IEC 61788 13 Edition 2 0 2012 07 INTERNATIONAL STANDARD NORME INTERNATIONALE Superconductivity – Part 13 AC loss measurements – Magnetometer methods for hysteresis loss in superconducting multifilamen[.]
Target uncertainty
The target uncertainty of this method is defined as coefficient of variation (COV; standard deviation divided by the average) The COV shall not exceed 5 %
Important variables and elements affecting the uncertainty of the results are specified as follows Introduction to the uncertainty is given in Annex C.
Uncertainty and uniformity of the applied field
The applied magnetic field system must maintain a relative standard uncertainty of no more than 0.5% Additionally, the magnetic field should exhibit a uniformity of 0.1% throughout the entire volume of the specimen.
VSM calibration
The objective of VSM calibration is to achieve a measurement of the specimen's moment with a relative combined standard uncertainty of no more than 1% Calibration must be conducted with all cryostats and any associated metal components in position, simulating the conditions of an actual measurement.
The magnetometer must be calibrated with a small nickel (Ni) sphere, specifically a 2.383 mm diameter sphere made from high-purity Ni wire, ensuring that the calibration is traceable to the National Institute of Standards and Technology (N.I.S.T., U.S.A.) standard reference material 772a.
The certified magnetic moment value, \( m \), is measured at \( (3.47 \pm 0.01) \, \text{mA m}^2 \) at 298 K in a field \( H \) of 398 kA/m (with \( H = 0.5 \, \text{T} \)) Calibration against this sphere involves applying field and temperature corrections, expressed by the formula \( m = 3.47 [1 + 0.0026 \ln(H/398)][1 - 0.00047(T - 298)] \, (\text{mA m}^2) \), where \( H \) is in kA/m (1 kA/m = 12.56 Oe) and \( T \) is in K For practical purposes, a calibration field of approximately
Temperature
Measurements shall be made at or near 4,2 K, the normal boiling point of liquid helium and the actual temperature of measurement reported to a combined standard uncertainty not exceeding 0,05 K
At temperatures other than 4,2 K, the temperature shall be known with a relative standard uncertainty not exceeding 1,2 %, which corresponds to the above combined standard uncertainty at 4,2 K.
Specimen length
To ensure accurate measurements of magnetization components, it is crucial to address the length dependence of specimens In short samples, the anisotropy of critical current density in both longitudinal and transverse directions can cause a significant "end effect," leading to a length dependence in the quality factor, Q h To mitigate this issue, it is recommended that superconducting components (filaments) have a length-to-diameter ratio exceeding 20 Additionally, the proximity effect in Cu/Nb-Ti multifilamentary composites is only anticipated when the filament spacing, d s, is less than approximately 1 µm.
PE contribution to magnetization will depend on sample length, L, and twist pitch, L p
Under this condition, these lengths will need to be taken into account in the following way when reporting the results:
– for d s < about 1 àm and the filaments are untwisted, Q h shall be measured as function of L and the results extrapolated to zero L;
– for d s < about 1 àm and the filaments are twisted, Q h shall be measured at L > 5 L p
Specimen orientation and demagnetization effects
Loss measurements will be conducted on strand specimens within a transverse magnetic field For fully penetrated fine filaments in multifilamentary Cu/Nb-Ti strands, demagnetization effects are negligible, which also applies to bundles with round, flat, or square cross-sections Nonetheless, it is important to report the specimen configuration for comprehensive results.
Normalization volume
It may be desirable to report hysteretic loss in terms of the superconductor volume To pursue this route, it is necessary to invoke a standard procedure for determining the matrix
(Cu)/superconductor volume ratio (see IEC 61788-5) For the purposes of this standard, these steps are eliminated, and AC loss is to be reported in terms of total composite volume
Volume should be measured with a relative combined standard uncertainty not to exceed
Mode of field cycling or sweeping
The applied field may be changed point-by-point over the field cycle starting and ending at
H max SQUID magnetometry is restricted to this mode of field change, and it is optional for the
VSM to be operated in point-by-point mode The VSM may also be operated semicontinuously, the M-H loop being constructed from 200 or so (M,H) data-pairs
5 The VSM method of measurement
General
For a full description of the application of VSM technique, the paper by Collings et al [1 1) ] is recommended.
