PRESENTATION ON S-N
BEHAVIOUR
BY
SATHVIK.B
1ST [Link](DESIGN ENGINEERING)
GENERAL SN BEHAVIOR
•Obtained under axial load or stress
control conditions with smooth specimens
•S- nominal stress(alternating stress)
•Nf- no of cycles to failure(fracture)
•Plotted on semi log or log-log coordinates
•Contain fewer data points at shorter
fatigue lives
•Torsion or bending load control test
condition- do not have data at shorter
fatigue lives
•This is due to significant plastic
deformation
FIG B
•Shows typical variability
•The variability in life for given stress
level can range from less than a
factor of 2 to more than one or two
orders of magnitude
•Shows discontinuity or “knee" in
the SN curve
•Knee –few materials
• b/w 10*6 and 10*7
• cycles
• non corrosive condn
•When sufficient data are taken at
several stress levels, SN curves are
usually drawn through median lives
and represent 50%expected failure.
•Assume a reasonable crack
nucleation life defined by crack
length of 0.25mm
•No of cycles required to form this
small crack in smooth notched
specimen- small% to entire life
•Illustrated in fig-app stress
amplitude v/s no of cycles to failure
and no cycles to crack nucleation
•Size of final crack depend on
fracture toughness and stress level
•Higher stress levels- shorter critical
crack size
Mean stress effects
on SN behaviour
Sm can have substantial
influence on fatigue
behaviour
It is seen that mean
tensile stresses are
detrimental and
compressive mean
stresses are beneficial
At intermediate or high stress levels under load
control test condition, substantial cyclic creep which
increases mean strain can occur in the presence of
mean stresses
This cyclic creep adds to the detrimental effects of
tensile mean stress on fatigue life and results in add
undesirable excess deformation
For Steels and Aluminium alloys
Sf- fully reversed( Sm=0, R=-1),fatigue limit of smooth
specimen
Substantial scatter exists, but general trend indicate tensile
mean stresses are detrimental
Small mean stresses have small effect.
Much of data fall between straight and curved line
Does not provide information on compressive mean stress
effects
Compressive mean stresses cause increase up to 50% in the
alternating fatigue strength
•Represent tensile mean
stress effect for uniaxial
state of stress.
•Used in fatigue design
when modified for notches,
size, surface finish,
environmental effects and
surface finish.
• Gerber eqn incorrectly
predicts a detrimental
effects of compressive
mean stresses and does not
properly represent notched
component tensile mean
stress behaviour. Thus
Gerber eqn for design is
not emphasized.
Modified Goodman and morrow equation for given
long life along with criterion for uniaxial yielding
Sy-monotonic tensile yield strength
Sy’-cyclic yield strength
In fatigue design with constant amplitude loading and
unnotched parts-fatigue failure should not occur prior
to given life.
If yielding is not to occur
If both fatigue failure and yielding not to occur-
Factors influencing SN behaviour
MICROSTRUCTURE
Metal fatigue is significantly influenced
This includes,
1. Chemistry,
2. Heat treatment,
3. Cold working,
4. Grain size,
5. Anisotrophy,
6. Inclusions,
7. Voids,
8. Delaminations
9. Discontinuties.
Chemistry, HT, and cold working – ultimate strength-fatigue limits
Fine grains- better SN fatigue resistance except at elevated temperature
Fine grains reduce localized strain along slip
bands decreasing the amounts of irreversible slip
and provide more grain boundaries to aid in trans-
crystalline crack arrest and deflection reducing
fatigue crack growth.
Anisotrophy caused by cold working gives increased
SN fatigue resistance
SIZE EFFECTS
UNNOTCHED BENDING CONDITION
If dia or t<10mm, then SN fatigue behaviour is
independent of dia or thickness
For larger size, SN fatigue resistance is decreased. the
decrease may differ slightly for rotating bending
compared to non rotating bending
As the dia or thickness increases to 50mm, fatigue
limit decreases to a limiting factor of about 0.7 to 0.8
of the fatigue limit for specimen less than 10mm in dia
or thickness.
UNNOTCHED AXIAL CONDITION
SN fatigue resistance is lower than for most bending condition
Fatigue limit- 0.75 to 0.9 of the bending fatigue limit for smooth specimens
In bending, for a given nominal stress, the stress gradient depends upon
specimen dia or thickness.
▲D OR T – ▼bending stress gradient region, ▲average stress in a local
region on the surface.
Average stress in local region – governing stress for fatigue.
For axial loading unnotched specimens- nominal stress gradient does not
exist and the average and maximum nominal stress have the same magnitude
resulting in less size effect than in bending
In bending and axial loading, larger specimens have higher probablity for
microstructural discontinuties in the highly stessed region that contribute to
decrease in fatigue resistance.
SURFACE FINISH
Since most fatigue failures originate at the surface, the
surface will have a substantial influence on fatigue behaviour.
Surface effects are caused by differences in surface roughness,
microstructure, chemical composition, and residual
stress.
This influence will be more pronounced at long lives where a
greater percentage of the cycles is usually involved with crack
nucleation.
Reference fatigue strengths are for highly polished smooth
specimens. Most engineering parts, however, are not highly
polished and grinding or machining, will cause
degradation in fatigue strength
The higher the ultimate tensile strength and
hardness, the greater the degradation of fatigue limits.
The decreases caused from grinding and machining are
more related to surface roughness and residual
stresses, while hot-rolled and as forged behaviour include
these two important aspects along with surface micro
structural and chemical composition changes such as
decarburization and hence surface hardenability.
Avoid the hot-rolled or as-forged surface conditions at
fatigue sensitive locations by removing the undesirable
surface by grinding or machining.
FREQUENCY
The influence of frequency on S - N behaviour of metals is complicated
because of synergistic effects of test temperature, corrosive
environment, stress-strain sensitivity to strain rate, and
frequency.
Specimen heating at higher test frequency due to internal hysteresis
damping can increase the specimen temperature and thus disguise the
true ambient temperature fatigue behavior.
Generation of heat due to cyclic loading depends on the volume of
highly stressed material. Axial loading will produce more heat than
bending or notched specimens. Thus, frequency effects can be different.
If heating and corrosion effects are negligible, frequencies from less than
1 Hz to 200 Hz have had only a small effect on S - N behavior for most
structural metals.
THANK YOU