WO2009098345A1

WO2009098345A1 - Superconducting current limiter integrated in the heat exchanger of a thermoacoustic refrigerator

Info

Publication number: WO2009098345A1
Application number: PCT/ES2009/070015
Authority: WO
Inventors: Manuel RODRÍGUEZ OSORIO; Maurice-Xavier FRANÇOIS; Adrien Betrancourt; Antonio Veira Suarez; Félix VIDAL COSTA
Original assignee: Universidade De Santiago De Compostela
Priority date: 2008-02-04
Filing date: 2009-02-02
Publication date: 2009-08-13
Also published as: ES2334616A1

Abstract

Superconducting current limiter of inductive, resistive or hybrid type, in which the superconducting sample is inserted in the cold exchanger of a thermoacoustic refrigerator that keeps it at cryogenic temperature.

Description

CURRENT SUPERCONDUCTOR LIMITER INTEGRATED IN THE HEAT EXCHANGER OF A THERMOACUSTIC REFRIGERATOR

SECTOR OF THE TECHNIQUE

Superconducting current limiter used to limit fault currents in electrical power production and transport systems.

STATE OF THE TECHNIQUE

Superconducting materials have the interesting property of having a strongly non-linear resistance and dependent on several magnitudes: the temperature, the current that circulates through them and the magnetic field to which they are subjected. These materials are characterized by a zero resistance state that is reached below a certain critical temperature, T _c , so they must be cooled by cryogenic liquids or electric refrigerators, and below a current I _c and a field B _c . In aspects of characterization of the superconductor it is more proper to refer to the critical current density, J _c , which is the I _c per unit area crossed by the current, J _c = I _c / A. Above these critical values, a superconductor undergoes a transition to a dissipative state with non-linear resistance, which can lead to an ohmic resistance regime (such as that of a metal) if dissipation heats the material sufficiently or the current circulating exceeds a second critical value, called J (see, for example, W. Klein et al., J. Low Temp. Phys. 61, 413 (1985); SG Doettinger et al., Phys. Rev. Lett. 73, 1691 ( 1994); ZL Xiao et al., Phys. Rev. B 59, 1481 (1999), and José María Viña Rebolledo, Contribution to the study of electrical transport in thin layers of superconducting cuprates: supercritical currents and paraconductivity, Doctoral Thesis, University of Santiago of Compostela (2003)).

This non-linear behavior of the resistance, function of the temperature, the current and the applied magnetic field, has suggested the use of superconductors in current limiting devices, aimed at reducing the pernicious effects of the high currents and voltages generated during a failure in a power distribution line, either in a production plant or a local network or branch (T. Verhaege and Y. Laumond 1998, Handbook of Applied Superconductivity, 2, ed B. Seeber (Bristol: Institute of Physics Publishing), p. 1691; WT Norris et al, Cryogenics, 37, 657 (1997); W. Paul et al, Physica C, 354, 27 (2001), and P. Tixador, IEEE Trans. Appl. Supercond., 4, 190 (1994)).

Depending on the way in which the superconductor is inserted in the circuit to be protected, two fundamental conceptions of limiter can be distinguished: resistive and inductive. The resistive limiter can simply be a superconducting element (in the form of a bar, thin film, coil, etc.) connected in series with the circuit to be protected. The limiter is designed so that, when the nominal current flows through the line, the critical value, I _c , is not exceeded, so that the resistance it offers is zero, and its presence is not perceived. However, when a fault occurs and the current grows to values greater than I _c , the transition to the dissipative (or even ohmic) state occurs, and a resistance arises that effectively limits the current. Note that, in principle, once the line fault is eliminated, the superconductor returns to its state of zero dissipation. The limiter is, therefore, self sufficient and does not need any additional element that detects the start or disappearance of a fault (the device itself is a detector). On the other hand, its reaction is almost instantaneous, and responds to a failure in times of the order of 1 ms or less.