VSM measurement principle
The Foner VSM operates by placing a specimen in a uniform magnetic field, which magnetizes it The specimen is then mechanically oscillated near pickup coils, leading to an oscillation in the magnetic field that induces an AC voltage This voltage is detected and converted into a magnetic moment value through electronic circuitry As a "substitution" device, the magnetometer's output requires calibration against a reference While custom-made VSMs are available, commercial versions are increasingly preferred, typically featuring a specimen mounted on a vertical rod that vibrates longitudinally with an amplitude of about 1 mm at a low frequency.
The magnetic field may be supplied by either a horizontally mounted iron-core electromagnet
In a typical experimental setup for Vibrating Sample Magnetometry (VSM), either an electromagnet (EM) or a vertically mounted superconducting solenoid (SCS) is used, influencing the vibration direction of the sample to be either perpendicular or parallel to the magnetic field The pickup coils are strategically positioned and paired to effectively cancel out external magnetic noise, ensuring that only the field oscillations generated by the specimen are detected.
The loss is determined from the numerically integrated area of the full M-H loop
The specimen is placed at the "sweet spot," a specific area within the pickup coil where the detected signal remains relatively stable despite minor adjustments in vertical or horizontal positioning To identify this sweet spot, a small calibrating specimen, such as nickel (Ni), is used to explore the specimen space, determining the volume where the response varies by no more than 2% In this context, Z represents the vertical direction, Y denotes the direction along the magnet-pole axis, and X indicates the direction perpendicular to the magnet-pole axis The center of the sweet spot is established through a method called "saddling," which involves maximizing the signal along Z and X while minimizing it along Y.
VSM specimen preparation
The size of the sweet spot in the typical VSM restricts specimen volume to less than about
For VSM measurement of Cu/Nb-Ti multifilamentary composite wires, three specimen configurations are permissible: a short straight specimen, a multiturn coil, and an oval coil The short straight specimen consists of one or more straight strand pieces, up to 1 cm in length, with finely ground flat ends For measuring long lengths of fine wire, a multiturn coil can be used, which should be oval and mounted with its long axis vertically for EM-VSM measurement, while the plane of the coil remains normal to the field direction In contrast, for SCS-VSM measurement, the multiturn coil must be round and positioned with its plane perpendicular to the vibration axis.
1) Numbers in square brackets refer to the Bibliography
To reduce interstrand coupling, it is essential to insulate the strands of the short straight bundle and the multiturn coil using varnish, potting, or other methods of electrical separation Positioned between the short straight sample and the multiturn coil is a helical coil, which, as suggested by Goldfarb et al [4], consists of a single strand wound along screw thread grooves The helix's axis aligns with the field direction, considered transverse to the specimen axis when the pitch angle is below 8° This helical technique allows for the measurement of a relatively long piece of moderately thick strand.
Figure 1 – A typical experimental setup of VSM measurement a) Short sample b) Helical coil c) Multiturn coil
Figure 2 – Three alternative specimen configurations for the VSM measurement
VSM measurement conditions and calibration
Field amplitude
The measuring field amplitude, to be determined by the application, shall be specified (see
Direction of applied field
The applied field will be oriented transversely to the strand axis, ensuring it is perpendicular to the axis of the short straight specimen, normal to the plane of the multiturn coil, and parallel to the axis of the helical coil.
Rate of change of the applied field (sweep rate)
To minimize the impact of coupling contributions, P_c, on AC loss, the sweep rate of the applied field must be kept low However, at extremely low sweep rates, such as during point-by-point measurements, the influence of strong coupling can resurface, leading to the phenomenon of eddy current decay.
Exponential creep must be considered when analyzing measurements at typical VSM sweep rates If detectable coupling is present, the quality factor \( Q_h \) should be extrapolated to zero \( dH/dt \), as it has been established that the measured \( Q \) is linear in \( dH/dt \) For specimens exhibiting a low \( n \) value in the voltage-current relationship at elevated temperatures, \( Q_h \) will also be determined through a similar extrapolation method.
In fine filament composites the measurer shall be alert to the possibility of a proximity effect
The contribution of PE significantly increases the hysteretic loss beyond what is anticipated from individual filaments This enhancement is a legitimate factor in the overall hysteretic loss and must be considered in calculations.
