A question now arises which value does the remaining carrying-capacity have, namely of the breathing girder webs being weakened by a fatigue crack. Since we have not observed any instabilized crack growth in the course of 1.test series, it remains to find a solution to the problem aforementioned, simply taking the equilibrium into consideration which several authors have resumed in the form of the so-called theory of tension field in the girder webs subject to shear. By virtue of test results and after introducing the tension field theory, the relationship was derived between the remaining shear carrying-capacity of the girders weakened by cracks and shear carrying-capacity of the unweakened ones, in the following form: Θ Θ+ Θ = − = − d sec h sin d cos V 1 d d V V 1 w c c u w ec u res (1) At that, the angle Θ of a diagonal of the tension field is equal to the two thirds of the angle Θd of the girder web diagonal. Eq. (1) be based on the assumption that the remaining shear carrying-capacity is linear in relation to the ratio dec/dw, ie in terms of the ratio of the effective depth of a fatigue crack dec to the web depth dw. 3. Metal fatigue considerations Before as early as 1900 some authors specified that the application of a mean tensile stress would decrease fatigue resistance, this approach being later rated by fatigue tests implemented on plain specimens at various mean stress levels, together with static tensile test particulars. The relationship on a given subject adopted from [2] is indicated in Fig. 1. Notwithstanding the philosophical difficulty of related static to dynamic fracture phenomena, the mentioned approach is widely accepted by industry though at high mean stress levels arising in many components (eg pre-stressed bolts) serious contradictions can be ascertained that may be explained merely by a crack growth analysis. Fig. 1 Goodman diagrams and parameter specifications. The reasons why a mean stress, σm can lessen fatigue resistance is that the maximum stress level in a constant ∆σ series of tests is increased and consequently the maximum crack tip opening displacement is enhanced; additionally the cyclic range of crack tip displacement is increased because crack closure on the elastic unloading part of a cycle can be excluded, and finally the environment has a much easier access and compensatory track to the plastically deformed zone at the crack tip. But still, giver cyclic torsion it is possible, with the crack sides of the Stage I shear cracks being left closed and thus the mean shear stress, τm, influence is not so outstanding and experimental results obey the depicted parabola more close at low R values (R = τmin/τmax). Of great attraction is the possibility of a periodic application of a tensile mean stress across the flanks of a shear directed crack since such loading conditions, frequently met in industrial components and structures (eg railway lines), may originate continuous Stage I crack growth without a transition to Stage II. To describe processes of crack initiation and propagation, six information entries are necessary: 1. Any sort of defect, no matter how small, represents a stress concentration (grain boundary, surface both inclusion and scratch) being at once a possible originator of the crack. 2. Fatigue crack are often watched below the fatigue limits of steels, these having grown but also subsequently arrested. 3. Micro- and macro-deformation responses with developed persistent slip bands may be immediately observed by optical and electron microscopes seeing that they are relatively large comparing to initial cracks which need only measure 0,1 x 0,1 microns but be difficult to find out.
RkJQdWJsaXNoZXIy MjM0NDE=