The inductive limiter has a transformer type design, in which the primary (usually metallic) is connected directly to the circuit to be protected, and the secondary is a ring-shaped superconductor or hollow cylinder. While the nominal current circulates in the circuit, the magnetic flux created by the primary in the magnetic core is canceled by the one generated by the superconductor, so that the effective inductance of the transformer is zero, and the only impedance present is the resistance of the primary coil (and components due to possible leakage inductances). When a fault occurs and I _c is exceeded, the transition of the superconductor causes the flow cancellation to cease, so an inductive and resistive impedance arises (depending on the design dominates one or another component) that limits the current. In a way, this can be seen as an ideal transformer whose secondary goes from being short-circuited to being in open circuit.

The hybrid limiter has characteristics common to the previous two. In this case the secondary of the transformer is a coil, usually metallic, to which a superconducting element (massive or thin film) is connected, indifferently (that is, it is not necessary that it be a ring or cylinder, as it happens in the inductive limiter). The beginning of operation is analogous to that of the inductive. In superconducting state the secondary coil cancels the flow created by the primary. When the superconducting element transits, its non-zero resistance breaks the equality of flows, and two impedance components arise: one resistive and one inductive. In all three cases, due to the high energy dissipated in the form of heat in the current limitation process, it is desirable that the limiter does not act more than a few network cycles, which is the time that the current switches need circuit to detect the fault and open the line (they must wait until a zero of the current is reached). While increasing the temperature of the superconductor can be beneficial in increasing the impedance of the device (which must happen in a similar time to the reaction, i.e. ~ 1 ms), it can also cause a very slow recovery of the device. once the fault was eliminated, since the superconductor would continue to dissipate and offer impedance in the line, something undesirable (see, for example, MR Osorio et al, Physica C, 372-376, 1635 (2002) and MR Osorio et al, Supercond Sci. Technol., 18, 739-746 (2005)). On the other hand, overheating can cause degradation of the superconducting element (including its melting), especially due to the presence of hot spots or weak areas, which are regions of the material with, for example, lower critical temperature, T _c , or lower critical current density, J _c . The so-called low critical temperature superconductors (SBT), whose T _c is lower than the boiling point of liquid nitrogen (77 K), are usually metallic, and the homogeneity is quite good. In addition, they have a high thermal conductivity (~ 100 W / mK), so that the heat generated in the transition can be easily evacuated to the cooling medium. These properties make them little sensitive to the problem of hot spots. However, critical high temperature superconductors (SAT) are ceramic and much more heterogeneous. On the one hand in regards to stoichiometry, a dominant phase may coexist with non-negligible proportions of precursors or intermediate products. On the other, they have a less uniform structure than a metal. Thus, granular SATs are formed by a multitude of small grains linked together, so that the transition to the dissipative state first affects the intergranular borders and then the interior of the grain. In addition, its thermal conductivity is much reduced (~ 1 W / mK), so if there is a weak zone and it overheats, its expansion would be slow and local damage could occur to the material that would worsen the performance of the limiter. The cooling of superconducting devices requires a system for the production of cryogenic liquids (generally it is helium or nitrogen) and a constant renewal of the cooling fluid. To achieve greater independence, it has been proposed to use compressors of mechanical type, capable of obtaining low temperatures using a thermal exchange gas (see, for example, OP Anashkin et al, Cryogenics, 42, 295 (2002) and M. Steurer et al, IEEE Trans. Appl. Supercond., 10, 840 (2000)). However, mechanical devices have a relatively high maintenance, fail due to the presence of moving parts that wear out and frequently use cooling fluids that are highly harmful to the environment. Therefore, it has been proposed to replace these devices with the so-called "thermoacoustic refrigerators" (G. Swift, "Termoacoustics: A unifying perspective for some engines and refrigerators", New York: Acoustic Society of America, 2001). These devices are based on the use of a longitudinal wave to transport heat from the point to be cooled to another in which thermal energy is emitted outside the machine. These devices are composed of four fundamental parts: thermoacoustic motor, resonator, thermoacoustic refrigerator and acoustic load. Figure 1 shows a diagram of the device and all its components, as well as the temperature profile throughout it. On the y-axis the temperature is given in arbitrary units, and on the x-axis the position along the thermoacoustic machine. Next, the role of each part will be detailed and the principle of operation will be succinctly explained.