In thick filament composites, it is crucial for the measurer to be vigilant about the potential occurrence of a flux jump, as this can interfere with the measurement of intrinsic magnetization The report must also contain a note regarding the flux jump, as specified in section 6.3 d).
Waveform of the field change
The field sweep rate shall be linear between the end-points ±H max , see 3.11 and 4.7 above.
Specimen size and shape correction
Calibration shall be performed as directed above under 4.2 Furthermore, consideration shall be given to the size and shape of the specimen with respect to those of the calibration sample
The specimen shall be centered on the sweet spot
For specimens smaller than the calibration sample, no size correction need be applied
For specimens exceeding the size of the calibration sample, two size correction methods are permissible: a) creating a replica of the specimen from nickel to serve as a secondary standard; b) mapping the sweet spot to develop a size and shape correction based on the measured response.
Allowance for addendum (background subtraction)
The measurer shall be alert to the possibility that the specimen holder and associated parts
(for example temperature sensor) may make a significant contribution to the loss Whenever this turns out to be the case, a correction shall be applied.
Data point density
In contemporary computer-controlled VSM measurements, users can choose from a wide range of data pairs to construct the M-H loop To accurately capture fine structures, such as those related to PE magnetization features, a high point density is essential Therefore, when conducting point-by-point measurements, the M-H loop should include at least 100 data pairs.
General
The report of the results of the AC loss testing shall include at least the following specifications The reason for any missing information shall be explained.
Initiation of the test
– Name of the laboratory performing the test
– Names of groups or persons requesting the test
– Other details concerning sponsorship of the test
Technical details
a) The superconducting composite strand – details when available
– Manufacturer and strand identification code
– Strand design, for example number of re-stacks
– Cu/superconductor volume ratio within filamentary bundle and overall
– Matrix residual resistance ratio, RRR
– Filament diameter b) The specimen – the strand as prepared for measurement
– Form of the specimen (bundle or coil)
• Dimensions of bundle, number of wires in bundle
– Total length of strand in sample
– Sample mounting – orientation with respect to the applied field c) Test facilities – apparatus and conditions
– Magnetometer calibration procedure and related details
– Uncertainty of field determination and calibration procedure
– Uncertainty of temperature determination and procedure used
– Specify whether point-by-point or continuous applied field change and, in the latter case, the field ramp rates used
– Number of data points taken in constructing the four segments of the M-H loop d) Results – final report and analysis
– The measured hysteretic loss, Q h , per unit volume of strand corrected to 4,2 K if necessary
– State whether proximity effect is noted
– State whether flux jumping is noted
– Discuss the dH/dt dependence of loss and if extrapolation to zero dH/dt was needed to determine the static Q h
– Discuss if creep-effect corrections to the low ramp-rate loss were needed
The SQUID method of measurement
A SQUID, or Superconducting Quantum Interference Device, consists of a superconducting ring interrupted by one (r.f SQUID) or two (d.c SQUID) weak links that significantly reduce superconductivity These devices demonstrate a macroscopically observable quantum interference effect that is closely linked to the magnetic flux within the ring By utilizing appropriate electronic circuitry, this quantum interference can be harnessed for highly precise measurements of magnetic flux Superconducting flux transformers play a crucial role in transferring the total magnetic flux from complex external superconducting pick-up coil configurations into the SQUID sensor's ring For an in-depth exploration of the physics, electronic circuits, and potential error sources associated with SQUID sensors, refer to [5].
In SQUID magnetometers, the magnetic moment of a specimen is determined by the magnetic flux it generates in a pick-up coil, which is accurately measured using a SQUID sensor Proper calibration of the instrument is essential, as it interprets the measured magnetic flux as resulting from a magnetic dipole moment To minimize magnetic flux noise and the background flux from the applied field, the pick-up coil is replaced with a gradiometer system The specimen's magnetic moment is calculated from the SQUID sensor's output voltage based on its position within the gradiometer coils Additionally, periodic movement of the sample enables monitoring of changes in the magnetic moment over extended time periods, helping to mitigate drift effects in the detection electronics.
