Arne Melander

A finite element study of short cracks with different inclusion types under rolling contact fatigue load

Abstract A finite element study was performed of driving forces of short cracks at inclusions in bearing steel exposed to rolling contact load. Five inclusion configurations were considered, namely a pore, a manganese sulphide inclusion, a through-cracked alumina inclusion, an alumina inclusion which was uncracked but which could debond from the matrix and finally a titanium nitride inclusion. Short cracks were allowed to grow from the inclusions. The inclusions were 20 μm in diameter. The cracks were allowed to grow from 2 μm up to 8 μm, away from the inclusion interface into the matrix. The cracks were oriented 45° to the rolling contact plane in agreement with the experimental observations of so-called butterfly cracks in bearing steels. The inclusions had different coefficients of thermal expansion than the matrix. This generated residual stresses during quenching corresponding to a hardening operation of the steel. Friction was taken into account on the crack faces and on the inclusion interface. The material was modelled as elastic. The driving forces of the cracks were studied in terms of crack-tip displacements and energy release rates. The ranges of energy release rates for the very shortest cracks were highest for the through-cracked alumina inclusion followed by the titanium nitride, the pore and finally the manganese sulphide inclusion and the debonding non-cracked alumina inclusion. If early crack growth controls fatigue life a danger index for these inclusions would fall in that order provided the inclusions all have the same size.

Influence of carbide and inclusion contents on the fatigue properties of high speed steels and tool steels

Abstract In this study the influence of carbide distribution, inclusion contents and the surface machining process on the fatigue properties of tool steels and high speed steels were investigated. Four different steels intended for cold work applications were included, of which three were powder metallurgically processed and one was conventionally ingot cast. One of the powder metallurgy steels was studied in four surface conditions namely hard turned, ground, polished and shot peened. Fatigue testing was performed on hour glass shaped specimens with a load ratio of R=0.05. The staircase method was used to determine the fatigue strength corresponding to a life of two million cycles. The causes of fatigue failures were determined from fracture surfaces. For the polished specimens of all steel grades it was found that internal inclusions and carbides caused the failures. Crack initiating internal carbides were found more frequently for one highly alloyed powder metallurgy steel and for the conventionally cast steel. For the other polished series, internal inclusions were controlling the fatigue failures. Surface crack initiation was encountered on the ground specimens as well as on the specimens that were hard-turned and subsequently shot peened as a final machining process. Fracture mechanics was used to describe the relation between inclusion and carbide sizes observed on fracture surfaces and the fatigue strength. A model was developed to predict the fatigue strength of the specimens. It was shown that the model managed to determine the fatigue life limiting factors, i.e. internal inclusions in “PM23” and ASP2014 and internal inclusions and internal carbides in VANADIS10 and M2. However, the predicted fatigue limits overall were lower for all steels than what was observed in practical experiments, hence the model was conservative.

Effect of material, heat treatment, grinding and shot peening on contact fatigue life of carburised steels

Contact fatigue resistance has been tested in a roller to roller rig. The test simulates the conditions in a gearbox with finite slip and with impurities in the lubricant. Two case hardened alloys were included in the study, different sequences of carburising and grinding and finally additional shot peening. The tested samples were characterised with respect to residual stress, microstructure, hardness and surface roughness. The contact fatigue tests show significant scatter in the amount of damage. The shot peened samples gave the least amount of contact fatigue damage. A slight difference in the amount of damage could be seen when comparing the ground and hardened and hardened and ground samples. No difference in the amount of damage could be seen between the two carburised steels.

An experimental and theoretical study of the fatigue properties of hot-dip-galvanized high-strength sheet steel

Abstract The mechanisms behind the reduction of the fatigue strength of high-strength steel owing to hot-dip galvanizing have been studied. Three coating thicknesses were considered, ranging from 80 to 220 μm. The steel has a proof stress of 680 MPa. Fatigue loading was performed at an amplitude below the fatigue limit of the uncoated steel. The coated steels gave lives of between 5×104 and 8×104 cycles. The thickest coating gave the shortest life. Metallographic observation of material not exposed to fatigue loading showed there was a crack network in the δ- and ζ- layers of the coatings. Some initial cracks penetrated through half or more of the coating depending on the coating thickness. Short cracks were found in the steel at the interface to the coating. The crack pattern after fatigue loading followed the original pattern in the coating. Additional cracks were created in the thickest coating. A higher frequency of cracks inside the steel was observed for the steels with thicker coatings. These cracks started at cracks in the coating. Finite-element calculations for the driving forces of the original cracks showed that the material with the thinnest coating was fatigue loaded close to the threshold for crack propagation. The two materials with thicker coatings had driving forces for their original cracks that exceeded the threshold. These results were in good agreement with the lives observed experimentally.

Fatigue strength of spot welded stainless sheet steels exposed to 3% NaCl solution

Fatigue properties of spot welded stainless sheets steels have been investigated in a 3% NaCl solution and, for comparison, also in air at ambient temperature. Corrosion fatigue tests have been con ...

Fatigue life of pressed steel sheet components

This paper adresses some of the aspects of fatigue of pressed steel sheet components. Experimental work on both small unnotched sheet specimens and a recently developed pressed laboratory specimen are presented for six low carbon sheet steels suitable for automotive applications. Some emphasis is also put on the fatigue life prediction of pressed sheet components using both the local strain approach based on smooth specimen data as well as fatigue crack growth analysis. The results indicate an increasing fatigue strength, both for fully reversed strain controlled smooth specimen testing and fully reversed four point bend testing of the pressed specimen, with increasing tensile strength. It is also shown that local strain approach life predictions are overly conservative but may be improved by either accounting for prestraining of the sheet material or by considering fatigue crack growth.

Low cycle constant amplitude fully reversed strain controlled testing of low carbon and stainless sheet steels for simulation of straightening operations

Abstract Ten sheet steel grades have been tested in fully reversed constant strain amplitude loading. Seven low carbon steels with proof stresses from 240 MPa to 1500 MPa were included together with one austenitic stainless steel and two duplex stainless steels. For each steel grade, tests at three different strain levels have been performed and hysteresis loops were recorded during twenty subsequent cycles. The development of cyclic hardening and softening was illustrated both as a function of strain amplitude and as a function of cycle number. Different constitutive models were evaluated to describe the cyclic flow data. It was found that the best fit was obtained with a constitutive model of the type σ a =σ 0 −C· e −k·e pl,a +σ x . σa and epl,a are stress amplitude and plastic strain amplitude respectively in the hysteresis loops. σ0, C and k are material constants. The variable σx is a function of imposed strain amplitude and cycle number.