Impedance detection of a consumer at a charging station
The active impedance measurement in charging stations differentiates between a connected consumer and a short circuit, addressing the risk of fires from foreign bodies by ensuring safe and efficient charging, even when the battery is not fully charged or switched off.
Patent Information
- Application Number
- DE102018214450
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2018-08-27
- Publication Date
- 2024-12-05
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Charging stations for robots and similar devices face the risk of short circuits due to metallic foreign bodies, which can lead to fires, especially when the devices dock autonomously and are not supervised, as existing passive detection methods fail to differentiate between the device and foreign objects based on battery voltage.
An active impedance measurement method is employed to distinguish between a connected consumer and a short circuit by applying a test signal and analyzing the complex-valued impedance, using both time and frequency domain measurements to ensure safe detection and charging.
This method reliably detects the presence of the consumer, reducing the risk of fire by minimizing energy input during detection and ensuring safe charging, even when the battery is switched off or not fully charged.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
[0001] The invention relates to a charging station (synonymously also called “base station”) for a consumer, in particular a robot, having the features of the preamble of claim 1. The invention further relates to an arrangement comprising the charging station and the robot, as well as a method.
[0002] Consumers, in particular energy-autonomous consumers, e.g., battery-operated handheld devices such as vacuum cleaners, etc., and in particular moving (household) robots, e.g., robot vacuum cleaners and robot lawnmowers, are powered by an energy storage device, in particular a battery, during operation and require a charging or base station at which they can charge their energy storage device. For this purpose, the base station generally has two charging contacts or electrodes, to which the charging voltage is applied and which can connect to the corresponding charging electrodes of a mating interface of the consumer. In the following, the present situation and the present invention are explained in part using a robot as an example, although all explanations can be applied analogously to other devices, in particular handheld devices, that dock at charging stations.The robot automatically and usually unattended returns to the charging station when the battery charge level drops significantly. The base station also typically serves as a resting station for the device when it has finished its work. For many devices, the presence of charging voltage at their charging electrodes is also an indication that they have reached their resting position at the base station.
[0003] Therefore, the charging voltage is usually continuously applied to the base station's usually exposed charging electrodes. If a foreign metal object (e.g., a paper clip, metal foil, or a piece of metallic cleaning wool) touches both electrodes, a short circuit can occur. A very low-resistance "deep" short circuit is usually prevented by a protective shutdown or current limiter in the base station's power supply, which is activated if the expected charging current is significantly exceeded. However, the charging currents when charging the consumer energy storage device are easily in the range of several amperes, so such a protective circuit may not activate if the short-circuiting metal part has a correspondingly not-so-low resistance value. However, a current of several amperes can be sufficient to heat a short-circuiting metal part to glowing red, thereby triggering or igniting a fire.This is especially true if a conductive foreign body still has highly flammable residue such as dirt, grease, oil, or plastic on it. Typical examples of such conductive foreign bodies are metallic cleaning wool or thin metal threads woven into textiles. The risk of a short circuit caused by metal parts falling or accidentally being "carried" to the base station by pets is particularly high if the base station's charging electrodes are at floor level. However, even with side-mounted contacts, a short circuit can occur, for example, due to metal threads woven into the user's blouse sleeves when the user handles the base station.
[0004] Many appliances have motors that contain permanent magnets. Their magnetic field often exhibits a significant field strength even on the underside of the appliance, so that the appliance itself may attract small metal particles, which it can then "pick up" as it moves through a room and carry to the base station. This effect significantly increases the likelihood that a foreign metal object will eventually land on the base station's charging electrodes.
[0005] EP 1 921 523 A2 describes a system in which, without a robot, a high-impedance voltage is applied to the charging electrodes of the base station via a precisely specified resistor, so that a short circuit would at most result in a harmlessly lower current flow. The robot contains the counterpart to the resistor in the base station, so that after the robot docks, a defined series resistance is created and a precisely defined current flows through the two resistors. Only when exactly this predetermined current flows do both the base station and the robot switch to charging mode, and the actual charging voltage with a high charging current is applied from the base station to the charging contacts.
[0006] DE 10 2010 042 227 A1 discloses a method for electrically charging an electrical energy storage device using a charging station, wherein electrical docking of the electrical energy storage device to the charging station is detected. The charging station supplies electrical charging energy to the docked energy storage device when communication signals exchanged between the charging station and the docked energy storage device are detected. The invention further relates to a charging station, an electrical energy storage device, and a mobile device.
[0007] EP 2 648 307 A1 discloses a robot control unit that includes a signal transmitting unit for electrical connection to a first docking terminal and for transmitting a predetermined detection signal, and a signal receiving unit for electrical connection to a second docking terminal. When the signal receiving unit receives a predetermined feedback signal corresponding to the predetermined detection signal, the robot control unit confirms that the robot with the docking terminal group has successfully docked to a docking station.
[0008] From EP 2 733 860 A1 it is known to connect an electric vehicle to a charging station via two charging cables, a control cable and a ground cable in order to supply it with power and to control the charging.
[0009] The invention is therefore based on the object of preventing a critical current flow through a short-circuiting foreign body in a charging station for a consumer, in particular a robot, and thus preventing the occurrence of a fire.
[0010] This object is achieved by a charging station for a designated consumer, in particular with the features of claim 1, by an arrangement comprising the charging station and the consumer, in particular a robot, with the features of claim 8, and by a method with the features of claim 9. Preferred or advantageous embodiments of the invention emerge from the subclaims, the following description, and the accompanying figures.
[0011] The subject of the invention is a charging station which is suitable and / or designed for an intended consumer, in particular a robot.
[0012] "Intended use" is to be understood here and below as follows: It is assumed that the basic conditions of the consumer are known for the charging station. This is possible because the charging station is designed for a specific consumer, consumer type, or certain basic conditions of consumers. Such consumers can then interact with the charging station. The charging station and consumer behave according to a "lock-and-key principle," i.e. their mutual basic conditions are coordinated with one another. For example, the charging station is designed for specific internal electrical properties (wiring, components, etc.) of the consumer.
[0013] The charging station can also be referred to as a base station, particularly in conjunction with a robot. The charging station is particularly stationary and serves to charge the consumer, in particular an energy storage device of the consumer. The energy storage device can be designed as a rechargeable battery, battery, capacitive energy storage device, etc. In addition, the charging station can also provide a garage for the consumer, in particular a robot. The consumer is particularly designed as a robot, particularly preferably as a household robot and / or mobile robot and / or autonomous robot. For example, the robot is implemented as a lawnmower robot or as a vacuum cleaner robot. The robot can also be designed as an aerial robot, in particular a drone. The consumer can dock or connect to the charging station in order to be charged.
[0014] The charging station has a charging interface formed by electrodes (hereinafter also referred to as “charging electrodes”). Particularly preferably, the charging interface has at least or exactly two electrodes, wherein one electrode is designed as a positive pole and the other electrode as a ground pole or negative pole. The consumer has (as intended, see above) a counter interface which is designed to be compatible and / or complementary to the charging interface. In particular, the charging interface and counter interface can be coupled in such a way that a charging current can be transmitted. The connection between the charging interface and the counter interface is contact-based and in particular not contactless. The charging current is thus transmitted in particular via cables, conductor tracks and other metallic conductors.
[0015] The consumer comprises a charging device and an energy storage device, wherein the counter interface is connected to the energy storage device via the charging device, so that the energy storage device can be charged via the counter interface. In particular, a charging voltage or a charging current can be applied to the counter interface, which serves to charge the energy storage device.
[0016] The charging station has a supply device for supplying the charging interface, in particular with energy and / or with charging current and / or charging voltage. The supply device can be set to a charging state and a non-charging state. In the charging state, a charging current and / or a charging voltage for charging the energy storage device is provided at the charging interface. In the non-charging state, the charging interface is de-energized, in particular by the supply device.
