Pulsed power source for a satellite propulsion system and corresponding methods

Abstract

A pulsed power source (200) for a spacecraft propulsion system, the pulsed power source comprising: a voltage input (210); a plurality of capacitor banks (226, 236, 246, 256) arranged to be charged by the voltage input; pulse discharge connections for electrically coupling each capacitor bank to a load (260); a discharge switch (228, 238, 248, 258) for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and an isolation switch (220, 230, 240, 250) for each capacitor bank, each isolation switch positioned on a charge path between the voltage input and the capacitor bank.

Description

[0001] SATELLITE PROPULSION SYSTEM AND METHODS

[0002] FIELD OF THE INVENTION

[0003] The present application relates to spacecraft propulsion systems and methods. In particular, the application relates to systems, apparatus and methods for creating pulsed power, which has particular applications for satellite propulsion.

[0004] BACKGROUND OF THE INVENTION

[0005] Pulsed Plasma Thrusters (PPT) are an example of an electric propulsion means. These types of thruster use stored energy to generate large bursts, or pulses, of electrical energy that are directed at a propellant fuel source to generate a plasma propellant. The plasma propellant can then be accelerated, e.g. by the Lorentz force, and emitted from the thruster (e.g. out of an exhaust of a spacecraft) to generate thrust.

[0006] PPTs have been in use since the 1960s, and although they typically have a low efficiency (typically around of less than 10%) they are used due to their generally good reliability. This is due to the absence of tanks, piping, and moving parts. They have found particular applicability in smaller satellites, such as a CubeSat (10 x 10 x 10 cm).

[0007] There is a desire to create a more efficient and reliable plasma thruster. One of the ways this could be done is by improving the production of the electrical energy pulses.

[0008] SUMMARY OF THE INVENTION

[0009] Aspects of the invention are set out in the independent claims and preferable features are set out in the dependent claims.

[0010] There is described herein a pulsed power source for a spacecraft propulsion system, the pulsed power source comprising: a voltage input; a plurality of capacitor banks arranged to be charged by the voltage input; pulse discharge connections for electrically coupling each capacitor bank to a load; a discharge switch for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and an isolation switch for each capacitor bank, each isolation switch positioned on a charge path between the voltage input and the capacitor bank.

[0011] Advantageously, by providing an isolation switch for each capacitor bank it is possible to isolate faulty capacitor banks from the rest of the circuit and prevent damage or malfunctioning of other parts of the circuit.

[0012] Activating the discharge switch for one or more of the banks can cause a high voltage current through the load and result in a pulse of electric power. The pulse of electric power can result in an arc of electricity. Preferably this pulse is used to generate and / or accelerate a plasma, which may be used for spacecraft propulsion.

[0013] The voltage input is preferably a high voltage input. The voltage input can have a voltage of at least 500V, preferably at least 1 ,000V or around 1 ,500 V. In some embodiments high voltage input is less than 50,000V or less than 20,000V, preferably less than 5,000V. The voltage input is generally a direct current, DC, voltage input.

[0014] The capacitor banks may each be provided on a separate branch of the circuit, or in other words in parallel to the other capacitor banks in the circuit. Where this is the case, the isolation switch and discharge switch for each capacitor bank would be positioned on the branch for the relevant capacitor bank.

[0015] In some embodiments the circuit may be arranged such that the capacitor banks can be charged by the voltage input in parallel and can be discharged across the load in series.

[0016] Preferably there is one set of discharge connections provided for each capacitor bank, however in alternative embodiment there may be a single set of discharge connections shared by a plurality, or even all, of the capacitor banks.

[0017] When the discharge switch is non-conducting (e.g. “off’, or open) it prevents current flowing from the capacitor bank to the load. When the discharge switch is in a conducting state (e.g. “on”, or closed) it allows current to flow from the capacitor bank to the load. Preferably the load has a very low resistance, so allowing current to flow, or discharge, from the capacitor bank to the load results in a large, short discharge current, which forms a pulse of electrical power.

[0018] The pulsed power source may further comprise the load, or the load may be provided separately and be connectable to the pulsed power source. The pulsed power source may further comprise: a controller configured to switch the isolation switch for each capacitor bank to a non-conducting state to isolate the respective capacitor bank from the voltage input.

[0019] The controller is configured to switch the isolation switches for different capacitor banks independently so that each capacitor bank may be isolated individually upon the controller sending a control signal to cause the isolation switch for a specific capacitor bank to switch to a non-conducting state. The respective capacitor bank refers to the capacitor bank for which the isolation switch is positioned between that capacitor bank and the voltage input.

[0020] Preferably, the controller is configured to switch the isolation switch to a non-conducting state to isolate the respective capacitor bank from the voltage input upon detection of a short circuit fault associated with the said respective capacitor bank, such as a short circuit fault in the discharge switch for the said respective capacitor bank. Short circuit faults can cause the entire circuit to malfunction, since when connected to a voltage input, instead of charging the capacitor banks the voltage input may discharge straight through the load. By isolating a capacitor bank that is associated with a short circuit fault from the voltage input, this can be avoided so the rest of capacitor banks in the circuit can function normally.

[0021] Optionally, the controller is configured to detect a short circuit fault during a capacitor bank charging mode in which the said respective capacitor bank is electrically connected to the voltage input for charging the said respective capacitor bank.

[0022] In the charging mode control signals may be sent from the controller to cause one or more of the capacitor banks to be charged by the voltage input. In some embodiments in the charging mode the capacitor banks are disconnected from the load, for example by opening the discharge switch for each capacitor bank so the discharge switch does not conduct electrical current. If the discharge switch has failed closed (which would result in a short circuit fault), the capacitor bank would remain electrically connected to the load even in a charging mode. A charging mode may be classified as a mode in which the control signals from the controller are intended to disconnect the capacitor banks from the load, for example by sending a control signal to open the discharge switches, or otherwise place the discharge switches in a non-conducting configuration. In some embodiments the controller may be configured to disconnect the capacitor bank from the load by failing to apply a trigger to switch the discharge switch into a conducting state. In some embodiments the controller is configured to detect a short circuit fault associated with a capacitor bank by monitoring the charging rate of the said capacitor bank.

[0023] In some embodiments the controller is configured to detect a short circuit fault associated with a capacitor bank if, during the capacitor bank charging mode, the voltage across the said capacitor bank fails to increase.

[0024] The controller may be configured to detect a short circuit fault associated with the said capacitor bank if, during the capacitor bank charging mode, the voltage across the said capacitor bank fails to increase above a threshold voltage. The controller may, for example, be configured to detect a short circuit fault associated with the said capacitor bank if, during the capacitor bank charging mode, the voltage across the said capacitor bank fails to increase within a threshold charge time.

[0025] The controller may be configured to detect a short circuit fault associated with a capacitor bank by monitoring current flowing from the said capacitor bank to the load, for example on the current through the discharge path between the said capacitor bank and the pulse discharge connections. Current flowing from the capacitor bank to the load during a charging mode indicates there is a short circuit fault that means the bank is electrically connected to the load and so will not charge even when electrically connected to the voltage input.

[0026] In some embodiments a short circuit fault associated with a capacitor bank may be detected by monitoring current flowing through the discharge switch for the respective capacitor bank.

[0027] Preferably the controller is additionally or alternatively configured to switch the isolation switch to a non-conducting state to isolate the respective capacitor bank from the voltage input upon detection of an open circuit fault associated with the said respective capacitor bank, optionally an open circuit fault in the discharge switch for the said respective capacitor bank. If there is an open circuit fault associated with a capacitor bank, charge may not be released from the capacitor bank and the capacitor bank may discharge unpredictably, causing damage to other components in the circuit.

[0028] In some embodiments, the controller is configured to detect an open circuit fault by monitoring the discharge rate of the said capacitor bank. The controller may be configured to detect a short circuit fault during a capacitor bank discharging mode in which one or more control signals are sent to cause the respective capacitor bank to be electrically connected to the load for discharging the respective capacitor bank. In the capacitor bank discharging mode the capacitor banks are intended to discharge charge stored on the capacitor to the load. Thus to initiate a discharging mode the controller may send a control signal to trigger the discharge switch for each capacitor bank to put the discharge switch in a conducting state (e.g. to close the discharge switch).

[0029] In some embodiments the controller is configured to detect an open circuit fault associated with the said respective capacitor bank if, during the capacitor bank discharging mode, the voltage across the said respective capacitor bank fails to decrease.

[0030] Preferably the controller is configured to detect an open circuit fault associated with the said respective capacitor bank if, during the capacitor bank discharging mode, the voltage across the said respective capacitor bank fails to decease by more than a threshold voltage decrease or fails to decrease to below a threshold voltage.

[0031] The controller may be configured to detect an open circuit fault associated with the said respective capacitor bank if, during the capacitor bank discharging mode, the voltage across the said respective capacitor bank fails to decrease in a threshold discharge time.

[0032] Optionally, the controller is configured to detect an open circuit fault associated with the said respective capacitor bank if, during the capacitor bank discharging mode, the voltage across the said respective capacitor bank is greater than the voltage across one or more of the other capacitor banks, preferably greater than the voltage across one or more of the other capacitor banks by more than a threshold voltage difference.

[0033] Optionally, the controller is configured to detect a short circuit fault and / or an open circuit fault associated with a capacitor bank by monitoring the voltage across said capacitor bank. In alternative embodiments, faults may be detected by monitoring the voltage differential between two or more of the plurality of capacitor banks. The controller may be configured to detect a fault (and preferably thus to open the isolation switch for a capacitor bank) if the voltage differential indicates the voltage across that capacitor bank is lower or higher than the voltage across one or more of the other capacitor banks in the plurality of capacitor banks.

[0034] The pulsed power source may further comprise: a diode for each capacitor bank, the diode positioned on the charge path between the voltage input and the capacitor bank and configured to allow a charging current to flow from the voltage input to the capacitor bank and to prevent current flow from the capacitor bank to other capacitor banks of the plurality of capacitor banks.

[0035] The term diode is intended to encompass any two-terminal rectifying device that conducts current mainly in one direction, such as a diode-coupled transistor. In preferred examples a Schottkey diode is used.

[0036] The capacitor banks will generally each be positioned on a separate branch of the circuit (in parallel to other capacitor banks), and the diode for each capacitor bank is preferably positioned on the branch for the relevant capacitor bank.

[0037] There is also described herein: a pulsed power source for a spacecraft propulsion system, the pulsed power source comprising: a voltage input; a plurality of capacitor banks arranged to be charged by the voltage input; pulse discharge connections for electrically coupling each capacitor bank to a load; a discharge switch for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and a diode for each capacitor bank, the diode positioned on the charge path between the voltage input and the capacitor bank and configured to allow a charging current to flow from the voltage input to the capacitor bank and to prevent current flow from the capacitor bank to other capacitor banks of the plurality of capacitor banks.

