Charging and discharging apparatus for a superconducting magnet, use of the charging and discharging apparatus, method of charging a superconducting magnet and method of discharging a superconducting magnet

Abstract

The present disclosure relates to a charge and discharge device for a superconducting magnet, use of the charge and discharge device, a method of charging a superconducting magnet and a method of discharging a superconducting magnet. A charge and discharge device (10) for a superconducting magnet (12) is disclosed, having at least one circuit block (14), each having a first and a second synchronous half-bridge converter (16, 18), wherein for each such converter (16, 18) a control signal generator (48) is assigned for outputting a control signal driving a switch (20) of the respective converter (16, 18) open and closed and outputting another control signal driving another switch (22) of the converter (16, 18) open and closed, wherein the switch (20) is open when the other switch (22) is closed and vice versa, in particular wherein each such converter (16, 18) comprises its own control signal generator (48). The charge and discharge device is simple in design and control and inexpensive.

Description

Technical Field

[0001] The present invention relates to a charging and discharging device for a superconducting magnet, the use of the charging and discharging device, a method for charging a superconducting magnet, and a method for discharging a superconducting magnet. Background Technology

[0002] Devices for charging and discharging superconducting magnets are commonly known. For example, IEEE Transactions on Power Electronics, pp. 1-1, 2020, “Design Optimization of Two-Quadrant High-Current Low-Voltage Power Supply”, E. Coulinge, S. Pittet, and D. Dujic disclose a DC / DC converter for charging and discharging a superconducting magnet. This DC / DC converter includes a full bridge consisting of four metal-oxide-semiconductor field-effect transistors (MOSFETs). The full bridge includes a series connection of first and second MOSFETs, which is connected in parallel with a series connection of third and fourth MOSFETs. A series connection of a resistor, inductor, and capacitor is connected between the connection point between the first and second MOSFETs and the connection point between the third and fourth MOSFETs. The inductor and capacitor form a low-pass filter. The superconducting magnet is connected in parallel with the capacitor to the DC / DC converter. When a voltage source is connected in parallel with the series connection of the MOSFETs, for a positive voltage output, the fourth MOSFET remains on and the third MOSFET remains off, while the first and second MOSFET pairs are driven by complementary pulse-width modulation (PWM) signals. For negative voltage output, the fourth MOSFET is off, the third MOSFET is on, and the first and second MOSFET pairs are driven by a PWM signal.

[0003] Because a full-bridge MOSFET is used, complex control logic is required to operate all four MOSFETs simultaneously and avoid short circuits. This makes full-bridge control complex and therefore expensive. Furthermore, the individual LC filter in the full-bridge topology must filter high voltage swings, which requires large inductors and / or capacitors. This increases the cost of the circuit.

[0004] US 11,693,074 B2 discloses a circuit arrangement for energizing a superconducting magnet in a magnetic resonance imaging apparatus. The circuit includes a first circuit section, a second circuit section, and a control device. The first circuit section is designed to generate a direct current voltage from an alternating current voltage as a DC link voltage, and the second circuit section is designed as a current source fed by the DC link voltage. The second circuit section includes a buck converter controllable by the control device, a transformer switchable by the control device, and a rectifier. A primary current is generated from the DC link voltage via the buck converter. The primary current is fed by a switching device, switched by the control device to the primary side of the transformer, and a secondary current for energizing the magnet is generated via a rectifier connected to the secondary side of the transformer. The buck converter is actuated based on measurements of the primary current and the secondary current.

[0005] US 5,159,541 discloses an asymmetric pulse-width modulation constant-frequency resonant DC / DC converter. The converter includes a bridge chopper component, a resonant circuit component, and a rectifier component. The bridge chopper component converts a DC input into a constant-frequency AC input. The resonant circuit component is connected to the bridge chopper component and includes a series resonant capacitor and a resonant inductor for sending the high-frequency AC to a high-frequency transformer component to generate transformer secondary AC. The rectifier component is connected to the high-frequency transformer component to rectify the transformer secondary AC to produce a stable DC output. The bridge chopper component includes at least two semiconductor switches, each with a snubber capacitor connected across it. The switches are alternately turned on by corresponding gating pulse signals at the constant high frequency applied thereto. The duty cycles of the gating pulse signals and thus the duty cycles of the switches are complementary. The duty cycles of the gating pulse signals can also be adjusted to maintain the stable DC output.

[0006] US 5,245,520 discloses a half-bridge asymmetric buck converter. The converter includes an input bridge transistor driven by a duty cycle D and an additional bridge transistor driven by a duty cycle (1-D). The converter also includes a power transformer, a rectifier, and an LC low-pass filter.

[0007] One object of the present invention is to provide a charging and discharging device that is simple and inexpensive in design and control. Furthermore, the use of this charging and discharging device, a method for charging a superconducting magnet, and a method for discharging a superconducting magnet will be described in detail.

[0008] This is achieved by using the charging and discharging device according to claim 1, the charging and discharging device according to claim 16, the method for charging a superconducting magnet according to claim 17, and the method for discharging a superconducting magnet according to claim 19. Summary of the Invention

[0009] According to the present invention, it is first recognized that a single buck converter or step converter is suitable for charging a superconducting magnet due to its unipolar output voltage, but not for discharging it. This is why the inventors turned to a bipolar topology (i.e., two buck converters). The output of one buck converter is connected to one pole of the superconducting magnet, and the output of the other buck converter is connected to the other pole of the superconducting magnet. Each of the two buck converters has two controllable switches, a common input voltage, and different output voltages due to the different duty cycles applied to the buck converters. In the prior art, one buck converter operates with a duty cycle symmetrical about 50% (central) to the duty cycle of the other buck converter.

[0010] The inventors further recognized that while centrally operating two buck converters may make the converters easier to control, this central operation results in high current ripple in the inductors of the low-pass filters passing through the buck converters. Consequently, the power losses of the inductors lead to high heat dissipation. Therefore, the charging / discharging device will require space-consuming and expensive cooling equipment. Current ripple through the inductors of each converter...

[0011]

[0012] It depends on the converter's duty cycle D and the converter's input voltage V. in The switching frequency F of the converter's switches sw And the filter inductance L.

[0013] The inventors further realized that, in order to reduce current ripple, D*(1-D) and / or V can be reduced. in and increase F sw And / or L. However, the inventors chose to optimize D because in order to reduce V in This will necessitate increasing the conductor cross-section of the printed circuit board (PCB) tracks, and copper is expensive. To increase the switching frequency F... sw More complex transistors will have to be used, which is expensive, and the converter's losses will increase. To increase the filter inductance, larger coils will have to be used, which is both space-consuming and expensive.

[0014] Claim 1 of the patent takes all these factors into account.

[0015] According to the present invention, a charging and discharging device for a superconducting magnet is provided. The device has at least one circuit block, each circuit block having first and second synchronous half-bridge converters. Each converter includes a series connection of a switch and another switch connected between an input point and a base point, wherein the switch and the other switch are connected at a junction point. Each converter also includes a series connection of a filter inductor and a filter capacitor. The filter inductor and the filter capacitor are connected at a junction point. The filter inductor is connected to the junction point, and the filter capacitor is connected to the base point. For each such converter, a control signal generator is provided for outputting control signals to drive the switches to open and close, and for outputting another control signal to drive the other switch to open and close. A duty cycle is applied to the control signals. The other control signal applies the inverse of the duty cycle, such that when the other switch is closed, the switch is open, and vice versa. In particular, each such converter includes its own control signal generator. In each circuit block, the duty cycle of the control signal of the first converter (also referred to as the first duty cycle D1) and the duty cycle of the control signal of the second converter (also referred to as the second duty cycle D2) are different from each other. Therefore, when a common input voltage U is applied between the input point and base point of the two converters... in At that time, the first output voltage U exists between the connection point and the base point of the first converter. out1 The second output voltage U between the connection point and the base point of the second converter out2 They are different. The first duty cycle D1 and the second duty cycle D2 are both at most 25% in case A), or at least 75% in case B).

[0016] The switching of one synchronous half-bridge converter can be controlled independently of the switching of the other synchronous half-bridge converter. Therefore, compared to prior art DC / DC converters that include a full bridge with four MOSFETs, the two independently controlled synchronous half-bridge converters enable simpler and more flexible control. Consequently, the control of the charging and discharging device is easier and less costly. Furthermore, each of the two synchronous half-bridge converters has a low-pass filter. Therefore, compared to prior art DC / DC converters that include a single low-pass filter for the entire full bridge, the low-pass filters of the synchronous half-bridge converters must each filter a smaller voltage swing. Therefore, the inductors and / or capacitors can be smaller. Consequently, the manufacturing cost of the charging and discharging device is lower.