[0017] The charging station has a detection device for detecting the connection of the consumer to the charging interface. In particular, the detection device can output a status of "consumer connected / docked" or "consumer not connected / docked." The detection device can be embodied as a digital data processing device, such as a microprocessor or the like; alternatively or additionally, the detection device can be embodied as an analog circuit.
[0018] Within the scope of the invention, it is proposed that the detection device is designed, in particular in terms of programming and / or circuitry, to apply a test signal (voltage and / or current) to the charging interface when the supply device is not charged and, in this context (i.e., before and / or during and / or after the test signal is applied), to detect an electrical quantity based on the complex-valued electrical output impedance of the charging interface as a detection quantity for detecting the coupling of the consumer to the charging interface. The electrical quantity is, in particular, the voltage between the electrodes or the current flowing through at least one of the electrodes. The detection device thus registers an electrical quantity that is generated or at least influenced by the electrical output impedance.The output impedance is the impedance seen at the charging interface from the outside of the charging station, i.e., toward the "consumer side." Applying the test signal then produces a specific test current or test voltage at the charging interface, depending on the situation: whether the consumer is connected or not.
[0019] Other detection methods, for example, incorrectly identify interfering objects, particularly metallic interfering objects, as consumers. In contrast, the detection device according to the invention performs a detection based on the electrical properties and / or impedance of the energy storage device and / or consumer, thus ruling out confusion with foreign bodies, passive objects, particularly metallic interfering objects. Applying the test signal may result in only a reduced energy input into a metallic interfering object, thus reducing the risk of fire based on a current flow through a metallic interfering object. Thus, the invention provides reliable detection of the coupling of the consumer to the charging interface, while simultaneously reducing the risk of fire.
[0020] A particularly advantageous feature is that the charging station can be retrofitted to existing devices without modifying the device itself, meaning it can be used simply by replacing the charging station. No specific modifications to the device are necessary.
[0021] It is also an idea of the invention to propose a supplement or alternative to passive load detection based on impedance measurement. Passive load detection is described in the patent application (reference document) "Charging station for a load, arrangement comprising the load and the charging station, and method," filed on the same filing date, to which reference is hereby made. Protection can also be sought for the features disclosed therein; these features contribute to the technical aim of the invention and thus to solving the technical problem underlying the invention according to the application. The features are therefore implicitly part of the description of the invention contained in the filed application and thus part of the disclosure content of this application. The features are precisely defined and identifiable in the entire technical information content of the reference document.
[0022] The fundamental problem and task remain unchanged compared to a passive detection method, and the basic structure of the charging or base station, consisting of a power supply that generates the charging voltage and a downstream switching element that connects it to the charging contacts when the consumer is present, remains the same. The switching element switches between the charging and non-charging states. The switching element is part of, or therefore also forms, a disconnecting device, as explained in more detail below.
[0023] The only difference is the principle of load detection used in each case, i.e., the method used to determine whether the load is located on the charging contacts or whether a conductive foreign body is short-circuiting them. The methods considered in the parallel patent application are purely passive methods based on detecting leakage currents through the input protection diode or the battery voltage measurable at the load's contact electrodes (also called "counter electrodes"). In each case, the variables to be detected are determined directly or indirectly by the output voltage of the load's battery.
[0024] The major advantage of all passive methods is that during the detection phase, no energy is introduced (by the charging station) into a system that contacts the charging electrodes. In particular, no energy is transferred to a conductive foreign body that short-circuits the charging electrodes, so even in the worst case scenario, this foreign body cannot catch fire. Furthermore, passive methods are very fast because they are based on simple voltage or current measurements of direct voltages or currents.
[0025] The disadvantage of passive methods is that they rely on the existing battery voltage of the consumer. If the consumer, especially a robot, drives onto the station independently, this voltage will naturally be present, because the robot cannot move autonomously without a sufficient power supply. Most robot batteries – particularly lithium-ion batteries – have their own battery management system (BMS). To protect against harmful or, in the case of lithium-ion batteries, even dangerous deep discharge, if the battery voltage falls below a protective threshold, this system electrically disconnects the battery from the robot electronics, preventing any further power from being drawn from the battery. Due to the nature of the method, a robot with a robot battery disconnected by the BMS cannot be detected using passive methods, for example if the user manually places the robot on the base station to charge it.However, without charging voltage, the battery is no longer connected by the BMS.
[0026] In contrast, the active detection methods described here, which use impedance measurement, are slower and require test voltages or currents (test signal), so they also introduce energy into a potential short circuit. However, appropriate circuit design must ensure that this energy input is so low that it cannot cause a fire. The energy input from the charging station, however, now makes it possible to detect a load completely independently of the battery's charge level and then charge it using the applied charging voltage.
[0027] The detection method is designed to clearly, reliably, and (in the case of a robot that docks independently) sufficiently quickly identify the load, and in particular to distinguish it from a short circuit across the charging electrodes. The characteristic used is the (complex or complex-valued) impedance of the load or a short circuit. A short circuit in the form of a metallic foreign body bridging the charging electrodes almost exclusively exhibits a purely ohmic impedance. Only very small inductive (in the nH range) or capacitive (only in the pF range) components are possible.
[0028] In contrast, most loads have a capacitor (usually in the range of 10-100 nF) connected directly across the contact electrodes to suppress bounce effects and RF interference when driving onto the charging electrodes. In addition, there is usually a much larger buffer capacitor (usually in the range of 100 uF) located directly at the battery's voltage output. However, depending on the load's circuitry, this may be found on the load's counter electrodes. Therefore, if a capacitance greater than 10 nF is detected across the charging electrodes of the charging station, it can be assumed that a foreign metal object is not short-circuiting the charging electrodes, but that the load is contacting the charging contacts with its counter electrodes.
[0029] There are many different methods available for measuring impedance. They can be broadly differentiated according to whether they attempt to measure impedance in the time or frequency domain. Time-domain measurements (i.e., over time) use pulsed stimulation and measure, for example, the step or impulse response. Frequency-domain measurements, on the other hand, stimulate with alternating voltages of different frequencies or another signal with a defined frequency spectrum, and attempt to determine the frequency response or the change in the spectrum in whole or in part.
[0030] The invention relates in particular to a system consisting of a charging station / base station and a mobile autonomous system, the robot. The term "autonomous" is important here because the entire system is thus unmonitored. It involves unattended operation, which requires higher safety requirements than, for example, the use of a handheld vacuum cleaner, where a user can intervene during operation whenever a safety issue is discovered.
[0031] The problem that the invention solves for a robot can, however, also arise in a similar form for manually operated cleaning devices. Specifically, these are battery-operated handheld devices. When charging with a plug-in power supply, which is connected to the counterpart in the vacuum cleaner via a cable with a plug-in connector, this problem does not arise, as there are no large-area exposed electrodes. However, a consumer or device, e.g. a wet vacuum cleaner, which has a type of charging station on the floor on which the device can be placed and charged, has similar problems. Mechanically, the device is hung on the charging station and supplied with charging current via relatively large exposed electrodes (not on the floor, but on the side near the floor). Whenever the device is removed from the charging station, e.g.To clean the room, the charging station's charging electrodes are exposed, and essentially the same danger exists as with a robot system. When the device is removed, e.g., the user is vacuuming another room or even leaving the vacuum cleaner unused in another room while still charged (not uncommon in multi-story apartments), the charging station is in unattended operation—exactly the same situation as with a robot system. This safety problem can be solved by the invention in exactly the same way as with a robot.
[0032] In a preferred embodiment of the invention, the detection variable or the electrical variable contains the imaginary part of the output impedance, i.e., its capacitive or inductive component. In particular, this component is used exclusively. In particular, this is the electrical output capacitance. Optionally, the real part of the output impedance, in particular the ohmic output resistance between the electrodes, is also taken into account in the detection variable. "Output" is again to be understood as above. The electrical variable is therefore any complex-valued impedance with a reactive and / or capacitive component. Equivalent circuits include, for example, the series or parallel connection of a capacitor and a resistor. It can be assumed that such impedances of the load and foreign bodies generally differ, and thus an effective distinction can be made between docking / undocking of the load and a foreign body.