[0038] Advantageously, by providing a diode it is possible to prevent the charge stored on a capacitor bank from discharging to the rest of the circuit and causing faults in other parts of the circuit and / or charging up other capacitor banks in the circuit.

[0039] Preferably, the diode for each capacitor bank is positioned between the isolation switch and the capacitor bank and is configured to prevent current flow from the capacitor bank to the isolation switch.

[0040] The pulsed power source may further comprise: a controller configured to control the state of the discharge switches for the capacitor banks. This controller may be the same controller that controls the isolation switches (where present), but in some embodiments separate means (e.g. a separate controller) for controlling any isolation switches may be provided.

[0041] Preferably, the controller is configured to control the state of each discharge switch independently of the other discharge switches of the plurality of discharge switches. There is also described herein: a pulsed power source for a spacecraft propulsion system, the pulsed power source comprising: a voltage input; a plurality of capacitor banks arranged to be charged by the voltage input; pulse discharge connections for electrically coupling each capacitor bank to a load; a discharge switch for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and a controller configured to control the state of the discharge switches for the capacitor banks, wherein the controller is configured to control the state of each discharge switch independently of the other discharge switches of the plurality of discharge switches.

[0042] Generally there is a plurality of discharge switches corresponding to the plurality of capacitor banks. The state of each switch is either conducting (e.g. “on”) or non-conducting (e.g. “off’). When the switch is on, it is in a conducting mode and capable of conducting current, and when it is off it is not capable of conducting current.

[0043] Preferably, the controller is configured to control the state of the discharge switches by triggering each discharge switch using a different trigger signal. Triggering a discharge switch involves putting the switch into a conducting mode, e.g. turning the switch on, or closing the switch.

[0044] Preferably, the controller is configured to trigger a first discharge switch for a first capacitor bank of the plurality of capacitor banks without triggering the discharge switch of any other capacitor banks of the plurality of capacitor banks.

[0045] Preferably, the controller is configured to trigger a first discharge switch for a first capacitor bank of the plurality of capacitor banks at a first time and to trigger a second discharge switch for a second capacitor bank of the plurality of capacitor banks at a second time, different from the first time.

[0046] Optionally, the controller is configured to control the pulsed power source to shape a pulse by triggering the discharge switches for each of the plurality of capacitor banks over a pulse trigger time period, the pulse trigger time period being the time interval between triggering the first triggered discharge switch and triggering the last triggered discharge switch.

[0047] Pulse shaping may also be described as modulating. By triggering the discharge switches at different times (e.g. consecutively), the energy stored in the capacitor banks can be released (or in other words, discharged) sequentially (or in other words, at different times). The shape of the pulse can thus be altered and this can have advantages in this application. For example, when used in a plasma propulsion system, a first initial part of the pulse may cause a plasma to form and a second part of the pulse may cause the plasma to be accelerated.

[0048] The pulse trigger time period may be at least 10ns and up to around 2ps, preferably at least around 100ns and / or up to around 1 ,5ps.

[0049] Preferably the controller is configured to control the pulsed power source to form pulses of at least a first pulse shape and a second pulse shape, different from the first pulse shape, by altering the relative timing of triggering of the discharge switches for two or more of the plurality of capacitor banks.

[0050] Any of the pulsed power sources described above may also comprise: a slow discharge path for each capacitor bank. This can allow capacitor banks with open circuit faults to discharge safely.

[0051] Any of the pulsed power sources described above may also may further comprise: sensing circuitry configured to measure the voltage across each capacitor bank. The measure of the voltage may be an absolute value of the voltage across the capacitor bank or may be a relative value of the voltage across the capacitor bank compared to the voltage across one or more of the other capacitor banks.

[0052] Where there is a controller, the controller may be configured to received signals indicative of the voltage across each capacitor bank. The sensing circuitry preferably feeds sends a signal indicative of the measured voltage to the controller, where. The controller can use the received measured voltage to detect faults, such as open- or short-circuit faults and / or it can be used to decide whether to open the isolation switch for the capacitor bank.

[0053] Any of the pulsed power sources described herein may further comprise: a charge resistor for each capacitor bank positioned between the voltage input and the capacitor bank. Such a charge resistor may have a resistance of at least 80 ohms, preferably at least 100 ohms or at least 500 ohms. The charge resistor may have a resistance as great as 1 gigaohm or greater. In some embodiments the resistance of the charge resistor is less than around 10 gigaohms or less than around 5 gigaohms.

[0054] In some embodiments, each capacitor bank comprises at least one capacitor, preferably at least five capacitors, more preferably at least 10 capacitors. Generally each bank will have fewer than 100 capacitors or fewer than 50 capacitors. In any of the pulsed power sources described herein the plurality of capacitor banks may comprise at least four capacitor banks, optionally at least 10 capacitor banks. Generally there will be fewer than 100 or fewer than 50 capacitor banks. In some embodiments fewer than or equal to 20 or fewer than or equal to 10 capacitor banks.

[0055] The pulsed power sources described herein may be configured such that each capacitor bank has a capacitance of at least around 0.1 pF and not more than around 10pF, preferably at least 0.5pF and / or not more than around 5pF. Preferably the capacitance of each capacitor bank is around 1 pF, for example between 0.5 pF and 2 pF.

[0056] The pulsed power sources described herein preferably have a total capacitance across the plurality of capacitor banks of at least 1 pF and not more than 100pF.

[0057] The voltage input is preferably a high voltage input. For example, the voltage input may be at least 500V, preferably at least 1 kV. Generally the voltage input would be less than 1 MV. The voltage input may be less than around 100kV or even less than 10kV.

[0058] Preferably the charging time constant of each capacitor bank is between around 0.5ms and 5ms. Additionally or alternatively, the time constant of each capacitor bank is at least around 1ms and / or not more than around 3ms or around 2ms. The time constant of a resistorcapacitor series combination is indicative of the time it takes for the capacitor bank to reach 63.2% of its maximum charge capacity (assuming it has no initial charge). The charging time constant, T, of a capacitor bank, can be determined from the capacitance, C, of the capacitor bank and the resistance in series with the capacitor bank in the charging path.

[0059] Any of the pulsed power sources described above may be configured to generate electrical pulses having an energy of at least 1J, preferably at least 3J, and / or not more than 50J, preferably not more than 25J. The energy of the electrical pulses may be the same as, or roughly equal to (within operational tolerances), the energy stored in the plurality of capacitor banks. The maximum energy, E, stored in a capacitor bank can be calculated from the voltage applied across the capacitor bank, V, and the capacitance of the capacitor bank, C. E =1 / 2CV2.

[0060] Any of the pulsed power sources herein may have discharge switches for each capacitor bank that can be triggered by an electrical current signal.

[0061] The discharge switch for each capacitor bank is preferably a transistor. The discharge switches described above are preferably semiconductor switches. Semiconductor switches such as thyristors, IGBTs or MOSFETs (metal-oxide-semiconductor field-effect transistors) may be used as the discharge switch.

[0062] In preferred embodiments the isolation switch is a semiconductor switch, preferably a MOSFET (metal-oxide-semiconductor field-effect transistor). Preferably the MOSFET is an n-channel MOSFET, configured to use NMOS logic. However in other embodiments the MOSFET may be a p-channel MOSFET, configured to use PMOS logic.

[0063] There is also described herein: a plasma thruster for a spacecraft propulsion system, the plasma thruster comprising: a pulsed power source according to any preceding claim; and a plasma generator coupled to the load and arranged to receive electric current pulses from the pulsed power source and to form plasma using one or more electric current pulses.

[0064] The plasma generator may be arranged to use solid material, such as solid metal, as a propellant. The propellant is used to create the plasma. Alternatively liquid or gaseous propellants may be used to generate the plasma. In some examples, such as for a cathode arc thruster, the propellant is the cathode itself. Preferably the plasma generator is also arranged to accelerate the plasma using one or more electric current pulses. The same pulse may be used to generate and accelerate the plasma.

[0065] A method of isolating capacitor banks in a pulsed power source for spacecraft propulsion is also described herein, the method comprising: sending a control signal to cause one or more of a plurality of capacitor banks to be charged by a voltage input; monitoring each of the one or more capacitor banks for one or both of a short circuit fault and an open circuit fault associated with the one or more capacitor banks; and isolating one of the one or more of the capacitor banks a short circuit fault or an open circuit fault associated with the said one of the one or more of the capacitor banks is detected by sending a control signal to switch an isolation switch for the said one of the one or more of the capacitor banks to a nonconducting state, the isolation switch for each capacitor bank being positioned on a charge path between the voltage input and the capacitor bank.

[0066] Sending a control signal to cause a capacitor bank to be charged by a voltage input may be a control signal to initiate a charging mode, for example to put the one or more capacitor banks into a charging mode.

[0067] The method may further comprise: sending a control signal to cause at least one of the one or more of the plurality of capacitor banks to be electrically connected to a load to generate an electric power pulse. This may be performed before or after the step of isolating the one of the one or more capacitor banks.

[0068] Preferably the control signal to cause the at least one of the one or more of the plurality of capacitor banks to be electrically connected to a load is a control signal to switch a discharge switch for each of the at least one of the one or more capacitor banks to a conducting state to electrically connect the respective capacitor bank to the load, each discharge switch positioned on a discharge path between the respective capacitor bank and the load.

[0069] In some embodiments, monitoring each of the one or more capacitor banks for a short circuit fault comprises monitoring a charging rate of each of the one or more capacitor banks.

[0070] In some embodiments, monitoring each of the one or more capacitor banks for an open circuit fault comprises monitoring a discharge rate of each of the one or more capacitor banks.

[0071] Additionally or alternatively, monitoring each of the one or more capacitor banks for an open circuit fault or a short circuit fault comprises monitoring the voltage across each of the one or more of the capacitor banks, preferably wherein a short circuit fault associated with a capacitor bank is detected if the voltage across said capacitor bank does not decrease upon a control signal for electrically connecting said capacitor bank to the load being sent.

[0072] The step of monitoring the voltage across each capacitor bank may continue into the discharge phase or mode, when the capacitor banks are electrically connected to the load.

[0073] There is also described herein: a method of generating pulsed power for spacecraft propulsion, the method comprising: sending a control signal to cause a plurality of capacitor banks to be charged by a voltage input; and generating an electric power pulse by closing a discharge switch for each capacitor bank to electrically connect the capacitor bank to a load, each discharge switch positioned on a discharge path between the capacitor bank and pulse discharge connections connected to the load, wherein a first discharge switch of the plurality of discharge switches for a first capacitor bank is closed at a first time and a second discharge switch of the plurality of discharge switches is closed at a second time, wherein the first time is different from the second time.

