Lifetime predictions during creep

The starting point for predicting lifetimes during creep is the Monkman-Grant equation, which states that the product of the time to failure tf and the strain rate is a constant, or... 

In many applications, plastic parts carry reasonably constant mechanical loads over periods up to few years. The polymer will creep during the lifetime of the part. At moderate load levels, long-term prediction of creep from short-term tests is possible, because the viscoelastic response of polymers (creep, stress relaxation) measured at different temperatures superimpose when shifted along the time axis [24]. 

All of the above conclusions have been drawn for isothermal conditions without considering a thermal fluctuation under creep conditions but which in practice represent the usual case during the service life of structural adhesive joints. Considering temperature effects in creep-dependent lifetime, prediction can either follow a worst case scenario in which the definition of the load-dependent fracture envelope or the test for compKance with predefined strain limits is carried out at the maximum temperature to he expected during service life. This conservative approach is likely to lead to excessive contingency reserves. 

In other words, the lifetime of a part should decrease exponentially with increasing temperature as well as with increasing applied stress both predictions are borne out by experiments. It is worth noting that Eqs. (12.42) to (12.44) are only valid if the rate of damage generation was controlled by the bulk creep response of the material and steady-state conditions are established during the experiment. 

Figure 6.21 (a) Creep lifetimes, measured experimentally on grades at different temperatures. Comparison with predictions of the necking model (lower and upper bounds). Temperature range 500—700°C (b) comparison between measured and predicted necking cross-sections during a test interrupted several times carried out on a grade 91 steel (350 MPa, 550°C,... 

---