[0033] Alternatively or additionally, it is proposed that the charging station have a load resistor which is arranged in the circuitry between the electrodes of the charging interface. This can, for example, trigger discharge processes in the consumer which are initiated by applying the test signal. In this way, dynamic processes or impedance curves can be triggered, which then have different and thus distinguishable characteristics for connected / undocked consumers and interfering objects. Preferably, the load resistor is at least high-resistance. High-resistance is understood in particular to mean a resistance value which is dimensioned such that at least 95%, preferably at least 98%, of the residual storage voltage applied to the electrodes of the charging interface drops. For example, the load resistor can be greater than 1 MOhm, preferably greater than 5 MOhm and in particular greater than 9 MOhm.
[0034] The load resistor is preferably dimensioned such that less than 95%, preferably less than 90%, and in particular less than 85% of the residual storage voltage applied to the electrodes of the charging interface drops across the load resistor. In contrast to the previously described embodiment, the use of a moderately high-resistance load resistor eliminates or at least reduces interference that can arise with very high-resistance load resistors due to an antenna effect of the load resistors during voltage measurement. Since such interference can arise, in particular, from electrostatic charging and discharging, the alternative embodiment results in greater interference immunity.
[0035] In an alternative embodiment of the invention, the charging device of the consumer has a blocking diode. The blocking diode can also be referred to as an input protection diode. This diode serves, on the one hand, as reverse polarity protection if the consumer is placed incorrectly on the electrodes, and, on the other hand, has the task of preventing a possible short circuit if the counter electrodes of the consumer are short-circuited by a metal part, for example if a consumer with side-mounted counter electrodes touches a metal chair leg with both electrodes. Furthermore, the blocking diode serves to prevent the consumer from discharging via the charging station if the consumer is on the charging station with a charged energy storage device, but the charging station is not currently receiving any mains voltage, for example due to a power failure.This diode is preferably a power diode, which must be able to switch through the consumer's charging current of several amperes with virtually no loss.
[0036] In a preferred development of the invention, the charging station has a control device. The control device can be designed as a separate device or form an integral part of the detection device. In the event that the control device is designed as a separate device, it can be designed as a digital processing device and / or circuit-based processing device. The control device is, in particular, designed in terms of programming and / or circuitry to switch the supply device between the charging state and the non-charging state based on the detection variable. For this purpose, the control device can receive the detection variable and / or a status based on the detection variable, such as "consumer connected / docked" or "consumer not connected / docked."In particular, the control device is designed to switch the supply device from the non-charging state to the charging state in the case of the status “consumer coupled / docked” or a corresponding detection variable.
[0037] In a preferred structural adaptation of the invention, the charging station has a disconnecting device for interrupting the charging current and / or the charging voltage. The disconnecting device is designed, in particular, to decouple the charging interface, in particular at least one electrode of the charging interface. For example, the disconnecting device is designed as a switch (the aforementioned switching element), e.g., as a relay or MOSFET. The control device is designed to control the disconnecting device to switch the supply device between the charging state and the non-charging state.
[0038] It is preferred that the charging station has a charging current testing device for recording the charging current. When detecting the detection variable, it must be taken into account that a self-holding effect occurs in the charging state once the consumer has been identified and the charging voltage and / or charging current have been applied. Depending on the method selected, an impedance measurement would be possible even with a charging voltage applied (e.g., through measurements in the frequency domain), but this would be technically difficult and complex. The charging current testing device monitors the charging current so that it can be detected if the consumer interrupts or terminates the charging process. In this case, the charging current will collapse and / or drop to 0. While a lack of charging current is a necessary condition for undocking the consumer, it is not a sufficient condition.It is entirely possible that a consumer drives up to the charging station with a partially discharged energy storage device, fully charges the energy storage device there, and then does not draw any further charging current, but nevertheless does not leave the charging station because its work is complete and it now remains in standby mode on the charging station. Whether a small supply current for the current consumption of the consumer electronics is drawn from the charging station or from the energy storage device in this standby state depends on the design of the consumer. Optionally, the current consumption in standby mode can, for example, be drawn exclusively from the energy storage device. Optionally, the charging voltage could also be repeatedly interrupted briefly and, during the interruption, a consumer detection process could be carried out as described above.However, constant interruptions during a charging process would increase the overall charging time, and these constant interruptions would also be detrimental to the service life of the energy storage device. Therefore, the charging current is additionally checked (a lack of charging current is a necessary criterion for disconnecting the load), and no new load is detected as long as the charging current is flowing after the initial load detection.
[0039] It is preferably provided that the control device controls the supply device to switch from the charging state to the non-charging state as soon as it is detected via the charging current and / or the charging current testing device that the consumer has assumed the status "consumer not charging." In particular, the detection device is configured, in terms of programming and / or circuitry, to detect a further electrical variable based on the charging current as a further detection variable for detecting the charging of the consumer at the charging interface when the supply device is in a charging state.
[0040] According to the invention, the charging station contains a test signal generator, in particular a voltage source (or current source), for applying a pulsed source voltage (or for impressing a specific current signal) in the form of the test signal between the electrodes before and / or during the detection of the electrical quantity, and the detection device is designed to detect the electrical quantity in the time domain. A voltage pulse or a voltage jump is specified as the test signal. The detection device then measures and evaluates the pulse / step response. In particular, a pulsed application of a constant voltage source takes place. The test signal generator can be an independent signal source; it can also be formed implicitly by other components such as the supply device and, if applicable, the disconnecting device. The pulse or jump in the voltage can also be generated by the charge state orThe charging of the connected load is interrupted to perform the impedance measurement. This involves switching off the charging voltage, which leads to a particularly falling voltage step. The step response is then measured and evaluated. The interruption is, or leads to, the generation of the voltage pulse (the falling edge of the pulse).
[0041] One of the concepts behind the invention is that an impedance measurement is performed in the time domain. The hardware setup can be very similar to that for a passive measurement method. If the method is implemented with a microcontroller, identical hardware can even be used, requiring only the respective control software to be adapted. The associated advantages will be discussed later.
[0042] For example, the impulse response of the system at the charging electrodes is measured periodically. For this purpose, a short stimulus / stimulation pulse (duration tp, e.g., 0.2 ms) is periodically applied to the charging electrodes as a test signal. This is achieved, for example, by briefly turning on a switching element that connects the power supply to the electrodes in the charging state. A relay is less suitable as a switching element for this purpose; a MOSFET transistor is preferably used. When the control voltage at the gate is high, controlled by a control unit, the MOSFET is turned on, and the charging voltage is briefly applied to the charging electrodes. The charging voltage can then be measured during the pulse duration tp of the stimulation pulse.If a foreign body short-circuits the electrodes, the signal curve will be similar, but the amplitude of the signal is usually significantly lower, down to almost zero in the case of a full (very low-resistance) short circuit.
[0043] If, on the other hand, the consumer is docked, a capacitor (known to be present in the intended consumer) on or between the counter electrodes of the consumer becomes noticeable. In this case, the voltage will rise for a slightly longer period when the stimulus pulse is applied due to the charging of the capacitor and then remain constant for the remainder of the pulse duration tp. After the end of the stimulus pulse, the capacitor is slowly discharged via an internal resistance in the charging station between the electrodes, e.g. the load resistor. This results in the voltage curve typical of an exponential discharge curve of an RC combination, which asymptotically approaches zero (RC charge / discharge curve). By now, for example, increasing the voltage at the electrodes with a time delay Δt (e.g.If the voltage is measured at a time interval (e.g., 5 ms) after the end of the stimulation pulse, i.e., at time tM, a very clear indication is available for distinguishing between free charging electrodes and a short circuit from a connected load. If a significant voltage level is measured at time tM, a capacitive element, the load with its input circuit, must be present at the electrodes, since a short circuit generally does not have a significant capacitive component.