In commercial SQUID systems, the diameter of the pick-up coil is typically a few centimeters, similar to the distance between the coils To achieve maximum signal amplitude, the movement of the specimen also spans a few centimeters, although this range may be reduced in newer models The magnetic field for specimen magnetization is generated by superconducting magnets aligned parallel to the specimen's movement direction.
Section 5.3 outlines the typical sizes and configurations of specimens If the specimen's dimension perpendicular to the axis of the pick-up coils exceeds 5 mm, a recalibration of the instrument, as detailed in section 4.2, is required to accommodate the specimen geometry specified in section 5.4.5.
(depending on the design of the pick-up coils in the specific instrument)
A.3 Specific SQUID measurement conditions and calibration
All specifications from version 5.4 are applicable, except that SQUID magnetometers require a point-by-point measuring mode, resulting in longer measurement times and reduced point density Due to the slow data acquisition speed of SQUIDs, a complete magnetization loop must include at least 50 data points, and often more are necessary to accurately capture fine structures in the hysteresis loop.
Extension of the standard to the measurement of superconductors in general
The standard magnetization method for assessing low-frequency AC loss in Cu/Nb-Ti superconducting composites has been broadened to encompass the measurement of superconductors in general and at temperatures beyond 4.2 K.
The methods described in IEC 61788-13 are extended to the measurement of unstabilized
Nb-Ti refers to niobium-titanium without copper, or to a copper/niobium-titanium composite where the copper has been eliminated through etching In this scenario, the resulting filamentary bundle can be reinforced with resin impregnation.
The IEC 61788-13 standard extends its measurement methods to various classes of superconductors, including both stabilized and unstabilized forms in round wire This encompasses low-temperature superconductors (LTSC) such as Nb-Ti-Ta and Nb3Sn, processed through techniques like the bronze route, powder-in-tube route, and internal-tin route, as well as Nb3Al Additionally, it includes the intermediate-temperature superconductor MgB2 and high-temperature superconductors (HTSC) like Bi-2212, Bi-2223, and YBCO.
All classes of wire may be measured in the form of a “short straight specimen” Ductile wires may also be measured as a “helical coil” or “multiturn coil”
To ensure accurate measurements, the sample size must be entirely within the sweet spot If a larger sample is necessary, a correction must be applied This correction can be implemented in two ways: first, numerically, by examining the sample space both within and outside the sweet spot using a small standard calibrating sample, such as a small nickel sphere; second, by recalibrating the magnetometer with a standard sample of nickel that matches the shape of the sample being measured.
B.5 Measurement at temperatures other than 4,2 K
Measurements using a vibrating sample magnetometer can be conducted across a broad temperature spectrum, including at 4.2 K, the normal boiling point of liquid helium, below 4.2 K by utilizing a vacuum on liquid helium, and above 4.2 K through a temperature-controlled helium gas stream Additionally, refrigeration methods can involve either liquid helium or a cryocooler.
The VSM's measuring components are positioned outside the cooled area, allowing a machine calibrated at room temperature with a standard Ni sample to conduct measurements across a range of temperatures.
In 1995, international standards organizations, including IEC, agreed to standardize statistical terminology by adopting the term "uncertainty" for all quantitative statistical expressions This decision led to the removal of the quantitative use of "precision" and "accuracy," although these terms can still be used qualitatively The evaluation methods and terminology for uncertainty are now standardized.
ISO/IEC Guide to the Expression of Uncertainty in Measurement (GUM) [1] 2
Each Technical Committee (TC) had the autonomy to determine whether to align existing and future standards with the new unified approach, a decision that poses challenges and can lead to confusion, particularly for those unfamiliar with statistics and the concept of uncertainty During the TC 90 meeting in Kyoto in June 2006, it was resolved to adopt these changes in upcoming standards.
To convert "accuracy" and "precision" values into corresponding "uncertainty" figures, it is essential to understand the source of these numbers The coverage factor of the original value may vary, being 1, 2, 3, or another figure Additionally, when a manufacturer's specification is represented by a rectangular distribution, the conversion factor becomes 1/3.
The appropriate coverage factor is applied when converting the original number to equivalent standard uncertainty This conversion process is not a compliance requirement for TC 90 standards, but it is explained to inform users about the changes made Importantly, converting to uncertainty terminology does not change the user's responsibility to assess their measurement uncertainty to ensure compliance with the standard's criteria.