Structural failures due to a single static loading are rare. The majority of structural failures are fatigue failures caused by fluctuating loads. Stress-life (SN) and strain-life (EN) are CAE fatigue analysis methodologies that can be used to predict service life in a virtual environment.  ANSYS nCode DesignLife has both SN and EN fatigue engines, but which method should be used? It depends upon the application.

Components subjected to stresses less than yield do not experience plastic deformation and have relatively long lives. This type of service is commonly referred to as high-cycle fatigue. For ductile metals, high-cycle fatigue is generally considered to be greater than 100,000 cycles of operation. Components subjected to stresses greater than yield experience plastic deformation and have relatively short lives. This type of service is commonly referred to as low-cycle fatigue. For ductile metals, low-cycle fatigue is generally considered to be between 100 and 100,000 cycles of operation.

SN was the one of the first analytic methods developed to predict fatigue damage.  It is limited to high-cycle fatigue. In SN, the calculated elastic stress range is used with a Wöhler S-N curve (log-log graph of stress range versus number of cycles to failure) to determine the damage per stress range. The damage is accumulated throughout the operational history to determine the total damage. The basic steps in SN are:

1. Calculate linear elastic stress histories.
2. Extract fatigue cycles (generally using the rainflow algorithm).
3. Assess damage caused using stress range and S-N curve.
4. Accumulate the damages (generally using Miner’s rule).

EN is a more recently developed method. It is applicable to both low-cycle and high-cycle fatigue. In EN, the calculated strain range is used with an EN damage curve developed from the Coffin–Manson relation. The basic steps in EN are very similar to those in SN, except the elastic–plastic strains are used to predict damage. The elastic–plastic strains can be directly calculated using a nonlinear FEA or estimated from a linear–elastic FEA using an elastic–plastic correction method, such as Neuber’s rule. The basic steps in EN are:

1. Calculate elastic–plastic stress/strain histories (either directly or by using an elastic analysis and an elastic-plastic correction).
2. Extract fatigue cycles (generally using the rainflow algorithm).
3. Assess damage caused by each cycle using stress/strain range and damage curve.
4. Sum damages (generally using Miner’s rule).

Even though EN is applicable to both low-cycle and high-cycle fatigue, SN is still widely used for many reasons.

• Most components are designed for high-cycle operation.
• Large amounts of applicable fatigue test data are available.
• Many commercial industrial design codes are based on the SN approach.
• Historical data exists to verify the acceptability of the SN approach.
• SN approach is simpler to implement than EN. It can often be performed using hand calculations. EN requires a software tool.

The choice of the appropriate approach depends upon the application. If the component is subjected to fluctuating stresses greater than yield (low-cycle), then EN is probably required. If the component is subjected fluctuating stresses less than yield (high-cycle), then EN is still applicable, but SN is also a valid approach.