The inside of the device is filled with an inert gas such as helium or nitrogen, at high pressure (tens of bars). When certain conditions occur, a wave is generated that resonates in the tube, and which results in a small pressure variation, called "dynamic pressure", which overlaps the static gas pressure. This wave will allow you to do work in the refrigerator and extract the heat from the point you want to cool. The role of each of the components of this device is detailed below:

Thermoacoustic motor (M): it is formed by two heat exchangers and, in between them, a stack of plates known as "stack" (G), in which a strong temperature gradient is created (several hundred degrees Celsius) . The heat is injected through the hot exchanger (H) (by means of Joule dissipation in a resistor, a solar panel, etc.) and the excess is evacuated after reaching the second exchanger (F), which is kept at room temperature thanks to a water flow circuit. The gradient thus established causes the amplification of the natural vibrations of the gas molecules, so that an acoustic wave is generated that resonates within the whole machine. This wave is stationary, that is, the offset between the dynamic pressure and the speed of the wave is 90 °. The frequency is therefore given by the length of the entire device. In general, the wavelength is approximately twice the length of the entire apparatus.

In this way, of the thermal power entering the engine, a part is used to generate the acoustic power (P), and the rest is evacuated through the exchanger at room temperature, or it is lost due to the natural processes of conduction in the walls and radiation. The thermodynamic cycle that allows the wave to be generated is called the "Brayton" cycle. For this to occur, the wave must be stationary, with the pressure and velocity quadrature (ie, at 90 °).

Resonator (RS): it is a tube, usually made of steel, whose length allows the frequency of the wave generated in the motor to be set. The diameter can be fixed or vary in different parts of the tube, depending on the frequency desired.

Thermoacoustic (RF) cooler: this component produces the inverse effect that the wave generated in the motor. The acoustic power is used to extract heat from the cold exchanger (C) (which can be connected to a circuit or cavity that you want to cool) and take it, through a porous medium known as "regenerator" (D) to another exchanger to Ambient temperature (E) which, as in the case of the motor exchanger, allows the removal of excess heat and is cooled by means of a water circuit. The thermodynamic cycle that occurs within the regenerator is called "Ericsson," and requires that the wave be progressive, not stationary (that is, the pressure and velocity must be in phase). Acoustic load (AC): it consists of a constriction, a long and thin tube

(called "inertia") and a cavity of diameter much larger than any of the tubes (which we will call "capacity"). These elements are used to locally modify the gap between pressure and velocity, and thus achieve that the wave is of a progressive type inside the refrigerator. It can be shown that their role is equivalent to that played by an electrical resistor, an inductor and a capacitor in an alternating current circuit.

The acoustic load is connected to the refrigerator through a new heat exchanger (A) and an intermediate tube (B). The exchanger allows to eliminate residual heat produced by the dissipation of energy in the acoustic load (especially in the constriction). The intermediate tube or "buffer tube" insulates the refrigerator from this exchanger, and may be internally coated with a material that is poorly conductive to heat. The opening of the hole that plays the role of resistance is given by a flow coefficient usually used by valve manufacturers in the US The smaller it is, the narrower the constriction.

Until now, the refrigerator's cold exchanger is connected by an isolated circuit with a cavity in which what you want to keep at low temperature is introduced. This can be done with the current limiter. However, the cooling efficiency is diminished, since the machine must cool not only the limiter, but also the complete circuit. In addition, the time required to extract heat from the superconducting element when a fault occurs is clearly greater. The present invention proposes a solution to these drawbacks.