[0017] Because both the first duty cycle D1 and the second duty cycle D2 are at most 25% in case A) or at least 75% in case B), the ripple current through each filter inductor is low. This improves the electromagnetic compatibility (EMC) of the charge / discharge device. Due to the low ripple current, the corresponding filter inductors have low power losses. Losses are reduced by approximately 22% compared to operation with duty cycles in the range of 40% to 60%. Due to the low power losses, heat dissipation from the filter inductors is low. Therefore, cooling equipment can be reduced or completely eliminated. In the latter case, a step of applying thermal paste is eliminated during the production of the charge / discharge device. Furthermore, without cooling equipment, the charge / discharge device is lighter.

[0018] In summary, the present invention provides a space-saving and lightweight charging and discharging device that is also inexpensive and easy to control in design.

[0019] Typically, a common input stage exists, to which the input point of each converter in each circuit block is connected. Correspondingly, a common base stage exists, to which the base point of each converter in each circuit block is connected. The common input voltage U... in Then it exists between the common input stage and the common base stage. The common input voltage U in It is a constant voltage, for example, selected as 12V, or selected as between 6V and 24V. Typically, the base point of each converter in each circuit block is connected to ground.

[0020] The control signal generator may include an analog-to-PWM signal converter. The analog-to-PWM signal converter may be an operational amplifier comparator with complementary outputs. The input to the analog-to-PWM signal converter may be an analog voltage, and the output may be a pulse-width modulated signal.

[0021] Each switch and the other switch are either on or off. In each case, the switch and the other switch can be selected as transistors. The transistor can be a MOSFET, particularly an N-channel enhancement-mode MOSFET, preferably a SiMOSFET. It is also conceivable that the transistor is a gallium nitride field-effect transistor (GaNFET). Each transistor can be driven by a gate driver. MOSFET integrated circuits (ICs) with gate driver MOSFET combinations can be provided. It is also conceivable to use GaNFET ICs with gate driver GaNFET combinations.

[0022] Each filter inductor and filter capacitor combination forms a low-pass filter. This low-pass filter is used to filter switching waveforms.

[0023] The phase of the second duty cycle of the second synchronous half-bridge converter can be shifted by approximately 180 degrees compared to the phase of the first duty cycle of the first synchronous half-bridge converter. This reduces ripple in the input and / or output ripple voltages.

[0024] The first and second synchronous half-bridge converters can be placed together on the same PCB.

[0025] The charging and discharging device of this invention constitutes an asymmetric four-quadrant (bidirectional) high-current converter. The first and second converters are controlled in a non-centralized manner, wherein the average values ​​of D1 and D2 are significantly different from 50%, i.e., below 25% (case A) or above 75% (case B). This device is suitable for charging and discharging simply by appropriately selecting D1 and D2, and particularly by selecting an appropriate sign for their difference. The charging or discharging speed can be set by selecting the absolute value of the difference between D1 and D2. Note that the difference between D1 and D2 can be dynamically changed to adapt to the charging and discharging speed, and the average value of D1 and D2 can also be dynamically adapted (“dynamic common mode”), particularly minimizing losses (by keeping the average values ​​of D1 and D2 as far away from 50% as possible) and enabling charging and discharging to be performed in a particularly safe manner. The device of this invention does not require a transformer and can be designed based on a non-isolated step-down converter. Preferred embodiments of the present invention

[0026] In a particular embodiment, a first transition point is provided, to which the connection point of each first converter of each circuit block is directly or indirectly connected, and a second transition point is provided, to which the connection point of each second converter of each circuit block is directly or indirectly connected, wherein a first load point is provided for connecting a first end of the superconducting magnet, and a second load point is provided for connecting a second end of the superconducting magnet, wherein the first transition point is connected to the first load point via a first load connection, and the second transition point is connected to the second load point via a second load connection. This design makes it particularly easy to parallelize several circuit blocks, especially several first converters and several second converters, and connect them together to the superconducting magnet.

[0027] In another specific embodiment, the charging and discharging device further includes a control unit that provides a first generator control signal to the control signal generator of each first converter in each circuit block, and a second generator control signal to the control signal generator of each second converter in each circuit block. Due to this design, centralized control of the duty cycle of each first converter and each second converter in each circuit block is particularly easy. The first generator control signal may include U... out1 The target value, and the second generator control signal can include U out2 The target value.

[0028] In another specific embodiment, a current measuring device is included in either the first or second load connection, wherein the signal output of the current measuring device is connected to the control unit via a current measuring line. Because of this current measuring device, the current currently operating in the superconducting magnet can be measured. Therefore, the current in the superconducting magnet can be controlled based on the measured current currently operating in the superconducting magnet, particularly to prevent quenching.

[0029] The current measuring device can be formed by a current sensing amplifier that connects its non-inverting input and its inverting input to different points of a first load connection or a second load connection. Between these different points, the first load connection or the second load connection can form a resistor, or a resistor can be inserted into the corresponding connection. The output of the current sensing amplifier can be connected to a control unit via a current measuring line.

[0030] The portion of the load connection between the corresponding load point and the current measuring device can be formed by a cable. Similarly, the end of another load connection (adjacent to its corresponding load point) can be formed by another cable. The length of each cable can be 1 to 3 meters. Preferably, the cable length can be 2 meters. Compared to longer cables, such short cables have relatively low resistance. Therefore, the cable loss is low. When none of the inductors in the converter of each circuit block have a ferromagnetic core, it is particularly likely that the charging and discharging device is positioned so close to the superconducting magnet and in its magnetic field.

[0031] In another specific embodiment, the control unit is configured to provide generator control signals depending on the charging and discharging progress of the superconducting magnet, wherein the charging and discharging progress of the superconducting magnet is monitored by a current measuring device. This design allows the charging and discharging process to be controlled based on the monitored current currently flowing through the superconducting magnet. This further supports the prevention of quenching.

[0032] In another specific embodiment, a first potential measurement line connects a first load point to the control unit, and a second potential measurement line connects a second load point to the control unit. Due to this design, the voltage currently applied to the superconducting magnet can be measured. Therefore, the measured voltage can be used to control the voltage applied to the magnet, particularly limiting the voltage applied to the magnet to prevent quenching.

[0033] In another specific embodiment, the control unit has a first setting input for setting a maximum magnet voltage to be applied to the superconducting magnet, and a second setting input for setting a current target for the superconducting magnet. The control unit is configured to generate a first generator control signal and a second generator control signal, taking into account the set current target, the set maximum magnet voltage, a signal from a current measurement line indicating the current currently operating in the superconducting magnet, and signals from a first potential measurement line and a second potential measurement line indicating the voltage currently applied to the superconducting magnet. Due to this design, the current flowing through the superconducting magnet can be controlled based on the current target and the measured current flowing through the superconducting magnet, and the voltage applied to the magnet can be controlled by setting the maximum magnet voltage and measuring the current voltage applied to the magnet. Queue loss can be effectively prevented by regulating the current and limiting the voltage during charging and discharging.

[0034] In another specific embodiment, a common mode value CM is defined, where CM = (U out1 +U out2 ) / 2, wherein the control unit is configured to provide a first generator control signal and a second generator control signal, such that the common mode value CM varies with time during the charging and discharging of the superconducting magnet using the charging and discharging equipment. By changing the common mode value CM, the charging or discharging speed can be balanced on the one hand, and the loss can be balanced on the other hand, to obtain an optimized overall charging or discharging process.

[0035] In another specific embodiment, during charging, the common modulus CM decreases over time in case A) or increases over time in case B), particularly depending on the measured charging progress; and / or during discharging, the common modulus CM increases over time in case A) or decreases over time in case B), particularly depending on the measured discharging progress. Because of this, the losses of the filter inductor are further dynamically reduced during charging, or the discharge rate can be dynamically increased during discharging. Generally, the switching of the converter (or the values ​​of D1 and D2, respectively) is controlled dynamically within a region where the losses of the filter inductor are as low as possible.

[0036] In another specific embodiment, a differential mode value DM is defined, where DM = U out1 -U out2 The control unit is configured to provide a first generator control signal and a second generator control signal, such that the differential mode value DM changes over time during the charging and discharging of the superconducting magnet using a charging and discharging device.

[0037] In another specific embodiment, during charging, the absolute value of the differential mode |DM| decreases over time, specifically depending on the measured charging progress; and / or during discharging, the absolute value of the differential mode |DM| increases over time, specifically depending on the measured discharging progress. This allows for a further dynamic reduction in the filter inductance losses during charging, while also preventing quenching. The charging and discharging rates can be optimized, specifically increasing the rate when the probability of quenching is low, or decreasing the rate when the probability of quenching is high.

[0038] In another specific embodiment, a differential mode value DM is defined, where DM = U out1 -U out2 The control unit is configured to provide a first generator control signal and a second generator control signal, such that during the charging and discharging of the superconducting magnet using a charging and discharging device, the absolute value of the differential mode |DM|applicable|DM|≤0.25*|U in | Preferably, |DM|≤0.20*|U in This allows for good charging and discharging speeds in practice, while maintaining magnet safety. in It is usually chosen as a positive value.

