Wear 258 (2005) 402–411 Comparison between weight loss of bends in a pneumatic conveyor and erosion rate obtained in a centrifugal erosion tester for the same materials T. Deng a,∗ , J. Li b , A.R. Chaudhry a , M. Patel c , I. Hutchings d , M.S.A. Bradley a a Wolfson Centre for Bulk Solids Handling Technology, Medway School of Engineering, University of Greenwich, Riverside House, Beresford street, Woolwich, London SE18 6BU, UK b Medway School of Engineering, University of Greenwich, Chatham Maritime, Kent, UK c Centre for Numerical Modelling and Process Analysis, University of Greenwich, Greenwich Maritime, London, UK d Department of Engineering, University of Cambridge, Cambridge, UK Available online 27 September 2004 Abstract In order to find a link between results obtained from a laboratory erosion tester and tests carried out on a pneumatic conveyor, a comparison has been made between weight loss from bends on an industrial-scale pneumatic conveyor and erosion rates obtained in a small centrifugal erosion tester, for the same materials. Identical test conditions have been applied to both experiments so that comparable test results have been obtained. The erosion rate of mild steel commonly used as the wall material of conveyor pipes and pipe bends was determined individually on both test rigs. A relationship between weight loss from the bends and erosion rate determined from the tester has been developed. A discussion based on the results and their applicability to the prediction of wear in pneumatic conveyors concludes the paper. © 2004 Elsevier B.V. All rights reserved. Keywords: Bend wear; Pneumatic conveyor; Centrifugal erosion tester; Erosion 1. Introduction Any attempt to develop a model for the prediction of bend failure in the past has been preoccupied with the need to vali- date a link between the results of laboratory test work and the real system’s performance. This remains a major challenge for the current investigation. The importance of such a link is apparent when the prac- ticalities of real system testing, as an alternative means of acquiring erosion data, are compared with that of laboratory testing. The former is expensive, protracted and very diffi- cult to implement as a repeatable procedure, whilst the latter offers the possibility of an inexpensive, controlled procedure with good prospects of repeatability and fast acquisition of data. ∗ Corresponding author. Tel.: +44 20 8331 8646; fax: +44 20 8331 8647. E-mail address: T.Deng@greenwich.ac.uk (T. Deng). Investigations into the phenomenon of erosion in pneu- matic conveying bends that were carried out by Mills et al. between 1970s and 1980s showed that the most significant variables with respect to erosion damage were conveying ve- locity, particle concentration, particle size, particle shape and bend geometry [1]. Since then, the importance of penetration rate in predicting the failure of pneumatic conveying bends has been realised and has been the subject of numerous re- search activities. In 1983, Shimoda [2] carried out tests on a pneumatic conveyor rig to investigate the behaviour of the flowing sus- pension in the pipe bend, and determined the most prevalent impact angle that occurred during the operation of the con- veyor. A gas-blast erosion tester was used to quantify the ero- sion performance of the material under conditions designed to simulate those that occurred in the pneumatic conveying test facility used. The maximum angle of impact found from the pneumatic conveyor trials was used to interpolate to a 0043-1648/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.wear.2004.02.012