Design for creep and life prediction
Need for extrapolation techniques
• There are different approaches to model the
creep life of materials
– Modeling life based on steady state creep rates
– Use the Larson-Miller parameter approach
– Grant Bucklin approach
– θ-projection method
Need for extrapolation techniques
• The design of engineering components should take into account
the stress level that the component would experience during
service. The stress level should be such that the component does
not rupture or experience excess deformation during the useful
service life of the component.
• The value of this stress level can be obtained by one of the
following two criteria: a) stress level at which rupture/failure
would be caused in 1,00,000 or 2,00,000 h, whichever period is
appropriate, and b) stress level which produces a nominal strain
of 0.1%, 0.2% or 0.5% in a certain period, say 1,00,000 h.
• An appropriate factor of safety is invoked to ensure that the
above mentioned stress level is not attained or surpassed during
service
Ref: R. K. Penny, D. L. Marriott, Design for creep, Chapman and Hall, 1995.
Example of a ruptured sample after creep
carried out under internal pressurization
Need for extrapolation techniques
• For the design of high temperature nuclear reactor pressure vessels,
the ASME high temperature Code Section III, Division I, Subsection NH
recommends that the allowable creep stress shall be the lowest of
(i) 67% of the stress to cause rupture
ii) 100% of the stress to produce 1% total strain
(iii) 80% of the stress to cause onset of tertiary creep.
• It is difficult to conduct many tests till 100,000 h to determine the
stress level to rupture, even for established engineering materials and
thus there is a need to extrapolate data from much shorter laboratory
tests, say 103-104 h.
• This is more important for new materials where there is a need to
understand their long term behavior within a short span of time so as
to decide on their suitability for a particular application. Hence the
extrapolation techniques become important.
Extrapolation techniques
• Penny and Marriott have reviewed the different
extrapolation techniques and discussed the
advantages and disadvantages associated with each
method. The extrapolation techniques can be
categorized into three main groups:
– parametric methods, e.g. Larson-Miller parameter
– graphical methods, e.g. Grant-Bucklin method
– algebraic methods, similar to parametric methods but this
method requires combining the effects of stress,
temperature, time into a single relation
– Other known methods are θ-projection concept etc
Extrapolation techniques
• Parametric approach: Larson-Miller parameter
– When stress is plotted against a function of time
and temperature, a single master curve will be
obtained.
– This master curve can be constructed by
performing short term tests at higher
temperatures. It is then assumed to be equally
valid for longer times and lower temperatures
thus allowing for extrapolation.
Extrapolation techniques
• The Larson-Miller equation is given by
LMP = T(C + lntr)
• Here LMP is the Larson Miller parameter, T is temperature in
Kelvins and tr is the time to rupture in hours. C is a constant
which is assumed to be 46*, although other values of C such as
20 and 31 have been found to work well for austenitic and
ferritic steels respectively.
• Plots of applied stress versus the LMP would then allow
extrapolation of short term data for long term predictions.
*
G E Dieter, Mechanical Metallurgy, Tata McGraw Hill
Application of the Larson Miller parameter
approach
• Below is the stress vs LMP plot for pure Ti. What will be the LMP value
for an applied stress of 150 MPa. What is the time to failure, if the test
temperature is 400 C and the constant C is 20?
Ref: Srikant et al., Mater Sci Engg A 2007
Solution-1
• Below is the stress vs LMP plot for pure Ti. What will be the LMP value for
an applied stress of 150 MPa. What is the time to failure, if the test
temperature is 300 °C and assume the constant C is 15 for Ti?
From the plot for applied stress
of 150 MPa,
LMP = 1.32 x 104
As LMP = T (ln t+C)
ln t = (LMP / T) – C
T = 300 °C = 573 K
ln t = (13200 / 573) – 15
ln t = 23.03-15 = 8.03
t = exp(8.03) = 3071 hours
Ref: Srikant et al., Mater Sci Engg A 2007
Larson Miller parameter- problem 2
• Below is the LMP vs stress plot for a Nickel based superalloy. What should
be the applied stress so that the service life of the alloy at 500 °C is 5000
hours. Assume C as 25
Ref: X Lu et al.,J Mater Res Tech. 2014
Solution 2
• Below is the LMP vs stress plot for a Nickel based superalloy. What should
be the applied stress so that the service life of the alloy at 500 °C is 5000
hours. Assume C as 25
LMP = T (ln t+C)
The LMP for 500 C and 5000
hours is
x 10-3
LMP = 773 (ln (5000) + 25)
= 25908
From the plot, LMP of
approximately 26000
corresponds to a log (stress) of
2.7
Therefore applied stress
= 501 MPa
Ref: X Lu et al.,J Mater Res Tech. 2014
