INTERFACE INSTABILITY IN THE BULK PROCESSING OF 2223 BSCCO POWDERS

Citation
M. Dao et al., INTERFACE INSTABILITY IN THE BULK PROCESSING OF 2223 BSCCO POWDERS, Philosophical magazine. A. Physics of condensed matter. Structure, defects and mechanical properties, 78(4), 1998, pp. 857-877
Citations number
23
Language
INGLESE
art.tipo
Article
Categorie Soggetti
Physics, Applied","Material Science","Physics, Condensed Matter","Metallurgy & Metallurigical Engineering
ISSN journal
1364-2804
Volume
78
Issue
4
Year of publication
1998
Pages
857 - 877
Database
ISI
SICI code
1364-2804(1998)78:4<857:IIITBP>2.0.ZU;2-5
Abstract
Interface instability (or ''sausaging') has been a major problem in th e oxide powder-in-tube tape-rolling process. The fast Fourier transfor m of core instabilities are first performed to study the rolled tapes and to obtain quantitative information on the wavelengths and magnitud es of the nonuniform profiles. The existing experimental observations are also re-examined and summarized. Treating this problem as the bima terial interface instability, both bifurcation analysis and finite-ele ment modelling are applied to study the influences of roll-gap geometr y, packing fill factor and clad material properties on the instability wavelength and magnitude. Good correlations between theoretical resul ts and experimental observations are found. The critical waveleneth/cu rrent core thickness ratio lambda/d(c) is found to be sensitive to th e fill factor only, and insensitive to the reduction/pass ratio, and t he core and clad material properties; that is, the relative tape geome try is the single dominating factor that affect the normalized critica l wavenumber lambda/d(c). Consistent with experimental observations, a smaller reduction/pass ratio, a higher initial core porosity, a high er hardening clad material and a larger core fill factor can reduce th e normalized instability magnitude Delta v/d(c) (i.e. the instability magnitude/current core thickness ratio) at the same tape reduction str ain level. The results suggest that the reason for the much smaller in terface variation magnitude with a small reduction/pass ratio (i.e. 5% per pass) compared with a large reduction/pass ratio (i.e. 25% per pa ss) is not because the small reduction/ pass ratio can eliminate or de lay the interface instability initiation; it is, however, most probabl y caused by the random disruption of the interface by the many rolling steps with critical wavelengths very close to each other between adja cent rolling steps.