[0074] Preferably the first and second times are separated by a predetermined time period. Optionally, the plurality of capacitor banks includes a first set of capacitor banks comprising the first capacitor bank and a second set of capacitor banks comprising the second capacitor bank, and wherein the first set of capacitor banks is operable to store a first total charge and the second set of capacitor banks is operable to store a second total charge, wherein the first total charge is smaller than the second total charge; and wherein generating an electric power pulse comprises electrically connecting discharge switches for each capacitor bank in the first set of capacitor banks at the first time and electrically connecting discharge switches for each capacitor bank in the second set of capacitor banks at the second time. Generally the first time is earlier than the second time.

[0075] Each set of capacitor banks may include one or more capacitor banks. In some embodiments the first set of capacitor banks includes fewer capacitor banks than the second set of capacitor banks, for example if the capacitance of each of the capacitor banks is the same. The total charge stored by a capacitor bank may be dependent on the capacitance, C, of the capacitor bank and the voltage input, V. The total charge, Q, a capacitor bank is operable to store may be determined by Q=CV.

[0076] By discharging a smaller charge (stored on the first set of capacitor banks) before discharging a larger charge (stored on the second set of capacitor banks) it is possible to provide a pulse with a smaller first burst of energy and a larger second burst of energy. This can be advantageous where the power source is used in a plasma thruster as the smaller first burst can be used to generate a plasma and the second larger burst can be used to accelerate the plasma.

[0077] There is also described herein: a controller for a pulsed power source, the controller arranged to receive one or more monitoring signals and send control signals; wherein the controller is operable to perform a method substantially as described above.

[0078] A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to cause a said pulsed power source to carry out any of the methods described above.

[0079] Any system feature as described herein may also be provided as a method feature, and vice versa. As used herein, means plus function features may be expressed alternatively in terms of their corresponding structure. Any feature in one aspect of the invention may be applied to other aspects of the invention, in any appropriate combination. In particular, method aspects may be applied to system aspects, and vice versa. Furthermore, any, some and / or all features in one aspect can be applied to any, some and / or all features in any other aspect, in any appropriate combination.

[0080] It should also be appreciated that particular combinations of the various features described and defined in any aspects of the invention can be implemented and / or supplied and / or used independently.

[0081] BRIEF DESCRIPTION OF THE FIGURES

[0082] Methods and systems for generating electrical pulses for pulsed plasma thrusters are described by way of example only, in relation to the Figures, wherein:

[0083] Figure 1 shows an example high voltage circuit for a pulsed power source;

[0084] Figure 2 shows another example high voltage circuit for a pulsed power source;

[0085] Figure 3 illustrates an example method of protecting a pulsed power source during a charging phase;

[0086] Figure 4 illustrates an example method of protecting a pulsed power source during a discharging phase;

[0087] Figure 5 illustrates an example method of discharging a pulsed power source;

[0088] Figure 6 illustrates an example of a pulse discharge from a pulsed power source;

[0089] Figure 7 illustrates another example of a pulse discharge from a pulsed power source;

[0090] Figure 8 illustrates another example of a pulse discharge from a pulsed power source; and

[0091] Figure 9 illustrates an example of a controller for a pulsed power source.

[0092] DETAILED DESCRIPTION Referring to Figure 1 , a high voltage circuit 100 for a pulsed power source will now be described. The high voltage circuit 100 may be used as part of a Pulsed Power System (PPS) for a pulsed plasma thruster (PPT), which may be used for spacecraft propulsion, for example for a satellite.

[0093] The high voltage circuit 100 comprises a high voltage input 110. The high voltage input 110 in this case is a 1500 Volt (V) power source. The high voltage circuit 110 also comprises at least one capacitor bank 126. The capacitor bank 126 comprises a plurality of capacitors. The capacitor bank 126 in this example has a capacitance of around 1 F. In the high voltage circuit 100 of Figure 1 a single capacitor bank 126 is shown, however in preferred embodiments there are a plurality of capacitor banks, for example at least three capacitor banks or as many as around 10 or 20 capacitor banks.

[0094] The capacitor bank 126 is connected to the high voltage input 110 so that the high voltage input 110 can charge the capacitor bank 126. Where there are multiple capacitor banks, they are preferably each connected to the high voltage input 110 in parallel.

[0095] The high voltage circuit 100 also comprises a charge resistor 124. The charge resistor 124 in this case has a capacitance of 1500(1 The charge resistor 124 provides resistance that allows the capacitor bank 126 to be charged when the capacitor bank 126 is electrically connected to the high voltage input 110. The charge resistor 124 is positioned on a charging path between the high voltage input 110 and the capacitor bank 126. Where there are multiple capacitor banks, each capacitor bank may have its own respective charge resistor.

[0096] The high voltage circuit 100 is connectable to a load 160. In this case the load 160 has a resistance of around 10(1 The load may be part of a plasma generator. The plasma generator 160 may, for example, have a cathode and an anode. When a pulse of electricity is released across the load 160 the circuit between the anode and cathode is completed and a plasma is formed. The capacitor bank 126 stores charge, which can be used to create the electrical pulse that is released across the load 160. Where there are multiple capacitor banks, each of the capacitor banks may be connected to the load 160 so that the charge stored in each capacitor bank can be used to generate the electrical pulse on the load 160. The plurality of capacitor banks may be connected in parallel across the load 160.

[0097] The capacitor bank 126 is connected to the load 160 via a discharge switch 128. In this case the discharge switch 128 is a high speed discharge switch, in the form of an active semiconductor device that can be switched to rapidly discharge the capacitor bank 126 through the load 160. The discharge switch 128 is on a discharge path from the capacitor bank 126 to the load 160.

[0098] When the discharge switch 128 is closed, or in a conducting state, the capacitor bank 126 is electrically connected to the load 160. This results in a discharge of the capacitor bank 126 across the load 160. When the discharge switch 128 is open, or in a non conducting state, the capacitor bank 126 is disconnected from the load 160 and charge may be stored on the capacitor bank 126, rather than discharging via the load 160. A controller (not illustrated) may be provided to provide control signals (also referred to as triggers) to switch the discharge switch 128 between a conducting state and a non-conducting state. The discharge switch 128 is preferably a solid state switch that can be switched by an electric current control signal.

[0099] Where there are multiple capacitor banks there is generally a discharge switch for each capacitor bank, the discharge switch being located on a discharge path between the respective capacitor bank and the load.

[0100] The inventors have found that discharge switches, such as the active semiconductor device 128 used in the high voltage socket 100, are susceptible to performance deterioration over time. In some cases the discharge switch 128 may fail open circuit, in a non-conducting state, meaning it has a fault that means it cannot be switched into a conducting state. However in other cases the discharge switch 128 can fail short-circuited, in a conducting state, and not be able to be switched into a non-conducting state.

[0101] In the event of a single short-circuited device discharge switch anywhere in the circuit 100 (e.g. discharge switch 128 being stuck in a conducting state), the energy from the high voltage input 110 will discharge through the load 160 via that short-circuited discharge switch. This prevents any of the capacitor banks charging up, and high power pulses of electricity cannot be generated.

[0102] In order to mitigate this, an isolation switch 120 has been added to the high voltage circuit 100. The isolation switch 120 is positioned between the high voltage input 110 and the capacitor bank 126. When in a conducting state, the isolation switch 120 allows current to flow from the high voltage input 110 to the capacitor bank 126 to charge the capacitor bank 126. When the isolation switch 120 is in a non-conducting state, the capacitor bank 126 is isolated from the high voltage input 110, so current cannot flow from the high voltage input to the capacitor bank 126, or to the load 160 (via the discharge switch 128, when it is in a conducting state). When the isolation switch is in a conducting state, the high voltage input 110 can be electrically connected to the capacitor bank 126 to charge the capacitor bank 126. The isolation switch 120 may be a semiconductor switch, and in this case the isolation switch 120 is a metal oxide semiconductor field effect transistor (MOSFET). The isolation switch 120 can be controlled by a controller (not shown). The controller that controls the isolation switch 120 may be the same controller as the controller that controls the discharge switch 128, although in alternative embodiments separate controllers may be used. The controller may send control signals (also referred to as triggers) to switch the isolation switch 120 between an open (non-conducting) state and a closed (conducting) state.

[0103] When there are multiple capacitor banks in the circuit 100, there may be provided an isolation switch for each capacitor bank, so as to allow isolation of the capacitor bank if the discharge switch (or another component) for the respective capacitor bank develops a fault. Where capacitor banks are arranged in parallel, the isolation switch for each capacitor bank may be placed on the branch of the circuit for the respective capacitor bank, so each capacitor bank can be isolated independently.

[0104] In order to detect a short circuit fault associated with a capacitor bank, for example a short circuit fault in a discharge switch for a particular capacitor bank, there may be provided a sensor (not shown). For example a current probe or sensor (such as a Hall effect sensor) may be used to detect if there is current flowing through the discharge switch 128 when the discharge switch 128 is meant to be in a non-conducting state. If a short circuit fault associated with capacitor bank 126 is detected, the isolation switch 124 associated with the capacitor bank 126 will be opened (or moved into a non-conducting state) and left in the non-conducting state. This will allow other capacitor banks in the circuit 100 to charge and provide pulsed power.

[0105] In the case of an open-circuit fault (for example caused by the discharge switch 128 being stuck in a non-conducting state), the capacitor bank 126 will charge as usual, but the energy stored in the bank of capacitors 126 will not be released into the load 160 on a trigger event. Although the rest of the system (e.g. any other capacitor banks in the circuit 100) should still operate as designed, the charge (and thus high voltage) will remain on the open-circuited bank 126 when the other capacitor banks in the circuit 100 discharge. The longer a damaged capacitor bank remains at a high voltage, the more likely it is that an arc flash will occur in this capacitor bank. Arc flashes are unpredictable and release a significant amount of energy, both as heat and in the electromagnetic spectrum. This dissipative energy may cause problems for other components in the circuit 100 and other parts of the system. For example, in extreme cases in the laboratory arc flashes have has dislodged components from the assembled board. Loose components are a particular issue because they can move around the unit internals unpredictably, especially in microgravity (such as would be experience in space), creating electrical shorts across other components or bumping into (and potentially damaging) fragile devices or components in the system. Heat dissipation is also undesirable because it adds to the thermal budget of the system. In the electromagnetic spectrum, these unpredictable discharge events can induce extremely large transients in sensitive devices, potentially causing other devices to fail in unexpected ways.