[0039] In another specific embodiment, the charging and discharging device includes multiple circuit blocks, wherein all input points of all converters in all circuit blocks are connected to a common input stage, and all base points of all converters in all circuit blocks are connected to a common base stage. This makes voltage supply to all circuit blocks particularly easy. Furthermore, by providing several circuit blocks, the power to be transferred can be distributed across several circuit blocks. Therefore, a single converter can be made from low-power components and is thus inexpensive.

[0040] In the case of multiple circuit blocks, a control signal generator can be used for multiple first converters and / or a control signal generator can be used for multiple second converters. In this case, a control signal generator can apply the same phase in the duty cycle in all first converters and / or a control signal generator can apply the same phase in the inverted duty cycle in all second converters. However, preferably, each converter has its own control signal generator, which can be considered part of the converter. Therefore, local feedback loops can be established in each converter and / or local phase shifts can be applied, particularly for each block.

[0041] Typically, the control signals for different circuit blocks with their corresponding duty cycles are phase-shifted by 2 to each other. / N, where N is the number of circuit blocks. Phase-shifting techniques used for first and second converters used in parallel help increase the ripple frequency and make the ripple less noticeable, thus requiring fewer passive components for filtering. For N half-bridges in parallel, the recombined ripple frequency is typically F. ripple = N * F sw .

[0042] In another specific embodiment, each such converter includes its own control signal generator, wherein each converter includes a feedback line connected to the connection point and the control signal generator, and the control signal generator is configured to determine the duty cycle of its control signal in consideration of the feedback signal received via the feedback line. This allows each first converter to be decoupled from other first converters in several blocks connected together, and each second converter to be decoupled from other second converters in several blocks connected together, thereby applying current parallelization without directly measuring the current. This makes each converter cheaper, since current measurement for high currents is expensive.

[0043] Each control signal generator may include an operational amplifier, particularly an operational amplifier interconnected as a PI controller. An analog voltage signal for the analog-to-PWM signal converter of the corresponding control signal generator may be present at the output of the operational amplifier. The output of the operational amplifier may be fed back to its inverting input via a feedback connection, in which a capacitor and a resistor are arranged. The capacitor may be directly connected to the output of the operational amplifier, and the resistor may be directly connected to the inverting input of the operational amplifier. A feedback line, one end of which is also connected to the inverting input of the operational amplifier, may be connected at the other end to the connection point of the corresponding converter, and a resistor may be arranged therein.

[0044] In another specific embodiment, each converter includes an input capacitor connected to the input point and the base point. The input capacitor acts as a current storage buffer, stores DC energy, and provides AC current between the input capacitor and the filter capacitor or decoupling capacitor.

[0045] In another specific embodiment, each converter further includes a series connection of a decoupling inductor and a decoupling capacitor, wherein the decoupling inductor and decoupling capacitor are connected at another connection point, and wherein the decoupling inductor is connected to the connection point and the decoupling capacitor is connected to the base point. Specifically, another connection point of each first converter in each circuit block is connected to a first transition point, and another connection point of each second converter in each circuit block is connected to a second transition point. Each combination of decoupling inductors and decoupling capacitors can form a low-pass filter for decoupling high currents. The decoupling inductor of each first converter allows for simple parallel connection of the first converters via impedance series decoupling. The decoupling inductor of each second converter allows for simple parallel connection of the second converters via impedance series decoupling. When the first and second converters are connected in parallel, each decoupling inductor acts as a balancing resistor. This makes it possible to decouple each first converter from other first converters in several blocks connected together, and to decouple each second converter from other second converters in several blocks connected together.

[0046] The present invention also relates to the use of a charging and discharging device for charging or discharging a superconducting magnet. A common input voltage Uin is supplied between the input point and the base point of each converter in each circuit block, such that a first output voltage U is generated between the connection point and the base point of each first converter in each circuit block. out1 And generate a second output voltage U between the connection point and the base point of each second converter block. out2 A superconducting magnet is connected to a charging and discharging device, causing voltage U to... out1 and U out2 The difference effectively charges the superconducting magnet. Specifically, the charging / discharging device is configured to provide a first transition point, the connection point of each first converter in each circuit block being directly or indirectly connected to this first transition point; and to provide a second transition point, the connection point of each second converter in each circuit block being directly or indirectly connected to this second transition point; to provide a first load point for connecting a first end of the superconducting magnet, and a second load point for connecting a second end of the superconducting magnet; and to connect the first transition point to the first load point via a first load connection, and the second transition point to the second load point via a second load connection; and to connect the superconducting magnet between the first load point and the second load point.

[0047] Because the charging / discharging device operates with a first duty cycle D1 and a second duty cycle D2, where both D1 and D2 are at most 25% in case A) or at least 75% in case B), the ripple current through the converter's impedance is low, resulting in low power loss through the impedance. Consequently, heat dissipation due to the impedance is low. For this reason, cooling equipment for the charging / discharging device can be reduced or eliminated. This makes the charging / discharging device occupy less space and is less expensive.

[0048] The present invention also relates to a method for charging a superconducting magnet.

[0049] In particular, methods for charging superconducting magnets using this charging and discharging equipment include:

[0050] - Utilizing at least one first synchronous half-bridge converter from a common input voltage U in Generate the first output voltage U out1 The first duty cycle D1 is applied.

[0051] - Utilizing at least one second synchronous half-bridge converter from the common input voltage U in Generate the second output voltage U out2 A second duty cycle D2 is applied.

[0052] - By choosing D1 and D2 differently, U out1 and U out2 They are different.

[0053] - Choose D1 and D2, where in case A) both D1 and D2 are at most 25%, or in case B) both D1 and D2 are at least 75%.

[0054] - will U out1 and U out2 Applying it to the superconducting magnet, so that the superconducting magnet is effectively controlled by U out1 and U out2 The differential charging, specifically, where U out1 and U out2 It is applied to different ends of the superconducting magnet.

[0055] As a result of choosing D1 and D2, both D1 and D2 account for at most 25% in case A), or both D1 and D2 account for at least 75% in case B, resulting in low ripple current through the converter impedance. For this reason, the heat dissipated by these impedances is low, which is why the cooling elements of the charging / discharging device can be minimized or omitted. This ensures that the charging / discharging device is space-saving and inexpensive.

[0056] In a specific variant, a common mode value CM is defined, where CM = (U out1 +U out2 The common-mode value CM varies with time during superconducting magnet charging, specifically decreasing with time in case A) or increasing with time in case B), depending particularly on the measured charging progress. Due to this design, the filter inductor loss can be further dynamically reduced during charging. In the charging case, U... out1 and U out2 The difference is positive.

[0057] This invention also relates to a method for discharging a superconducting magnet, and particularly a method for discharging a superconducting magnet using a charging and discharging device, comprising:

[0058] - Utilizing at least one first synchronous half-bridge converter from a common input voltage U in Generate the first output voltage U out1 The first duty cycle D1 is applied.

[0059] - Utilizing at least one second synchronous half-bridge converter from the common input voltage U in Generate the second output voltage U out2 A second duty cycle D2 is applied.

[0060] - By choosing D1 and D2 differently, U out1 and U out2 They are different.

[0061] - Choose D1 and D2, where in case A) both D1 and D2 are at most 25%, or in case B) both D1 and D2 are at least 75%.

[0062] - will U out1 and U out2 Applying it to the superconducting magnet, so that the superconducting magnet is effectively controlled by U out1 and U out2 The differential discharge, in particular, where U out1 and U out2 It is applied to different ends of the superconducting magnet.

[0063] Because D1 and D2 are chosen such that, in case A), both D1 and D2 are at most 25%, or in case B), both D1 and D2 are at least 75%, the ripple current through the converter's filter inductor is relatively low, resulting in relatively low power loss of the filter inductor. Based on this, the converter's cooling equipment can be minimized or omitted, making the charging and discharging equipment space-saving and inexpensive.

[0064] In a specific embodiment, a common mode value CM is defined, where CM = (U out1 +U out2 ) / 2, where during the discharge of the superconducting magnet, the common-mode value CM varies with time; specifically, the common-mode value CM increases with time in case A), or decreases with time in case B), particularly depending on the measured discharge progress. Discharge can proceed with good discharge rate and low loss under safe magnet operation conditions. In the discharge condition, U out1 and U out2 The difference is negative. Attached Figure Description

[0065] Other advantages and features of the invention will become apparent from the following description of embodiments of the invention with reference to the accompanying drawings, which illustrate important details. Each feature described above or explained below can be implemented individually or in any useful combination in variations of the invention.

[0066] Figure 1 A circuit diagram illustrating an exemplary embodiment of the charging and discharging device according to the present invention is shown;

[0067] Figure 2 Enlarged to show Figure 1 The left half of the circuit diagram;

[0068] Figure 3 Enlarged to show Figure 1 The right half of the circuit diagram;

[0069] Figure 4 Enlarged and displayed in different presentation methods Figure 3 The region marked with X in the middle;

[0070] Figure 5a The input signal and in-phase output signal of the analog-to-PWM converter of the first synchronous half-bridge converter of the charging and discharging device are shown in a coordinate system with the x-axis representing time and the y-axis representing voltage.

