JP7796103B2 - Systems and methods for charging and discharging electrochemical devices - Google Patents

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This PCT (Patent Cooperation Treaty) application is related to and claims priority from U.S. Patent Application No. 63/059,044, entitled "Systems and Methods for Electrochemical Device Charging and Discharging," filed July 30, 2020, the entire contents of which are incorporated herein by reference. This application is also related to co-pending U.S. Patent Application No. 17/232,975, entitled "Systems and Methods for Battery Charging," filed April 16, 2021, the entire contents of which are incorporated herein by reference.

Embodiments of the present invention generally relate to systems and methods for charging batteries, and more particularly to systems and methods for generating a high-efficiency and/or high-speed charging signal for charging a battery.

Many electrically powered devices, such as power tools, vacuum cleaners, any number of different portable electronic devices, and electric vehicles, use rechargeable batteries as a source of operating power. Rechargeable batteries are constrained by a finite battery capacity and must be recharged when depleted. Recharging batteries can be inconvenient because it often requires shutting down the device for the required time. In the case of vehicles, recharging can take several hours. For this reason, considerable effort has been expended to develop fast-charging technologies that reduce the time required to recharge batteries. However, fast-recharging systems typically have low efficiency, while slow-recharging systems result in lengthy recharging operations, defeating the fundamental purpose of fast usage.

As shown in FIG. 1A, at perhaps the simplest level, charging a battery involves passing a DC charging current through the battery. However, various types of batteries have a limited current they can accept before being damaged. FIG. 1A is a schematic diagram of a simple circuit 100 for recharging a single-cell battery. Other circuit components, such as an ammeter, voltmeter, and controller, are not shown. The battery 104 may be recharged by application of a recharge power signal from a controllable power source 102. Various embodiments involving charging and discharging, as discussed herein, are applicable to electrochemical devices such as batteries. The term "battery" in the art can be used in various ways to refer to individual cells having an anode and a cathode separated by an electrolyte, as well as collections of such cells connected in various configurations. Batteries typically include repeating units of a counter charge source and a first electrode layer separated by an ion-conducting barrier, often a liquid or polymer membrane saturated with the electrolyte. Because these layers are configured thinly, multiple units can occupy the battery's volume, increasing the battery's available power with each stacked unit. While many examples are discussed herein as applicable to batteries, cells, or battery cells, it should be understood that the described systems and methods may also apply to batteries including many different types of cells, as well as different possible interconnections of cells, such as cells coupled in parallel, series, and series-parallel. For example, the systems and methods discussed herein may also apply to battery packs including multiple cells configured to provide a specified pack voltage, output current, and/or capacity. Furthermore, the embodiments discussed herein may also apply to different types of electrochemical devices, such as, by way of example, various different types of lithium batteries (including, but not limited to, lithium metal batteries and lithium ion batteries), lead-acid batteries, various types of nickel batteries, and solid-state batteries. The various embodiments discussed herein may also apply to different structural battery configurations, such as button or “coin” type batteries, cylindrical cells, pouch cells, and prismatic cells. Application of a power signal to the electrodes of the battery 104 causes electrons to flow back through the battery, thereby replenishing the stored concentration of charge carriers (e.g., lithium ions) at the anode. In one particular example, the power source 102 may be a direct current (DC) voltage source that provides a DC charging current to the battery cells 104. Alternatively, different types of power sources may be used, such as a controlled current source.

Pulse charging is considered in some fast charging scenarios. FIG. 1B shows a graph 110 of a prior art DC voltage signal 122 generated by a power source 102 and applied to a battery cell 104 to recharge the battery. The graph shows the input voltage 112 of the charging signal 122 versus time 114. In general, the power source 102 can be controlled to provide repetitive pulses 122 to the electrodes of the battery cell 104 to recharge the battery cell. In particular, the power source 102 can be controlled to provide a repetitive square wave (shown as pulse 116 followed by pulse 118) signal to the battery cell 104. The peaks of the square wave pulses 116, 118 can be below a voltage threshold 120 corresponding to the operating constraints of the voltage source 102. A typical charging signal used to recharge the battery cell 104 can result in the charging signal being applied during the charging period, with rest periods of some duration between applications of the charging signal. Operation of circuit 100 in this manner produces a recharge signal 122 with a repeating square wave pattern, as shown in Figure 1B.