[0106] In order to prevent the capacitor bank 126 discharging through the isolation switch 120, the circuit 100 is also provided with a diode 122. The diode 122 in this case is a reverse blocking Schottkey diode (although other types of diode could be used). The diode 122 is positioned on a current path between the isolation switch 120 and the capacitor bank 126. The diode 122 allows current to flow from the high voltage input 110, through the isolation switch 120 (when it is in a conducting state) and towards the charge resistor 124 and the capacitor bank 126. However the diode 122 does not allow current flow in the opposite direction, from the capacitor bank 126 through the charging resistor 124 and towards the isolation switch 120.

[0107] If a capacitor bank has failed open-circuit (e.g. when the discharge switch for the respective capacitor bank is stuck in a non-conducting state), with the present arrangement it is difficult to detect the fault until the system has attempted to release the energy stored in the capacitor bank(s).

[0108] An open-circuit fault associated with the capacitor bank 126 may be detected by monitoring the voltage across the capacitor bank 126, e.g. using a voltage sensor (not shown) that may be connected in parallel across the capacitor bank 126. Once the discharge switch 128 for the capacitor bank 126 has been triggered (e.g. by sending a control signal from a controller to switch the discharge switch 128 into a conducting state), in the absence of an open-circuit fault the voltage across the capacitor bank 126 should drop rapidly as the capacitor bank 126 discharges to the load 160 to provide a pulse of electrical power. If the voltage across the capacitor bank 126 fails to drop as expected, an open-circuit fault is detected. The voltage measurement may be provided to a controller (which may be the same as, or different from, the controllers mentioned above) that can make the determination as to whether an open-circuit fault associated with the capacitor bank is present. Where there are multiple capacitor banks, a separate voltage sensor may be provided for each capacitor bank. In some arrangements instead of measuring the absolute voltage across a capacitor bank, the relative voltage between two or more capacitor banks may be determined.

[0109] The diode 124 will prevent the charge stored on the capacitor bank 126 from discharging via the isolation switch 120 and causing faults in other parts of the circuit 100 and / or charging up other capacitor banks in the circuit. The high-voltage circuit 100 comprises a high impedance slow discharge path (not shown) and the failed capacitor bank 126 will safely discharge through this path. Where there are multiple capacitor banks, each bank may be provided with its own high impedance slow discharge path.

[0110] It is also possible to isolate the capacitor bank 126 that has an open-circuit fault using the isolation switch 120 so the faulty capacitor bank 126 is not charged again by the high voltage input 110. This may be done using a control signal from a controller (not shown), which may be the same as or different from one or more of the controllers mentioned above. If the isolation switch 120 is in a conducting state when the open-circuit fault is detected, the controller can send a control signal, or trigger, to place the isolation switch 120 in a nonconducting state. If the isolation switch 120 is in a non-conducting state when the open circuit fault is detected, the controller need not send such a control signal, or trigger.

[0111] Once a short circuit or open-circuit fault associated with the capacitor bank 126 has been detected and the isolation switch is in a non-conducting state, the controller would normally not send a further control signal to switch the isolation switch 120 to a conducting state, so the relevant capacitor bank circuit would be isolated from the rest of the high-voltage circuit 100. This can prevent the capacitor bank 126 from charging up again once it has discharged via the high impedance, slow discharge path.

[0112] Thus the isolation switch 120 provides means for isolating the capacitor bank 126 from the rest of the circuit 100 if a fault associated with the capacitor bank 126 is detected, such as an open-circuit fault or a short-circuit fault. Where there are multiple capacitor banks, this allows the remaining system to function as usual, albeit at a slightly reduced overall power level (due to one of the capacitor banks not contributing to the power).

[0113] Although the discharge switch 128 described above in relation to Figure 1 is a solid state switch that can be switched between conducting and non-conducting states by electrical current signals, in alternative embodiments, the discharge switch 128 could be another type of high speed switch, such as a gas switch.

[0114] Referring to Figure 2, another high voltage circuit 200 for a pulsed power source will now be described. As with the high voltage circuit 100 of Figure 1 , the high voltage circuit 200 of Figure 2 may be used as part of a Pulsed Power System (PPS) for a pulsed plasma thruster (PPT), which may be used for spacecraft propulsion, for example for a satellite. The high voltage circuit 200 of Figure 2 functions in a similar manner to the high voltage circuit 100 of Figure 1. Identical or equivalent components may be used in both circuits 100, 200.

[0115] The high voltage circuit 200 includes a high voltage input 210. The high voltage input 210 has a voltage of between around 100V and 10,000V, preferably between 1 ,000V and 2,000V. The high voltage input 210 provides a direct current (DC).

[0116] Connected in parallel to the high voltage input 210 are a plurality of capacitor banks. In this case there are four capacitor banks 226, 236, 246, 256 shown. Each capacitor bank 226, 236, 246, 256 is positioned on a parallel branch of the circuit 200. The capacitor banks 226, 236, 246, 256 are each configured to be charged by the high voltage input 210. In other words, the circuit 200 is configured such that current can flow on a charging path from the high voltage input 210 to each of the capacitor banks 226, 236, 246, 256 in order to charge the capacitor banks 226, 236, 246, 256.

[0117] Each capacitor bank 226, 236, 246, 256 is formed of a plurality of individual capacitors. Each capacitor bank 226, 236, 246, 256 has a capacitance, C, of between around 0.1 pF and 10pF, preferably between around 0.5 pF and 3 pF. When connected to the high voltage input 210, each capacitor bank 226, 236, 246, 256 will charge up to store a maximum charge.

[0118] The maximum charge stored by a capacitor bank can be calculated by charge, Q = CV (where C is the capacitance and V is the voltage applied across the capacitor bank). For example, for a case where the voltage of the high voltage input is 1 ,500V and the capacitance of a capacitor bank is 1 pF, the maximum charge stored by the capacitor bank will be Q = 1x1 O'6x 1500 = 0.0015 Coulombs. In preferred embodiments, the maximum charge stored by each capacitor bank is between around 0.0001 Coulombs and 0.01 Coulombs. The maximum energy stored by a capacitor bank can be calculated by energy, E = % CV2(where C is the capacitance and V is the voltage applied across the capacitor bank). For example, for an embodiment where the voltage of the high voltage input is 1 ,500V and the capacitance of a capacitor bank is 1 F, the maximum energy stored by the capacitor bank will be E = % x 15002x 1x10’6= 1.13 Joules (J). In preferred embodiments, the maximum energy stored by each capacitor bank is between around 0.1 J and 10J, more preferably between around 0.5J and 5J.

[0119] A load 260 is also provided in the high voltage circuit 200. The capacitor banks 226, 236, 246, 256 are connectable in parallel to the load 260. All the capacitor banks 226, 236, 246, 256 are connectable to the same load 260. The load 260 has a resistance of between around 1 Q and 100Q, preferably between around 5Q and 20(1 When the capacitor banks 226, 236, 246, 256 have been charged, they can be electrically connected to the load 260 to release the energy stored in the capacitor banks to the load 260. This energy is released in the form of a pulse of electrical power.

[0120] As with the load 160 of circuit 100, the load 260 of circuit 200 may be part of a plasma generator. The plasma generator may, for example, have a cathode and an anode. When a pulse of electricity is released across the load 260 the circuit between the anode and cathode is completed and a plasma is formed. The charge stored in the capacitor banks 226, 236, 246, 256 can be used to create the electrical pulse that is released across the load 160.

[0121] There is also provided a charge resistor 224, 234, 244, 254 for each capacitor bank 226, 236, 246, 256. The charge resistors 224, 234, 244, 254 each have a resistance of between around 500Q and 5,0000, preferably between around 1 ,0000 and 2,0000. The resistance of the charge resistors 224, 234, 244, 254 is preferably significantly greater than the resistance of the load 260, for example at least one or two orders of magnitude larger. Thus the capacitor banks 226, 236, 246, 256 will charge up from the high voltage input 210 relatively slowly in comparison to the length of time taken to discharge to the load 260.

[0122] Assuming no other significant resistance in the charging paths between the high voltage input 210 and the capacitor banks 226, 236, 246, 256, the resistance of the charge resistor 224, 234, 244, 254, in combination with the capacitance of the capacitor bank 226, 236, 246, 256, can determine the time constant for charging the capacitor bank. The time constant, T, equates to the time required to charge the capacitor bank from an initial zero charge to 63% of its maximum charge at the applied voltage. The time required to bring the charge to about 99% of the maximum charge is approximately 5 T. Assuming the majority of the resistance when charging is provided by the charge resistor 224, 234, 244, 254, the time constant, T, can be calculated by T = RC (where R is the resistance of the charge resistor 224, 234, 244, 254 and C is the capacitance of the capacitor bank 226, 236, 246, 256). For example, where the resistance of the charge resistor is 1 ,5000 and the capacitance of a capacitor bank is 1 F, the charging time constant, T = 1500 x 1 x10’6= 0.0015 s. In preferred embodiments, the charging time constant for each capacitor bank is between around 0.0001s and 0.01s, more preferably between around 0.0005s and 0.005s.

[0123] Circuit 200 also comprises a discharge switch 228, 238, 248, 258 for each capacitor bank 226, 236, 246, 256 positioned between the respective capacitor bank 226, 236, 246, 256 and the load 260. In other words, the discharge switch 228, 238, 248, 258 for each capacitor bank 226, 236, 246, 256 is positioned on a discharge path between the capacitor bank 226, 236, 246, 256 and the load 260. The discharge switches 228, 238, 248, 258 are operable to selectively electrically couple the respective capacitor banks 226, 236, 246, 256 to the load 260. The discharge switches 228, 238, 248, 258 are switchable between a conducting and a non-conducting state.

[0124] When in the conducting state each discharge switch 228, 238, 248, 258 allows current to flow from the respective capacitor bank 226, 236, 246, 256 to the load 260. In this manner, an electrical power pulse can be generated, whereby energy stored in the capacitor banks 226, 236, 246, 256 is suddenly released into the load 260 upon the discharge switches 228, 238, 248, 258 being placed in a conducting state. Due to the relatively low impedance or resistance across the load, the capacitor banks 226, 236, 246, 256 discharge across the load far more quickly than they charge. For example, where the resistance of the load is 10Q and the capacitance of a capacitor bank is 1 F, the discharge time constant, T = 10 x 1 x10’6= 10 ps.

[0125] When one of the discharge switches 228, 238, 248, 258 is in a non-conducting state, current cannot flow from the respective capacitor bank 226, 236, 246, 256 to the load 260. Thus, in normal operation when the respective discharge switches 228, 238, 248, 258 are in a nonconducting state and are electrically connected to the high voltage input 210, the capacitor banks 226, 236, 246, 256 can charge.