[0071] Figure 5b The diagram illustrates the input signal and in-phase output signal of the analog-to-PWM converter of the second synchronous half-bridge converter of a charging and discharging device, plotted in a coordinate system where the x-axis represents time and the y-axis represents voltage. The output signal of the analog-to-PWM converter of the second synchronous half-bridge converter is shown relative to... Figure 5a The output signal of the analog-to-PWM converter of the first synchronous half-bridge converter undergoes a phase shift;

[0072] Figure 6 The diagram shows the in-phase output signals of each analog-to-PWM converter of a charging and discharging device, which consists of six parallel first synchronous half-bridge converters and six parallel second synchronous half-bridge converters, plotted in a coordinate system with the x-axis representing time and the y-axis representing voltage. The output signals of the analog-to-PWM converters of the first synchronous half-bridge converters are phase-shifted relative to each other, and the output signals of the analog-to-PWM converters of the second synchronous half-bridge converters are phase-shifted relative to each other.

[0073] Figure 7a The process of adjusting the maximum and minimum magnet voltage during the charging and discharging of a superconducting magnet using a charging and discharging device is shown in a coordinate system with the x-axis representing time and the y-axis representing voltage.

[0074] Figure 7bThe diagram illustrates the progress of the voltage at one end of a superconducting magnet and the voltage at the other end of the superconducting magnet during charging and discharging using a charging and discharging device, plotted in a coordinate system where the x-axis represents time and the y-axis represents voltage.

[0075] Figure 7c The graph illustrates the progression of the common-mode value (the average of the voltages at one end and the other end of the superconducting magnet) during charging and discharging of a superconducting magnet using a charging and discharging device, plotted in a coordinate system with the x-axis representing time and the y-axis representing voltage.

[0076] Figure 7d The graph illustrates the process of current flowing through a superconducting magnet and voltage (differential mode value) applied to it during charging and discharging using a charging and discharging device, plotted in a coordinate system with time on the x-axis, current on the left (y-axis), and voltage on the right (y-axis). Detailed Implementation

[0078] Figures 1 to 4 A circuit diagram of an exemplary embodiment of the charging and discharging device 10 according to the present invention is shown. The charging and discharging device 10 is used to charge and discharge a superconducting magnet 12.

[0079] The charging / discharging device 10 shown in the figure has two circuit blocks 14; note that, typically, the charging / discharging device 10 according to the present invention has at least one circuit block 14. Each circuit block 14 has a first synchronous half-bridge converter 16 ( Figure 3 ) and the second synchronous half-bridge converter 18.

[0080] Each synchronous half-bridge converter 16, 18 includes a series connection of switch 20 and another switch 22. Here, switch 20 is formed as a transistor in the form of MOSFET 24, and the other switch 22 is formed as another transistor in the form of another MOSFET 26. Each MOSFET 24, 26 is either closed or open at a certain time. The drain connection of MOSFET 24 is connected to input point 28. The source connection of MOSFET 24 is connected to the drain connection of another MOSFET 26 at confluence point 30. The source connection of the other MOSFET 24 is connected to base point 32. An input capacitor 34 is arranged between input point 28 and base point 32. Figure 4 The input capacitor 34 serves as a current storage buffer, storing DC energy and providing AC current between the input capacitor 34 and the filter capacitor 36.

[0081] In addition, each synchronous half-bridge converter 16, 18 has a filter inductor 38 and a filter capacitor 36 connected in series. One end of the filter inductor 38 is connected to the junction point 30 ( Figure 3 Furthermore, the other end of the filter inductor 38 is connected to one end of the filter capacitor 36 at connection point 40. A first output voltage U exists between connection point 40 and base point 32 of each first converter 16. out1 A second output voltage U exists between connection point 40 and base point 32 of each second converter 18. out2 The other end of filter capacitor 36 is connected to base point 32. The combination of filter inductor 38 and filter capacitor 36 forms a low-pass filter for filtering switching waveforms.

[0082] If such exists Figure 1 , 3 In the case of at least two circuit blocks 14 shown (the second and another circuit blocks are marked Y), each synchronous half-bridge converter 16, 18 also includes a series connection of a decoupling inductor 42 and a decoupling capacitor 44. One end of the decoupling inductor 42 is connected to connection point 40, and the other end of the decoupling inductor 42 is connected to one end of the decoupling capacitor 44 at another connection point 46. The other end of the decoupling capacitor 44 is connected to base point 32. The combination of the decoupling inductor 42 and the decoupling capacitor 44 forms another low-pass filter for decoupling high current. The decoupling inductor 42 of each first converter 16 allows for simple parallel connection of the first converter 16 through impedance series decoupling. The decoupling inductor 42 of each second converter 18 allows for simple parallel connection of the second converter 18 through impedance series decoupling. When the first converter 16 and the second converter 18 are connected in parallel, each decoupling inductor 42 acts as a balancing resistor. This allows the first converter 16 to be decoupled from the first converter 16 of the several circuit blocks 14 connected together, and the second converter 18 to be decoupled from the second converter 18 of the several circuit blocks 14 connected together.

[0083] Furthermore, each synchronous half-bridge converter 16, 18 includes a control signal generator 48 ( Figure 3 Each control signal generator 48 is configured to output a control signal 50 for driving the switch 20 to open and close. Figure 5a ), and output another control signal (not shown) to drive another switch 22 to open and close.

[0084] Here, each control signal generator 48 has a first operational amplifier 54, specifically a first operational amplifier 54 interconnected as a PI controller. The output of the first operational amplifier 54 is fed back to its inverting input via a feedback connection in which a first capacitor 56 and a first resistor 58 are arranged. The first capacitor 56 is directly connected to the output of the first operational amplifier 54, and the first resistor 58 is directly connected to the inverting input of the first operational amplifier 54. One end of the feedback line 60 is also connected to the inverting input of the first operational amplifier 54, and the other end of the feedback line 60 is connected to the connection point 40, where a second resistor 62 is provided. An output voltage U exists between the connection point 40 and the base point 32 of each first converter 16. out1 The feedback and the output voltage U present between connection point 40 and base point 32 of each second converter 18 out2 The feedback is sent to the corresponding first operational amplifier 54. Due to current parallelization, the first converter 16 can be decoupled from the first converter 16 of the several circuit blocks 14 connected together, and the second converter 18 can be decoupled from the second converter 18 of the several circuit blocks 14 connected together. The first operational amplifier 54 of the first converter 16 outputs an analog voltage signal 64. Figure 5a The first operational amplifier 54 of the second converter 18 outputs another analog voltage signal 65. Figure 5b ).

[0085] Each control signal generator 48 also includes an analog-to-PWM signal converter 66. The analog-to-PWM signal converter 66 is configured as a second operational amplifier 68 in the form of a comparator. Figure 4 The second operational amplifier 68 has a non-inverting input, an inverting input, and two complementary outputs. The non-inverting output of the second operational amplifier 68 is connected to the gate of the MOSFET 24 via a control signal connection 72 in which a gate driver 74 is arranged. The inverting output of the second operational amplifier 68 is connected to the gate of another MOSFET 26 via another control signal connection 76 in which another gate driver 78 is arranged.

[0086] The output of the first operational amplifier 54 of the first converter 16 is connected to the non-inverting input of the second operational amplifier 68 of the first converter 16 for transmitting the analog voltage signal 64. Figure 5a Triangular wave signal 70 ( Figure 5aA control signal 50 is applied to the inverting input of the second operational amplifier 68 of the first converter 16. The non-inverting output of the second operational amplifier 68 of the first converter 16 outputs a control signal 50, which drives the MOSFET 24 of the first converter 16 to open and close. The inverting output of the second operational amplifier 68 of the first converter 16 outputs another control signal (not shown), which drives another MOSFET 26 of the first converter 16 to open and close. This other control signal of the second operational amplifier 68 of the first converter 16 is inverted relative to the control signal 50 of the second operational amplifier 68 of the first converter 16, such that control signal 50 applies a duty cycle D1, and the other control signal applies the inversion of said duty cycle D1.

[0087] The output of the first operational amplifier 54 of the second converter 18 is connected to the non-inverting input of the second operational amplifier 68 of the second converter 18 for transmitting another analog voltage signal 65. Figure 5b Another triangular wave signal 71 is applied to the inverting input of the second operational amplifier 68 of the second converter 18. The non-inverting output of the second operational amplifier 68 of the second converter 18 outputs a control signal 52, which drives the MOSFET 24 of the second converter 18 to open and close. The inverting output of the second operational amplifier 68 of the second converter 18 outputs another control signal (not shown), which drives another MOSFET 26 of the second converter 18 to open and close. This other control signal of the second operational amplifier 68 of the second converter 18 is inverted relative to the control signal 52 of the second operational amplifier 68 of the second converter 18, such that control signal 52 applies a duty cycle D2, and the other control signal applies the inversion of said duty cycle D2.

