Page 141 - Biaxial Multiaxial Fatigue and Fracture

126 R.€? KAUFMAN AND T.H. TOPPER

‘Alternating
Periodic Overload

Fig. 1. Schematic of crack face interference free conditions. The magnitude of the periodic
overload stress is significantly larger than the clamping force stress (spring force stress).

the portion of the alternating shearing stress acting on the crack tips is dependent on the
magnitude of the Mode 1 periodic overloads and the spring force of the material.
Fig. I shows that the application of a large enough periodic Mode I overload causes an
interference free condition to result. The magnitude of the periodic overload affects the amount
of the total shearing force acting at the crack tips. Large periodic compressive overloads
compress the asperities in the crack wake, thereby reducing crack closure/interference and
increase the crack growth rate. Large periodic tensile overloads (yield point magnitude) leave
shear cracks fully open. Once closure free crack growth is achieved, further increasing the
magnitude of the tensile or compressive applied periodic overload (Mode I) does not decrease
the fatigue life or increase the crack growth rate.
Data from a test machine that uses internal and external pressure and axial loading on a
tubular specimen to produce a wide range of static compressive and tensile mean stresses normal
to the maximum alternating shear stress planes is presented in the following. It is expected that
tensile mean stresses normal to one or both maximum shear planes will decrease the frictional
force or surface interference on the shear plane which will result in a decrease in the fatigue life
for a given applied alternating shear stress range. Once the normal static tensile mean stresses
(Si,, ) are sufficiently large to fully separate the crack faces, the fatigue life will remain constant
for larger tensile static mean stresses (ostaue M~~~). The proposed model of the effect of static
mean stresses together with alternating shear stress for a given fatigue life is illustrated in Fig. 2.

Material
The material used during this program of study was SAE 1045 steel with two Brinell hardness
numbers (BHN 456 and 203). The first hardness level, BHN 456, was chosen because its ratio of
yield strength to fracture strength is high, and the onset of ratcheting in load control is delayed