[0126] In this case the discharge switches 228, 238, 248, 258 are active semiconductor devices. The discharge switches 228, 238, 248, 258 can be ‘triggered’ (or in other words switched, e.g. from a non-conducting to a conducting state) by an electrical control signal, or pulse. In some embodiments, upon receiving a trigger or control signal to switch the discharge switch 228, 238, 248, 258 from a non-conducting to a conducting state the discharge switch 228, 238, 248, 258 remains in the conducting state until a further trigger or control signal is received to switch the discharge switch 228, 238, 248, 258 back to a non-conducting state. In other embodiments, upon receiving a trigger or control signal to switch the discharge switch 228, 238, 248, 258 from a non-conducting to a conducting state the discharge switch 228, 238, 248, 258 transitions into a conducting state for a predetermined period of time and subsequently returns to a non-conducting without requiring a control signal to be received.

[0127] The state of the discharge switches 228, 238, 248, 258 is controlled by a controller (not shown). Each discharge switch 228, 238, 248, 258 may be controlled by its own controller or the same controller may be used to control all the discharge switches 228, 238, 248, 258. The controller(s) can send the trigger, or control signal, to switch the discharge switches 228, 238, 248, 258 from a non-conducting to a conducting state (and optionally back to a non-conducting state).

[0128] In some embodiments a single control signal, or trigger, can be used to switch the state of all the discharge switches 228, 238, 248, 258. Thus the discharge switches 228, 238, 248, 258 would be triggered simultaneously and the capacitor banks 226, 236, 246, 256 would also discharge across the load 260 at the same time.

[0129] Alternatively, or additionally, the discharge switches 228, 238, 248, 258 can be controlled independently. For example, the circuit 200 can be arranged such that different control signals, or triggers, are provided for each discharge switch 228, 238, 248, 258. In these embodiments, the discharge switches 228, 238, 248, 258 can be switched, or triggered, at different times. Advantageously, this can allow the discharge switches 228, 238, 248, 258 to be tested for malfunctions (such as open-circuit faults) independently of one another, which can allow faults to be identified easily and the faulty switch can be isolated from the rest of the circuit 200, or can be repaired or replaced. Where the pulsed power circuit 200 is used as part of a spacecraft propulsion system this can be useful for pre-flight testing, for example to ensure each capacitor bank is performing correctly. In addition, independent switching of the discharge switches 228, 238, 248, 258 can allow the shaping of the pulses generated by the circuit 200 to be altered, for example by slightly offsetting the time at which one or more of the capacitor banks 226, 236, 246, 256 discharges to the load 260. Where the pulsed power circuit 200 is used as part of a plasma generation device this can be particularly beneficial as a first part of a pulse can be used to generate a plasma and a second part of the pulse to accelerate the plasma. Where the discharge switches 228, 238, 248, 258 are switched by electrical pulse triggers and multiple discharge switches 228, 238, 248, 258 are controlled by a single controller, independent switching may be achieved by providing a separate electrical connection between the controller and each discharge switch 228, 238, 248, 258.

[0130] As with the circuit 100 of Figure 1 , described above, the circuit 200 is provided with an isolation switch 222, 232, 242, 252 on the charging path of each capacitor bank 226, 236, 246, 256. An isolation switch 222, 232, 242, 252 is provided between the high voltage input 210 and each capacitor bank 226, 236, 246, 256. In other words, each capacitor bank 226, 236, 246, 256 has an isolation switch 222, 232, 242, 252 that is configured to selectively connect or disconnect the respective capacitor bank 226, 236, 246, 256 to or from the high voltage input 210.

[0131] When the isolation switch 222, 232, 242, 252 for a particular capacitor bank 226, 236, 246, 256 is in a conducting state (in other words, when the isolation switch is “closed”) the isolation switch 222, 232, 242, 252 allows current to pass from the high voltage input 210 to the respective capacitor bank 226, 236, 246, 256. When the isolation switch 222, 232, 242, 252 for a particular capacitor bank 226, 236, 246, 256 is in a non-conducting state (in other words, when the isolation switch is “open”) the isolation switch 222, 232, 242, 252 does not allow current to pass from the high voltage input 210 to the respective capacitor bank 226, 236, 246, 256, and also does not allow current to flow from the capacitor bank 226, 236, 246, 256 to the rest of the circuit 200. Thus when an isolation switch associated with a capacitor bank is placed in a non-conducting state the capacitor bank cannot be charged by the high voltage input 210. Each isolation switch 222, 232, 242, 252 is positioned on a circuit path between the high voltage input 210 and the discharge switch 228, 238, 248, 258 of the respective capacitor bank 226, 236, 246, 256. This means when an isolation switch associated with a capacitor bank is in a non-conducting state current cannot flow from the high voltage input 210 to the load 260 via the discharge switch 228, 238, 248, 258 associated with the respective capacitor bank 226, 236, 246, 256.

[0132] One advantage of the isolation switches 222, 232, 242, 252 is that if a failure associated with a capacitor bank 226, 236, 246, 256 is detected, that capacitor bank 226, 236, 246, 256 can be isolated from the rest of the circuit 200. Faults associated with a capacitor bank 226, 236, 246, 256 can prevent the rest of the circuit 200 functioning properly and / or can cause damage to the rest of the circuit 200, and these may be prevented by isolating the capacitor bank by placing and leaving the relevant isolation switch in a non-conducting state. The isolation switches 222, 232, 242, 252 may be semiconductor switches, and in this case the isolation switches 222, 232, 242, 252 are metal oxide semiconductor field effect transistors (MOSFETs). The isolation switches 222, 232, 242, 252 can be controlled by a controller (not shown). In some embodiments one controller is used to control the state of each of all the isolation switches 222, 232, 242, 252 in the circuit, whereas in alternative embodiments multiple controllers may be used to control the state of the isolation switches 222, 232, 242, 252 (e.g. one controller for each isolation switch 222, 232, 242, 252).

[0133] The controller(s) that controls the isolation switches 222, 232, 242, 252 may be the same controller as the controller(s) that controls the discharge switches 228, 238, 248, 258, although in alternative embodiments separate controllers may be used. For example, a first controller could be used to control the state of the isolation switches 222, 232, 242, 252 and a second controller could be used to control the state of the discharge switches 228, 238, 248, 258. The controller may send control signals (also referred to as triggers) to switch the isolation switches 222, 232, 242, 252 between an open (non-conducting) state and a closed (conducting) state.

[0134] The circuit 200 also comprises a diode 222, 232, 242, 252 associated with each capacitor bank 226, 236, 246, 256, positioned between the capacitor bank 226, 236, 246, 256 and the isolation switch 222, 232, 242, 252 for that capacitor bank 226, 236, 246, 256.

[0135] In the circuit 200 shown in Figure 2 each diode 222, 232, 242, 252 is positioned between the isolation switch 222, 232, 242, 252 for that capacitor bank 226, 236, 246, 256 and the charge resistor 224, 234, 244, 254 for that capacitor bank 226, 236, 246, 256. However in alternative embodiments the diodes 222, 232, 242, 252 may each be positioned between the charge resistor 224, 234, 244, 254 and the relevant capacitor bank 226, 236, 246, 256.

[0136] The diodes 222, 232, 242, 252 in this case are reverse blocking Schottkey diodes (although other types of diode could be used). The diodes 222, 232, 242, 252 each allow current to flow from the high voltage input 210 through the respective isolation switch 222, 232, 242, 252 and to charge the respective capacitor bank 226, 236, 246, 256. However the diodes 222, 232, 242, 252 do not allow current flow in the opposite direction, from the capacitor banks 226, 236, 246, 256 through the charging resistors 124 and towards the isolation switches 222, 232, 242, 252.

[0137] In order to detect a short circuit fault associated with a capacitor bank 226, 236, 246, 256, for example a short circuit fault in a discharge switch for a particular capacitor bank 226, 236, 246, 256, there may be provided a sensor (not shown). For example a current probe or sensor (such as a Hall effect sensor) may be provided for each capacitor bank 226, 236, 246, 256 and be used to detect if there is current flowing through the discharge switch 228, 238, 248, 258 when the discharge switch 228, 238, 248, 258 is intended to be in a nonconducting state. If a short circuit fault associated with capacitor bank 226, 236, 246, 256 is detected, the isolation switch 222, 232, 242, 252 associated with the capacitor bank 226, 236, 246, 256 can be opened (or moved into a non-conducting state) and left in the nonconducting state. Thus if only one capacitor bank, such as the second capacitor bank 236, experiences an associated short circuit fault (such as the second discharge switch 238 being failed in a permanently conducting state), the capacitor bank 236 can be isolated by maintaining the respective isolation switch 230 in a non-conducting state. This will allow the three other capacitor banks 226, 246, 256 in the circuit 200 to charge and provide pulsed power.

[0138] If an open-circuit failure has occurred in the circuitry of one of the capacitor banks 226, 236, 246, 256 (e.g. when the discharge switch for the respective capacitor bank is stuck in a nonconducting state), it may be difficult to detect the fault until the system has attempted to release the energy stored in the capacitor bank(s).

[0139] An open-circuit fault associated with one of the capacitor banks 226, 236, 246, 256 may be detected by monitoring the voltage across the capacitor banks 226, 236, 246, 256, e.g. using a voltage sensor (not shown) for each capacitor banks 226, 236, 246, 256. The voltage sensor for each capacitor bank 226, 236, 246, 256 can be connected in parallel across the respective capacitor bank 226, 236, 246, 256. Once the respective discharge switch 228, 238, 248, 258 for the capacitor bank 226, 236, 246, 256 has been triggered (e.g. by sending a control signal from a controller to switch the respective discharge switch 228, 238, 248, 258 into a conducting state), in the absence of an open-circuit fault the voltage across the respective capacitor bank 226, 236, 246, 256 should drop rapidly as the capacitor bank 226, 236, 246, 256 discharges to the load 260 to provide a pulse of electrical power. If the voltage across the capacitor bank 226, 236, 246, 256 fails to drop as expected, this is indicative of an open-circuit fault associated with the respective capacitor bank 226, 236, 246, 256.

[0140] The voltage measurement may be provided to a controller (which may be the same as, or different from, the controllers mentioned above) that can make the determination as to whether an open-circuit fault associated with the capacitor bank is present. In this embodiment, a separate voltage sensor is provided for each capacitor bank 226, 236, 246, 256. In some arrangements instead of measuring the absolute voltage across a capacitor bank, the relative voltage between two or more capacitor banks may be determined in order to detect an open circuit fault.

[0141] The diodes 222, 232, 242, 252 prevent the charge stored on a capacitor bank 226, 236, 246, 256 from discharging via the bank’s respective isolation switch 220, 230, 240, 250 and causing faults in other parts of the circuit 200 and / or charging up other capacitor banks in the circuit. The high-voltage circuit 200 comprises a high impedance slow discharge path (not shown) for each capacitor bank 226, 236, 246, 256. Thus in the event of an open-circuit fault associated with one of the capacitor banks 226, 236, 246, 256, the failed capacitor bank 226, 236, 246, 256 will safely discharge through this high impedance slow discharge path.