However, in some cases, recharging the battery cells 104 by applying the square wave charging signal 122 may shorten the lifespan of the battery cells being recharged or may reduce the efficiency of the battery recharge. For example, the abrupt application of a charging current to an electrode (typically the anode) of the battery cell 104 (i.e., the sharp leading edge 124 of the square wave pulse 116) may create a large initial impedance across the battery terminals. In particular, FIG. 1C illustrates a graph of the estimated real impedance value of the battery cell 104 versus the corresponding frequency of the recharging signal applied to the battery cell 104, according to one embodiment. In particular, graph 150 illustrates a plot of the real impedance value (axis 154) versus a logarithmic frequency axis (axis 152) of the frequency of the input signal to the battery cell 104. Plot 150 illustrates the real impedance value between the electrodes of the battery cell 104 at different frequencies of the recharging power signal used to recharge the battery. The shape and measurements of plot 150 may vary based on the type of battery, the state of charge of the battery, the operating constraints of the battery, battery heat, etc. However, the characteristics of the battery during charging can be generally understood from plot 158. In particular, the real impedance value at the electrodes of the battery cell 104 varies based on the frequency of the charging signal supplied to the battery, and generally, there may be a sharp increase in real impedance value 328 at higher frequencies. For example, an input power signal to the battery cell 104 at frequency f _Sq 162 may introduce a high real impedance 160 at the electrodes of the battery cell 104.

Referring again to the square-wave charging signal 122 in FIG. 1B, significant frequency signals may be present at the corners of the square-wave pulse 116. In particular, rapid changes in the charging signal to the battery cell 104 (such as the leading edge 124 of the pulse 116) can introduce noise consisting of high-frequency harmonics, such as at the leading edge of the square-wave pulse, the trailing edge of the square-wave pulse, and when using a conventional reverse pulse scheme. As shown in graph 150 in FIG. 1C, these harmonics create a large impedance at the battery electrodes. This high impedance can lead to many inefficiencies, including capacity loss, heat generation, and imbalances in electrokinetic activity across the battery cell, undesirable electrochemical responses at the charging interface, as well as degradation of materials within the battery cell 104, which can damage the battery and shorten its lifespan. Furthermore, cold-starting a battery with a fast pulse initiates capacitive charging and diffusion processes, thereby limiting faradaic action. During this time, the proximal lithium reacts and is quickly consumed, leaving a period of unwanted side reactions and diffusion-limited conditions that adversely affect the health of the cell and its components. These and other inefficiencies are particularly detrimental during fast recharges of the battery cell 104, which often involve relatively large currents.

It was with these observations in mind, among other things, that various aspects of the present disclosure were conceived and developed.

Aspects of the present disclosure include a charging system including a charging signal shaping circuit. The system further includes a controller in operative communication with the charging signal shaping circuit to control the charging signal shaping circuit to define a charging signal for the electrochemical device based on harmonics associated with a value representative of current flow to the electrochemical device. The system further includes a power converter operatively coupled to the electrochemical device to supply power to a load.

In another aspect, the power converter is in operative communication with a controller. The controller is configured to control the power converter to generate a discharge waveform from the electrochemical device based on harmonics associated with a value representative of current flow from the electrochemical device. In another aspect, the charge signal includes a series of regulated charge pulses, the discharge signal includes a series of regulated discharge pulses, and the controller controls the charge signal shaping circuit and the power converter to interleave the series of regulated discharge pulses with the series of regulated charge pulses.

These and other aspects of the disclosure are described in more detail below.