[0088] Because of the phase reversal, switch 20 of the corresponding converters 16 and 18 is open when the other switch 22 of the corresponding converter 16 and 18 is closed, and vice versa. To prevent the two switches 20 and 22 of the corresponding converters 16 and 18 from closing simultaneously, an appropriate dead-time is provided.

[0089] The triangular wave signal 71 of the second converter 18 ( Figure 5b ) can be relative to the triangular wave signal 70 of the first converter 16 ( Figure 5a Phase shift, especially a 180-degree phase shift. Figure 5a (b) If this is the case, then the phase shift of the second duty cycle D2 relative to the first duty cycle D1 is the corresponding degree.

[0090] The first duty cycle D1 and the second duty cycle D2 are both at most 25% in case A), or at least 75% in case B). The first duty cycle D1 and the second duty cycle D2 are different from each other. Therefore, when a common input voltage U is applied between the input point 28 and the base point 32 at the two converters 16 and 18... in At that time, the first output voltage U exists between the connection point 40 and the base point 32 of the first converter 16. out1 The second output voltage U between the connection point 40 and the base point 32 of the second converter 18 out2 They are different from each other. They share a common input voltage U. in Typically, it is a constant voltage between 6 volts and 24 volts, especially 12 volts. Each base point 32 is connected to ground 83.

[0091] If such exists Figure 1 , 3 As shown, at least two circuit blocks 14 provide a common input stage (not shown), with the input point 28 of each converter 16, 18 of each circuit block 14 connected to this common input stage. Furthermore, a common base point stage (not shown) exists, with the base point 32 of each converter 16, 18 of each circuit block 14 connected to this base point stage. The common input voltage U... in It exists between a common input level and a common base level.

[0092] The triangular wave signals 70 of the first converter 16 of multiple circuit blocks 14 can be phase-shifted towards each other by 2. / N degrees, where N is the number of circuit blocks 14. When this is the case, the first duty cycle D1 of the first converter 16 of circuit block 14 can also be phase-shifted 2 degrees towards each other. / N degrees. Correspondingly, the triangular wave signal 71 of the second converter 18 of circuit block 14 can be phase-shifted 2 degrees towards each other. / N, so that the second duty cycle D2 of the second converter 18 of circuit block 14 can also be phase-shifted 2 towards each other. / N. Figure 6 The diagram shows control signals 50, 50', 50'', 50''', 50''', 50'''', 50''''', and 50'''''' for the second operational amplifier 68 of the first converter 16 of the charging / discharging device 10, which has six (N=6) circuit blocks 14, and control signals 52, 52', 52''', 52'''', 52''''', and 52'''''' for the second operational amplifier 68 of the second converter 18, which has a second duty cycle D2. Here, the first duty cycles D1 of the first converter 16 of the circuit blocks 14 are shifted by 60 degrees from each other, and the second duty cycles D2 of the second converter 18 of the circuit blocks 14 are shifted by 60 degrees from each other.

[0093] Another connection point 46 of each first converter 16 of each circuit block 14 is connected to the first output transition point 84, and another connection point 46 of each second converter 18 of each circuit block 14 is connected to the second output transition point 86.

[0094] The first output transition point 84 is connected to the first load point 88 via the first load connection 90, and the second output transition point 86 is connected to the second load point 92 via the second load connection 94. The end section of the first load connection 90 directly connected to the first load point 88 and the other end section of the second load connection 94 directly connected to the second load point 92 are each formed as a cable with a resistance 95.

[0095] One end of the superconducting magnet 12 is connected to a first load point 88, and the other end is connected to a second load point 92. A voltage U exists between the first load point 88 and the base point 32. out1' U out1' Basically corresponds to U out1 There is a voltage U between the second load point 92 and the base point 32. out2' U out2' Basically corresponds to U out2 .

[0096] Based on this, the differential voltage value DM is defined as the voltage U existing between the first load point 88 and the base point 32. out1' Subtract the voltage U that exists between the second load point 92 and the base point 32 out2' That is, DM=U out1' - U out2' Furthermore, the common-mode voltage CM is defined as the voltage U present between the first load point 88 and the base point 32. out1' And the voltage U existing between the second load point 92 and the base point 32 out2' The mean. That is, CM = (U out1' +U out2' ) / 2.

[0097] An internal phase-shift technique is very useful for reducing input capacitor current, and it synchronizes U out1' and U out2' The ripple voltage between them. In the voltage DM across the magnet, if U out1' and U out2' If they are equal, then the synchronization ripple is partially or completely canceled out. The phase-shifting technique of the several first converters 16 and second converters 18 used in parallel helps to increase the ripple frequency, thus requiring fewer passive components for filtering.

[0098] In addition, the charging and discharging device 10 includes a control unit 96 ( Figure 1The control unit 96 is configured to send control signal generators 48 to each of the first converters 16 of each circuit block 14. Figure 3 The system outputs a first generator control signal and a second generator control signal to the control signal generator 48 of each second converter 18 in each circuit block 14. The first generator control signal provides a voltage U present between the first load point 88 and the base point 32. out1' The target value. The second generator control signal provides the voltage U present between the second load point 92 and the base point 32. out2' The target value.

[0099] The control unit 96 here includes four interconnected electrical components arranged 102, 104, 106, and 108. Figure 1 ).

[0100] The first electrical component arrangement 102 can be configured to output the maximum magnet voltage V adjusted by the first electrical component 102. lim ( Figure 2 The first electrical component arrangement 102 may generate its output depending on the maximum magnet voltage 112. The maximum magnet voltage is set and changed by the user of the charging / discharging device 10. It is also conceivable that the maximum magnet voltage is initially set once by the user of the charging / discharging device 10, and then automatically changed by a suitable control device from then on.

[0101] The first electrical component arrangement 102 has a third resistor 114, one end of which is connected to a first setting input 116 in the form of a human-machine interface for inputting a maximum magnet voltage 112 set by the user of the charging / discharging device 10. The other end of the third resistor 114 is connected to the non-inverting input of a third operational amplifier 118. The third resistor 114 can be used as an offset for the maximum magnet voltage 112, specifically, this offset can be approximately 100 millivolts. The first electrical component arrangement 102 has a fourth resistor 120 and a fifth resistor 122, one end of which is connected to the non-inverting input of the third operational amplifier 118. The fourth resistor 120 and the fifth resistor 122 together form a voltage divider, which can be used to adjust to the corresponding analog-to-PWM converter 66 (…). Figure 3 The voltage of the fourth resistor 120 is connected to the voltage source 123, where approximately 2.5 volts can be present. The other end of the fifth resistor 122 is connected to ground 83. The third operational amplifier 118 can be interconnected as a voltage follower. The output of the third operational amplifier 118 is fed back to its inverting input. At the output of the third operational amplifier 118, there may be a maximum magnet voltage V adjusted by the first electrical component 102. lim .

[0102] The first electrical component arrangement 102 has a sixth resistor 124, one end of which is connected to the output of the third operational amplifier 118, and the other end of which is connected to the non-inverting input of the fourth operational amplifier 126. The sixth resistor 124 can generate another voltage offset. The first electrical component arrangement 102 has a seventh resistor 128 and an eighth resistor 130, one end of which is connected to the non-inverting input of the fourth operational amplifier 126. The seventh resistor 128 and the eighth resistor 130 can together form another voltage divider. The other end of the seventh resistor 128 is connected to another voltage source 131, where approximately 2.5 volts can be present. The other end of the eighth resistor 130 is connected to ground. The fourth operational amplifier 126 can be interconnected as a voltage follower. The output of the fourth operational amplifier 126 is fed back to its inverting input. A target common-mode value (TCM) can be present at the output of the fourth operational amplifier 126. The sixth resistor 124 to the eighth resistor 130 and the fourth operational amplifier 126 together form TCM > (adjusted maximum magnet voltage V). lim The limitation is ) / 2.

[0103] Second electrical component arrangement 104 ( Figure 1 The second electrical component arrangement 104 can be configured to draw in (pull in) or release (push out) current at its output. The arrangement can depend on the adjusted maximum magnet voltage V. lim and the voltage U existing between the first load point 88 and the base point 32 and between the second load point 92 and the base point 32. out1' U out2' Generate its output.

[0104] The second electrical component arrangement 104 has a ninth resistor 132, one end of which is connected to the second load point 92 via a second potential measurement line 134. Figure 3The output of the fifth operational amplifier 136 is fed back to its inverting input via a feedback connection in which a tenth resistor 138 is arranged. One end of the eleventh resistor 140 is connected to the first load point 88 via a first potential measurement line 142, and the other end is connected to the non-inverting input of the fifth operational amplifier 136. One end of the twelfth resistor 144 is connected to the non-inverting input of the fifth operational amplifier 136, and the other end is connected to ground 83. The output of the fifth operational amplifier 136 is connected to the non-inverting input of the sixth operational amplifier 148 via a thirteenth resistor 146. The sixth operational amplifier 148 can be interconnected as a PI controller. The output of the sixth operational amplifier 148 is fed back to its non-inverting input via a feedback connection in which a second capacitor 150 and a fourteenth resistor 152 are arranged. The second capacitor 150 is directly connected to the output of the sixth operational amplifier 148, and the fourteenth resistor 152 is directly connected to the non-inverting input of the sixth operational amplifier 148. The inverting input of the sixth operational amplifier 148 is connected to the output of the third operational amplifier 118. One end of the first diode 154 is connected to the output of the sixth operational amplifier 148. The forward direction of the first diode 154 is the same as the direction of the output of the sixth operational amplifier 148.