[0044] To improve noise immunity, it is recommended to perform not just one measurement sample, but several, and average them, or even to include the falling edge of the discharge curves in the evaluation, i.e., to verify that successive measurements must each exhibit monotonically decreasing values. Incidentally, the optional load resistor is not absolutely necessary, since discharging the known internal capacitance of the consumer is not necessarily required for the measurement. Without a load resistor, however, the measurement is very high-impedance and is therefore sensitive to electrostatic charges, which can easily occur near the floor, especially with synthetic fiber carpets.
[0045] How often the stimulus pulse must or should be generated depends on how quickly a load needs to be detected. If the load approaches the station independently, detection must occur quickly (e.g., in less than 100 ms) so that the switching element can switch on the charging voltage in time (transition from non-charging to charging state) before the load loses contact with the charging electrodes. In this case, the stimulus pulses must be applied to the charging electrodes at short intervals (e.g., 30 ms).
[0046] If, however, only the user's touching of the load needs to be detected, a stimulus pulse interval of a few seconds is sufficient. On the other hand, the product of the number of pulses per second and the pulse duration tp determines the energy introduced per second into a potential short circuit. For safety reasons, this energy should be as low as possible to reduce the risk of fire.
[0047] As with passive methods, a "self-holding effect" occurs here too: Once the load has been detected during docking and the charging voltage has been applied via the switching element, only the charging voltage can be determined at the electrodes, meaning that impedance measurement can no longer be carried out. Therefore, as with passive methods, undocking detection is required here too. This can be achieved, for example, as with passive methods, by interrupting the charging voltage from time to time (e.g., every few seconds) and then repeating the detection process during the interruption. In particular, the stimulus pulse during the interruption is even superfluous, since the charging voltage was still present immediately before the interruption and was charging the capacitor in the load.As with passive methods, the charging current can be measured and monitored to prevent interruptions in the charging process that could be detrimental to the battery. The corresponding considerations for passive methods can be applied analogously here.
[0048] In a preferred embodiment of the invention, the detection device is therefore designed to only determine detection of the load if the electrical variable in the time domain corresponds to an RC charging and / or discharging curve between the electrodes that is consistent with the input capacitance of the intended load, i.e. correspondingly expected values known or dimensioned from the intended load are determined. In particular, one or two or more measuring points, in particular a complete curve profile, in particular an edge steepness and / or edge shape of the profile of the electrical variable, e.g. the voltage between the electrodes, are recorded and evaluated. The typical RC curve goes exponentially asymptotically to zero when the test signal is removed.
[0049] During the determination, the output capacitance at the electrodes and thus the input capacitance of the load in the docked state are determined. The load is detected if a capacitance of at least 5, 8, 10, 15, or 20 nF is detected. Otherwise, non-detection occurs, meaning that no intended load is currently connected. Detection is therefore determined in particular by the presence of at least a predefined residual voltage between the electrodes at a predefined time after the test signal is switched off. The values are dimensioned according to the electrical characteristics of the charging station and the intended load.
[0050] In a preferred embodiment of the invention, the charging station contains a test signal generator, in particular a voltage source for applying a test signal between the electrodes before and / or during the detection of the electrical quantity, and the detection device is designed to detect the electrical quantity in the frequency domain. A test signal is therefore applied to the electrodes, where an electrical quantity is registered. The spectrum of the registered signal is then examined in whole or in part, and conclusions are drawn from this regarding the presence of the load, in particular a robot. In the simplest case, this is a simple sine signal with a variable frequency or, for example, a multi-tone signal (a superposition of several sine waves of different frequencies). However, a non-periodic signal can also be used. A typical example would be a noise signal.This is not only theoretically interesting, but also of practical importance. Pseudo-noise signals (pn sequences) can be easily generated using a microcontroller. The signal is then simply recorded at the electrodes and transformed into the frequency domain using an FFT. This allows one to observe, for example, whether the spectrum of what is essentially white noise is attenuated with increasing frequency (by a capacitor).
[0051] In other words, an alternating voltage, e.g. of different frequencies, is applied and the frequency response of all or part of the entire arrangement (charging station and, if applicable, consumers and / or foreign bodies) is determined and evaluated.
[0052] One of the concepts behind the invention is that, in addition to measurements in the time domain, measurements in the frequency domain are also possible. In this case, one does not measure the impulse or step response of the system located at the charging electrodes, but rather, for example, the frequency response. An example is presented below. In contrast to passive methods and measurements in the time domain, a voltage measurement input to or on the microcontroller is superfluous here - if present. Instead, the microcontroller also has a pulse input connected to a frequency counter. In general, a frequency counter must be provided.
[0053] The frequency counter or an associated circuit, for example, is an oscillator circuit from which a square-wave output signal is derived. The core is a parallel resonant circuit formed by a capacitor and the coil. The value of the capacitor should be in the order of magnitude of the expected load capacitance (e.g. 33 nF), and the coil should then be selected so that the resonant frequency of the resonant circuit is between 30 kHz and 500 kHz. Lower frequencies are unfavorable because they result in long settling times; higher frequencies can lead to problems with parasitic elements and also cause unwanted high-frequency interference. A comparator connected to the resonant circuit excites the resonant circuit to oscillate at the resonant frequency via feedback resistors. A comparator with complementary outputs, for example, is used.Such comparator ICs are known to those skilled in the art, for example, as IC LT1713 / LT1714 from Linear Technology. A (second) comparator stage connected to the output of the (first) comparator serves only to convert the oscillator signal into a clean square wave, which can then be fed to the frequency counter. The reference level for this (second) comparator is generated via resistors or a resistive voltage divider. The resonant circuit is connected to the electrodes via the coupling capacitors. In particular, the charging electrodes are thus connected in parallel with the resonant circuit in terms of AC current, while the coupling capacitors simultaneously provide galvanic isolation. The capacitance values of the coupling capacitors should be large compared to the value of the resonant circuit capacitor, i.e., 2.2 µF in this example. Unipolar ceramic capacitors are preferred for this purpose.
[0054] The capacitance measurement itself can only be carried out when the switching element is open (non-charging state), because when the switching element is closed, the circuit would primarily “see” a very large smoothing capacitor of the charging voltage source and would not be able to detect the additional contribution of the consumer capacitance (input capacitance).
[0055] While it would be possible to decouple the large smoothing capacitor from the signal using a choke, this would be very complex in the specified frequency range. Thus, this method also automatically results in a "self-locking effect," as previously described for the time-domain measurement. Only this time, it's not the charging voltage itself that's causing the interference, but rather the smoothing capacitor.
[0056] Frequency counting can easily be performed by the controlling microcontroller itself. To do this, the output of the aforementioned (second) comparator stage is connected to a pulse input of the microcontroller, which is internally connected to a counter stage. If the counter is first reset and then allowed to run for a predefined time, it will register all pulses at the microcontroller's input. After the predefined time, the counter reading is proportional to the oscillation frequency of the resonant circuit.
[0057] With free charging electrodes, the resonant circuit will oscillate at its resonant frequency. The two coupling capacitors to the resonant circuit have virtually no influence, as the parasitic capacitance between the charging electrodes will be a few pF at most. However, if the load is docked, its input capacitance is in series with the two coupling capacitors and parallel to the resonant circuit capacitor. Since the capacitance values of the coupling capacitors are significantly greater than the input capacitance of the load, the latter's value will dominate the series connection of the three capacitors. Consequently, with the load docked, its input capacitance is de facto parallel to the resonant circuit capacitance. This reduces the oscillation frequency significantly, but not drastically. The reduction in frequency can be easily evaluated using the connected frequency counter.
[0058] In the case of a metallic foreign body short-circuiting the charging electrodes, the series connection of the two coupling capacitors and the (very low) impedance of the foreign body are parallel to the resonant circuit. This means that a very large capacitance (around 1 µF in the above example) is parallel to the resonant circuit capacitor, thereby drastically reducing the oscillation frequency of the resonant circuit.