1 is a schematic diagram of a conventional circuit for charging a battery. FIG. 1 is a signal diagram of a prior art DC voltage or current signal for recharging a battery. 1 is a graph of estimated real impedance values of a battery versus corresponding frequencies of a charging signal applied to the battery, according to one embodiment. FIG. 1 is a schematic diagram illustrating a circuit for charging a battery utilizing a charging signal shaping circuit, according to one embodiment. 1 is a graph of a sinusoidal cell charging signal having a frequency corresponding to a determined minimum real impedance value of a battery cell, according to one embodiment. 1 is a graph of measured real impedance values of a battery versus corresponding frequencies of a charging signal applied to the battery cells, according to one embodiment. FIG. 2 is a schematic diagram illustrating a circuit for shaping a battery charging signal based on a frequency corresponding to a minimum impedance value, according to one embodiment. 1 is a flow chart illustrating a method for generating a charging signal for a battery based on a frequency corresponding to a minimum impedance value, according to one embodiment. 1 is a graph of superimposed square wave and sine wave pulses of a battery charging signal, according to one embodiment. 1 is a graph of measured real impedance values of a battery versus corresponding frequencies of a charging signal applied to the battery at specified maximum and minimum frequencies, according to one embodiment. FIG. 1 is a signal diagram of a shaped battery charge pulse having multiple frequencies corresponding to the highest and lowest frequency real impedance values within a range of acceptable values based on the battery cell's displayed impedance, according to one embodiment. 1 is a flow chart illustrating a method for generating a charging signal for a battery based on a range of frequencies corresponding to maximum and minimum real impedance values of a battery cell, according to one embodiment. FIG. 4 is a signal diagram of a first shaped charging pulse train generated from a battery charging circuit, according to one embodiment. FIG. 10 is a signal diagram of a second shaped charging pulse train generated from a battery charging circuit, according to one embodiment. FIG. 10 is a signal diagram of a charging signal showing real and imaginary impedance values of a battery as applied to the battery over time, according to one embodiment. 1 is a graph of measured real, imaginary, and reference impedance values of a battery cell versus corresponding frequencies of a charging signal applied to the battery cell, according to one embodiment. FIG. 10 is a signal diagram of a shaped battery cell charging signal including a leading edge and a body portion generated from a battery charging circuit, according to one embodiment. 1 is a plot of the measured voltage drop and current across a battery as the battery is charged in response to a charging signal applied to the battery, according to one embodiment. 1 is a plot of the measured voltage drop and current across a battery as the battery is charged in response to a charging signal applied to the battery, according to one embodiment. 1 is a plot of measured current and voltage across a current sense resistor at a battery versus time in response to a charging signal applied to the battery, according to one embodiment. FIG. 1 illustrates an example of a computer system that can be used to implement embodiments of the present disclosure. FIG. 1 is a diagram of a circuit for defining a charging signal and providing a power conversion function to power a load while charging the load's battery, in one example. FIG. 1 is a diagram of a circuit for defining a charging signal and providing a step-down function to power a load while charging the load's battery, in one example. FIG. 1 is a diagram of a circuit for defining a charging signal and providing a boost function to power a load while charging the load's battery, in one example. FIG. 16 shows an example of a harmonically adjusted charging pulse produced by a control signal applied to the circuit of FIG. 15. 18B shows an example of a control pulse of a control signal that, when applied to the circuit of FIG. 15, produces the harmonically adjusted charging signal of FIG. 18A. FIG. 16 illustrates an example of control pulses for driving a charge forming power converter such as through the circuitry shown in FIG. 15, in one example. FIG. 2 illustrates an example of a PWM signal for driving a step-down or step-up power converter. FIG. 19D illustrates an example of a harmonic shaped output current waveform generated from a duty cycle such as that shown in FIG. 19D, in one example. FIG. 10 illustrates an example of changing the duty cycle of a buck or boost power converter to shape the discharge pulse according to harmonics with a transition from an initial short on cycle to a long on cycle in an example. FIG. 1 is a diagram illustrating an example of a charging signal forming circuit including a parallel boost circuit. FIG. 1 illustrates an example of a charge signal shaping circuit including a parallel step-down circuit, according to an example embodiment. FIG. 1 is a diagram illustrating an example of a charging signal forming circuit including a parallel boost circuit. FIG. 1 illustrates an example of a charge signal shaping circuit including a parallel step-down circuit, according to an example embodiment.

Disclosed herein are systems, circuits, and methods for charging (recharging) and discharging batteries. The terms charging and recharging are used synonymously herein. The systems, circuits, and methods discussed herein allow batteries to be charged or discharged with energy more efficiently than conventional charging circuits and methods. As discussed herein, in addition to energy efficiency, several other benefits are realized, alone or in combination with the efficiency. For example, the charging and/or discharging techniques described herein can reduce the rate at which anodes are damaged, reduce heat generated during charging or discharging (or provide a method for controlling heating), and provide multiple benefits, such as reduced anode and cell damage and reduced risk of fire or short circuit. In another example, the charging techniques described herein can enable faster charging by applying a high charge rate to the cell. Over what are considered standard charge or discharge rates, the techniques described herein can provide relative improvements in cycle depth and/or cycle life. In one example, during periods considered to be "slow charging" of a battery, the disclosed systems and methods result in longer battery life and charging energy efficiency. In another example, during periods considered to be "fast charging," the disclosed systems and methods improve the balance between charge rate and battery life while reducing heat generation. While conventional charging circuits have focused on the electronic devices of the charging circuit to address charging circuit efficiency, the disclosed systems, circuits, and methods provide an efficient battery charging signal for battery charging applications.

