Unit-2

Fatigue Failure

Ravi Varma Penmetsa

 It has been found experimentally that when a material is
subjected to repeated stresses, it fails at stresses below the yield point
stresses. Such type of failure of a material is known as fatigue. The failure is
caused by means of a progressive crack formation which are usually fine
and of microscopic size. The failure may occur even without any prior
indication. The fatigue of material is effected by the size of the component,
relative magnitude of static and fluctuating loads and the number of load
reversals
1. The stresses which vary from one value of compressive to
the same value of tensile or vice versa, are known as completely reversed or
cyclic stresses.
2. The stresses which vary from a minimum value to a maximum value are
called fluctuating or alternating stresses. The stress can be tensile or
compressive or partly tensile and partly compressive.
3. The stresses which vary from zero to a certain maximum value are called
repeated stresses.
 A little consideration will show that if the stress is kept
below a certain value as shown by dotted line in Fig, the material will
not fail whatever may be the number of cycles. This stress, as
represented by dotted line, is known as endurance or fatigue limit
(σe). It is defined as maximum value of the completely reversed
bending stress which a polished standard specimen can withstand
without failure, for infinite number of cycles (usually 107 cycles)
Causes of stress concentration i) Abrupt changes in cross-section like in
keyway, steps, grooves, threaded holes results in stress concentration. ii) Poor
surface finish -The surface irregularities is also one of the reason for stress
concentration. iii) Localized loading -Due to heavy load on small area the stress
concentration occurs in the vicinity of loaded area. iv)Variation in material
properties- Particularly defects like internal flaws, voids, cracks, air holes,
cavities also results in stress concentration.
• In static loading, stress concentration in ductile materials is not so
serious as in brittle materials, because in ductile materials local
deformation or yielding takes place which reduces the concentration.
In brittle materials, cracks may appear at these local concentrations
of stress which will increase the stress over the rest of the section. It
is, therefore, necessary that in designing parts of brittle materials
such as castings, care should be taken. In order to avoid failure due
to stress concentration, fillets at the changes of section must be
provided.
• In cyclic loading, stress concentration in ductile materials is always
serious because the ductility of the material is not effective in
relieving the concentration of stress caused by cracks, flaws, surface
roughness, or any sharp discontinuity in the geometrical form of the
member. If the stress at any point in a member is above the
endurance limit of the material, a crack may develop under the
action of repeated load and the crack will lead to failure of the
member
Methods of Reducing Stress Concentration
1. Additional Notches and Holes in Tension Member
A flat plate with a V-notch subjected to tensile force is shown in
Fig.(a). It is observed that a single notch results in a high degree of
stress concentration. The severity of stress concentration is reduced
by (I) Use of multiple notches as shown in fig (b)
(II) Drilling additional holes as shown in fig (c)
(III) Removal of undesired material as shown in fig (d)
2.Fillet Radius, Undercutting and Notch for Member in Bending
A bar of circular cross-section with a shoulder and subjected to bending moment is
shown in Fig. (a). Ball bearings, gears or pulleys are mounted against this shoulder.
The shoulder creates a change in cross-section of the shaft, which results in stress
concentration. There are three methods to reduce stress concentration at the base
of this shoulder. (I) By providing fillet radius at shoulder as shown Fig (b).
(II) By undercutting the shoulder as illustrated in Fig. (c).
(III) By additional notch at the top of shoulder as shown in Fig. (d).
3.Drilling Additional Holes for Shaft:
A transmission shaft with a keyway is shown in Fig. (a). The keyway is a
discontinuity and results in stress concentration at the corners of the keyway ,
therefore holes are drilled to minimize the stress concentration at the corners of
keyway as shown in fig (b).
4.Reduction of Stress Concentration in Threaded Members
A threaded component is shown in Fig. (a). There are two methods to reduce stress
concentration (I) Undercutting as shown in Fig. (b)
(II) Reduction in Shank Diameter shown in Fig. (c)
Theoretical or Form Stress Concentration Factor
It is defined as the ratio of the maximum stress in a member
(at a notch or a fillet) to the nominal stress at the same section based
upon net area. Mathematically,

The value of kt depends upon the material and geometry of the part

Fatigue Stress Concentration Factor

When a machine member is subjected to cyclic or fatigue
loading, the value of fatigue stress concentration factor shall be
applied instead of theoretical stress concentration factor.
Notch Sensitivity
It is observed that there is an actual reduction in fatigue
limit of a material due to stress concentration, which is less than the
amount indicated by the theoretical stress concentration
factor(kt).Notch sensitivity may be defines as the degree to which the
theoretical effect of stress concentration is reached. The notch
sensitivity depends on stress gradient (which depends on the radius
of the notch, hole or fillet) and on the grain size of the material .
Fatigue Failure Theories
The failure points from fatigue tests made with different steels and
combinations of mean and variable stresses are plotted in Fig. as functions of
variable stress (σv) and mean stress (σm). The most significant observation is that,
in general, the failure point is little related to the mean stress when it is
compressive but is very much a function of the mean stress when it is tensile. In
practice, this means that fatigue failures are rare when the mean stress is
compressive (or negative)
1.Gerber Method for Combination of Stresses
• The relationship between variable stress (σv) and mean stress (σm)
for axial and bending loading for ductile materials are shown in
above Fig.
• The point σe represents the fatigue strength corresponding to the
case of complete reversal (σm = 0) and the point σu represents the
static ultimate strength corresponding to σv = 0.
• A parabolic curve drawn between the endurance limit (σe) and
ultimate tensile strength (σu) was proposed by Gerber.
• Generally, the test data for ductile material fall closer to Gerber
parabola .
• Goodman line and Soderberg lines are usually preferred in designing
machine parts.
2. Goodman Method for Combination of Stresses
A straight line connecting the endurance limit (σe) and the ultimate
strength (σu), as shown by line AB in Fig, follows the suggestion of Goodman. A
Goodman line is used when the design is based on ultimate strength and may be
used for ductile or brittle materials. line AB connecting σe and σu is called
Goodman's failure stress line. If a suitable factor of safety (F.S.) is applied to
endurance limit and ultimate strength, a safe stress line CD may be drawn parallel
to the line AB
 The expression does not include the effect of stress concentration.
It may be noted that for ductile materials, the stress concentration
may be ignored under steady loads. Since many machine and
structural parts that are subjected to fatigue loads contain regions of
high stress concentration, therefore equation must be altered to
include this effect. In such cases, the fatigue stress concentration
factor (Kf) is used to multiply the variable stress (σv). So the equation
may now be written as

