A Novel Calorimetric Method for Measurement of AC ... - IEEE Xplore

IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY, VOL. 24, NO. 5, OCTOBER 2014

9002104

A Novel Calorimetric Method for Measurement of AC Losses of HTS Tapes by Optical Fiber Bragg Grating Jing Shu Dai, Yin Shun Wang, Wei Jie Zhao, Li Meng Xia, and Di Sun

Abstract—This paper presents an innovative calorimetric method for measurements of ac losses of high-temperature superconducting (HTS) tapes by using the effect of optical fiber Bragg grating (FBG) wavelength shift resulting from temperature change caused by ac losses, where it is adopted as a temperature sensor. The HTS specimen on which the FBG is attached on are adiabatically processed before measuring. Compared with conventional electric and calorimetric methods, FBG temperature sensor has the advantage of rapid response and anti-electromagnetic interference due to its thin geometrical sizes and optical characteristics; it is thus very suitable to ac loss measurement of HTS applications in much more complicated electromagnetic circumstance. By measuring the effect of temperature on wavelength shift in low temperature range, it is indicated that FBG has good-enough repeatability of dependence of wavelength shift on temperature measurement around liquid nitrogen temperature of 77 K. After calibration, ac loss of BSCCO HTS tape in self-field is measured by the FBG sensor, and the measured loss is also measured by conventional electric method using a lock-in amplifier. It is shown that the measured ac loss is in agreement with the results measured by electric method and those based on Norris theory so that the presented calorimetric method is available for measuring the ac loss of superconductors and expected to be generalized to ac loss or thermal measurements of other bulk superconductors. Index Terms—AC loss, calorimetric method, fiber Bragg grating (FBG), high-temperature superconducting (HTS).

I. I NTRODUCTION

T

HERE will be some electromagnetic energy loss in superconductors that carrying alternating current (AC) or in alternating magnetic field. AC loss is one of the important parameters to measure the performance of practical superconductor and becomes an important aspect in the research of superconductivity. There are several ways of measuring AC loss in superconductor, such as magnetic method, electric method and calorimetric method [1]. Magnetic method and electric method respond fast and of high accuracy, however, they cannot be used under complicated electromagnetic circumstance. AC loss in superconductor will cause temperature rise in the superconductor. Manuscript received May 8, 2014; accepted June 12, 2014. Date of publication June 19, 2014; date of current version July 9, 2014. This work was supported in part by the National Natural Science Foundation of China under Grant 51077051 and in part by Beijing Education Commission. The authors are with the State Key Laboratory for Alternate Electric Power System with Renewable Energy Sources, High Voltage and EMC Beijing Area Major Laboratory, North China Electric Power University, Beijing 102206, China (e-mail: [email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/TASC.2014.2332009

There are two kinds of calorimetric method: the method using cryogenic thermometer or thermocouple to measure temperature and the method that measuring evaporation of cryogenic medium. The former applies to the measurement of AC loss in short superconducting sample, the latter is applicable to the AC loss measurement in large sample or superconducting coil [2], and even the superconducting power device [3]. Comparing with electric method, the accuracy and measurement speed of conventional calorimetric method are low [4]. However, calorimetric method can be used for the measurements in any complicated electromagnetic circumstance [5]. Traditional calorimetric method commonly used thermocouple as the thermometer [6]. Thermocouple requires external power supply and cannot be used in some high-voltage situations. FBG temperature sensor can be used to replace thermocouple on that occasion due to its thin geometrical sizes and optical characteristics. In recent years, with the development of the technology of fiber Bragg grating (FBG), FBG has been used more and more widely in power system. Many scientists have begun to study the characteristics of FBG at low temperature [7] and the applications of FBG in superconducting technology [8]. Since FBG temperature sensor has advantage of rapid response and anti-electromagnetic interference [9], the calorimetric method based on FBG is very suitable to AC loss measurement of HTS applications in much more complicated electromagnetic circumstance comparing with conventional electric and calorimetric methods. II. FBG P ROPERTIES AT C RYOGENIC T EMPERATURE A. Principle of Temperature Sensing of FBG In general a fiber Bragg grating can be characterized by its Bragg wavelength (λB ), which is the center wavelength of the light reflected from the grating [10]. The Bragg wavelength is given as λB = 2neff Λ

(1)

where neff is the effective refractive index of the fiber core and Λ is the grating period. When only considering the change of external environment temperature (T ), neff will change due to thermo optic effect while Λ will change due to thermal expansion effect. They both make the FBG wavelength shift. Take the derivative of (1), we get dΛ dneff Λ + 2neff dλB = 2 dT. (2) dT dT

1051-8223 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.