In one example, various embodiments discussed herein charge and/or discharge a battery by generating pulses of a charge or discharge signal corresponding to one or more frequencies (which may be one or more harmonics) associated with optimal energy transfer based on real and/or imaginary values of energy transfer to and/or from the battery cells. In one example, the frequency may be associated with a minimum real impedance value of the battery. In another example, the pulses of the charge signal correspond to harmonics associated with both the real and imaginary impedance values of the battery. In yet another example, the pulses of the charge signal may correspond to harmonics associated with one or both of the conductance and susceptance of the battery cell's admittance. More specifically, systems and circuits are described that determine the frequency corresponding to the minimum impedance value. In some examples, the techniques discussed herein enable the minimum impedance frequency to be reevaluated because the frequency resulting in minimum impedance may change with state of charge, temperature, and other factors. These circuits can shape or generate pulses of the charge signal (e.g., charge current) that correspond to harmonics or frequencies associated with minimum impedance. As introduced above, as the state of charge and temperature fluctuate during recharge and discharge, the frequency corresponding to the minimum impedance value may change due to changes in material properties, chemical processes, and electrochemical processes within the battery. Accordingly, circuitry may optionally perform an iterative process of monitoring or determining the frequency corresponding to the battery's minimum impedance value and adjusting charge pulses to and/or discharge pulses from the battery. This iterative process may improve the efficiency of the charge or discharge signals, thereby reducing battery recharge times, extending battery life (e.g., increasing the number of charge/discharge cycles that can be performed), optimizing the amount of current to and from the battery, and avoiding energy loss due to various inefficiencies, among other benefits.

To generate charge pulses with appropriate harmonic content, the battery recharging circuit may include one or more charge pulse shaping circuits, impedance measurement circuits (including both hardware and/or software components), and/or application-specific integrated circuits. In one particular embodiment, the charge pulse shaping circuit may include a filter circuit controllable by a pulse control signal. The filter circuit may prevent rapid changes in the charge pulse sent to the battery cells. In particular, the filter circuit may shape the input current square wave based on Z = jωL so that high frequencies limit current flow and low frequencies allow current to flow through the circuit. Selection of filter circuit components can shape the leading edge of the charge pulse to maximize power delivered to the battery cells while limiting inefficient harmonics present in conventional square wave power signals. The pulse control signal to the filter circuit can also set the duration of each frequency-adjusted charge pulse delivered to the battery cells. The charge signal shaping circuit may also include a current shaping circuit controllable by a current shaping control signal. In one embodiment, the current shaping circuitry may modify the magnitude of the charge pulse by removing or draining current from the pulse prior to its application to the battery cell. The shaping component may also be responsible for defining the trailing edge of the pulse, the pulse duration, the voltage level between pulses, and other functions.

The systems, circuits, and methods disclosed herein are applicable to charging any type of battery, which may include multiple cells connected in some manner to achieve the desired capacity, voltage, and output current range for any application. Additionally, various embodiments discussed herein are believed to enable fast charging. In one or both situations, the circuitry may be controlled to provide a recharge pulse with a shaped leading edge rather than the sharp edges associated with a traditional square wave. In one example, the leading edge of the charge pulse may be based on a determined frequency (harmonic) corresponding to a harmonic associated with a minimum or near-minimum real impedance value of the battery cell. The charge pulse may also be based on a combination of the minimum real and imaginary impedance of the cell being charged. In another example, the charge pulse may be based on the conductance and/or susceptance of the battery cell being charged, or any other admittance factor, alone or in combination. Still other aspects of the battery cell are contemplated and can be used to shape the charge pulse. Generally speaking, when real and imaginary impedance values are considered, the techniques evaluate the harmonic values where these values, alone or in combination, result in relatively low impedance. These techniques evaluate the admittance values where the conductance and susceptance of that admittance, alone or in combination, result in relatively high harmonics.