[0059] This now provides a feature that allows for easy differentiation between three situations based on the frequency measured by the counter stage: free electrodes, a short circuit across the charging electrodes, and a connected load. The isolation device can then be controlled in a manner analogous to passive methods or impedance measurement in the time domain. The "self-holding effect" must also be considered (without decoupling by a choke), and the charging current should also be monitored advantageously.
[0060] Another way to determine the system's potential capacitive impedance at the charging electrodes (air, short-circuiting foreign body, load) is to couple galvanically isolated, low-amplitude AC signals with different frequencies to the charging electrodes and then measure the current flowing in each case. When a load is connected, the current flow must be frequency-dependent due to the load capacitance (its input capacitance), whereas in a short circuit, it is not. Noise or pseudo-noise signals, such as white noise with a frequency-independent spectrum, can also occur instead of the AC voltage signals.
[0061] In a preferred development of the invention, the charging station therefore contains an LC resonant circuit whose capacitor is connected in parallel with the contacts, and a frequency counter for determining the current oscillation frequency of the resonant circuit. The detection device is designed to detect the load only if the oscillation frequency is within a predeterminable value range below the resonant frequency of the resonant circuit. This resonant frequency is the actual natural frequency of the isolated resonant circuit, in particular without parallel connection of additional capacitances. The predeterminable value range (with knowledge of the properties of the intended load to be detected) is therefore - as explained above - accompanied by the intended input capacitance of the load being connected to the charging interface and thus connected in parallel with the resonant circuit capacitor.
[0062] A further subject matter of the invention relates to an arrangement comprising the charging station, as described above or according to one of the preceding claims, and the consumer. In a preferred development of the invention, the charging device of the consumer has the blocking diode described above, wherein the blocking diode is arranged in circuitry between the counter interface and the energy storage device.
[0063] In a preferred embodiment, the consumer is a robot. The invention is particularly suitable for robots, e.g., due to the higher risk of short-circuiting objects being carried around compared to other consumers.
[0064] A further subject matter of the invention relates to a method for detecting the consumer by the charging station, as described above or according to one of the preceding claims. In particular, the charging station is part of the arrangement according to the invention.
[0065] The method comprises a step for detecting the coupling of the consumer to the charging interface, wherein when the supply device is not in a charging state, a test signal is applied to the charging interface and, in this context, an electrical variable based on the electrical output impedance of the charging interface is detected as the detection variable for detecting the coupling of the consumer to the charging interface, as previously described. Preferably, the voltage between the electrodes or the current flowing through at least one of the electrodes is used as the electrical variable. If the detection variable has been detected by the detection device and evaluated by the detection device and / or the control device, and the consumer has been identified, the supply device is switched from the non-charging state to the charging state.
[0066] In a preferred development of the invention, the charging current is detected by the charging current testing device, in particular cyclically and / or regularly, with a measurement being carried out every 5 seconds, for example. If the charging current is detected positively, the charging process is continued. In the event that the charging current is less than a predeterminable limit value for a predeterminable period, the control device and / or the detection device concludes that the charging process has ended or that the consumer is uncoupled or undocked from the charging station. In this case, the supply device is preferably set to the non-charging state and a test signal is optionally applied to the charging interface and docking detection is carried out via the electrical variable based on the electrical output impedance of the charging interface as a detection variable for detecting the coupling of the consumer to the charging interface.
[0067] In a preferred embodiment of the invention, the method according to the invention is carried out as an active method during an interruption of a passive method for detecting the load, as mentioned above, in which, among other things, no test signal is applied to the charging interface when the supply device is not in the charging state. In particular, the passive method is carried out cyclically comparatively frequently, e.g., every ms, whereas the active method according to the invention is carried out cyclically comparatively rarely, e.g., every 3 s.
[0068] As mentioned above, time-domain impedance measurement can be performed using the same hardware as the aforementioned passive methods, provided the control and evaluation are performed using a microcontroller. Therefore, both methods can be effectively combined. This provides the advantage of the relatively fast and reliable passive method for detecting autonomous docking of the load, but at the same time, albeit at a significantly reduced speed, time-domain impedance measurement can also be used to detect and charge a load with a battery disconnected from the BMS, which the user places on the charging / base station. For this purpose, a passive measurement is continuously performed every millisecond during docking detection.Every few seconds (in the example, every 3 seconds), the passive measurement is briefly interrupted, a stimulus pulse is applied, and the impulse response is briefly measured after the time Δt, as described above. If no significant capacitance is detected, the passive measurement is repeated. Compared to pure impedance measurement, this combination offers the advantage of significantly reduced energy input into a potential short circuit, since an active measurement is now performed only every few seconds instead of, for example, at 30 Hz.
[0069] The described passive method obviously works if and only if the energy storage device can provide the consumer with a supply voltage. It therefore fails if the energy storage device has switched off its output voltage to protect against deep discharge. Lithium-ion batteries, in particular, as energy storage devices have their own battery management system for safety reasons, which disconnects the battery from the environment if the battery voltage falls below a safe level. Therefore, as explained above, such a battery could not be charged with the charging station described. With the battery switched off, the consumer will obviously not be able to drive onto the charging station independently; instead, the user will place the battery onto the charging station.In this case, it is not unattended operation; rather, the user is present and could detect and remove a foreign object short-circuiting the electrodes. In this case, it would be perfectly acceptable to optionally apply the charging voltage briefly, independently of the load detection, so that a battery that has been switched off to protect against deep discharge can also be charged. The duration of the charging voltage application should be such that the battery is then charged sufficiently to cancel its shutdown. The charging voltage application can be initiated manually, for example via a switch that is to be operated by the user, particularly one that is concealed (concealed so that the load cannot accidentally activate it through movement). Another option would be to apply the charging voltage for a few seconds after the mains voltage has been switched on.The user can then charge the battery by briefly unplugging and replugging the charging station's power cord.
[0070] Further features, advantages, and effects of the invention will become apparent from the following description of a preferred embodiment of the invention and the accompanying figures. These show: Fig. 1 is a schematic block diagram of an arrangement comprising a charging station and a consumer as an embodiment of the invention; Fig. 2 a pulse diagram (a) and curves of the voltage U over time at (a) no-load, (b) short-circuit and (c) connection of a consumer to the charging interface; Fig. 3 a schematic state diagram of a method for detecting a consumer by a charging station. Fig. 4 a circuit for impedance measurement in the frequency domain.
[0071] The Fig. Figure 1 shows a highly schematic representation of an arrangement 1 with a charging station 2 and a load 3, in this case a robot. The load 3 is designed as a mobile load 3 that can move independently. For this purpose, the load 3 has a driving device (not shown), which, for example, comprises driven wheels. The load 3 is designed, for example, as a robot vacuum cleaner or a robot lawnmower, but can also be realized as an aerial robot. The charging station 2 is designed as a base station for the load 3 and enables the load 3 to be charged with electrical current.
[0072] For the electrical coupling between the charging station 2 and the consumer 3, the charging station 2 has a charging interface 4, and the consumer 3 has a counter-interface 5, which can be coupled to one another by means of electrical contact. The charging station 2 has electrodes 6a, b in the charging interface 4, and the consumer 3 has counter-electrodes 7a, b in the counter-interface 5. During the electrical coupling, the electrode 6a is contacted to the counter-electrode 7a, and the electrode 6b is contacted to the counter-electrode 7b.
[0073] The charging station 2 has a supply device 8, which comprises a switching power supply 9 for providing a charging voltage UL and a charging current I, which provides a direct voltage of, for example, 19 V. A capacitor C1 is provided for smoothing. The supply device 8 supplies the charging interface 4 or the electrodes 6 a, b with the charging voltage and / or the charging current I. To charge a battery, a constant charging current is preferably supplied; in this operating case, the supply device 8 is therefore preferably a constant current source. The corresponding charging method is known to those skilled in the art as CC (constant current) charging.
