑎 16(220×103 ) 1120×103

Variable shear stress, 𝜏𝑎 = = = MPa
𝜋𝑑3 𝜋𝑑3 𝑑3

𝝉𝒚 =Yield strength in shear is 0.5𝝈𝒚 = 𝟎. 𝟓 × 𝟒𝟏𝟎 = 𝟐𝟎𝟓 𝑴𝑷𝒂.

A=0.55 for bending, B=0.85, C=0.62, (given)

𝜏𝑦 𝐾𝑡𝑓 𝜏𝑎
𝜏𝑒𝑞 = 𝜏𝑚 + ( )
𝜎𝑒𝑛 𝐴𝐵𝐶

560×103 205 1 1120×103

𝜏𝑒𝑞 = +( )
𝑑3 275 0.55×0.85×0.62 𝑑3

𝝉𝒆𝒒 = 𝟑𝟒𝟒𝟎×𝟏𝟎𝟑
MPa
𝒅𝟑
For bending moment,

𝑀(max) +𝑀(𝑚𝑖𝑛) 440+(−220)

𝑀𝑚 = = = 110𝑁. 𝑚 = 110 × 103 N-mm
2 2

𝑀(max) −𝑀(𝑚𝑖𝑛) 440−(−220)

𝑀𝑎 = = = 330𝑁. 𝑚 = 330 × 103 N-mm
2 2

32𝑀𝑚 32(110×103 ) 1120×103

Mean bending stress,𝜎𝑚 = = = MPa
𝜋𝑑3 𝜋𝑑3 𝑑3

32×𝑀𝑎 32(330×103 ) 3360×103

and Variable bending stress, 𝜎𝑎 = = = MPa
𝜋𝑑3 𝜋𝑑3 𝑑3

A=1 for bending, B=0.85, C=0.62 (given)

 𝜎𝑦 𝐾𝑡𝑓 𝜎𝑎
𝜎𝑒𝑞−𝑛 = 𝜎𝑚 + ( )
𝜎𝑒𝑛 𝐴𝐵𝐶

110 × 103 205 1 × 330 × 103

σeq−n = + ( )
d3 275 1 × 0.85 × 0.62 × d3

𝟏𝟎𝟔𝟐𝟔
𝝈𝒆𝒒−𝒏 = MPa
𝒅𝟑

A. Determine the diameter of the shaft using the Maximum shear stress theory.

2
1
𝜏𝑒𝑞 𝑚𝑎𝑥 = √( 𝜎𝑒𝑞−𝑛 ) + (𝜏𝑒𝑞 2 )
2

1 10626×103
2
3440×103 𝟔𝟏𝟕𝟕×𝟏𝟎𝟑
𝝉𝒆𝒒 𝒎𝒂𝒙 = √( 2 ) +( )=
𝑑3 𝑑3 𝒅𝟑

𝜏𝑦 𝟔𝟏𝟕𝟕×𝟏𝟎𝟑 205
𝜏𝑒𝑞 𝑚𝑎𝑥= =
𝑛 𝒅𝟑 2

d=39.2 mm

B. Determine the diameter of the shaft using the Maximum normal stress theory.
2
1 1
𝜎𝑒𝑞 𝑚𝑎𝑥 = 𝜎𝑒𝑞−𝑛 + √( 𝜎𝑒𝑞−𝑛 ) + (𝜏𝑒𝑞 2 )
2 2
2
1 10626 1 10626 × 103 3440 × 103
𝜎𝑒𝑞 𝑚𝑎𝑥 = + √( ) + ( )
2 𝑑3 2 𝑑3 𝑑3

𝜎𝑦 400
𝜎𝑒𝑞 𝑚𝑎𝑥 = =
𝑛 2

11490×103
= 200
𝑑3

d= 38.5 mm, choose the bigger value; d= 39.2 mm or 40 mm

Cumulative Fatigue Damage:

In certain applications, the mechanical component is subjected to different stress levels for
different parts of work cycle. The life of such a component is determined by MINER’S rule.
Suppose that a component is subjected to completely reversed stresses σ1 for n1 cycles, σ2
for n2 cycles and so on. Let N1 be the number of stress cycles before fatigue failure when
only σ1 is acting. One stress cycle will consume (1/N1) of the failure life and since there are
n1 such cycles at this stress level, the percentage damage of fatigue life will be (1/N1)n1 or
(n1/ N1). Similarly, the proportionate damage at stress level σ2 is (n2/ N2).