[0142] It is also possible to isolate the capacitor bank 226, 236, 246, 256 that has an open-circuit fault using the isolation switch 220 so the faulty capacitor bank 226, 236, 246, 256 is not charged again by the high voltage input 210. This may be done using a control signal from a controller (not shown), which may be the same as or different from one or more of the controllers mentioned above. If the isolation switch 220, 230, 240, 250 is in a conducting state when the open-circuit fault is detected, the controller can send a control signal, or trigger, to place the isolation switch 220, 230, 240, 250 in a non-conducting state. If the isolation switch 220, 230, 240, 250 is in a non-conducting state when the open circuit fault is detected, the controller need not send such a control signal, or trigger. Once an open-circuit fault associated with the capacitor bank 226, 236, 246, 256 has been detected and the isolation switch is in a non-conducting state, the controller would normally not send a further control signal to switch the isolation switch 120 to a conducting state, so the relevant capacitor bank circuit would be isolated from the rest of the high-voltage circuit 100. This can prevent the capacitor bank 126 from charging up again once it has discharged via the high impedance, slow discharge path.

[0143] In alternative embodiments where no isolation switch 222, 232, 242, 252 is provided, the diode 222, 232, 242, 252 for each capacitor bank 226, 236, 246, 256 could be positioned between the capacitor bank 226, 236, 246, 256 and the high voltage input 210. Preferably the diode 222, 232, 242, 252 for each capacitor bank 226, 236, 246, 256 is positioned on a current path between the respective capacitor bank 226, 236, 246, 256 and other capacitor banks 226, 236, 246, 256 in the circuit 200. Although the discharge switches 228, 238, 248, 258 of circuit 200 are active semiconductor devices that can be triggered, or switched, by electrical pulse signals, in alternative embodiments other types of switches may be used for the discharge switches 228, 238, 248, 258. For example gas switches, which may be triggered by a laser signal, may be used.

[0144] Figure 3 illustrates an example method 300 of protecting a pulsed power source during a charging phase. The pulsed power source has one or more capacitor banks which, in a charging phase, can be charged up using a power source and then discharged comparatively quickly (compared to the charging phase) to generate a pulse of electrical power. The pulsed power source may be a pulsed power source having circuitry 100 or 200 of the form shown in Figure 1 or Figure 2 and described above.

[0145] The method 300 is preferably performed by a controller capable of sending control signals to components in the pulsed power source and of monitoring the state of components in the pulsed power source. However in some embodiments sending signals and monitoring the state of components may be performed by separate controllers.

[0146] At step 302 a signal is sent to cause the capacitor bank(s) to charge. This commences a charging phase. In some embodiments sending a signal to cause the capacitor banks to charge may comprise sending a signal to electrically connect the voltage input 110, 210 to the capacitor banks. For example, the step may involve sending a signal to move the isolation switches 120, 220, 230, 240, 250 in the circuit 100 or 200 into a conducting state (if those isolation switches were in a non-conducting state prior to step 302). However other means of connecting the voltage input to the capacitor banks are also envisaged. The signal may take the form of an electronic pulse directed to each of the isolation switches.

[0147] Additionally, or alternatively, step 302 may comprise sending a signal to switch the discharge switches 120, 220, 230, 240, 250 in the circuit 100 or 200 into a non-conducting state, so that current does not discharge from the capacitor banks 126, 226, 236, 246, 256 to the load 160, 260 whilst the capacitor bank(s) are being charged. If step 302 also involves switching the isolation switches 120, 220, 230, 240, 250 into a conducting state, preferably switching the discharge switches 120, 220, 230, 240, 250 into a non-conducting state is performed slightly before, or at the latest at the same time as, switching the isolation switches 120, 220, 230, 240, 250 into a conducting state. This prevents the high voltage input 210 short circuiting across the load 160, 260 via the conducting discharge switches 120, 220, 230, 240, 250.

[0148] At step 304 the capacitor bank(s) are monitored as they are charged, or during the charging phase. In particular, the capacitor bank(s) can be monitored to detect a short circuit fault associated with the capacitor bank. As described above, a short circuit fault may prevent the capacitor bank from charging because current flows directly from the voltage input 110, 210 to the load 160, 260 via the discharge switches 120, 220, 230, 240, 250, for example because the discharge switches 120, 220, 230, 240, 250 have become fixed or stuck in a conducting state.

[0149] Step 304 of monitoring the capacitor bank(s) whilst in the charging phase can be performed by monitoring the current flowing from each discharge switch to the load 160, 260. Preferably the capacitor banks are each independently monitored so any fault can be attributed to a particular capacitor bank. For example, a current probe or sensor (such as a Hall effect sensor) may be positioned to measure current on each discharge path between one of the capacitor bank(s) 126, 226, 236, 246, 256 and the load 160, 260.

[0150] If current (e.g. non-zero current) is detected on the discharge path then this is indicative of a short circuit fault associated with the respective capacitor bank 126, 226, 236, 246, 256. A short circuit fault associated with just one of the capacitor banks would result in the output of the high voltage generator discharging straight through the load via that short circuit, energising the downstream system early and none of the capacitor banks in the system would charge up. Thus step 304 may comprise detecting whether a fault is present with any of the capacitor bank(s), and in particular detecting whether a short circuit fault is present.

[0151] At step 306, if a fault has been detected at step 304, then a signal is sent to isolate the capacitor bank that is associated with the fault. Generally this is done by sending a signal to move the isolation switch for the respective capacitor into a non-conducting mode, or in other words opening the respective isolation switch. For example, if the first capacitor bank 226 is detected to have a short circuit fault then a signal can be sent to the first isolation switch 220 to switch it from a conducting state to a non-conducting state. The first capacitor bank 226 will then be isolated from the rest of the circuit 200 so the high voltage input will no longer discharge through the load 260 via the short circuit. The signal may take the form of an electronic pulse directed to each of the isolation switches. Figure 4 illustrates an example method 400 of protecting a pulsed power source during a discharging phase. The pulsed power source has one or more capacitor banks which, in a charging phase, can be charged up using a power source and then discharged comparatively quickly (compared to the charging phase) to generate a pulse of electrical power. The pulsed power source may be a pulsed power source having circuitry 100 or 200 of the form shown in Figure 1 or Figure 2 and described above.

[0152] The method 400 is preferably performed by a controller capable of sending control signals to components in the pulsed power source and of monitoring the state of components in the pulsed power source. However in some embodiments sending signals and monitoring the state of components may be performed by separate controllers.

[0153] The method 400 shown in Figure 4 may be performed directly after the method 300 shown in Figure 3.

[0154] At step 402 a signal is sent to cause the capacitor bank(s) to discharge. This commences a discharging phase. The signal may cause the capacitor bank(s) to be electrically connected to the load. For example the signal may be sent to a discharge switch 128, 228, 238, 248, 258 connecting each capacitor bank 126, 226, 236, 246, 256 to the load 160, 260 to switch the discharge switch into a conducting state. In some embodiments a single signal is sent to cause more than one of the capacitor banks to discharge, which can result in the more than one capacitor banks being discharged switched into a discharging phase simultaneously. For example the same signal may be sent to all the discharge switches 226, 236, 246, 256 in a circuit in order to place all the discharge switches in a conducting state at the same time; in other words all the discharge switches are “triggered” simultaneously. Alternatively, separate signals may be sent to each of the discharge switches 226, 236, 246, 256 to cause the switches to transition into a non-conducting state. Thus one or more of the discharge switches 226, 236, 246, 256 may be switched, or triggered, at a different time to one or more other discharge switches in the circuit 200.

[0155] The signal(s) may take the form of electronic pulses directed to each of the discharge switches.

[0156] In some embodiments as part of step 402 a signal may also be sent to disconnect the capacitor bank(s) from the high voltage input 110, 210. For example, the signal may be sent to each isolation switch 220, 230, 240, 250 to switch it to a non-conducting state or the high voltage input 110, 210 may be disconnected from the rest of the circuit 100, 200. In normal operation when the signal to cause the capacitor bank(s) to discharge is received by the relevant component(s) the capacitor bank(s) will discharge rapidly across the load, causing a high voltage pulse of electric power. However if circuitry associated with one of the capacitor bank(s) is malfunctioning the capacitor bank may fail to discharge to the load.

[0157] At step 404 the capacitor bank(s) are monitored during the discharging phase. Preferably the capacitor bank(s) are each independently monitored so any fault can be attributed to a particular capacitor bank. The capacitor bank(s) may be monitored during the discharging phase by monitoring voltage across each capacitor bank. The voltage measured may be absolute voltage or may be voltage relative to one or more of the other capacitor bank(s).

[0158] Alternatively, or additionally, current flowing from a capacitor bank to the load may be measured, for example by a current probe or sensor (such as a Hall effect sensor). The current may be measured along the discharge path between the capacitor bank and the load. For example, current on one or both sides of the discharge switches 128, 228, 238, 248, 258 or the current going through the discharges switches 128, 228, 238, 248, 258 may be measured.

[0159] When the discharging phase begins, under normal operation the capacitor bank(s) should discharge rapidly through the load. Thus a current would be expected to flow in the discharge path between each capacitor bank to the load via the respective discharge switch and the voltage across each capacitor bank would be expected to decrease rapidly, normally to around zero.

[0160] However, if there is an open-circuit fault associated with one of the capacitor banks then that capacitor bank will not discharge to the load and the voltage across the capacitor bank will remain roughly the same. An open-circuit fault may, for example, be caused by one of the discharge switches being faulty and failing in a non-conducting state such that it cannot be switched into a conducting state.

[0161] If step 404 comprises monitoring the voltage across the capacitor bank(s), an open-circuit fault associated with a capacitor can be detected based on the voltage across the capacitor bank failing to decrease within a time threshold after the signal is sent to cause the capacitor bank to discharge. For example, it may be detected that there is an open-circuit fault associated with a capacitor bank if the voltage across the capacitor bank fails to decrease by more than a threshold voltage decrease within the time threshold. The threshold voltage may be a percentage of the starting voltage across the capacitor when the signal to cause the capacitor bank(s) to discharge is sent, or may be a percentage of the voltage of the high voltage input. The threshold may, for example, be set to at least 10%, at least 20% or at least 50% of the starting voltage or of the input voltage. Alternatively an open circuit fault may be detected if the voltage across the capacitor bank fails to decrease to below a low voltage threshold within a time threshold. The low voltage threshold may be a voltage value close to zero, for example the low voltage threshold may be less than around 100V, less than around 50V or less than around 10V. In some embodiments the low voltage threshold may be less than around 5V. The low voltage threshold may be, for example, at least around 0.1V.