9002104


Fig. 1. Acrylate-coated FBG. (a) Schematic figure of coated FBG. (b) Photo of coated FBG.

Fig. 3.

Fig. 2. Experimental setup for response of FBG on temperature. (a) Schematic arrangement of FBG and cryocooler. (b) Overview of experiment facilities.

Both sides of (2) divide λB , and take in (1), we get 1 dneff dλB 1 dΛ + = dT. λB neff dT Λ dT

(3)

If we define ξ = (1/neff ) dneff /dT as thermo optical coefficient and α = (1/Λ) dΛ/dT as thermal expansion coefficient, and put them into (3), we can obtain ΔλB = αT ΔT

(4)

where αT is temperature sensitivity coefficient and αT = λB (ξ + α), ΔT is variation in temperature. At room temperature, for bare silica FBG doped with germania, α and ξ equal to 0.55 × 10−6 /K and 7.1 × 10−6 /K, respectively. To increase temperature sensitivity of FBG and to protect bare FBG, usually a layer of material of relatively large thermal expansion coefficient was coated outside bare FBG, such as acrylate or polyimide. A kind of acrylate-coated FBG is shown in Fig. 1. The length of the area of grating is 15 mm and the wavelength of FBG is 1540 nm. B. Experiment Study of Response of FBG at Cryogenic Temperature The temperature response of FBG at low temperature was studied by an experiment. Three acrylate-coated FBGs used in the experiment were of the same kind and produced by the same manufacturer. The FBGs and a Cernox resistance thermometer sensor produced by Lake Shore Cryotronics, Inc. were pasted to a copper sheet so that they can measure temperature of the copper sheet at the same time, as shown in Fig. 2. Each of the FBGs was only one side pasted so that the FBGs could measure

Plot of FBG wavelength against temperature.

the temperature without influenced by the strain of the copper sheet, since strain of the copper sheet cannot be transmitted to FBG when one side of FBG is free. During the experiment, temperature of the copper sheet was lowered from 130 K to 6 K (the lowest temperature that the experimental facilities could get) by cryocooler and then raised back. The wavelength of FBG was measured by optical sensing interrogator. Use the wavelength of FBG at 6 K as a reference value to calculate the FBG wavelength shiftment. The measurement result was shown in Fig. 3. As is shown in Fig. 3, FBG has good-enough repeatability of dependence of wavelength shift in temperature range of 60 K through 120 K. At temperature lower than 40–60 K, temperature sensitivity of FBG reduced as temperature dropped. When temperature was lower than 40 K, dependence of FBG wavelength on temperature is very weak. This makes the applications of FBG at liquid helium temperature become limited. III. E XPERIMENT OF AC L OSS M EASUREMENT A. Experiment Techniques To measure self-field AC loss of a BSCCO HTS tape, an experiment device was designed and fabricated, as shown in Fig. 4. A copper box without cover was put into a foam chamber to reduce heat leakage. Firstly, pour liquid nitrogen into the foam chamber and make sure that the liquid level was not higher than the copper box. Each end of the HTS tape was welded onto a connecting copper block, the end of which was conical. The tape and connecting copper blocks were placed in the copper box. Then, Pour liquid nitrogen into the copper box to a marked height that was neither higher than the HTS tape nor lower than the bottom of the connecting copper blocks. Current leads were connected to the connecting copper block and immersed into liquid nitrogen in the foam chamber. Voltage leads were connected to the HTS tape. Three optical FBGs were one side pasted on the surface of the tape. A small foam plate was placed above the tape and the lid of the foam chamber was covered. The liquid nitrogen could make sure that experiment was performed under a circumstance of low temperature. The tape was surrounded by nitrogen of poor heat conducting properties, which was produced by the volatilization of liquid nitrogen. Along with the heat preservation function of foam, the tape would have very little heat exchange with the environment. It should be noted that liquid nitrogen in the foam chamber and

DAI et al.: NOVEL CALORIMETRIC METHOD FOR MEASURING AC LOSSES OF HTS TAPES BY OPTICAL FBG

9002104

Fig. 6. Calibration curve of energy loss against FBG wavelength in perpendicular DC magnetic field.