According to this hypothesis, the rupture occurs when the sum of fractions of damage (C) ,
defined only by the consumed cycles (ni/Ni), at various load levels, reaches unity.
𝑛1 𝑛2 𝑛3 𝑛𝑥
C=𝑁1 + 𝑁2 + 𝑁3 + − − − 𝑁𝑥 = 1; this is known as Miner’s equation.

Sometimes, the number of cycles n1, n2, n3---- at stress cycles σ1, σ2, σ3 are unknown.
Suppose that α1, α2, α3 ---- are the portions of the total life that will be consumed by the
stress levels σ1, σ2 , σ3 --- etc.

Let N be the total life of the component.

Then, n1= α1 N, n2= α2 N, n3= α3 N.

[If a part is stressed for 3,000 cycles at a stress level which would cause failure in 100,000
cycles, 3 percent of the fatigue life would be expended.
3000
∴ 𝛼1 = 100000 = 0.03
Repeated stress at another stress level would consume another similarly calculated portion of
the fatigue life.]
Substituting the values in Miner’s equation,

𝛼1 𝛼2 𝛼3 𝛼𝑥 1
+ + +------------ =
𝑁1 𝑁2 𝑁3 𝑁𝑥 𝑁

Also,
𝛼1 + 𝛼2 + 𝛼3 +----- 𝛼𝑥 = 1

With the help of the above equation, the life of the components subjected to different stress
levels can be determined.

Problem:
8
Tests show that the median life of bearings operating at high frequency to be 2x10
7
cycles under 1 kN load and 3x10 cycles under 2 kN load. How many cycles the we can
expect the bearing to last if 1 kN load operates 90% of the time and 2 kN load operates
during the remaining 10% of the time?

Solution:
Let the total number of cycles (life) be N.
α1= Portion of the total life consumed by 1 kN load = 0.9
The number of cycles n1 for 1kN load= α1xN= 0.9N
α2 = Portion of the total life consumed by 2 kN load = 0.1

The number of cycles n2 for 2kN load= α2xN= 0.1N

According to the Miners Equation,

𝒏𝟏 𝒏𝟐 𝒏𝟑 𝒏𝒙
+ + +−−− =𝟏
𝑵𝟏 𝑵𝟐 𝑵𝟑 𝑵𝒙

𝒏𝟏 𝒏𝟐
In this case, 𝑵𝟏 + 𝑵𝟐= 1

8
Where N1= Failure life of the bearing when only 1 kN load acts= 2x10 cycles
7
N2= Failure life of the bearing when only 2 kN load acts= 3x10 cycles

Substituting the above values in Miners equation,

0.9𝑁 0.1𝑁
+ =1
2×108 3×107

8
N= 1.3 x10 cycles
PROBLEMS FOR PRACTICE:

4. A connecting rod is subjected to an axial load that fluctuates between 120 kN

in tension to 60 kN in compression. The material has a yield strength of 360 MPa,
and normal endurance stress of 300 MPa. Taking the factor of safety as 2.1,
find a suitable diameter of the connecting rod.
5. A round rod of diameter 1.2d is reduced to a diameter d with a fillet radius of 0.1d.
This stepped rod is to sustain a twisting a twisting moment that fluctuates between
+2.5 kN-m and +1.5 kN-m together with a bending moment that fluctuates between
+1kN-m to – 1kN-m. The rod is made of carbon steel (𝜎𝑦 =330 MPa and 𝜎𝑢 =620 MPa)
Determine the diameter ‘d’ of the rod. Take the load factor =1 for bending 0.6 for torsion.
Size and surface finish factor= 0.85. Factor of safety=2.