A model to predict the solid particle erosion rate of metals and its assessment using heat-treated steels

• Published in: Wear 2001-03, Vol.248 (1), p.162-177
• Main Authors: O’Flynn, D.J, Bingley, M.S, Bradley, M.S.A, Burnett, A.J
• Format: Article
• Language: English
• Subjects: Erosion; Heat-treated steels; Martensite
• ISSN: 0043-1648; 1873-2577
• DOI: 10.1016/S0043-1648(00)00554-8

A new predictive model for the wear rate of metals during solid particle impact erosion is presented. The model proposes that erosion rate is related to the product of toughness ( U T) and uniform strain ( ε U). Predictions for the variation of erosion rate with impact angle are also made. The validity of the model was assessed using an extensive set of new experimental data generated for heat-treated steels. Two steels were heat treated to form a total of 12 different microstructures, each having distinctly different mechanical behaviour. Erosion tests were carried out at a combination of three impact velocities and three angles of particle impingement in a rotating disc accelerator erosion tester. Fine olivine sand was used as the abrasive at one feed rate. Tensile tests were carried out on all the heat-treated steels over a range of temperatures from room temperature to 400°C. The model predictions were not satisfied by mechanical property measurements made at room temperature. However, for each given erosion test condition, a good linear relationship was found between room temperature erosion rate and 1/ U T ε U when mechanical properties were measured at elevated temperatures. The elevated temperature chosen to give the best-fit was between 200 and 300°C depending on the impact velocity. It is believed that the significance of the elevated temperature property measurements is that they account for localised heating occurring at the impacting particle during the high strain/strain-rate deformation typical of erosion. Certain heat-treatments gave a poorer fit to the relationship and explanations for this are proffered. The model was also able to account for changes in erosion rate with impact angle. Suggestions are made for improving the model and to refine its predictive capability.