[0074] The charging station 2 or the supply device 8 can be set to a charging state, wherein the charging interface 4 is supplied with the charging voltage and optionally additionally with the charging current I. The supply device 8 can also be set to a non-charging state, wherein the charging voltage and / or the charging current I are not applied to the charging interface 4. Switching between the charging state and the non-charging state is implemented by a separating device 10, which is arranged between the supply device 8 and the charging interface 4, in particular the electrode 6a. In the exemplary embodiment, the separating device 10 is implemented as a controllable switching element.
[0075] A load resistor 12 (also designated "R1") is arranged between the electrodes 6a, b, via which the electrodes 6a, b are connected to one another, particularly inseparably and / or permanently. In this embodiment, the load resistor 12 has a resistance of 100 kΩ.
[0076] The charging station 2 has a detection device 11 which measures the voltage drop across the load resistor 12 or between the electrodes 6a, b or the voltage drop or the voltage U (between the potentials V+ and V-, here ground GND) as an electrical quantity G or detection quantity.
[0077] Furthermore, the charging station 2 has an optional charging current test device 14, which taps or detects the charging current I of the supply device 8 flowing via the charging interface 4. The detection device 11 is connected to the charging current test device 14 via signaling or circuitry, so that the charging current I is detected by the detection device 11. The charging current test device 14 is implemented, for example, as a shunt or a galvanically isolated Hall sensor. The charging current I is tapped between the supply device 8 and the isolating device 10, and the voltage U across the load resistor 12 is tapped between the isolating device 10 and the charging interface 4.
[0078] The charging station 2 has a control device 16, which can form an integral part of the detection device 11 or be designed as a separate component and connected to it via signaling. The control device 16 controls the isolating device 10 to switch the supply device 8 or the charging station 2 between the charging state and the non-charging state.
[0079] The load 3 has a charging device 17 and an energy storage device 18, such as a rechargeable battery, wherein the charging device 17 is electrically arranged between the counter interface 5 and the energy storage device 18, so that the charging current I is conducted from the counter interface 5 to the energy storage device 18. Furthermore, the charging device 17 has a blocking diode 19, which is connected between the counter electrode 7a and the energy storage device 18 to prevent a return current from the energy storage device 18 to the counter interface 5.
[0080] From a functional point of view, the arrangement 1, in particular the charging station 2, can detect a coupling or docking of the consumer 3 to the charging station 2, wherein the counter interface 5 is in electrical contact with the charging interface 4.
[0081] For this purpose, the supply device 8 is set to a non-charging state via the detection device 11 and / or the control device 16 by opening the separating device 10.
[0082] For detection, an impedance measurement is carried out in the time domain. Here, the impulse response of the system or arrangement 1 is periodically measured at the charging electrodes 6a, b. For this purpose, a short stimulation pulse 20 (duration, for example, 0.2 ms) is periodically applied to the charging electrodes 6a, b as a test signal 13 from a voltage source 28. This occurs by briefly switching on the switching element in the isolating device 10, here a MOSFET. The voltage source 28 is therefore formed here by the switched-mode power supply 9. A test signal generator 29 is realized through the interaction of the switched-mode power supply 9, the isolating device 10, and the control device 16 that controls them, and is symbolically shown in the Fig. 1 shown.
[0083] Fig. Figure 2 shows a pulse diagram (four individual diagrams, each plotted against time t) that explains the detection function. Fig. Figure 2a shows the control of the switching element or the isolating device 10 (gate voltage at the MOSFET). Also shown is the voltage curve of the voltage U applied to the charging electrodes 6a, b and measurable there, once with exposed electrodes 6a, b ( Fig. 2b), once with a (high-resistance) ohmic short circuit ( Fig. 2c) of the electrodes 6a, b and finally with the load 3 docked to the electrodes 6a, b ( Fig. 2d).
[0084] When the control voltage at the gate of the MOSFET is high (H), controlled by the control unit or control device 16, the MOSFET is turned on and the charging voltage is briefly applied as voltage U (until the transition to low L) to the charging electrodes 6a, b. Then, with the electrodes 6a, b open-circuited, the full charging voltage UL can be measured as voltage U during the pulse duration tp of the stimulation pulse ( Fig. 2b).
[0085] In the case of a foreign body short-circuiting the electrodes 6a, b (not shown), the signal curve will be similar, but the amplitude of the signal is usually significantly lower than the charging voltage UL, down to almost zero in the case of a full (very low-resistance) short circuit ( Fig. 2c).
[0086] If, however, consumer 3 is docked ( Fig. 2d), a capacitor C3 becomes noticeable as its decisive input capacitance, which is connected in the consumer 3 between its counter electrodes 7a, b. In this case, the voltage U between the electrodes 6a, b will increase somewhat longer or slower when the stimulus or stimulation pulse 20 is applied due to the charging of the capacitor C3 than in the Fig. 2b, c (as in the form of an RC charging curve 25) and then remain constant for the remainder of the pulse duration tp (e.g., approximately at the value of the charging voltage UL). After the end of the stimulus pulse 20, the capacitor C3 is slowly discharged via the load resistor 12 or R1. This results in the voltage profile of the typical exponential discharge curve 25 of an RC combination, which asymptotically approaches zero.
[0087] By measuring the voltage U at the electrodes 6a, b with a time delay Δt (e.g. 5 ms) after the end of the stimulation pulse 20, i.e. at the measuring time tM (indicated by measuring points of the current voltage U in Fig. 2b-d), one has a very clear indication of free charging electrodes 6a, b ( Fig. 2b) as well as a short circuit ( Fig. 2c) - namely U(tM) = 0 - from a docked consumer ( Fig. 2d) - namely (U(tM) not equal to zero). If a significant voltage level not equal to zero is measured at the measuring time tM, a capacitive element, in this case the load 3 with its input circuit or the capacitance C3, must be present at the electrodes 6a, b, since a short circuit ( Fig. 2c) basically has no significant capacitive component.
[0088] To improve the interference sensitivity, it is recommended alternatively to carry out not only one measurement sampling at the measurement time tM, but several (for example additional measurement times t1, 2, 3) and to average the measured voltages U over these or even to include the falling edge 22 of the RC discharge curves 25 in the evaluation, i.e. to check that successive measurements (tM, t1-3) must each have monotonically decreasing values of the voltage U.
[0089] The load resistor 12 or R1 can optionally be omitted, since a (rapid) discharge of the capacitance C3 is not necessarily required for the measurement. However, without the load resistor 12 or R1, the measurement is carried out with very high resistance and is therefore sensitive to electrostatic charging of the charging interface 4, which can easily occur near the floor, especially with synthetic fiber carpets.
[0090] The stimulation pulses 20 can be applied to the charging electrodes 6a, b at short intervals (e.g. less than 100ms, e.g. 30ms) in order, for example, to apply the charging voltage UL in time to a consumer 3 that is driving independently onto the charging station 2, before the consumer 3 loses contact with the charging electrodes 6a, b.
[0091] If, however, only the placement of the consumer 3 by the user needs to be detected, an interval of 20 stimulation pulses in the range of seconds is sufficient.
[0092] As soon as a suitable voltage U etc. is detected by the detection device 11, it is concluded that the consumer 3 is correctly docked to the charging station 2 and the isolating device 10 is closed in order to switch the supply device 8 from the non-charging state to the charging state.
[0093] Once load 3 has been detected during docking and the charging voltage UL has been applied via the switching element or the isolating device 10, only the charging voltage UL can be determined, meaning the impedance measurement can no longer be performed. Therefore, undocking detection is required. For this purpose, the charging voltage UL (isolating device 10 open) is interrupted from time to time (e.g., every few seconds), and the detection process is then repeated during the interruption.
[0094] A dedicated stimulus pulse 20 during the interruption is unnecessary because immediately before the interruption, the charging voltage UL was still applied to electrodes 6a, b and charged capacitor C3. This results in the falling edge 24 of stimulation pulse 20.