DESCRIPTION OF THE INVENTION

The present invention consists of a current superconductive, resistive, inductive or hybrid type limiter (see an example in Figure 3), characterized by being constituted by a superconducting element (SP, in Figures 2 and 3) that is inserted directly in the cold heat exchanger (C, in Figures 1, 2 and 3) of a thermoacoustic refrigerator. The superconductor may be adhered on the outer surface of the heat exchanger (following its perimeter, as in Figure 2 (b)), or on one of its faces (as in Figure 2 (a), preferably the left side in the scheme of Figure 1), without obstructing the gas passage channels as far as possible (O, in Figures 2 and 3). In the latter case, the other surface (located on the right in Figure 1) may be in good thermal contact with the regenerator (D, in Figure 1). The superconductor adheres to the cold heat exchanger by means of a substance that is a good conductor of heat but behaves at the same time as an electrical insulator. In this way, the heat generated by the dissipation of electric power in the superconducting element is quickly transmitted by conduction to the cold heat exchanger and from there it is transported along the regenerator to the heat exchanger at room temperature (E, in the Figure 1), where it is evacuated by an auxiliary water circuit that, for simplicity, is not described in the Figure 1. This procedure allows an important simplification in the construction of the refrigeration system, since it avoids the need to use an external cavity (within which the limiter would be) connected by an auxiliary circuit to the refrigerator. In addition, since the superconducting element is directly in contact with the cold heat exchanger, the process of eliminating the heat produced during the operation of the limiter is accelerated, and it is facilitated that it is again operational in a time around a second. , as required in electric power distribution systems (V. Meerovich and V. Sokolovsky, Supercond. Sci. Technol., 20, 457 (2007)).

The superconducting element can be of the massive type (either a granular sample or a monodomain), although preferably it is a thin wire or a superconducting film, the thickness of which is typically between 100 nm and lμm, so the heat produced must travel a distance much smaller than in a massive sample, in which the characteristic dimensions are millimeters or centimeters. Because of this, heat can be evacuated much faster. The films are grown on a suitable substrate, generally of strontium or sapphire titanate (SrTiO ₃ and Al ₂ O ₃ , respectively), which have a high thermal conductivity, but are electrical insulators. This last property allows the use of an adherent substance that simply meets the requirement of having a high thermal conductivity. In this way, there is no electrical contact between the heat exchanger metal (usually copper) and the superconducting film and, at the same time, heat travels easily from the film to the substrate and from there to the cold heat exchanger. The latter initially acts as a thermal mass, absorbing the heat produced in the superconducting element during the fault. Subsequently, this heat is transferred to the regenerator (D) by means of the gas that fills the entire thermoacoustic refrigerator, and from there it passes to the heat exchanger at room temperature (E).

EXAMPLE 1

By means of the thermoacoustic device simulation software of the SOCIÉTÉ HEKYOM (Orsay, France), a preliminary calculation of the dimensions of an 80 K refrigerator, capable of extracting about 50 W, as well as the corresponding acoustic load for a power was performed acoustic input of about 500 W at a frequency of 70 Hz. The most important dimension in this example is the diameter of the regenerator (D in Figure 1), since it determines the size of the cold heat exchanger (C in Figures 1 , 2 and 3) and therefore the useful space to insert the superconductor to be refrigerated. Specifically, the regenerator we use has a diameter of 5.6 cm.

At 80 K it is complicated to have an extraction capacity that is greater than a few tens of watts. As for the input power, it is a reasonable value for a thermoacoustic motor, unless an amplifier system is used (D. L. Gardner and G. W. Swift, J. Acoust. Soc. Am., 114, 1905 (2003)). In this example the motor characteristics have not been calculated. It has simply been assumed that the acoustic power is generated by some external element.

Subsequently, and taking this into account, we proceeded to calculate a prototype current limiter that could meet the following requirements:

1- Having a superconducting element of reduced size, such that it can be coupled on the surface of the exchanger without significantly diminishing the removable power. It can be considered that this is achieved as long as the porosity of the exchanger is not greatly reduced.

This means that the superconducting element must be adhered so as to obstruct the smallest possible number of heat exchanger channels, through which the gas that fills the thermoacoustic machine flows. 2- Be useful at medium / high power and offer good performance during failures of varying amplitude.

3- Do not dissipate during a typical fault (about 100 ms) more energy than the refrigerator can evacuate.

An inductive limiter is difficult to implement, since the transformer takes up too much space in the exchanger (it may be viable for a low power limiter, since then the core can be small in size). Therefore, the best options are the resistive limiter and the hybrid (resistive / inductive). In this example, a hybrid prototype will be used, since the resistive limiter has two drawbacks that, although not as important as the one that allowed the inductive limiter to be discarded, are relevant: 1- The only origin of the impedance is the resistance of the superconducting element itself. This may force the use of excessive length for the space available within the heat exchanger.