.(Correction factor for bending)

a
3. Soderberg Method for Combination of Stresses
A straight line connecting the endurance limit (σe) and the yield
strength (σy), as shown by the line AB in Fig, follows the suggestion
of Soderberg line. This line is used when the design is based on yield
strength. The line AB connecting σe and σy, as shown in Fig, is
called Soderberg's failure stress line. If a suitable factor of safety
(F.S.) is applied to the endurance limit and yield strength, a safe
stress line CD may be drawn parallel to the line AB
a
1. A machine component is subjected to a flexural stress which fluctuates between +
300 MN/m2 and – 150 MN/m2. Determine the value of minimum ultimate strength
according to 1. Gerber relation; 2. Modified Goodman relation; and 3. Soderberg
relation. Take factor of safety = 2.
2. A bar of circular cross-section is subjected to alternating tensile
forces varying from a minimum of 200 kN to a maximum of 500 kN.
It is to be manufactured of a material with an ultimate tensile
strength of 900 MPa and an endurance limit of 700 MPa. Determine
the diameter of bar using safety factors of 3.5 related to ultimate
tensile strength and 4 related to endurance limit and a stress
concentration factor of 1.65 for fatigue load. Use Goodman straight
line as basis for design.
3.A circular bar of 500 mm length is supported freely at its two ends. It
is acted upon by a central concentrated cyclic load having a
minimum value of 20 kN and a maximum value of 50 kN. Determine
the diameter of bar by taking a factor of safety of 1.5, size effect of
0.85, surface finish factor of 0.9. The material properties of bar are
given by ultimate strength of 650 MPa, yield strength of 500 MPa
and endurance strength of 350 MPa.
4. A 50 mm diameter shaft is made from carbon steel having ultimate
tensile strength of 630MPa. It is subjected to a torque which
fluctuates between 2000 N-m to –800 N-m. Using Soderberg method,
calculate the factor of safety. Assume suitable values for any other
data needed.
5. A cantilever beam made of cold drawn carbon steel of circular cross-
section as shown in Fig, is subjected to a load which varies
from – F to 3 F. Determine the maximum load that this member can
withstand for an indefinite life using a factor of safety as 2. The
theoretical stress concentration factor is 1.42 and the notch
sensitivity is 0.9. Assume the following values :

Ultimate stress = 550 Mpa

Yield stress = 470 Mpa
Endurance limit = 275 Mpa
Size factor = 0.85
Surface finish factor = 0.89
6. A steel cantilever is 200 mm long. It is subjected to an
axial load which varies from 150 N (compression) to 450
N (tension) and also a transverse load at its free end
which varies from 80 N up to 120 N down. The cantilever
is of circular cross-section. It is of diameter 2d for the
first 50 mm and of diameter d for the remaining length.
Determine its diameter taking a factor of safety of 2.
Assume the following values

Yield stress = 330 MPa

Endurance limit in reversed loading = 300 MPa
Correction factors = 0.7 in reversed axial loading
= 1.0 in reversed bending
Stress concentration factor = 1.44 for bending
= 1.64 for axial loading
Size effect factor = 0.85
Surface effect factor = 0.90
Notch sensitivity index = 0.90
7. A hot rolled steel shaft is subjected to a torsional moment that varies
from330 N-m clockwise to 110 N-m counterclockwise and an applied
bending moment at a critical section varies from 440Nm to –220 Nm.
The shaft is of uniform cross-section and no keyway is present at the
critical section. Determine the required shaft diameter. The material
has an ultimate strength of 550 MN/m2 and a yield strength of 410
MN/m2. Take the endurance limit as half the ultimate strength,
f(s)=2, size factor of 0.85 and a surface finish factor of 0.62.
8. A pulley is keyed to a shaft midway between two bearings. The shaft
is made of cold drawn steel for which the ultimate strength is 550
MPa and the yield strength is 400 MPa. The bending moment at the
pulley varies from – 150 N-m to + 400 N-m as the torque on the shaft
varies from – 50 N-m to + 150 N-m. Obtain the diameter of the shaft
for an indefinite life. The stress concentration factors for the keyway
at the pulley in bending and in torsion are 1.6 and 1.3 respectively.
Take the following values:
Factor of safety = 1.5
Load correction factors = 1.0 in bending, and 0.6 in torsion
Size effect factor = 0.85 ,Surface effect factor = 0.88