PRZEGLĄD ELEKTROTECHNICZNY (Electrical Review), ISSN 0033-2097, R. 88 NR 8/2012 29 Mariusz WOŹNIAK 1 , Simon C. HOPKINS 1 , Bartłomiej A. GŁOWACKI 1,2 University of Cambridge (1), Institute of Power Engineering (2) Characterisation of a MgB 2 wire using different current pulse shapes in a pulsed magnetic field Abstract. A pulse field - pulse current system for critical current measurements has been modified to allow control of the current pulse shape. Regression analysis of the voltage vs. current behaviour has been used to determine the critical current and n value using a specified electric field criterion from a single measurement cycle, in place of the previous qualitative approach from a series of pulses. The results for a superconducting magnesium diboride wire agree well with conventional DC measurements in a constant magnetic field. Streszczenie. System do pomiarów prądu krytycznego przy użyciu impulsowego prądu i pola magnetycznego został zmodyfikowany, aby umożliwić kontrolę kształtu impulsu prądu. Prąd krytyczny i parametr n określono z przebiegu napięcia od prądu używając analizę regresji i kryterium pola elektrycznego z pojedynczego cyklu pomiarowego w zastępstwie poprzednio używanego jakościowego sposobu opartego na serii impulsów. Wyniki otrzymane dla przewodu z dwuborkiem magnezu dobrze zgadzają się z konwencjonalnymi pomiarami stałym prądem i polem magnetycznym. (Charakteryzacja przewodu MgB 2 w impulsowym polu magnetycznym przy użyciu różnych kształtów impulsu prądu). Keywords: pulse measurements, magnesium diboride, critical current. Słowa kluczowe: pomiary impulsowe, dwuborek magnezu, prąd krytyczny. Introduction The pulse field – pulse current (PFPC) technique is a promising method for the transport critical current I c (B) characterisation of short superconductor samples, allowing high fields and high currents to be achieved economically and without excessive sample heating. However, for some samples discrepancies in measured I c (B) between this and the standard constant field direct current (CFDC) technique arise. The PFPC method often slightly overestimates the CFDC results in higher magnetic fields, as reported recently for an MgB 2 /Cu wire [1]. Larger discrepancies, typically an underestimate of I c at low fields, can occur in some conductor architectures (e.g. those with a large number of filaments) due to flux jumps arising from the non-uniform current distribution resulting from the transient magnetic self-field [2]. A model for the intrinsic stability of a multifilamentary Nb-Ti/Cu-Mn/Cu wire was recently shown to describe quite well the PFPC measurement [3]. The PFPC technique may also underestimate I c (B) due to difficulties in current transfer from the current leads to the superconducting region through the resistive matrix of short samples, as has been shown by finite element analysis [4]. In all reported measurements [1-4] the critical current I c (B) was determined by comparing qualitatively the shape of the voltage vs. time response of the sample from a series of approximately sinusoidal current pulses of increasing amplitude, each delivered during a plateau of magnetic field. The qualitative nature of this analysis and the relatively high noise level for high sampling rate data acquisition meant that a low and strictly defined electric field criterion could not be applied. The near-sinusoidal current pulse shape also meant that the rate of change of current, and hence the induced voltage, was time dependent, further complicating analysis. For the measurements reported here, the capacitive discharge current source used previously has been replaced by one capable of delivering accurately any arbitrary shape and length current pulse (Fig.1). In this contribution, regression analysis is applied to the voltage vs. current response to allow the critical current to be determined using a 1 µV cm -1 criterion for sinusoidal and linear current pulse shapes, and the differences between them and CFDC measurements are discussed. Experimental methods A 0.58 mm diameter wire with the MgB 2 core occupying 47 % of its cross-sectional area was manufactured using the powder-in-tube (PIT) method with an oxide dispersion Fig.1 Temperature profiles of the heat treatments of the wire. Insert shows an optical microscope micrograph of the wire cross-section. strengthened copper (Glidcop™) sheath (Fig.1). The starting core composition was Mg+2B+0.09Cu, i.e. with copper powder additions to minimise the diffusion of copper from the Glidcop sheath and thus maximising the fraction of MgB 2 in the core [5,6]. A 7 cm long wire sample was heat treated with a 20 °C min -1 heating rate at 700 °C for 5 min in a flowing 95 % Ar + 5 % H protective atmosphere (Fig.1). Sample 18 mm in length were measured perpendicular to the magnetic field and in liquid helium by the CFDC technique in a Bitter magnet (type Bitter 100, BM1 [7]) in the ILHMFLT (Wrocław, Poland), with a voltage tap spacing of 6 mm and a current ramp rate up to 2 A s -1 (Fig.2 a)). Fig. 2 Schematic diagrams showing the mounting of wire samples for: a) constant field direct current (CFDC) and b) pulse field pulse current (PFPC) measurements. In fig. b), the coil below the sample is used to measure the voltage induced by the pulsed magnetic field, so that this contribution to the sample voltage can be cancelled. Temperature sensors and magnetic field coils are not shown to scale.