2- All the power must dissipate in the superconducting element itself, so its temperature can become excessive, and that reduces the ability to recover the limiter in a short time.

An axial view of the hybrid exchanger / limiter assembly is shown in Figure 3. The first is represented as a circular piece (C) with multiple perforations (O), through which the gas must pass during the expansions and rarefactions forced by the wave. The limiter is composed of an external core (N) that surrounds the exchanger, although it does not have to do so, being able to be simply on one side, two windings (in principle metallic), the primary (P) on which the external generator imposes a voltage (V), and the secondary (S) in which a superconducting element (SP) is connected in the form of a meander path (it can have any shape and be, for example, a semicircle). The primary winding is connected to the circuit to be protected. The secondary winding is magnetically coupled with the primary winding and as already said, it is directly connected to the superconducting element, which is the only one inside the cooling device.

In this example, the characteristics of the limiter are as follows:

- Nominal power: 220 V x IO A.

Core: supposed to be made of steel with 3% silicon. It is a typical alloy of power transformers. The saturation field is 1.7 T and the maximum relative permeability is around 5000. The length and section of the magnetic path are 0.6 m and 0.14 x 0.14 m ² , respectively. Windings: they are made of copper, with diameters of 3 and 4 mm for primary and secondary, with the number of turns equal to 60 and 15, respectively. - Superconducting path: it is a 5 mm wide YBCO film,

10 cm long and lμm thick. The substrate is SrTiO ₃ and has 0.5 mm of thickness. The critical temperature is about 90 K, the critical current density, J _c , is 4x10 ⁶ A / cm ² and the normal state resistivity is equal to 43 μΩ-cm at 100 K.

The core section was calculated to prevent the core from entering the region of magnetic saturation by applying the nominal voltage. Therefore, the equation was applied:

Vicacious -

A> ef J 9 πvN _p B _sat

wherein A is the arm section of the core, V _and ficaz the applied voltage v frequency (assumed equal to 50 Hz), N _p the number of turns of the primary and 5 _sat field core saturation.

To study the operation of the limiter, different faults were simulated by reducing the value of a load impedance that is supposed to be connected in series with the circuit. The nominal voltage did not change throughout the process. Figure 4 (a) shows the limited current (left axis, in a continuous line) and the one that would circulate in the circuit if there were no limiter (right axis, in a broken line), for a large nominal amplitude fault. The expected value is approximately IKA, but the current is strongly reduced even below the nominal value (10 A), up to about 3 A, in a time less than 1 ms. Figure 4 (b) shows the instantaneous power dissipated in the superconducting element (left axis, in a continuous line) and the corresponding energy (right axis, in a broken line), obtained by integrating the instantaneous power. Except for an extremely fine initial peak, the dissipated power is rapidly reduced below 800 W. This gives an energy approximately equal to 30 J, which is less than the maximum value of the extraction capacity of the thermoacoustic refrigerator that had been designed. As for the temperature increase in the superconducting path, Figure 4 (c) shows that the critical temperature has been exceeded, but the maximum value is about 130 K. This value is not too high and does not represent any risk of damage to The superconducting sample.

Claims

L- Current superconductive, resistive, inductive or hybrid type limiter, characterized by being constituted by a superconducting element that adheres to the cold heat exchanger of a thermoacoustic refrigerator.

2. Superconducting current limiter according to claim 1, characterized in that the superconducting element can be of the massive type (granular or monodomain), although preferably it is a thin film or a thin superconducting wire.

3. Superconducting current limiter according to claim 1, characterized in that the superconducting element adheres to the surface of the cold heat exchanger by means of a substance that is a good heat conductor.

4. Superconducting current limiter according to claim 1, characterized in that the superconducting element is arranged so as not to prevent the passage of gas through the heat exchanger channels.

5. Superconducting current limiter according to claim 1, characterized in that it is scalable to higher powers when the diameter of the cooling device is increased.