Fig. 6 presents the calibrating relations of loss with wavelength shift for three optical FBGs, which indicates that the loss sensitivity of the presented calorimetric method by FBG reaches to 10−3 W/m. C. AC Loss Measurement Fig. 4. Experimental arrangements. (a) Schematic experiments. (b) Photo of experimental arrangements.

Fig. 5.

Schematic arrangements in calibration.

the copper box should be replenished to the marked heights after each measurement, so that the experiment environment and temperature keeps same during testing. In this experiment, the temperature of experimental environment was 81 K. The BSCCO HTS tape specimen with cross section of 4.8 × 0.3 mm2 used in the experiment was manufactured by AMSC. By measuring, the critical current in self-field was 115 A and 100 A at 77 K and the experiment temperature 81 K, respectively. The two values are agreement with the published data [11]. B. Calibration Between Loss and FBG Response in DC Magnetic Field Dependences of the heat generated on wavelength shift can be obtained by supplying a DC current. A DC magnetic field with 0.4 T produced by NdFeB magnets was perpendicularly applied to sample in order to ensure that the sample quench because the irreversible field of Bi2223 superconductor is less than 0.25 T, as shown in Fig. 5. Thus, a voltage in sample creates when it transports a DC current lower than the critical current in self-field. The DC current and voltage, that is the loss, are measured and then the loss is calculated by taking the product of both DC voltage and current. At the same time, temperature rise resulting from the loss was measured by FBG. Then a calibration curve between loss and wavelength shift of FBG response was finished.

AC loss measuring experiment was performed in self-field by using both of calorimetric and electric methods. The main experiment instrument of electric method was lock-in amplifier (LIA) that measuring the voltage generated along the HTS sample in-phase with the current. The LIA method requires very accurate phase setting and control that can become difficult particularly for higher critical current tapes. During the experiment, the voltage and temperature would change as the AC current that passed along the HTS tape changed. The AC loss per unit length (P ) can be obtained by measuring FBG wavelength shiftment and referring to the calibration curve, as shown Fig. 6. It can be also measured by LIA technique and calculated by P =

Irms Vrms l

(5)

where Irms refers to root mean square of the transport current, Vrms denotes the root mean square value of voltage with conventional four-probe technique [12], l is the length between the voltage leads. IV. R ESULTS AND D ISCUSSION Generally, self-field AC losses can be calculated by Norris equations based on Bean Critical State model (CSM). As for geometrical structure of a thin strip, the self-field AC loss is given by [13] f μ0 Ic2 (1−iac ) ln(1−iac ) (6) Psf = +(1+iac ) ln(1+iac )−i2ac iac < 1 π 2 ln 2−1 iac ≥ 1. If the cross section of superconductor is circular, the self-field AC loss can be determined by [13] f μ0 Ic2 (1−iac ) ln(1−iac )+(2−iac ) iac iac < 1 2 Psf = (7) 1 iac ≥ 1 π 2

9002104


V. C ONCLUSION

Fig. 7. Self-field AC losses of HTS tape at frequency 120 Hz.

Fig. 8. Self-field AC losses of HTS tape at frequency 180 Hz.

where f is the frequency of the AC current, Ic is the DC critical current, Im is the peak AC current, iac is the ratio of the peak AC current to the DC critical current, namely iac = Im /Ic . The experiment measuring AC loss was preformed using experimental facilities described above. In order to avoid the interference of power frequency 50 Hz, the experiment was performed of frequency 120 Hz and 180 Hz. Fig. 7 shows the losses due to AC transport current of frequency 120 Hz and Fig. 8 shows the losses of frequency 180 Hz. Data are shown for both the calorimetric method and electric method. We also show the calculating results of Norris (6) and (7) in Figs. 7 and 8 as comparisons with the measurement results. As we can see from Figs. 7 and 8, the measurement results of the presented calorimetric method and conventional electric method agree well. The measurement results conform to the Norris theory as the measured losses are mainly larger than the calculating results of Norris strip equation and smaller than those of Norris rod equation, which indicates the validity of the calorimetric measurement technique [14]. Therefore, it is concluded that the presented calorimetric method is available for measuring the AC loss of superconductors. Additionally, it should be noted that the influence of strain on FBG must be eliminated during measuring temperature since FBG is also sensitive to strain.

