| The propagation process of short fatigue crack can occupy 70%～90% of the service life for smooth surface structure. While the relevant theories are still not mature and complete, research on this topic is of important scientific and practical significance. Based on the replica test observations on surface short fatigue cracks of smooth hourglass shaped specimens with and without rolling maintence technology of railway LZ50 axle steel, following studies are performed:1. The fatigue damage mechanism of two kinds of micro-structural barriers for LZ50 axle steel is revealed through the experimental study on specimens without surface rolling.Research shows that the short fatigue cracks initiate in the ferrite grain on specimen surface. In the micro-structural short crack (MSC) stage, the crack growth rate of the dominant effective short fatigue crack (DESFC) exhibits decelerations twice clearly. After the initiations of short cracks, the first deceleration occurs while crack tips are close to the ferrite grain boundary. The second deceleration happens while DESFC breaks the limit of grain boundary, and the crack tips meet the pearlite banded structure. Once DESFC overcomes the resistance of the pearlite banded structure, its predominance is established quickly and the propagation of short fatigue crack enters the physical short crack (PSC) stage. From this point on, the crack growth rate increases continuously till the long crack that can be observed by conventional methods occurs, and keeps this trend to the final fracture of specimen. No obvious micro-structural barriers can be observed. Above two decelerations are corresponding to the average DESFC size of 14.77μm and 107.11μm, respectively. These two values are separately close to the average diameter of ferrite grain (14.61μm) and the average interval of rich pearlite banded structures (109.09μm) for present material. Thus it is clear that the fatigue damage of LZ50 axle steel includes two kinds of micro-structural barriers mechanisms.2. By introducing a resistance coefficient function of micro-structural barriers, a new short fatigue crack growth model that includes multiple micro-structural barriers is presented. The probabilistic description of this model is also proposed. Combined with the result of long crack test, a unified growth model and its probabilistic equation for short and long crack are developed. An equivalent stress intensity factor is applied in the unified model.The resistance coefficient function can reflect the relationship between the crack size and the resistance of micro-structural obstacle. That is the shorter the distance between DESFC tip and the micro-structural barrier is, the stronger the constraint force is. Once DESFC breaks the previous obstacle, it enters the influence range of the next barrier. Above process repeats again and again, and has a periodic feature. With the increase of crack size, the growth driving force becomes greater and greater. At the same time, the effect of micro-structural barrier starts to weaken continuously. Present short fatigue crack growth model inherits the idea of effective short fatigue crack criterion. The product of the total cyclic strain energy density of remote fields,△Wt, and the DESFC size, a, acts as the crack driving force of the model; characteristic size parameters of different micro-structural barriers are also included, and the resistance coefficient function is utilized to reflect the periodic deceleration phenomena during the propagation process. Probabilistic parameters of present model and corresponding short fatigue crack growth curve are obtained through fitting the test data. Description to the test results indicates the availability of the model. Furthermore, considering that the behavior of short and long crack propagation is a continuous and evolving physical process, the relationship between J-integration and stress intensity factor under the condition of small scale yielding is extended. An elastoplastic driving force, equivalent stress intensity factor△Keq, is introduced to describe the growth of short and long crack. Based on previous short fatigue crack growth model, a unified short and long crack growth model is developed. Fitting effects to the short crack and the long crack test results indicate the availability of the unified model.3. Comparative study is carried out on specimens with rotary turning and surface rolling at different moments and specimens without surface rolling. The results reveal that to obtain the better prolonging effect of fatigue life, surface rolling should be applied to specimens without rolling as early as possible. A rolling effect function, M(f), is introduced to revise the short crack growth model of specimens without surface rolling. The new model that can consider the effect of different rolling moments on crack growth rate is then developed.About 235～100 MPa circumferential residual compressive stress and 316～132 MPa axial residual compressive stress are engendered in specimen surface and sub-surface after rolling treatment, which can decrease the effective fatigue stress during test. Meanwhile, surface rolling increases the micro-hardness of material surface. For example, the hardness of ferrite and pearlite increases 10.81% and 3.15% than that of specimens without surface rolling. This treatment can restrain the initiation and growth of short cracks. Comparisons to the test results of five group specimens with different turning and rolling moments, indicate that the prolonging rate of life increases from 379% to 641% while the treating moments are advanced from f=0.7 to f=0.0. This shows that to obtain longer fatigue life, rotary turning and surface rolling should be applied to specimens without rolling as early as possible. The rolling effect function, M(f), reflects the relationship between the prolonging rate of life and the rolling moments. It can be used to revise previous short fatigue crack growth model and indicate the significant impact of rolling moment on crack growth rate.
|
PDF Full Text Request