HAL Id: jpa-00223732
https://hal.archives-ouvertes.fr/jpa-00223732
Submitted on 1 Jan 1984
HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.
EXPERIENCE IN STANDARDIZING SUPERCONDUCTOR MEASUREMENTS
A. Clark, L. Goodrich, F. Fickett
To cite this version:
A. Clark, L. Goodrich, F. Fickett. EXPERIENCE IN STANDARDIZING SUPERCONDUC- TOR MEASUREMENTS. Journal de Physique Colloques, 1984, 45 (C1), pp.C1-379-C1-382.
�10.1051/jphyscol:1984176�. �jpa-00223732�
JOURNAL DE PHYSIQUE
Colloque Cl, supplément au n° 1, Tome 45, janvier 1984 page Cl-379
EXPERIENCE IN STANDARDIZING SUPERCONDUCTOR MEASUREMENTS
A.F. Clark, L.F. Goodrich and F.R. Fickett
National Bureau of Standards, Boulder, Colorado, U.S.A.
Résumé - La recherche menant aux mesures étalon caractérisant les supracon- ducteurs pratiques est décrite. Une attention particulière est portée à la mesure du courant critique.
Abstract - The research leading to standard measurement techniques for characterizing practical superconductors is described. Special attention is given to measuring critical current.
I - INTRODUCTION
Several years ago, the U.S. National Bureau of Standards (NBS) with support from the Department of Energy began a program to develop standard measurement methods for characterizing practical superconductors. The ultimate goal of the program was the adoption of voluntary standards by the American Society for Testing and Mater- ials (ASTM). Reported in this paper are results of the research at NBS in support of this effort, in particular, results on critical current measurements that will be useful to magnet designers.
II - RESEARCH
To assure reliable and reproducible measurement techniques, the effects of more than twenty parameters on the measurement of critical current were determined.
Such varied items as the thermal contraction of the sample holder material and ripples or spikes coming from the current supply were studied for their effect on critical current values. Others studied were magnetic field homogeneity and angle with respect to the sample, the sample geometry, the placement and method of attaching current and voltage leads, the voltage criterion used to define the critical current, conductor properties such as the volume percent of stabilizer, and many more /l/. Reported here are data on just a few of these effects having impact on parameters that are of interest to magnet designers and builders, and some that are deviously subtle and are noted as a caution. In addition, for the benefit of our colleagues outside of the U.S., the standards adopted by ASTM are briefly described in hopes that they may achieve wider use if their derivation and content are known.
Critical Current Criteria. One of the first problems addressed was the definition of critical current in a practical superconductor HI. This definition depends strongly on the detection capability of the measurement system and the shape of the voltage current characteristic (V-I curve) as shown in Fig. 1. The voltage increase, V, is approximately proportional to I , where I is the current. The value of n depends upon the material and is generally between 20 and 100. It can also depend on the magnetic field. In addition, the electric field value chosen as the criterion can dictate the geometry of the measurement apparatus (Fig. 2 ) . The final choice of a criterion for critical current was electric field, partially because it can be used in magnet design as a measure of the power loss required.
The point to be made is that the method and criteria used in measurement of criti- cal current can affect the power loss calculation in a magnet operating near the critical current. For this reason, more than one value of electric field criterion is allowed by the ASTM standard.
Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1984176
C1-380 J O U R N A L DE PHYSIQUE
Lap Joint Resistance. The effect of lap joint resistance was also studied / 3 / . The joint resistance is a strong function of both magnetic field and current value relative to the critical current. However, it can be shown that this resistance arises from three contributions, each of which can be characterized intrinsically.
The first is an interface resistance which depends,only on the interface itself and its area, and can be characterized by a value pa which is the slope of a plot of joint resistance versus the inverse area of the joint. A typical value for a thin Pb-Sn eutectic solder layer between bulk copper contacts is p i = 4.5 nil cm2. The second is the resistance of the copper layer between the joint and the supercon- ducting filaments. As seen in Fig. 3, this layer is also subject to the magneto- resistance typical of copper. The third is the complex resistance of the composite region as the current distributes into the filaments. This component has a magneto- resistance and depends strongly on the copper-superconductor interface. The sum of the last two components manifests itself as a current transfer resistance discussed further in the next section. The total lap joint resistance will result in a power loss which can be calculated. It can also result in erroneous measurements of critical current characterized by irreversible V-I curves.
Current Transfer. As the current enters a superconductor or gets redistributed among the superconducting filaments when the conductor enters a higher magnetic field region, the current must pass through filament-stabilizer interfaces and some of the bulk stabilizer as well. The voltages generated by this current transfer can significantly affect a critical current measurement /I/ and can cause a pover loss as well. Figure 4 shows the electric field versus distance from a contact joint for three different joint areas. The joint area dramatically affects the size of this voltage, and the length necessary to reduce these voltages depends strongly on the resistivity of the matrix material between the filaments 141.
conductors this current transfer length can be as much as 30 times the For wire
NbaSn
iameter.A more subtle but very interesting effect takes place when the current in a multi- filamentary superconductor leaves a low magnetic field region and enters a higher
one / 5 / . In this case, current transfer voltages can be generated deep within a
loo 1 1 1 1
,
1 1 1 1,
1 1Nb,Sn +3
Short
-
00
-
-
-
-
-.