[0105] Because the fifth operational amplifier 136 may have a supply voltage ranging from 0 volts to 5 volts, its output has a voltage higher than 0 volts and a maximum of 5 volts during charging, and 0 volts during discharging. Therefore, the fifth operational amplifier 136 only amplifies the positive voltage difference between its inputs, which exists only during charging. The voltage at the output of the fifth operational amplifier 136 minus the voltage drop across the thirteenth resistor 146 exists at the non-inverting input of the sixth operational amplifier 148, where the adjusted maximum magnet voltage V, as a target value, exists at its inverting input. lim If the voltage difference at the input of the sixth operational amplifier 148 is positive during charging, it means that the target value V... lim If the voltage is less than DM, then the voltage at the output of the sixth operational amplifier 148 is less than the voltage at the other end 155 of the first diode 154. In this case, current is drawn through the first diode 154 in the direction of the sixth operational amplifier 148. Due to this design, the maximum magnet voltage is limited. In summary, the portion of the second electrical assembly 104 including the ninth resistors 132 to the fourteenth resistors 152, the fifth operational amplifier 136 and the sixth operational amplifier 148, and the first diode 154 is "active" only during charging.

[0106] The second electrical component arrangement 104 has a fifteenth resistor 156, one end of which is connected to a first load point 88 via a first potential measurement line 142, and the other end of which is connected to the inverting input of a seventh operational amplifier 158. The output of the seventh operational amplifier 158 is fed back to its inverting input via a feedback connection in which a sixteenth resistor 160 is arranged. One end of a seventeenth resistor 162 is connected to a second load point 92 via a second potential measurement line 134, and the other end of which is connected to the non-inverting input of the seventh operational amplifier 158. One end of an eighteenth resistor 164 is connected to the non-inverting input of the seventh operational amplifier 158, and the other end of which is connected to ground 83. The output of the seventh operational amplifier 158 is connected to the non-inverting input of an eighth operational amplifier 166. The eighth operational amplifier 166 can be interconnected as a PI controller. The output of the eighth operational amplifier 166 is fed back to its inverting input via a feedback connection in which a third capacitor 168 and a nineteenth resistor 170 are arranged. The third capacitor 168 is directly connected to the output of the eighth operational amplifier 166, and the nineteenth resistor 170 is directly connected to the inverting input of the eighth operational amplifier 166. One end of the twentieth resistor 172 is connected to the output of the third operational amplifier 118, and the other end is connected to the inverting input of the eighth operational amplifier 166. One end of the second diode 174 is connected to the output of the eighth operational amplifier 166, and the other end 175 is connected to the other end 155 of the first diode 154. The forward direction of the second diode 174 is the same as that of the first diode 154.

[0107] Similar to the charging period, if the target value is negative during the discharging period (negative adjusted maximum magnet voltage -V)... lim If the voltage at the output of the eighth operational amplifier 166 is higher than the voltage at the other end 175 of the second diode 174, then the current is pushed through the second diode 174 in the direction of the first diode 154. Due to this design, the minimum magnet voltage is limited. In summary, the portion of the second electrical component 104 including the fifteenth resistor 156 to the twentieth resistor 172, the seventh operational amplifier 158 and the eighth operational amplifier 166, and the second diode 174 is "active" only during discharge.

[0108] The third electrical component is arranged in 106 ( Figure 1 This can be used to control the current currently flowing through the superconducting magnet 12. Thermal effects can be corrected through current control. The third electrical component arrangement 106 can be configured to output a target differential mode value (TDM). The third electrical component 106 can generate its output based on the current currently flowing through the superconducting magnet 12 and the current target 178. The current target 178 is set and changed by the user of the charging / discharging device 10. It is also conceivable that the current target 178 is set and changed by a suitable control device.

[0109] The third electrical component 106 arrangement here includes a ninth operational amplifier 176. The ninth operational amplifier 176 can be interconnected as a PI controller. A second setting input 180, in the form of another human-machine interface, is connected to the non-inverting input of the ninth operational amplifier 176 for inputting a current target 178. The output of the ninth operational amplifier 176 is fed back to its inverting input via a feedback connection in which a fourth capacitor 182 and a twenty-first resistor 184 are arranged. The fourth capacitor 182 is directly connected to the output of the ninth operational amplifier 176, and the twenty-first resistor 184 is directly connected to the inverting input of the ninth operational amplifier 176. One end of a twenty-second resistor 186 is connected to the inverting input of the ninth operational amplifier 176. The other end of the twenty-second resistor 186 is connected via a current measurement line 192 to the signal output 188 of a current measuring device 190. Figure 3 ).

[0110] A current measuring device 190 is connected to a first load connection 90. The current measuring device 190 includes a tenth operational amplifier 194 in the form of a current sensing amplifier, with its non-inverting and inverting inputs connected to different points on the first load connection 90. Between these different points, the first load connection forms a twenty-third resistor 196, or a twenty-third resistor 196 is arranged within the first load connection 90.

[0111] The twenty-fourth resistor is 198 ( Figure 2 One end of the resistor is connected to the output of the ninth operational amplifier 176. The target differential mode value TDM is located at the other end of the twenty-fourth resistor 198. The other ends 155 and 175 of the first diode 154 and the second diode 174 are connected to the other end of the twenty-fourth resistor 198. The twenty-fourth resistor 198 can be used for isolation and / or has a value of one kiloohm.

[0112] If the second electrical component arrangement 104 draws current into the first diode 154, the target differential mode voltage value TDM decreases. Conversely, if the second electrical component arrangement 104 discharges current through the second diode 174, the target differential mode voltage value TDM increases.

[0113] Fourth electrical component arrangement 108 ( Figure 1 This can be used for single-ended to differential signal conversion. The fourth electrical component arrangement 108 can be configured to output a first generator control signal and a second generator control signal. The fourth electrical component arrangement 108 can generate its output depending on the target differential mode value TDM and the target common mode value TCM.

[0114] The fourth electrical component 108 includes an eleventh operational amplifier 200 ( Figure 2Eleventh operational amplifier 200 can be interconnected as a voltage follower. The other ends 155 and 175 of the twenty-fourth resistor 198 and diodes 154 and 174 are connected to the non-inverting input of eleventh operational amplifier 200. The output of eleventh operational amplifier 200 is fed back to its inverting input via a feedback connection. A first generator control signal is present at the output of eleventh operational amplifier 200. The first generator control signal may include a voltage U present between the first load point 88 and the base point 32. out1' The target value. The output of the eleventh operational amplifier 200 is connected to the first operational amplifier 54 of each first converter 16 in each circuit block 14. Figure 3 () is an in-phase input.

[0115] The fourth electrical component 108 includes a twelfth operational amplifier 202. The twelfth operational amplifier 202 can be interconnected as an inverter. The output of the fourth operational amplifier 126 is connected to the non-inverting input of the twelfth operational amplifier 202. The output of the twelfth operational amplifier 202 is fed back to its inverting input via a feedback connection in which a twenty-fifth resistor 204 is arranged. The inverting input of the twelfth operational amplifier 202 is connected to the output of the eleventh operational amplifier 200 via a twenty-sixth resistor 206. A second generator control signal is present at the output of the twelfth operational amplifier 202. The second generator control signal may include a voltage U present between the second load point 92 and the base point 32. out2' The target value. The output of the twelfth operational amplifier 202 is connected to the non-inverting input of the first operational amplifier 54 of each second converter 18 in each circuit block 14.

[0116] A method is provided for charging a superconducting magnet 12 using a charging and discharging device 10. This depends on a common input voltage U. in The voltage U existing between the first load point 88 and the base point 32 out1' It is generated by at least one first synchronous half-bridge converter 16 applying a first duty cycle D1. It also depends on the common input voltage U. in The voltage U existing between the second load point 92 and the base point 32 out2' Generated by at least one second synchronous half-bridge converter 18 with a second duty cycle D2 applied. The duty cycles of D1 and D2 are selected differently, such that U... out1' and U out2' They are different from each other. Furthermore, D1 and D2 are chosen such that, in case A), both are at most 25%, or in case B), both are at least 75%. The voltage U present between the first load point 88 and the base point 32... out1' And the voltage U present between the second load point 92 and the base point 32 out2' The voltage U is applied to the superconducting magnet 12, so that the superconducting magnet is effectively controlled by the voltage U.out1' U out2' The difference (i.e., charged by the differential-mode voltage value DM). The differential-mode voltage value DM is positive. During charging, the differential-mode voltage value DM decreases with time and / or the common-mode voltage value CM decreases with time in case A) or increases with time in case B). The voltage U present between the first load point 88 and the base point 32 is... out1' A voltage U is applied to one end of the superconducting magnet and exists between the second load point 92 and the base point 32. out2' It is applied to the other end of the superconducting magnet.