[0095] Alternatively or additionally, the charging current I can also be measured and monitored in order to prevent an interruption of the charging process that is technically disadvantageous with regard to the battery or energy storage device 18.
[0096] Optionally, this method can also be combined with a passive detection method (see above). This provides the advantage of a relatively fast and reliable passive method for detecting the autonomous docking of the robot or consumer 3, but at a significantly reduced speed, using impedance measurement in the time domain, it is also possible to detect and charge a robot or consumer 3 with a battery that has been switched off by the battery management system (BMS) and that is placed by the user on the base station, i.e., charging station 2. For this purpose, a passive measurement is continuously performed every millisecond during docking detection. Every few seconds (every 3 s in the example shown), the passive measurement is briefly interrupted, a stimulus pulse 20 is applied, and as previously described, the impulse response (voltage U) is briefly measured after the time Δt, etc.If no significant capacitance is detected (i.e., no capacitance C3), the passive measurement is repeated. Compared to pure impedance measurement, this combination offers the advantage of significantly reduced energy input into a potential short circuit, since an active measurement is now performed only every few seconds instead of, for example, at 30 Hz.
[0097] When the energy storage device 18 is charged by the supply device 8 following the successful detection of the load 3, the charging current I is monitored via the detection device 11. As soon as the charging current I falls below a predetermined value, it is assumed either that the load 3 has been undocked from the charging station 2 or that the charging process has been completed. To distinguish between these two states, the isolating device 10 can be opened and the detection process can be performed again to check whether the load 3 is still docked to the charging station 2.
[0098] In the Fig. Figure 3 shows a state diagram for monitoring the method for detecting the load 3 in the charging station 2: The symbol "10: -" means that the isolating device 10 is open, and "10: +" means that the isolating device 10 is closed. In this embodiment, charging occurs with a constant current I of 2.2 A to 2.5 A. The switching power supply 9 is designed as a constant current source.
[0099] In a step 100, a charging current and / or a charging voltage is provided by the supply device 8 for initialization, but at the same time the separating device 10 ( Fig. 2: 10: -) is opened so that the charging interface 4 is de-energized by the charging station 2.
[0100] In a step 200, with the separating device 10 open ( Fig. 2: 10: -) docking detection occurs, and stimulation pulse 20 is generated, and the voltage drop or voltage U across load resistor 12 is measured at measurement time tM. If the voltage drop is less than 2V, the measurement is repeated cyclically, in this embodiment every 3 seconds (deltat7 = 3 s). A passive measurement is also repeated cyclically at a time interval of deltat1 = 1 ms. Fig. 3 thus represents the combination of the active method in the time domain with the passive method with leakage current measurement (as described in the parallel patent application, see above).
[0101] As soon as a voltage drop greater than 2 V for a predeterminable duration, in this example 25 ms (deltat2 > 25 ms) is detected by the detection device 11, the control device 16 closes the isolating device 10 ( Fig. 2: 10: +), so that a charging voltage is applied to the charging interface 4 and a charging current can flow. The supply device 8 is set to the charging state. The charging state is maintained for at least a predeterminable period (wait), in this exemplary embodiment of 25 seconds (deltat3 = 25 s). In the exemplary embodiment shown, this predeterminable fixed time with the charging voltage applied serves, on the one hand, to give the consumer 3, which has docked to the charging station 2, sufficient time to initiate the charging process, and, on the other hand, to create better conditions for the next docking attempt for the consumer 3, which attempted to dock and in doing so triggered the consumer detection, but did not stop in time and passed over the charging interface 4.
[0102] In a step 400, the charging process takes place or is continued. The charging current is detected by the detection device 11 via the charging current testing device 14 and checked cyclically, in this exemplary embodiment every 5 seconds (delta4 = 5 s). If the charging current is above a predeterminable value, in this exemplary embodiment greater than 2 A, the charging process continues. If the charging current I remains below the predeterminable value for a predeterminable period of time, in this exemplary embodiment for 30 seconds (delta5 = 30 s), it is initially concluded that the charging process of the energy storage device 18 is complete.
[0103] In a step 500, the separating device 10 ( Fig. 2: 10: +) remains closed, so that the charging voltage U is applied and the charging current I is at least provided. Since the consumer 3 needs the charging voltage U as an indicator (that it is located on the charging station 2), the charging voltage U remains switched on. Therefore, the control device 16 will briefly interrupt the charging voltage U at short time intervals (e.g., a few seconds) and carry out a voltage measurement, thus using the implicit (negative) step response (switching off the voltage during the transition from state 500 to state 600). If the detection is successful, the control device 16 assumes that the consumer 3 is still located on the charging station 2 and switches the charging voltage U on again. Otherwise, the consumer 3 has obviously left the charging station 2 and the charging voltage U is switched off by the control device 16 via the isolating device 10 for safety reasons.Optionally, a stimulation pulse 20 can be applied repeatedly.
[0104] Thus, in this embodiment, undocking detection is carried out cyclically, every 30s (deltat6=30s), in a step 600. In the embodiment shown, undocking detection is carried out via the stimulation pulse 20 and the load resistor 12, wherein the separating device 10 is opened by the control device 16 and / or the detection device 11 ( Fig. 2: 10: -) and the voltage drop U across the load resistor 12 is detected by the detection device 11.
[0105] If the voltage drop U is above the predeterminable value, in this example 2 V, it is assumed that the consumer 3 is docked at the charging station 2. The detection device 11 and / or control device 16 then closes the isolating device 10 again in a step 700, and a charging current measurement is carried out by the charging current testing device 14 and the detection device 11. If the charging current I is above the predeterminable value, in this case 2 A, the process returns to step 400. If the charging current is below the predeterminable value, the process returns to step 500.
[0106] In the event that in step 600 the voltage drop U is below the predeterminable value, it is assumed that the consumer 3 is uncoupled or undocked from the charging station 2, and the process returns to step 200.
[0107] In addition to the time-domain measurement, a frequency-domain measurement is also possible. In this case, one does not measure the impulse or step response of the system located at the charging electrodes 6a, b, but rather, for example, the frequency response, i.e., a frequency as an electrical quantity G.
[0108] In Fig. 1. The voltage measurement input "U" to the microcontroller or to the detection device 11 is then superfluous. For this purpose, the microcontroller additionally has a pulse input EC connected to a frequency counter 15.
[0109] In addition, the Fig. The circuit shown in Figure 4 is required. The circuit is greatly simplified; in particular, the power supply lines and start and stop signals have been omitted. It is an oscillator circuit that outputs a square-wave signal. Its core is a parallel resonant circuit 26 formed by capacitor C10 and coil L10 (outlined in dashed lines). The value of capacitor C10 should be in the order of magnitude of the expected load capacitance (capacitor C3), i.e., 33 nF in the example shown, and L10 should then be selected so that the resonant frequency of the resonant circuit is between 30 kHz and 500 kHz.
[0110] The first comparator K1 excites the resonant circuit via the feedback resistors R10 and R11 to oscillate at its inherent resonant frequency fR. In this example, a comparator K1 with complementary outputs is used. Such comparator ICs are known, for example, as IC LT1713 / LT1714 from Linear Technology. The downstream (second) comparator stage K2 serves only to convert the oscillator signal into a clean square wave signal, which is then fed to the frequency counter 15 (via the pulse input EC). The reference level for this second comparator of the comparator stage K2 is generated by the resistors R12 and R13. The coupling or coupling capacitors C11 and C12 connect the resonant circuit to the Fig. 1 (connection points VA, VB). In particular, the charging electrodes 6a, b are thus connected in parallel with the parallel resonant circuit 26 in terms of alternating current, while the coupling capacitors C11, 12 simultaneously provide galvanic isolation. The capacitance values of the coupling capacitors C11, 12 should be large compared to the value of capacitor C10, i.e., 2.2 µF in this example. Unipolar ceramic capacitors are preferred for this purpose.