[0162] Additionally or alternatively, where current flow is monitored at step 404, current flow would not be detected on the discharge path between each capacitor bank and the load if there is an open-circuit fault associated with the capacitor bank. Thus an open-circuit fault associated with a capacitor can be identified based on a current on a discharge path between the capacitor bank and the load not being detected, or the current on the discharge path failing to increase above a current threshold.

[0163] If a fault is detected during the discharging phase at step 404, then the method 400 proceeds to step 406. At step 406 a signal is set to isolate the capacitor bank that is associated with the fault. Generally this is done by moving the isolation switch for the respective capacitor into a non-conducting mode, or in other words opening the respective isolation switch. For example, if the first capacitor bank 226 is detected to have an open circuit fault then a signal can be sent to the first isolation switch 220 to switch it from a conducting state to a non-conducting state. Due to the presence of the first diode 222, energy stored in the first capacitor bank 226 will be prevented from flowing back around the circuit 200 into the other capacitor banks. Instead the faulty capacitor bank 226 will discharge slowly through the high impedance slow discharge path (not shown). Closing the first isolation switch 220 (or placing it in a non-conducting state) means the faulty first capacitor bank 226 will be isolated from the rest of the circuit 200 so the high voltage input will no longer be charged by the high voltage input 210 so the rest of the circuit can function normally (albeit at slightly lower power).

[0164] Figure 5 illustrates an example method 500 of discharging a pulsed power source. The pulsed power source has at least two capacitor banks which, in a charging phase, can be charged up using a power source and then discharged comparatively quickly (compared to the charging phase) to generate a pulse of electrical power. The pulsed power source may be a pulsed power source having circuitry 100 or 200 of the form shown in Figure 1 or Figure 2 and described above.

[0165] The method 500 is preferably performed by a controller capable of sending control signals to various components in the pulsed power source. However in some embodiments sending signals to different components in the pulsed power source may be performed by separate controllers.

[0166] This method 500 may be performed as part of steps 302 and 402 of methods 300 and 400 described above. However the method 500 may also be performed independently of methods 300 or 400, without requiring monitoring or isolating of a faulty capacitor banks as in methods 300 or 400.

[0167] At step 502 a signal is sent to cause a plurality of capacitor banks to charge. This step 502 can be performed in the same manner as step 302 of method 300, described above, or may be performed differently. The plurality of capacitor banks comprises at least two capacitor banks, but preferably includes more than two capacitor banks, such as at least four capacitor banks or at least 10 capacitor banks. Additionally or alternatively, in preferred embodiments there are fewer than 100 capacitor banks, more preferably fewer than 80 capacitor banks or fewer than 50 capacitor banks.

[0168] In some embodiments step 502 of sending a signal to cause the capacitor banks to charge may comprise sending a signal to electrically connect the voltage input 110, 210 to the capacitor banks. For example, the step may involve sending a signal to move the isolation switches 220, 230, 240, 250 in the circuit 200 into a conducting state (if those isolation switches were in a non-conducting state prior to step 502). However other means of connecting the voltage input to the capacitor banks are also envisaged. The signal may take the form of an electronic pulse directed to each of the isolation switches.

[0169] Additionally, or alternatively, step 502 may comprise sending a signal to switch the discharge switches 220, 230, 240, 250 in the circuit 200 into a non-conducting state, so that current does not discharge from the capacitor banks 226, 236, 246, 256 to the load 260 whilst the capacitor bank(s) are being charged. If step 502 also involves switching the isolation switches 220, 230, 240, 250 into a conducting state, preferably switching the discharge switches 220, 230, 240, 250 into a non-conducting state is performed slightly before, or at the latest at the same time as, switching the isolation switches 220, 230, 240, 250 into a conducting state. This prevents the high voltage input 210 short circuiting across the load 160, 260 via the conducting discharge switches 120, 220, 230, 240, 250.

[0170] At step 504 a signal is sent to cause a first capacitor bank 226 of the plurality of capacitor banks 220, 230, 240, 250 to discharge to the load. This may be performed in the same manner as in step 402 of method 400, described above.

[0171] The signal may cause the first capacitor bank 226 to be electrically connected to the load. For example the signal may be sent to the first discharge switch 228 that connects the first capacitor bank 226 to the load 260 to switch the discharge switch 228 into a conducting state.

[0172] In some embodiments a single signal is sent to cause more than one of the plurality of capacitor banks 226, 236, 246, 256 to discharge, which can result in the more than one capacitor banks being discharged switched into a discharging phase simultaneously. However, according to the method 500, at least two of the capacitor banks in the plurality of capacitor banks are caused to discharge at different times. Thus the signal at step 504 would cause only a subset of the plurality of discharge switches to move into a conducting state, and thus only a corresponding subset of the plurality of capacitor banks 226, 236, 246, 256 would be caused to discharge.

[0173] In some embodiments, as part of step 504 separate signals may be sent simultaneously, each signal to cause a subset of the discharge switches 226, 236, 246, 256 to transition into a non-conducting state. Thus one or more of the plurality discharge switches 226, 236, 246, 256 may be switched, or triggered, at the same time as one or more other discharge switches in the circuit 200.

[0174] The signal(s) may take the form of electronic pulses directed to each of the discharge switches.

[0175] In some embodiments as part of step 402 a signal may also be sent to disconnect the capacitor bank(s) from the high voltage input 110, 210. For example, the signal may be sent to each isolation switch 220, 230, 240, 250 to switch it to a non-conducting state or the high voltage input 110, 210 may be disconnected from the rest of the circuit 100, 200.

[0176] In normal operation when the signal to cause the capacitor bank(s) to discharge is received by the relevant component(s) the capacitor bank(s) will discharge rapidly across the load, causing a high voltage pulse of electric power. Step 506 comprises waiting for a predetermined time period after step 504. The predetermined time period may, for example, be between around 1 ns and around 1 s. In some embodiments the predetermined time period is between around 1 ns and 100ns, preferably between around 1 ns and around 50ns, or even between around 1 ns and around 20ns. In some scenarios the predetermined time period is between around 200ns and around 1 s, or between around 100ns and around 250 ps.

[0177] The length of the predetermined time period may change according to the reason for triggering capacitor banks to discharge at different times. For example, if the reason for triggering discharging at different times is for changing the shape of the electrical pulse then the predetermined time period may be shorter compared to if the purpose is to test whether each of the capacitor banks is discharging correctly, e.g. in a pre-flight test.

[0178] At step 508, once the predetermined time period has elapsed a signal is sent to cause a second capacitor bank 236 of the plurality of capacitor banks 226, 236, 246, 256 to discharge. The step 508 comprises sending a signal to discharge one or more of the remaining capacitor banks that had not been caused to discharge in step 504. The manner in which the signal is sent to cause the second capacitor bank 236 to discharge can be the same as in step 504. For example, an electrical signal may be sent to the second discharge switch 238 to place the second discharge switch 238 in a conducting state. This in turn will cause the second capacitor bank 236 to be electrically connected to the load 260 and to discharge its stored charge through the load 260 at a time later than the first capacitor bank 226 discharged its stored charge through the load 260. In other words, if the first capacitor bank 226 is caused to discharge at a first time and the second capacitor bank 236 is caused to discharge at a second time, the first time and the second time are separated by an amount of time equal to the predetermined time period.

[0179] By changing the relative timing of discharging capacitor banks in the circuit it is possible to change the shape of the electrical pulse generated. This may have particular advantages in pulsed plasma thrusters in that it is possible to adjust the shape of the pulse to improve the quality of the plasma generated, e.g. to make a “cleaner” plasma.

[0180] Pulse forming of this kind is possible because the circuit 200 is configured so that individual switches (or different groups) of the discharge switches 228, 238, 248, 258 can be switched, or triggered, independently. In some embodiments it may be preferable to discharge a small group of capacitor banks initially, followed by discharge of a larger group of capacitor banks. This can provide a small first burst, followed by a relatively larger second burst of energy. The smaller first burst of energy could be used to generate plasma, and the second larger burst of energy could be used to accelerate the plasma generated by the first burst (e.g. so it can be used to propel a spacecraft).

[0181] Figure 6 illustrates a graph 600 showing pulse discharge voltage over time for a pulsed power source in which the switches for all the capacitor banks in the circuit are triggered simultaneously. As can be seen, the voltage of the pulse rises rapidly to a peak 602, before tailing off (or reducing in size) gradually.

[0182] Figure 7 illustrates a graph 700 showing pulse discharge voltage over time for a pulsed power source in which the switches for the capacitor banks in the circuit are triggered at different times to shape the pulse. As can be seen, the voltage of the pulse rises rapidly to a first peak 702, and then rises rapidly again to a second peak 704, higher than the first peak 702. Such a pulse shape may be produced if a first group of one or more capacitor banks is discharged at a first time, followed by a second group of one or more capacitor banks being discharged at a second time, later than the first time.

[0183] In alternative embodiments (not shown), the voltage at the earlier first peak 702 may be higher than the voltage at the later second peak 704, for example if a greater number of capacitor banks (or banks with a greater capacity) are discharged at a first time compared to the number of capacitor banks (or capacity of capacitor banks) discharged at a second time, later than the first time.

[0184] In some embodiments altering the relative timing of discharging the capacitor banks can be used to cause the size of the pulse to ramp up slowly. For example, several different capacitor banks may be triggered to discharge consecutively, e.g. by at least three capacitor banks (or groups of capacitor banks) being caused to discharge each at different times. Providing a gradual increase in the energy or voltage of the pulse can be advantageous in plasma generation and acceleration for spacecraft propulsion; for example it may reduce the likelihood of causing a shock or producing shock waves in the plasma. This may improve control of the spacecraft propulsion.

[0185] Figure 8 illustrates a graph 800 showing pulse discharge voltage over time for a pulsed power source in which the switches for the capacitor banks in the circuit are triggered at different times to shape the pulse. As can be seen, the voltage of the pulse rises gradually over a ramp-up period 802 to a peak 804, before reducing again. The rise during the ramp- up period is slower than in either of the examples of Figure 6 or 7. A pulse shape as in Figure 8 may be produced if a plurality of capacitor banks are triggered to discharge consecutively.

[0186] Although the graphs shown in Figures 6 to 8 show voltage produced by the pulsed power source, similar graphs could be shown using current, power or energy of the pulse produced.

[0187] Figure 9 illustrates a hardware architecture of an example controller 900 for a pulsed power source. The controller 900 may be used to control the circuits 100, 200 described above.

