Crack Paths 2009

condition, an initial micropitting is commonly observed, but then it stops being the

failure caused by subsurface initiated spalling [7]. Nevertheless, these two damage

mechanisms can have sometimes a synergic effect and correspondent cracks join

together.

As inclusions play a fundamental role in R C Fof hardened steels [8], design criteria

against R C Fcan be proposed which consider them as equivalent early formed cracks

and define the R C Flimit in terms of their propagation threshold, according to the short

cracks theory [9]. The material production process becomes therefore crucial for these

applications and its goodness in this sense can be judged by an analysis of the inclusion

content. It is however quite complex to find a quantitative relationship between

inclusion content and R C Fresistance and in particular to predict which improvement in

terms of R C Flife can be achieved throughout a certain process refinement. This paper

deals with this latter aspect, studying experimentally the R C Fbehaviour of typical steel

for gears, obtained with three slightly different industrial processes. For this aim, some

ring specimens were constructed and subjected to R C Ftests. The damage evolution on

the specimens has been studied and their life determined. A statistical analysis of the

inclusion content has been also carried out for the three processes and a relationship

between R C Flife and maximumexpected inclusion size has been searched for.

R C FTESTS

The tests were carried out on a high performances PC controlled bi-disk machine

equipped with two independent mandrels, one of which translating on linear slides by

means of a servo-hydraulic actuator. The specimens are ring shaped, machined from the

same tubes used for the gears productions, made of hardened and tempered S A E5135

steel. They present an inner diameter of 148 mm,an outer diameter of 175 m mand a

thickness of 10 mm.Three series of specimens were constructed, hereafter named ‘A’,

and therefore

‘B’ and ‘C’, obtained with different facilities and steel-practices

containing three different grades of inclusion, as will be reported later. A counteracting

disk with diameter of 65 mm, made of quenched and tempered 100Cr6 steel having

hardness of 60 HRC, is mounted during the tests against the ring specimens,

constituting a commonreference. The R C F tests were carried out in pure rolling

condition with water lubrication. Three hertz contact pressure levels po was considered,

namely 2000, 2200 and 2400 MPa. The rolling speed of the counteracting disc was set

to 500 r.p.m., to which corresponds a ring-specimen rolling speed of 185 r.p.m.. Each

test was continued up to the occurrence of severe R C Fphenomena. At the test end, the

specimens were cut along transversal sections (perpendicular to the ring axis) and

observed by optical and electron microscope in order to analyze the damage under the

surface. E D Sanalyses of the most significant inclusions were also carried out.

Each specimen presented macro damage phenomena, like spalling, after a cycles

number NF reported in Table 1.

784