The Paris, Walker, Forman, Hartman-Schijve and tabular options can include temperature dependency. For the Walker option, three different methods are available to define the behaviour for stress ratios less than zero:

• use the standard Walker equation
• use a modified form of the Walker equation
• use a Paris equation

For users with proprietary data or other non-standard data, a number of user subroutine options are available for definition of the fatigue crack growth data as a function of K (or G), R and T.

These user subroutines also allow for the application of more advanced elastic-plastic propagation laws during a fatigue crack growth analysis. For example, users have successfully implemented:

• Kujawski-Ellyin model
• UNIGROW model
• Antolovich-Saxena model
• Rice-Klingbeil model

In addition to the crack growth data, options are available for threshold definition and for input of fracture toughness values. The appropriate threshold option depends partly on the growth law which has been defined, but options available include a single deltaKth value and deltaKth as a function of stress ratio. Temperature dependency is also allowed in threshold definitions and, as with crack growth data, user subroutine options allow for completely general definition of a threshold condition.

As well as being able to define temperature dependent fatigue crack growth data, there are a number of additional issues related to the use of that data during the integration process.

For a general fatigue cycle there may be different temperatures associated with the maximum and minimum conditions for the cycle. The temperature at the maximum condition may not be the peak temperature in the entire cycle. Therefore, several options are available to define the method for temperature handling when calculating da/dn for such a cycle:

• use the temperature at the Kmax condition for the particular fatigue cycle (whether it is a major or minor cycle)
• use the maximum temperature in the entire load cycle
• use the mean da/dn from the above two methods
• use the average temperature at the Kmax and Kmin conditions

For coefficient or tabular data supplied at discrete temperatures, there is a requirement during the integration process to interpolate to some intermmediate temperature. Zencrack provides alternative interpolation methods to allow the user to choose the method most appropriate to their requirements.