[0188] The controller 900 comprises a control interface 910, a monitoring interface 920, a memory 930 and a processor 940. These components are interconnected by a data bus (this may in practice consist of several distinct buses such as a memory bus and I / O bus).

[0189] The memory 930 may comprise volatile I random access memory for storing temporary data and software code being executed. The memory 930 may additionally or alternatively include persistent storage arranged to store control and programming information, along with other software and data, such as an operating system, device drivers, software configuration data, historical data, measurement data and the like.

[0190] The memory 930 may store instructions to cause the processor 940 to perform one or more steps from any of the methods 300, 400, 500 described above in relation to Figures 3 to 5.

[0191] The monitoring interface 920 is a communications interface for receiving one or more signals indicative of the state of a pulsed power source circuit, or one or more components thereof. For example, the monitoring interface 920 may be configured to receive signals indicative of the voltage across one or more of the capacitor banks in a circuit and / or signals indicative of the current flowing from one or more of the capacitor banks in the circuit. Such signals may be used for monitoring the capacitor banks during charging or discharging, as in steps 304 and 404 in the methods described above.

[0192] The control interface 910 is arranged to send control signals to one or more components in a pulsed power source circuit. The control interface 910 may be capable of sending control signals to one or more switches in the circuit. For example, the control interface 910 may be configured to send control signals to one or more discharge switches to switch the switches from a non-conducting to a conducting state (in other words to “trigger” the discharge switch), and optionally from a conducting to a non-conducting state. Additionally or alternatively, the control interface 910 may be configured to send control signals to one or more isolation switches to switch the isolation switches from a conducting to a nonconducting state, and optionally from a non-conducting to a conducting state.

[0193] While a specific architecture is shown, any appropriate hardware / software architecture may be employed.

[0194] The above embodiments and examples are to be understood as illustrative examples. Further embodiments, aspects or examples are envisaged. It is to be understood that any feature described in relation to any one embodiment, aspect or example may be used alone, or in combination with other features described, and may also be used in combination with one or more features of any other of the embodiments, aspects or examples, or any combination of any other of the embodiments, aspects or examples. Furthermore, equivalents and modifications not described above may also be employed without departing from the scope of the invention, which is defined in the accompanying claims.

Claims

CLAIMS1. A pulsed power source for a spacecraft propulsion system, the pulsed power source comprising: a voltage input; a plurality of capacitor banks arranged to be charged by the voltage input; pulse discharge connections for electrically coupling each capacitor bank to a load; a discharge switch for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and an isolation switch for each capacitor bank, each isolation switch positioned on a charge path between the voltage input and the capacitor bank.

2. A pulsed power source according to claim 1 , further comprising: a controller configured to switch the isolation switch for each capacitor bank to a non-conducting state to isolate the respective capacitor bank from the voltage input.

3. A pulsed power source according to claim 2, wherein the controller is configured to switch the isolation switch to a non-conducting state to isolate the respective capacitor bank from the voltage input upon detection of a short circuit fault associated with the said respective capacitor bank, preferably a short circuit fault in the discharge switch for the said respective capacitor bank.

4. A pulsed power source according to claim 2 or 3, wherein the controller is configured to detect a short circuit fault during a capacitor bank charging mode in which the said respective capacitor bank is electrically connected to the voltage input for charging the said respective capacitor bank, preferably wherein the controller is configured to detect a short circuit fault by monitoring the charging rate of the said respective capacitor bank.

5. A pulsed power source according to claim 4, wherein the controller is configured to detect a short circuit fault associated with a said capacitor bank if, during the capacitor bank charging mode, the voltage across the said capacitor bank fails to increase, preferably wherein the controller is configured to detect a short circuit fault associated with the said capacitor bank if, during the capacitor bank charging mode, the voltage across the said capacitor bank fails to increase above a threshold voltage.

6. A pulsed power source according to any preceding claim, wherein the controller is configured to detect a short circuit fault by monitoring current flowing from the respective capacitor bank to the load.

7. A pulsed power source according to any of claims 2 to 6, wherein the controller is configured to switch the isolation switch to a non-conducting state to isolate the respective capacitor bank from the voltage input upon detection of an open circuit fault associated with the said respective capacitor bank, preferably an open circuit fault in the discharge switch for the said respective capacitor bank, optionally wherein the controller is configured to detect an open circuit fault by monitoring the discharge rate of the said capacitor bank.

8. A pulsed power source according to claim 7, wherein the controller is configured to detect a short circuit fault during a capacitor bank discharging mode in which control signals are sent to cause the respective capacitor bank to be electrically connected to the load for discharging the respective capacitor bank, preferably wherein the controller is configured to detect an open circuit fault associated with the said respective capacitor bank if, during the capacitor bank discharging mode, the voltage across the said respective capacitor bank fails to decrease.

9. A pulsed power source according to claims 8, wherein the controller is configured to detect an open circuit fault associated with the said respective capacitor bank if, during the capacitor bank discharging mode, the voltage across the said respective capacitor bank is greater than the voltage across one or more of the other capacitor banks, preferably greater than the voltage across one or more of the other capacitor banks by more than a threshold voltage difference.

10. A pulsed power source according to any preceding claim, further comprising: a diode for each capacitor bank, the diode positioned on the charge path between the voltage input and the capacitor bank and configured to allow a charging current to flow from the voltage input to the capacitor bank and to prevent current flow from the capacitor bank to other capacitor banks of the plurality of capacitor banks.

11. A pulsed power source for a spacecraft propulsion system, the pulsed power source comprising: a voltage input; a plurality of capacitor banks arranged to be charged by the voltage input;pulse discharge connections for electrically coupling each capacitor bank to a load; a discharge switch for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and a diode for each capacitor bank, the diode positioned on the charge path between the voltage input and the capacitor bank and configured to allow a charging current to flow from the voltage input to the capacitor bank and to prevent current flow from the capacitor bank to other capacitor banks of the plurality of capacitor banks.

12. A pulsed power source according to claim 10 or claim 11 , wherein the diode for each capacitor bank is positioned between the isolation switch and the capacitor bank and is configured to prevent current flow from the capacitor bank to the isolation switch.

13. A pulsed power source according to any preceding claim, further comprising: a controller configured to control the state of the discharge switches for the capacitor banks, wherein the controller is configured to control the state of each discharge switch independently of the other discharge switches of the plurality of discharge switches.

14. A pulsed power source for a spacecraft propulsion system, the pulsed power source comprising: a voltage input; a plurality of capacitor banks arranged to be charged by the voltage input; pulse discharge connections for electrically coupling each capacitor bank to a load; a discharge switch for each capacitor bank, each discharge switch positioned on a discharge path between the capacitor bank and the pulse discharge connections; and a controller configured to control the state of the discharge switches for the capacitor banks, wherein the controller is configured to control the state of each discharge switch independently of the other discharge switches of the plurality of discharge switches.

15. A pulsed power source according to claim 13 or claim 14, wherein the controller is configured to control the state of the discharge switches by triggering each discharge switch using a different trigger signal.

16. A pulsed power source according to any of claims 13 to 15, wherein the controller isconfigured to trigger a first discharge switch for a first capacitor bank of the plurality of capacitor banks at a first time and to trigger a second discharge switch for a second capacitor bank of the plurality of capacitor banks at a second time, different from the first time.

17. A pulsed power source according to any of claims 13 to 16, wherein the controller is configured to control the pulsed power source to shape a pulse by triggering the discharge switches for each of the plurality of capacitor banks over a pulse trigger time period, the pulse trigger time period being the time interval between triggering the first triggered discharge switch and triggering the last triggered discharge switch.

18. A pulsed power source according to any preceding claim, wherein the discharge switch for each capacitor bank is a transistor, preferably a MOSFET (metal-oxide- semiconductor field-effect transistor), and / or wherein the isolation switch is a semiconductor switch, preferably a MOSFET (metal-oxide-semiconductor field-effect transistor).

19. A plasma thruster for a spacecraft propulsion system, the plasma thruster comprising: a pulsed power source according to any preceding claim; and a plasma generator coupled to the load and arranged to receive electric current pulses from the pulsed power source and to form plasma using one or more electric current pulses.

20. A method of isolating capacitor banks in a pulsed power source for spacecraft propulsion, the method comprising: sending a control signal to cause one or more of a plurality of capacitor banks to be charged by a voltage input monitoring each of the one or more capacitor banks for one or both of a short circuit fault and an open circuit fault associated with the one or more capacitor banks; and isolating one of the one or more of the capacitor banks if a short circuit fault or an open circuit fault associated with the said one of the one or more of the capacitor banks is detected by sending a control signal to switch an isolation switch for the said one of the one or more of the capacitor banks to a non-conducting state, the isolation switch for each capacitor bank being positioned on a charge path between the voltage input and the capacitor bank.21 . A method according to claim 20, further comprising:sending a control signal to cause at least one of the one or more of the plurality of capacitor banks to be electrically connected to a load to generate an electric power pulse, wherein the control signal to cause the at least one of the one or more of the plurality of capacitor banks to be electrically connected to a load is a control signal to switch a discharge switch for each of the at least one of the one or more capacitor banks to a conducting state to electrically connect the respective capacitor bank to the load, each discharge switch positioned on a discharge path between the respective capacitor bank and the load.

22. A method according to claims 20 or 21 , wherein monitoring each of the one or more capacitor banks for a short circuit fault comprises monitoring a charging rate of each of the one or more capacitor banks, and / or wherein monitoring each of the one or more capacitor banks for an open circuit fault comprises monitoring a discharge rate of each of the one or more capacitor banks.

23. A method according to claim 20, wherein monitoring each of the one or more capacitor banks for an open circuit fault or a short circuit fault comprises monitoring the voltage across each of the one or more of the capacitor banks, preferably wherein a short circuit fault associated with a capacitor bank is detected if the voltage across said capacitor bank does not decrease upon a control signal for electrically connecting said capacitor bank to the load being sent.

24. A method of generating pulsed power for spacecraft propulsion, the method comprising: sending a control signal to cause a plurality of capacitor banks to be charged by a voltage input; and generating an electric power pulse by closing a discharge switch for each capacitor bank to electrically connect the capacitor bank to a load, each discharge switch positioned on a discharge path between the capacitor bank and pulse discharge connections connected to the load, wherein a first discharge switch of the plurality of discharge switches for a first capacitor bank is closed at a first time and a second discharge switch of the plurality of discharge switches is closed at a second time, wherein the first time is different from the second time.

25. A controller for a pulsed power source, the controller arranged to receive one or more monitoring signals and send control signals; wherein the controller is operable to perform the method of any of claims 20 to 24.

26. A computer-readable medium comprising instructions which, when executed by a computer, cause the computer to carry out the method of any of claims 20 to 24.

Images