Issue 21

D. Croccolo et alii, Frattura ed Integrità Strutturale, 21 (2012) 13-20 ; DOI: 10.3221/IGF-ESIS.21.02 14 I NTRODUCTION ailure of mechanical components is mainly caused by the fatigue stresses especially in the presence of geometric discontinuities such as the holes machined in order to join two different parts through, for instance, bolts or pins [1]. In the case of bolted connections the holes have, normally, a clearance coupling with the shank of bolts because the connection leverages the friction forces generated between the mating parts [2]. On the other hand the pins connections (shaft-hub connections) can be realised with different amount of interference [3-5]. The holes cause a geometrical discontinuity, which leads to stress concentration during the loading; furthermore, the drilling operation creates rough surfaces or damages so that the fatigue life of the component may be drastically reduced. Such situation forces a compensation of the fatigue life reduction that can be obtained with different techniques such as the cold expansion [6, 7] or the interference fit connections. The present paper aims at investigating the effect of the interference fit level on the fatigue life of holed plates, which can be used in riveted connections schematically sketched in Fig. 1. Since some catastrophic failures may occur in this type of joints, it was decided to investigate the relation between the amount of interference and the fatigue life. There are a lot of studies concerning the effect of interference fit on the fatigue life [8- 12]; however, these papers are mainly devoted to the study of aluminium alloys (2xxx and 7xxx series) or the direct effect of interference fit is shadowed by either cold expansion or bolt clamping effect. On the opposite the material investigated in this paper is high strength steel so that no previous tests or results can be found concerning this application. Figure 1 : Example of rivet connection. E XPERIMENTAL METHODS AND TOOLS he specimens are reported in the draft of Fig. 2. They were machined in order to obtain the actual dimension (160mm high, 18mm width and 4mm of thickness with a hole diameter D=5mm). The material properties are the following: Ultimate stress 1075MPa, Yield point 990MPa, Young’s modulus 209GPa, slope of the plastic curve 578MPa, Poisson’s ratio 0.3, density 7.850kg/m 3 . In order to analyse the interference effect on the fatigue strength, some pins made of the same material have been machined. Two different pin diameters were investigated: d=5.03mm and d=5.1mm in order to obtain a specific interference of 0.6% (low level) and 2% (high level), respectively, the specific interference being calculated as I%=(d-D)/D . A set of 7 specimens has been tested for each of the three different conditions: i) open hole ( OH ), ii) low interference level ( I06 ) and iii) high interference level ( I2 ). An additional specimen has been used for the I2 level in order to find the run out point. Therefore a total of 22 specimens have been machined and tested. The testing machine was an hydraulic press with a load cell of 100kN (frequency up to 25Hz) manufactured by Giuliani s.r.l.. The pins have been press fitted into the plates by the same standing press while standard clamps provided by the press manufacturer have been used to lock the specimens; the pin insertion, the specimen lock system and an example of fracture surface are shown in Fig. 3a, 3b and 3c respectively. The white circle of Fig 3c indicates the starting point of the crack. The maximum remote stresses ( RS=F/A ) were set in the range of 725MPa and 200MPa: such nominal stresses have been calculated as the ratio between the external force applied by the standing press and the specimen gross section A=18*4=72mm 2 . The OH specimens have been tested with steps of 75MPa from 650MPa to 200MPa whereas the IXX specimens have been tested with steps of 37.5MPa (from 650MPa to 425MPa for I06 specimens and from 725MPa to 500MPa plus a single point at 425MPa for I2 specimens). The stress ratio for all the specimens was taken equal to R=0.1. F T