This paper presents an innovative calorimetric method for measurements of AC losses of HTS tapes based on FBG technique. The response of FBG in low temperature range was measured. It is indicated by the measurement result that FBG has good repeatability on temperature measurement around liquid nitrogen temperature of 77 K. After calibration between energy loss and FBG response, AC loss of BSCCO HTS tape in selffield is measured by the FBG sensor and conventional electric method using LIA, respectively. The measurement results of the two methods agree well and are in the calculating results range of Norris theory with rod and strip models. Therefore, the presented calorimetric method based on FBG technique is available for measuring the AC loss of superconductors. Comparing with conventional electric and calorimetric methods, FBG temperature sensor has advantage of rapid response and anti-electromagnetic interference due to its thin geometrical sizes and optical characteristics. The calorimetric method based on FBG is expected to be generalized to AC loss measurements in sophistical electromagnetic environment such as the HTS tape transport AC current and simultaneously exposed to AC magnetic field with same phase or different phase. Also, the method can be applied to AC loss measurements or thermal measurements of other bulk superconductors. R EFERENCES [1] Y. S. Wang, The Basis of Superconducting Power Technology, 1st ed. Beijing, China: China Press, 2011, pp. 127–143, 170–179. [2] J. P. Murphy et al., “Experiment setup for calorimetric measurements of losses in HTS coils due to AC current and external magnetic fields,” IEEE Trans. Appl. Supercond., vol. 23, no. 3, pp. 1–5, Jun. 2013. [3] Z. Janu et al., “Experimental setup for precise measurement of losses in high-temperature superconducting transformer,” Cryogenics, vol. 46, no. 10, pp. 759–761, Oct. 2006. [4] Y. J. Tang, L. Ren, and J. Shi, The Basis of Superconducting Power, 1st ed. Beijing, China: China Press, 2012, pp. 193–194. [5] T. Hardono, C. Cook, and J. X. Jin, “Measurements of ac loss in HTSC wires exposed to an alternating field using calorimetric methods,” IEEE Trans. Appl. Supercond., vol. 9, no. 2, pp. 813–816, Jun. 1999. [6] S. P. Ashworth and M. Suenaga, “Experimental determination of the losses produced by the interaction of AC magnetic fields and transport currents in HTS tapes,” Phys. C, Supercond., vol. 329, no. 3, pp. 149–159, Feb. 2000. [7] R. RajiniKumar et al., “Fiber Bragg gratings for sensing temperature and stress in superconducting coils,” IEEE Trans. Appl. Supercond., vol. 16, no. 2, pp. 1737–1740, Jun. 2006. [8] Q. Wang et al., “Fiber Bragg gratings for strain sensing in high temperature superconducting magnet,” IEEE Trans. Appl. Supercond., vol. 17, no. 2, pp. 2377–2380, Jun. 2007. [9] R. Rajini-Kumar, M. Suesser, K. G. Narayankhedkar, G. Krieg, and M. D. Atrey, “Performance evaluation of metal-coated fiber Bragg grating sensors for sensing cryogenic temperature,” Cryogenics, vol. 48, no. 3/4, pp. 142–147, Mar./Apr. 2008. [10] F. P. Deng et al., “Study on temperature characteristics of optical fiber Bragg grating under the 77 K environment,” J. Optoelectron. Laser, vol. 18, no. 4, pp. 404–406, 2007. [11] M. A. Young et al., “Measurements of the performance of BSCCO HTS tape under magnetic fields with a cryocooled test rig,” IEEE Trans. Appl. Supercond., vol. 13, no. 2, pp. 2964–2967, Jun. 2003. [12] A. Kiihle, C. Trzeholt, S. Kriiger Olsen, C. Rasmussen, and O. Ternnesen, “Measuring AC-loss in high temperature superconducting cable-conductors using four probe methods,” IEEE Trans. Appl. Supercond., vol. 9, no. 2, pp. 1169–1172, Jun. 1999. [13] F. Gomory et al., “Partitioning of transport AC loss in a superconducting tape into magnetic and resistive components,” IEEE Trans. Appl. Supercond., vol. 11, no. 1, pp. 2967–2970, Mar. 2001. [14] R. Mele et al., “Analysis of AC loss behavior in BSCCO tapes with different core geometries,” IEEE Trans. Appl. Supercond., vol. 7, no. 2, pp. 1351–1354, Jun. 1997.