PREDICTION OF BEHAVIOR DURING WIRE DRAWIN G OF HIGH-CARBON WIRE

Citation
M. Jallon et M. Hergesheimer, PREDICTION OF BEHAVIOR DURING WIRE DRAWIN G OF HIGH-CARBON WIRE, Revue de métallurgie, 90(10), 1993, pp. 1303-1309
Citations number
NO
Categorie Soggetti
Metallurgy & Mining
Journal title
Revue de métallurgie
ISSN journal
00351563 → ACNP
Volume
90
Issue
10
Year of publication
1993
Pages
1303 - 1309
Database
ISI
SICI code
Abstract
Iron and steelmaker suppliers of wire rod for drawing need to know and to predict the behaviour of their rod during work hardening. This is useful for developing products in iron and steelmaking and for advisin g customers. To satisfy this need, Unimetal Recherche exploits a wire drawing research station (fig. 1). The results of work done there are recorded in a data bank from which a statistical model of the evolutio n of some mechanical properties during wire drawing was developed for high carbon steels. The wire drawing research station is equipped with 2 instrumented single pass drawing blocks running at industrial speed s. This equipment, contrary to multiple pass machines, makes it possib le to : work on limited amounts of material, ensure the reproducibilit y of testing conditions (fig. 2), take samples at all passes, interven e on the rod between passes: marking, etc. During wire drawing, the fo llowing items are measured and recorded permanently:speed, drawing for ce and wire temperature increases. Samples taken at each pass undergo mechanical testing: YP, TS, E, RA, torsion and bending. The wire rods and the drawn wire undergo metallographic study and, if necessary, are subject to further examination. The results are entered into a data b ase (table I and fig. 3). It is then easy to make comparison (fig. 4). On basis of the above mentioned experimental results, we have develop ed a model making it possible to : estimate the evolution of mechanica l properties during wire drawing of a rod with an untested grade or di ameter; define the wire rod necessary for obtaining a drawn wire havin g given characteristics. This statistical model, established using mul tiple linear regression, was designed for our high carbon steels (C > 0.250 %), for diameters between 5.5 mm and 14 mm and for work hardenin g rate Ln(S-0/S)< 3.