>
-
-
-
50 I00 I50 200
1,
*
Fig. 1. Typical V-I characteristic for Fig. 2. Representative I-V character- a commercial superconductor (multifila- istics (multifilamentary Nb Sn at 7 T)
mentary Nb Sn at 7 T) at different 3
for different sample holders.
voltage de?ection levels.
magnet structure. An experiment devised to determine this behavior showed that only when the current first enters or last leaves the high field region does it generate significant transfer voltages. This indicates that the equipotential lines within a superconductor can be very long indeed. The power losses generated by these voltages is probably quite low but the implication for magnet design is quite remarkable; putting each end of a magnet wire through a high field region could homogenize the current distribution in the wire, and this distribution would remain throughout the entire winding.
There are many other effects on critical current which have been studied at NBS /1/, but they are generally less important to magnet design. Of crucial importance, however, is the effect of strain on critical current. This is the subject of a whole field of study which is well summarized by Ekin 161.
The development of measurement standards for practical superconductor parameters has two primary goals, the adoption of voluntary standards for commodity exchange in the commercial markets and the application of these same standards for intercom- parison in research and development. Based upon the research at NBS and at the four U.S. wire manufacturers, a broad-based committee of ASTM wrote and, after many iterations, adopted a measurement technique as a standard for small superconductor critical current ( ~ 6 0 0 amperes). It does not spell out a specific apparatus but establishes limits for all of the pertinent variables. Similarly, the input on definitions of terms relating to superconductivity from more than 60 research labor- atories around the world was condensed into a set of definitions for use with prac- tical superconductors. These two standards are now published by ASTM as B713-82, Standard Definitions of Terms Relating to Superconductors, and B714-82, Standard Test Method for D-C Critical Current of Composite Superconductors, and are available either from ASTM or the authorn. Any constructive comments on their applicability would be useful in the continuing development of these standards.
Resistance
u- f
Jolnt Interface Reststance
-1
Fig. 3. Plot of the lap-joint resis- tance extrapolated to zero current as a function of magnetic field.
Fig. 4. Electric field due to current transfer as measured at various distances, x, from the current contact. Each curve shows data for a different length current contact. The wire is multifilamentary Nb Sn measured at 4 T and 160 A.
3
Cl-382 JOURNAL DE PHYSIQUE
The next major effort in the standards program is to develop a standard reference material (SRM). This will be a length of very well characterized superconductive NbTi, multifilamentary wire. A production length wire was measured for reproduci- bility in both short and long range homogeneity. That is, the variability in critical current was determined over both small distances, 2-20 cm, and large distances, 0.1-1 km. The short range homogeneity which gives a measure of the point to point variability in critical current is less than 1% as shown in Fig. 5 but in other conductors and over long distances can be much more. These critical current SRM's will be available soon from NBS as a 2 m length of wire with critical current characterized at 2, 4, 6, and 8 T to within 2%.
IV - FUTURE RESEARCH
The program is continuing to study critical current behavior in larger conductors and addressing such problems as self-field, aspect ratio, cabling, and current transfer in superconductors up to 3,000 amperes. For exzmple, Fig. 6 shows the effect of field orientation on the critical current of an aspected conductor.
Other parameters under study are ac losses, critical field, and stability.
-3
1 2 3 4 6
Sample number
I , , I
0
A x
2
-0
-
O = Tap 1 X O=Tap 2
- A = T a p 4 X
X = Tap 5
I , , I ,
Fig. 5. Short range homogeneity in Ic of the five SRM candidate materials
(percentage differences relative to center tap).
ANGLE, d ~ g r a r s
Fig. 6. Dependence of I on angle of field for two rectangular conductors at 7 T.
The zero angle is defines where the field is parallel to the wide face of the conductor.
IV
-
REFERENCES111 GOODRICH, L. F. and FICKETT, F. R., Cryogenics 22 (1982) 225.
;2] CLARK, A. F. and EKIN, J. W., IEEE Trans. Magn. MAG-13 (1977) 38.
: 3 ] GOODRICH, L. F. and EKIN, J. W., IEEE Trans. Magn. MAG-17 (1981) 69.
:4] EKIN, J. W., CLARK, A. F., and HO, J. C., J. Appl. Phys. 49 (1978) 3406, 3410.
: 5 ] GOODRICH, L. F., IEEE Trans. Magn. MAG-19 (1983) 244.
: 6 ] EKIN, J. W., in Materials at Low Temperatures, R. P. Reed and A. F. Clark,
eds. (American Society for Metals, Ohio) 465 (1983).