Discussing pulses based on real impedance minimums, applying a rising leading edge that corresponds approximately to the minimum real impedance value can eliminate inefficient or harmful high-order harmonic content in the charging signal. Furthermore, the duration of the charging pulse can be controlled by the circuit to maximize or increase the amount of power applied to the battery in the pulse without exceeding one or more upper thresholds in the magnitude of the charging pulse, among other things, damaging the battery and affecting its capacity or lifespan. Applying such pulse-shaped charging signals under circuit control can deliver an optimized amount of power to the battery in each pulse while simultaneously eliminating degradation-causing high-frequency harmonics from the signal. Such shaped charging signals can improve the efficiency and speed of battery cell recharging by lowering the impedance across various interfaces within the battery, including the electrodes, as the battery cells are charged.

FIG. 2 is a schematic diagram illustrating a circuit 200 for recharging a battery cell 204 utilizing a charge pulse shaping circuit 206 and an impedance measurement circuit 208, according to one embodiment. Generally, the circuit 200 may include a power source 202, which may be a voltage source or a current source. In one particular embodiment, the power source 202 is a direct current (DC) voltage source, although an alternating current (AC) source is also contemplated. More specifically, the power source 202 may include a DC power source that provides unidirectional current, an AC power source that provides bidirectional current, or a power source that provides ripple current (e.g., an AC signal with a DC bias to make the current unidirectional). Generally, the power source 202 provides a usable charging current that is shaped to charge the battery cell 204. In one particular embodiment, the circuit 200 of FIG. 2 may include a charge signal shaping circuit 206 that shapes one or more pulses of a charging signal for use in charging the battery cell 204. In one example, a circuit controller 210 may provide one or more inputs to the charge signal shaping circuit 206 to control the shaping of the charging signal. These inputs may be used by shaping circuit 206 to modify the signal from power source 202 into a more efficient charging signal for battery cells 204. The operation and configuration of charging signal shaping circuit 206 is described in more detail below.

In some cases, the charge signal shaping circuit 206 may modify energy from the power source 202 to generate a charge pulse that corresponds at least in part to a harmonic associated with a minimum real impedance value of the battery cell 204. Alternatively, the cell may be characterized so that the impedance is known at any given charge current, voltage level, charge level, number of charge/discharge cycles, and/or temperature, among other factors, and may be looked up from a memory instead of directly measuring the impedance. In one example, the circuit 200 may include an impedance measurement circuit 208 connected to the battery cell 204 to measure the cell voltage and charge current as well as other cell characteristics, such as temperature, and to measure or calculate the impedance between the electrodes of the cell 204. In one example, the impedance may be measured based on the applied pulse. Alternatively, the impedance may be measured as part of a routine to characterize the cell by applying signals of different frequency characteristics to generate a range of impedance values associated with the cell's different frequency characteristics, which may be performed before, during, or periodically during charging, and may be used in combination with lookup and other techniques. The cell impedance may include a real value and an imaginary value, i.e., a reactance value. The impedance of the battery cell 204 may vary based on many physical or chemical properties of the cell, such as the cell's state of charge and/or temperature. Therefore, the impedance measurement circuit 208 may be controlled by the circuit controller 210 to, among other things, determine various impedance values of the battery cell 204 as the cell is recharged and provide the measured impedance values to the circuit controller 210. In some cases, the circuit controller may provide the real component of the measured impedance of the battery cell 204 to the charge signal shaping circuit 206 so that energy from the power source 202 can be transformed into one or more charge pulses corresponding to harmonics associated with the lowest real impedance value of the battery cell 204. In another example, the circuit controller 210 may generate one or more control signals based on the received real impedance values and provide these control signals to the charge signal shaping circuit 206. The control signals may, among other functions, shape the charge pulse to include harmonic components corresponding to the real impedance values. In yet another example, the charge signal shaping circuit 206 may modify energy from the power source 202 to generate a charge pulse that corresponds at least in part to a harmonic associated with the conductance or susceptance component of the admittance of the battery cell 204 or any other factor associated with the impedance of the battery cell. Thus, although described herein in terms of real or imaginary components of impedance, these systems and methods may similarly measure or consider other characteristics of the battery cell, such as the conductance or susceptance component of the admittance of the battery cell.

