The effect of microstructure and composition on the rolling contact fatigue behaviour of cast bainitic steels

• Published in: Wear Vol. 263; no. 1; pp. 756 - 765
• Main Authors: Green, M.R., Rainforth, W.M., Frolish, M.F., Beynon, J.H.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Lausanne Elsevier B.V 10.09.2007; Amsterdam Elsevier Science; New York, NY
• Subjects: Applied sciences; Back-up roll; Bainite; Exact sciences and technology; Fracture mechanics (crack, fatigue, damage...); Friction, wear, lubrication; Fundamental areas of phenomenology (including applications); Machine components; Mechanical engineering. Machine design; Physics; Rolling contact fatigue; Solid mechanics; Structural and continuum mechanics; Rolling contact fatigue; Back-up roll; Bainite; Roll contact fatigue; Heat treatment; Fretting fatigue; Rolling contact; Durability; Composition effect; Mechanical contact; Fatigue limit; Cast steel; Experimental study; Modeling; Cast metal; Bainitic steel; Rolling mill roll; Hot rolling; Granular material; Austenite; Contact pressure; Microstructure; Tribology
• ISSN: 0043-1648; 1873-2577
• DOI: 10.1016/j.wear.2007.01.070

The operating life of back-up rolls used in hot rolling mills is limited by rolling contact fatigue (RCF), and by wear. Bainitic steels are used as they offer the appropriate balance of resistance to RCF and wear. Recently, Frolish et al. have proposed a model for the RCF of back-up rolls, which includes a microstructural parameter σ ′ y d ¯ that must be optimised, where σ ′ y is the yield stress of the material and d ¯ is the sheaf width within the bainite. The present work sought to verify the model and thereby optimise the microstructure for this application through testing of bainitic steels with a wide range of microstructure, generated through changes in composition (a base composition of nominal content 0.4% C, 5% Cr, 0.85% Mo was cast, with V contents systematically varied up to 0.35%) and heat treatment. RCF testing was undertaken using a maximum Hertzian contact pressure of 1500 MPa. Samples without V exhibited limited crack initiation and no propagation. All V containing samples failed. No effect of prior austenite grain size was found, in line with the model of Frolish et al. A granular bainitic structure promoted extended time to initiation but rapid subsequent crack propagation and failure, while the lower bainitic structures exhibited the reverse (early initiation, but slow propagation). The effect of bainite dimensions on crack behaviour are discussed.