[0117] Furthermore, a method for discharging a superconducting magnet 12 using a charging / discharging device 10 is provided. This depends on a common input voltage U. in The voltage U existing between the first load point 88 and the base point 32 out1' It is generated by at least one first synchronous half-bridge converter 16 applying a first duty cycle D1. It also depends on the common input voltage U. in The voltage U existing between the second load point 92 and the base point 32 out2' Generated by at least one second synchronous half-bridge converter 18 with a second duty cycle D2 applied. The duty cycles of D1 and D2 are selected differently, such that U... out1' and U out2' They are different from each other. Furthermore, D1 and D2 are chosen such that, in case A), both are at most 25%, or in case B), both are at least 75%. The voltage U present between the first load point 88 and the base point 32... out1' And the voltage U present between the second load point 92 and the base point 32 out2' The voltage U is applied to the superconducting magnet 12, so that the superconducting magnet 12 is effectively controlled by the voltage U. out1' U out2' The difference (i.e., the differential mode voltage DM) is discharged. The differential mode voltage DM is negative. During discharge, the differential mode voltage DM increases with time and / or the common mode voltage CM increases with time in case A) or decreases with time in case B). The voltage U present between the first load point 88 and the base point 32 is... out1' A voltage U is applied to one end of the superconducting magnet 12 and exists between the second load point 88 and the base point 32. out2' It is applied to the other end of the superconducting magnet 12.

[0118] exist Figures 7a to 7d In an exemplary variant, the basic voltage V during the charging and discharging of the superconducting magnet 12 is... lim -V lim U out1' U out2' CM and DM (in volts) and the current I through the superconducting magnet 12magnet (Unit: Amperes) are plotted on the y-axis relative to time in seconds on the x-axis. For the following graph, both the first duty cycle D1 and the second duty cycle D2 are at most 25%, and the inductance of the superconducting magnet 12 is 100 Henry. The x-axis is divided into 20,000 seconds. Charging begins at 20,000 seconds and ends at 80,000 seconds. Discharging begins at 120,000 seconds and ends at 180,000 seconds. Note that in practice, regular charging and discharging will not typically occur so closely one after another, but they are shown in the same graph here for simplicity.

[0119] exist Figure 7a In the middle, the minimum magnet voltage -V lim (Dashed line) and the adjusted maximum magnet voltage V lim (Dotted line) (y-axis) is plotted relative to time (x-axis). Minimum magnet voltage -V lim The maximum adjusted magnet voltage V may be negative. lim Adjusted maximum magnet voltage V lim Depends on the maximum magnet voltage of 112. From Figure 7a It can be seen that at the start of charging at 20,000 seconds, the maximum magnet voltage V lim The voltage was set from 0.2 volts to 2 volts, and then back to 0.2 volts after 30,000 seconds. This maximum magnet voltage curve is due to the fact that the superconducting magnet 12 had no risk of quenching at the start of charging. Therefore, rapid loading is possible. Thus, the maximum magnet voltage V at the start of charging... lim Set to 2 volts. The desired / maximum current flowing through the superconducting magnet 12 is Imagnet ( Figure 7d After establishing 60% to 75% (66% in this case), the superconducting magnet 12 becomes highly sensitive to quench loss. Therefore, the voltage V... lim It drops to 0.2 volts. Correspondingly, before the discharge ends, the maximum magnet voltage V... lim The voltage was set from 0.2 volts to 2 volts at 170,000 seconds and then set back to 0.2 volts at the end of the discharge at 180,000 seconds.

[0120] exist Figure 7b In the middle, voltage U out1' (dashed line), U out2' (Dotted line) (y-axis) is plotted relative to time (x-axis). From Figure 7b It can be seen that the voltage U starts at the beginning of charging after 20,000 seconds. out1' The voltage rises from 0.1 volts to 2.1 volts, decreases from 2.1 volts to 0.3 volts at 30,000 seconds, and then decreases again from 0.3 volts to 0.1 volts at 80,000 seconds. Furthermore, the voltage U at the start of discharge at 120,000 seconds...out2' The voltage rises from 0.1 volts to 0.3 volts, and then from 0.3 volts to 2.1 volts over 170,000 seconds. At the end of the discharge at 180,000 seconds, the voltage at the second load point drops from 2.1 volts to 0.1 volts.

[0121] exist Figure 7c In the graph, the common-mode voltage value CM (y-axis) is plotted relative to time (x-axis). From... Figure 7b voltage U out1' U out2' Export, Figure 7c The common-mode voltage CM rises from 0.1 volts to 1.1 volts at the start of charging (20,000 seconds), falls from 1.1 volts to 0.2 volts at 30,000 seconds, and falls again from 0.2 volts to 0.1 volts at 80,000 seconds. Furthermore, the common-mode voltage CM rises from 0.1 volts to 0.2 volts at the start of discharging (120,000 seconds), and rises again from 0.2 volts to 1.1 volts at 170,000 seconds. At the end of discharging (180,000 seconds), the common-mode voltage CM falls from 1.1 volts to 0.1 volts. Note that all CM values ​​here are relative to U. in (Here, 12V) The maximum is approximately 9%; it should be noted that, according to the present invention, it is generally preferred that CM ≤ 25% or CM ≤ 20% (Case A) or CM ≥ 75% or CM ≥ 80% (Case B).

[0122] exist Figure 7d In the middle, the differential mode voltage value DM (dashed line) and the current I flowing through the superconducting magnet 12 are shown. magnet (Solid line) (y-axis) is plotted relative to time (x-axis). From Figure 7b voltage U out1' U out2' Export, Figure 7d The differential mode voltage DM rises from 0 volts to 2 volts at the start of charging (20,000 seconds) to initiate rapid charging, drops from 2 volts to 0.2 volts at 30,000 seconds to initiate slow charging, and drops again from 0.2 volts to 0 volts at 80,000 seconds, thus stopping charging. Similarly, at the start of discharging (120,000 seconds), the differential mode voltage DM drops from 0 volts to -0.2 volts to initiate slow discharging, and drops again from -0.2 volts to -2 volts at 170,000 seconds to initiate rapid discharging. At the end of discharging (180,000 seconds), the differential mode voltage DM rises from -2 volts to 0 volts. The current I flowing through the superconducting magnet 12 during charging... magnetThe current rapidly increases from 0 amps at 20,000 seconds to 200 amps at 30,000 seconds, and then further but more slowly increases to 300 amps at 80,000 seconds. During discharge, the current slowly decreases from 300 amps at 120,000 seconds to 200 amps at 170,000 seconds, and then further and more rapidly decreases to 0 amps at 180,000 seconds. Note that all DM values ​​here adhere to approximately |DM| ≤ 9% (with U) at all times. in (In comparison); note that typically |DM|≤25% or |DM|≤20%.

[0123] In summary, a charging and discharging device 10 for a superconducting magnet 12 is disclosed, comprising at least one circuit block 14, each circuit block 14 having first and second synchronous half-bridge converters 16, 18, wherein for each such converter 16, 18, a control signal generator 48 is provided for outputting control signals for driving the switches 20 of the corresponding converter 16, 18 to open and close, and for outputting another control signal for driving another switch 22 of the converter 16, 18 to open and close, wherein the control signals apply duty cycles D1, D2, and the other control signal... The switch 20 is opened when the other switch 22 is closed, and vice versa, by applying the inversion of the duty cycles D1 and D2, respectively. Specifically, each such converter 16, 18 includes its own control signal generator 48, wherein in each circuit block 14, the duty cycle D1 (also referred to as the first duty cycle D1) of the control signal of the first converter 16 and the duty cycle D2 (also referred to as the second duty cycle D2) of the control signal of the second converter 18 are different from each other, such that when a common input voltage U is applied between the input point 28 and the base point 32 at the two converters 16, 18... in At that time, the first output voltage U exists between the connection point 40 and the base point 32 of the first converter 16. out1 The second output voltage U between the connection point 40 and the base point 32 of the second converter 18 out2 They are different from each other, and the first duty cycle D1 and the second duty cycle D2 are both at most 25% in case A) or at least 75% in case B). The charging and discharging device is simple and inexpensive in design and control.

