Metal working – Method of mechanical manufacture – Electrical device making
Patent
1993-01-05
1994-12-06
Arbes, Carl J.
Metal working
Method of mechanical manufacture
Electrical device making
505928, 505929, 505930, H01L 3924
Patent
active
053698738
DESCRIPTION:
BRIEF SUMMARY
This invention relates to a method of fabricating an elongated artefact wherein the artefact comprises a matrix containing spaced parallel filaments along the direction of elongation.
An example of such an artefact is a superconducting wire. The superconducting component should not be thick normal to the current-carrying direction, i.e. it should be merely a filament, otherwise magnetic fields will set up wasteful eddy currents in the component. At the same time, a mere filament would be able to carry only a small current, and therefore a superconducting wire conventionally consists of many parallel non-touching filaments of superconductor embedded in a matrix which is conveniently an ohmic conductor such as bronze or copper.
The theoretical potential of for example A15 superconductors such as Nb.sub.3 Sn has been known since 1960, but due mainly to their brittleness, in thirty years no ideal way of mass-producing them into wires has been found. Contributing to the difficulty is the requirement for the wires to include a continuous phase of pure copper, to act as a normal electrical conductor, heat sink and mechanical support in case the Nb.sub.3 Sn is accidentally warmed above its superconducting range.
Most conventional ways rely on forming some precursor of the superconductor to the final required shape, then converting the precursor. For example, in the so called bronze route, rods of pure niobium are drawn down in a tin bronze to the extent that a fine wire is produced with filaments of niobium embedded in it. This precursor is then heated such that the niobium filaments are largely converted to niobium tin by reaction with the tin in the bronze. The main disadvantage of this route is that if there is more than 13% tin in the bronze it becomes progressively brittle during drawing until It finally breaks. This means that the mean current density in the final conductor is much reduced by the large volume of bronze required. (For clarity hereafter, the niobium is referred to as "filaments" even when thick enough, as at the outset, to count as "rods".)
The so called internal tin route attempts to avoid the requirement for this large volume of bronze by including the tin separately in the precursor in the form of pillars which are usually more than two orders of magnitude larger than the niobium filaments. There can be problems with drawing down even such precursors (the tin melts) and, as disclosed in UK Patent Application GB 2201830A, this can be partly mitigated by using aluminium in place of tin; at a later stage in the method, when the cross-section of the composite has been substantially reduced by drawing down or extrusion, the aluminium is removed from the composite and replaced with tin.
The composite (conventionally of circular cross-section) is extruded or drawn down, then an array of extruded composites is bundled together and further extruded, and so on, with as many of these stages as necessary. To avoid rupture of niobium filaments during the first extrusion, a relatively stout outer layer of copper is often left around the filaments. It will be seen that this leads to opposing design considerations. At each of these stages of bundling, these stout layers of copper or subsequent copper extrusion cans become part of the volume of the final conductor. Since copper competes with niobium for tin, this wastefully increases the amount of tin which must be provided and also increases the volume proportion of non-superconducting material. To minimise this effect, the number of extrusion/bundling stages can be reduced. This entails cramming many niobium filaments into each single starting composite while, for reasons of manufacturing practicability, the tin remains present in one rather thick pillar per composite. On heat treatment to react the niobium filaments with the tin, exchange of copper and tin between the regions thereof occurs via relatively long tortuous diffusion paths through the stack of filaments and predisposes towards the formation of Kirkendahl voids caused by the different rates of
REFERENCES:
patent: 3925882 (1975-12-01), Sannbrook
patent: 4055887 (1977-11-01), Meyer
patent: 4084989 (1978-04-01), Meyer
patent: 4224735 (1980-09-01), Young et al.
patent: 4490439 (1984-12-01), Stekly et al.
patent: 4531982 (1985-07-01), Dubots
patent: 4723355 (1988-02-01), Both et al.
patent: 4863804 (1989-09-01), Whitlow et al.
patent: 5088183 (1992-02-01), Kanithi
patent: 5223348 (1993-06-01), Wong et al.
Proceeding of the Seventh Symposium on Engineering of Fusion Research, Knoxville, Tenn., 25th 28th Oct. 1977, pp. 1278-1281, IEEE, New York, US; W. A. Fietz: "Conductors for Tokamak toroidal field coils" * p. 1280, right-hand column; p. 1281, FIG. 7*.
IEEE Transactions on Magnetics, vol. Mag-21, No. 2, Mar. 1985, pp. 157-160, IEEE, New York, US; Y. Takahashi et al.: "Development of 12 T-10 kA A1-stabilized Nb3Sn conductor for TMC-II" p. 157, full page.
Proceedings of the IEEE, vol. 77, No. 8, Aug. 1989, pp. 1110-1123, IEEE, New York, US; E. Gregory et al.: "Conventional Wire and cable Technology" * pp. 1116-1120, full pages*.
Evetts Jan E.
Farmer Francis J. V.
Hawksley Thomas J.
Walters Colin R.
Arbes Carl J.
British Technology Group Ltd.
LandOfFree
Method of fabricating an elongated artefact does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method of fabricating an elongated artefact, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method of fabricating an elongated artefact will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-206080