[0111] The capacitance measurement itself can or may only be carried out when the switching element in the isolating device 10 is open; when the switching element is closed, the circuit would be Fig. 4 in the direction of the connection points V1, B, namely, they primarily "see" the very large smoothing capacitor C1 and cannot measure the additional contribution of the load capacitance (capacitor) C3. While it would be possible to decouple the capacitor C1 from it using a choke, this would be very complex in the specified frequency range. Thus, this method also automatically results in a "self-holding effect," as previously described for the time-domain measurement. Only this time, it is not the charging voltage UL itself that is causing the interference, but rather the smoothing capacitor C1.
[0112] Frequency counting can easily be performed by the controlling microcontroller, i.e., the detection device 11, itself. For this purpose, the output of the comparator stage K2 (i.e., the pulse input EC) is connected to a pulse input of the microcontroller or the detection device 11, which is internally connected to a counter stage. If the counter is first reset and then allowed to run for a predefined time, it will register all pulses at the pulse input EC. After the predefined time, the counter reading is proportional to the current oscillation frequency or oscillation frequency fa of the oscillating circuit 26 as a quantity G.
[0113] With the charging electrodes 6a, b free, the resonant circuit 26 will oscillate at its resonant frequency fa = fR. The coupling capacitors C11 and C12 have virtually no influence, since the parasitic capacitance between the charging electrodes 6a, b will be a few pF at most.
[0114] However, if load 3 is connected, capacitance C3 is connected in series with coupling capacitances C11 and C12 in parallel with resonant circuit capacitor C10. Since C11 and C12 are significantly larger than C3, the value of C3 will dominate the series connection of the three capacitors C11, C3, and C12. Consequently, when docked, C3 is de facto connected in parallel with resonant circuit capacitance C10. This reduces load 3's oscillation frequency fa of resonant circuit 26 significantly, but not drastically, compared to the resonant frequency fR. The reduction in frequency fa can be easily evaluated using the connected frequency counter 15.
[0115] In the case of a metallic foreign body short-circuiting the charging electrodes 6a, b, the series connection of the two coupling capacitors C11 and C12 as well as the (very low) impedance of the foreign body is parallel to the resonant circuit 26. This means that a very large capacitance (in the example mentioned around 1 µF) is parallel to the resonant circuit capacitor C10 and thereby drastically reduces the oscillation frequency fa of the resonant circuit 26 compared to the resonance frequency fR.
[0116] This provides a feature that can be easily distinguished based on the frequency fa measured by the counter stage or frequency counter 15 between, firstly, free electrodes 6a, b, secondly, a short circuit across the charging electrodes 6a, b, and thirdly, a connected load 3. The control of the switching element in the isolating device 10 can then be carried out in a manner analogous to the passive methods or to impedance measurements in the time domain. The "self-holding effect" must also be taken into account (without decoupling by a choke), and the charging current I should also advantageously be monitored here. Due to the strong similarity to the other methods, the explicit representation in a state diagram according to Fig. 4 omitted here.
[0117] Another way to determine the possible capacitive impedance of the system at the charging electrodes 6a, b (air, short-circuiting foreign body, load) is to couple galvanically isolated, low-amplitude alternating voltage signals with different frequencies to the charging electrodes 6a, b and then measure the current flowing through one of the electrodes 6a, b. When the load 3 is connected, the current flow must be frequency-dependent due to the load capacitance C3, whereas in the case of a short circuit, it is not. List of reference symbols 1 arrangement 2 charging stations 3 consumers 4 Charging interface 5 Counter interface 6a, b Electrodes 7a, b Counter electrodes 8 Supply facility 9 Switching power supply 10 Separator 11 Detection device 12 Load resistance 13 Test signal 14 Charging current test device 15 frequency counters 16 Control device 17 Charging device 18 energy storage 19 Blocking diode 20 stimulation pulses 22 descent flank 24 flank 25 Charge / discharge curve 26 resonant circuit 28 Voltage source 29 Test signal generator G size C1 capacitor R1 load resistance I Charging current U voltage t time GND Ground BMS battery management system V+,- potential H High L Low tp pulse duration UL charging voltage C3 capacitor Δt time delay tM measurement time t1 - 3 time points EC pulse input C10 capacitor L10 coil K1 Comparator (first) R10, 11 resistor fR resonance frequency fa current oscillation frequency R12, 13 resistor K2 comparator stage (second) C11, 12 coupling capacitors VA, B connection point
Claims
[1] Charging station (2) for an intended consumer (3), with a charging interface (4) with electrodes (6a, b) for contacting the consumer (3), wherein the consumer (3) has a counter-interface (5), a charging device (17) and an energy store (18), wherein the counter-interface (5) is connected to the energy store (18) via the charging device (17) so that the energy store (18) can be charged via the counter-interface (5), with a supply device (8) for supplying the charging interface (4), wherein the supply device (8) can be set to a charging state and a non-charging state, with a detection device (11) for detecting the coupling of the consumer (3) to the charging interface (4), characterized by , that the detection device (11) is designed to apply a test signal (13) to the charging interface (4) when the supply device (8) is not in a charging state and, in this context, to detect an electrical quantity (G) based on the complex-valued electrical output impedance of the charging interface (4) as a detection quantity for detecting the coupling of the consumer (3) to the charging interface (4), wherein the charging station (2) contains a test signal generator (29) for applying the pulse-shaped test signal (13) between the electrodes (6a,b) before and / or during the detection of the electrical quantity (G), wherein a voltage pulse or a voltage jump is specified as the test signal (13), and the detection device (11) is designed to carry out the detection of the electrical quantity (G) in the time domain, wherein the pulse jump response is then measured and evaluated by the detection device (11). [2] Charging station (2) according to claim 1, characterized by that the detection variable is designed as an electrical output capacitance and / or ohmic output resistance between the electrodes (6a, b). [3] Charging station (2) according to claim 1 or 2, characterized by that it contains a load resistor (12) between the electrodes (6a, b) of the charging interface (4). [4] Charging station (2) according to one of the preceding claims, characterized by that the detection device (11) is designed to detect the consumer (3) only if the electrical variable (G) in the time domain corresponds to an RC charging and / or discharging curve (25) between the electrodes (6a,b) which is consistent with the input capacitance of the consumer (3) connected as intended. [5] Charging station (2) according to one of the preceding claims, characterized bythat the charging station (2) contains a test signal generator (29) for applying a test signal between the electrodes (6a, b) before and / or during the detection of the electrical quantity (G), and the detection device (11) is designed to carry out the detection of the electrical quantity (G) in the frequency domain. [6] Charging station (2) according to claim 5, characterized by that the charging station (2) contains an LC oscillating circuit (26), the capacitor (C1) of which is connected in parallel to the contacts, and a frequency counter (15) for determining the current oscillating frequency (fa) of the oscillating circuit (26), and the detection device (11) is designed to detect the consumer (3) only if the oscillating frequency (fa) is in a predeterminable value range below the resonance frequency (fR) of the oscillating circuit. [7] Arrangement (1) comprising the charging station (2) according to one of the preceding claims and the consumer (3). [8] Method for detecting the consumer (3) by the charging station (2) according to one of claims 1 to 6 and / or in an arrangement (1) according to claim 7, characterized by that when the supply device (8) is not in a charging state, a test signal (13) is applied to the charging interface (4) and in this context an electrical variable (G) based on the electrical output impedance of the charging interface (4) is detected as a detection variable for detecting the coupling of the consumer (3) to the charging interface (4). [9] Method according to claim 8, characterized by that it is carried out as an active method in an interruption pause of a passive method for detecting the consumer (3), in which no test signal (13) is applied to the charging interface (4) when the supply device (8) is not in the charging state.
Citation Information
Patent Citations
Method of Docking an Autonomous Robot
EP1921523A2
Method and charging station for electrically charging an electrical energy storage device
DE102010042227A1
Automatic walking device, charging station, docking system and docking method
EP2648307A1
Communication system
EP2733860A1
