FATIGUE LIFE CYCLES

Definition: Fatigue is the condition whereby a material cracks or fails as a result of repeated (cyclic) stresses applied below the ultimate strength of the material.

1. Introduction to fatigue
Basic fatigue testing involves the preparation of carefully polished test specimens (surface flaws are stress concentrators) which are cycled to failure at various values of constant amplitude alternating stress levels. The data are condensed into an alternating Stress, S, verses Number of cycles to failure, N, curve which is generally referred to as a materials S-N curve. As one would expect, the curves clearly show that a low number of cycles are needed to cause fatigue failures at high stress levels while low stress levels can result in sudden, unexpected failures after a large number of cycles. Fatigue failures generally involve three stages: Crack Initiation, Crack Propagation, and Fast Fracture

2. Fluctuating stress cycles

Fig 9.1 Schematic Illustrating Cyclic Loading Parameters The following parameters are utilized to identify fluctuating stress cycles:

Mean Stress (m): Stress Range (r): Stress Amplitude (a):

m =
r = Smax-Smin

a = R=

Stress Ratio (R):

Tensile stresses are normally considered positive and compressive stresses are considered negative.

The Fatigue Life (Nf) of a component is defined by the total number of stress cycles required to cause failure. Fatigue Life can be separated into three stages where Nf = Ni + Np 1.) Crack Initiation (Ni) - Cycles required to initiate a crack. Generally results from dislocation pile-ups and/or imperfections such as surface scratches, voids, etc. 2.) Crack Growth (Np) - Cycles required to grow the crack in a stable manner to a critical size. Generally controlled by stress level since most common materials contain flaws, the prediction of crack growth is the most studied aspect of fatigue. 3.) Rapid Fracture - Very rapid critical crack growth occurs when the crack length reaches a critical value, ac. Since Rapid Fracture occurs quickly, there is no Rapid Fracture term in the Fatigue Life expression. S-N CURVE A plot of stress against number of cycles to failure. It is usually plotted alternating stress S, versus cycles to failure N. it is usually plotted S vs log N, but a plot of log S vs log N is sometimes used. The most \common procedure for generating the S-N data is the rotating bending test. S-N curves constitute design information of fundamental importance for machined parts subjected to repeated loading. Because of the scatter of fatigue life data at any given stress level, it must be recognized that there is not only one S-N curve for a given material, but a family of S-N curves with probability of failure as a parameter. To develop an S-N plot in the fatigue laboratory by standard methods, one would proceed the following ways. 1. Select a large group carefully prepared polished fatigue specimens of the material of interest and subdivide them into 4 or 5 smaller groups of at least 15specimens each. 2. Select 4 or 5 stress levels, perhaps judged by a few exploratory tests that span the stress range of S-N curve. 3. Run an entire subgroup at each of the selected stress levels following the procedures to be outlined here.

4. To make each test run, mount a specimen in the testing machine, using due care to avoid spurious stresses. Set the machine for the desired stress amplitude, with cycle counter set on zero. 5. Start the machine and run at constant stress amplitude until the specimen fails or the machine reaches a predetermined run out criterion, for example 5X105 cycles. 6. Record the stress amplitude used and the cycle count at the time of failure or run out. 7. Using a new specimen, repeat the procedure again recording the stress level and life at failure or run out. Continue to repeat this procedure until all specimens designated for the selected stress level have been tested. 8. Change to a new stress level and repeat the proceeding procedure until all specimens designed for this second stress level have been tested. Repeat this procedure until all selected stress levels have been tested. 9. Plot all data collected on a stress vs long life. 10. Simply construct a visual mean curve through the data.
1.2E+100 1E+100 8E+99 6E+99 4E+99 2E+99 0 1 2 3 4 5 6 Series1 Series2