[0124] List of reference numerals

Claims

1. A charging and discharging device (10) for a superconducting magnet (12). It has at least one circuit block (14), each circuit block (14) having a first and a second synchronous half-bridge converter (16, 18). Each converter (16, 18) includes - A series connection of switch (20) and another switch (22) between input point (28) and base point (32), wherein switch (20) and another switch (22) are connected at confluence point (30). - A series connection of filter inductor (38) and filter capacitor (36), wherein filter inductor (38) and filter capacitor (36) are connected at connection point (40), and wherein filter inductor (38) is connected to confluence point (30) and filter capacitor (36) is connected to base point (32). And for each of these converters (16, 18), there is a... - A control signal generator (48) is used to output control signals for driving the switch (20) to open and close, and to output another control signal for driving the other switch (22) to open and close, wherein the control signal applies a duty cycle (D1, D2), and the other control signal applies the inverse of the duty cycle (D1, D2), such that the switch (20) opens when the other switch (22) is closed, and vice versa. In particular, each of these converters (16, 18) includes its own control signal generator (48). In each circuit block (14), the duty cycle D1 (also referred to as the first duty cycle D1) of the control signal of the first converter (16) and the duty cycle D2 (also referred to as the second duty cycle D2) of the control signal of the second converter (18) are different from each other, such that when a common input voltage U is applied between the input point (28) and the base point (32) of the two converters (16, 18)... in At that time, the first output voltage U exists between the connection point (40) and the base point (32) of the first converter (16). out1 The second output voltage U between the connection point (40) and the base point (32) of the second converter (18) out2 They are different from each other, and the first duty cycle D1 and the second duty cycle D2 are both at most 25% in case A) or at least 75% in case B).

2. The charging and discharging device (10) according to claim 1, characterized in that... A first transition point (84) is provided, wherein the connection point (40) of each first converter (16) of each circuit block (14) is directly or indirectly connected to the first transition point (84), and a second transition point (86) is provided, wherein the connection point (40) of each second converter (18) of each circuit block (14) is directly or indirectly connected to the second transition point (86). A first load point (88) is provided for connecting the first end of the superconducting magnet (12), and a second load point (92) is provided for connecting the second end of the superconducting magnet (12). The first transition point (84) is connected to the first load point (88) via the first load connection (90), and the second transition point (86) is connected to the second load point (92) via the second load connection (94).

3. The charging and discharging device (10) according to claim 1 or 2, characterized in that... The charging and discharging device (10) also includes a control unit (96) that provides a first generator control signal to the control signal generator (48) of each first converter (16) of each circuit block (14) and a second generator control signal to the control signal generator (48) of each second converter (18) of each circuit block (14).

4. The charging and discharging device (10) according to claim 2 and further according to claim 3, characterized in that... A current measuring device (190) is included in either the first load connection (90) or the second load connection (94). Furthermore, the signal output (188) of the current measuring device (190) is connected to the control unit (96) via the current measuring line (192).

5. The charging and discharging device (10) according to claim 4, characterized in that... The control unit (96) is configured to provide generator control signals depending on the charge and discharge progress of the superconducting magnet (12), wherein the charge and discharge progress of the superconducting magnet (12) is monitored by a current measuring device (190).

6. The charging and discharging device (10) according to claim 4 or 5, characterized in that... The first potential measurement line (142) connects the first load point (88) to the control unit (96), and the second potential measurement line (134) connects the second load point (92) to the control unit (96). The control unit (96) has a first setting input (116) for setting the maximum magnet voltage (112) to be applied to the superconducting magnet (12) and a second setting input (180) for setting the current target (178) of the superconducting magnet (12). Furthermore, the control unit (96) is configured to generate the first generator control signal and the second generator control signal by taking into account the following: - Set the current target (178). - Set the maximum magnet voltage (112). - The current (I) currently operating in the superconducting magnet (12) is indicated by the current measurement line (192). magnet ) signals, and - Signals from the first potential measurement line (142) and the second potential measurement line (134) indicating the voltage DM currently applied to the superconducting magnet (12).

7. The charging and discharging device (10) according to any one of claims 3 to 6, characterized in that... Define the common modulus value CM, where CM = (Uout1 + Uout2) / 2. Furthermore, the control unit (96) is configured to provide a first generator control signal and a second generator control signal such that the common mode value CM changes over time during the charging and discharging of the superconducting magnet (12) using the charging and discharging device (10).

8. The charging and discharging device (10) according to any one of claims 3 to 7, characterized in that... Define the differential modulus value DM, where DM = Uout1 - Uout2. Furthermore, the control unit (96) is configured to provide a first generator control signal and a second generator control signal such that the differential mode value DM changes over time during the charging and discharging of the superconducting magnet (12) using the charging and discharging device (10).

9. The charging and discharging device (10) according to any one of claims 3 to 8, characterized in that... Define the differential modulus value DM, where DM = Uout1 - Uout2. Furthermore, the control unit (96) is configured to provide a first generator control signal and a second generator control signal, such that during the charging and discharging of the superconducting magnet (12) using the charging and discharging device (10), the absolute value of the differential mode |DM| is applied. |DM|≤0.25*|Uin|, preferably |DM|≤0.20*|Uin|.

10. The charging and discharging device (10) according to any one of the preceding claims, characterized in that... The charging and discharging device (10) includes multiple circuit blocks (14). All input points (28) of all converters (16, 18) of all circuit blocks (14) are connected to a common input stage, and all base points (32) of all converters (16, 18) of all circuit blocks (14) are connected to a common base stage.

11. The charging and discharging device (10) according to any one of the preceding claims, characterized in that... Each of these converters (16, 18) includes its own control signal generator (48). Each converter (16, 18) includes a feedback line (60) connected to the connection point (40) and the control signal generator (48), and the control signal generator (48) is configured to take into account the feedback signal received via the feedback line (60) to determine the duty cycle D1, D2 of its control signal.

12. The charging and discharging device (10) according to any one of the preceding claims, characterized in that... Each converter (16, 18) also includes a series connection of a decoupling inductor (42) and a decoupling capacitor (44), wherein the decoupling inductor (42) and the decoupling capacitor (44) are connected at another connection point (46), and wherein the decoupling inductor (42) is connected to the connection point (40) and the decoupling capacitor (44) is connected to the base point (32). Specifically, another connection point (46) of each first converter (16) of each circuit block (14) is connected to a first transition point (84), and another connection point (46) of each second converter (18) of each circuit block (14) is connected to a second transition point (86).

13. The use of the charging / discharging device (10) according to any one of claims 1 to 12 for charging or discharging a superconducting magnet (12), The common input voltage U in A first output voltage U is generated between the input point (28) and the base point (32) of each converter (16, 18) in each circuit block (14), such that a first output voltage U is generated between the connection point (40) and the base point (32) of each first converter (16) in each circuit block (14). out1 And a second output voltage U is generated between the connection point (40) and the base point (32) of each second converter (18) in each circuit block (14). out2 , Furthermore, the superconducting magnet (12) is connected to the charging and discharging device (10), causing the voltage U to... out1 and U out2 The difference DM effectively charges the superconducting magnet (12). Specifically, the charging and discharging device (10) is configured according to claim 2, and the superconducting magnet (12) is connected between the first load point (88) and the second load point (92).

14. A method for charging a superconducting magnet (12), Specifically, the charging and discharging device (10) according to any one of claims 1 to 12 is used. include: - Utilizing at least one first synchronous half-bridge converter (16) from the common input voltage U in Generate the first output voltage U out1 The first duty cycle D1 is applied. - Using at least one second synchronous half-bridge converter (18) from the common input voltage U in Generate the second output voltage U out2 A second duty cycle D2 is applied. - By choosing D1 and D2 differently, U out1 and U out2 They are different. - Choose D1 and D2, where in case A) both D1 and D2 are at most 25%, or in case B) both D1 and D2 are at least 75%. - will U out1 and U out2 Applying it to the superconducting magnet (12) so that the superconducting magnet (12) is effectively subjected to U out1 and U out2 The difference in DM charging, specifically, where U out1 and U out2 It is applied to different ends of the superconducting magnet (12).

15. The method according to claim 14, characterized in that... Define the common modulus value CM, where CM = (Uout1 + Uout2) / 2. Furthermore, during the charging of the superconducting magnet (12), the common-mode value CM changes with time. Specifically, the common-mode value CM decreases with time in case A) or increases with time in case B), depending particularly on the measured charging progress.

16. A method for discharging a superconducting magnet (12), Specifically, the charging and discharging device (10) according to any one of claims 1 to 12 is used. include: - Utilizing at least one first synchronous half-bridge converter (16) from a common input voltage U in Generate the first output voltage U out1 The first duty cycle D1 is applied. - Using at least one second synchronous half-bridge converter (18) from the common input voltage U in Generate the second output voltage U out2 A second duty cycle D2 is applied. - By choosing D1 and D2 differently, U out1 and U out2 They are different. - Choose D1 and D2, where in case A) both D1 and D2 are at most 25%, or in case B) both D1 and D2 are at least 75%. - will U out1 and U out2 Applying it to the superconducting magnet (12) so that the superconducting magnet (12) is effectively subjected to U out1 and U out2 The difference in DM discharge, particularly, where U out1 and U out2 It is applied to different ends of the superconducting magnet (12).

17. The method according to claim 16, characterized in that... Define the common modulus value CM, where CM = (U out1 +U out2 ) / 2, Furthermore, during the discharge of the superconducting magnet (12), the common-mode value CM changes with time. Specifically, the common-mode value CM increases with time in case A) or decreases with time in case B), depending particularly on the measured discharge progress.

Images