FIG. 3A is a graph 302 of an example of a sinusoidal charging signal that may be generated by circuit 200 of FIG. 2, having a frequency corresponding to the determined minimum real impedance value of battery cell 204. In this example, the frequency of the sinusoidal signal itself is the frequency corresponding to the minimum real impedance of the battery cell being charged. More specifically, graph 302 shows a plot 314 of the input voltage axis 304 versus the time axis 306 of the charging signal delivered to battery cell 204. In contrast to the square wave charging signal described above, the charging signal generated by circuit 200 may include a repetitive sinusoidal charging signal delivered to battery cell 204. While only two pulses (pulses 308, 310) are shown in FIG. 3A, it should be appreciated that such a pulse train may be delivered to the battery cell for a period sufficient to charge the battery cell to a certain level. The frequency of the sinusoidal wave may, and likely will, vary over time depending on the impedance of the battery cell and the control scheme implemented. As discussed herein, the frequency of the shaped pulses and sine waves may be set to the minimum impedance or near the minimum impedance (above the minimum impedance, below the minimum impedance, or both), depending on the implementation. Thus, the frequency need not be set exactly at the minimum impedance. The sine wave pulses 308, 310 of the charging signal 314 may continue to be generated and delivered to the battery cell 204 during the recharging operation of the circuit 200. The sinusoidal nature of the charging signal 314 may filter out high frequency noise components typically present in charging signals with a square wave profile, thereby lowering the impedance at the battery cell 204 and improving the efficiency of the recharging operation. Additionally, the charging signal 314 may include a settling or depolarization period 316 of some duration between the pulses 308, 310. The duration of the settling period 316 may be adjustable or controllable by the circuit controller 210 and may be based on various aspects of the recharging operation of the battery cell 204, including, but not limited to, the total power delivered by previous pulses 308 of the charging signal 314, the state of charge of the battery cell 204, the measured or estimated temperature of the battery cell 204, the measured impedance of the battery cell 204, and/or the hardware components used in the charging circuit. For example, the duration of the settling period 316 may be based on the processing speed of the circuit controller 210 to allow adequate time for the control circuit 210 to determine one or more target values for control of the charging circuit 200. Additionally, the magnitude of the pulses 308, 310 may be below a voltage threshold 312. The voltage threshold 312 may be based on several factors of the battery cell 204 and/or the power source 202, such as an upper voltage or current threshold of the power source and/or thermodynamic boundaries associated with the voltage, temperature, and current of the battery cell 204. In some cases, the voltage threshold 312 may be controlled by the circuit controller 210, as described in more detail below.

In one particular example, the frequency or harmonics of the sinusoidal pulses 308 of the charging signal 314 generated by the circuit 200 to recharge the battery cell 204 may be selected by the circuit controller 210 so that application of the charging pulses minimizes the impedance at the battery cell 204. For example, FIG. 3B illustrates a graph 322 of measured real impedance values of the battery cell 204 versus the corresponding frequency of the charging signal applied to the battery cell, according to one embodiment. In particular, graph 322 illustrates a plot of real impedance values (axis 324) versus the logarithmic frequency axis (axis 326) of the charging signal. Plot 328 illustrates real impedance values between the electrodes of the battery cell 204 at different frequencies of the sinusoidal charging signal. As shown, the real impedance values 328 may vary based on the frequency of the charging signal, with the real impedance values 328 generally increasing rapidly at the highest frequencies. However, plot 334 of the real impedance values of the battery cell 204 also illustrates a minimum real impedance value 330 corresponding to a particular charging signal frequency, labeled f _Min . The plot of real impedance values 334 of a battery cell 204 may depend on many factors of the cell, such as battery chemistry, state of charge, temperature, configuration of the charging signal, etc. Thus, the frequency f _Min 332 corresponding to the minimum real impedance value 330 of a battery cell 204 may likewise depend on the characteristics of the particular battery cell 204 being charged. The frequency f _Min 332 may correspond to other factors of the battery cell 204, such as the configuration of the cells in the pack and the connections between the cells in the pack.

Due to potential inefficiencies in the impedance of the battery cell 204, such as converting received power to heat, generating sinusoidal charging pulses 308, 310 at or near a frequency 332 corresponding to a minimum real impedance value 330 of the battery cell 204 may improve the efficiency of application of charging energy to the battery cell 204. In other words, shaping the pulses 308, 310 of the charging signal 314 to include harmonics at or near the frequency f _Min 332 may improve the efficiency of the charging signal 314 to the battery cell 204 by reducing wasted energy converted to heat by the impedance of the battery cell 204. To this end, one embodiment of the recharging circuit 200 of FIG. 2 may include an impedance measurement circuit 208 coupled to the battery cell 204 to determine various real impedance values of the battery cell over a range of frequencies of the charging signal. The impedance measurement circuit 208 may include any known or later-developed circuit configured to measure the impedance between the electrodes of the battery cell 204, such as a voltage sensor and a current sensor. Multiple impedance values of the battery cell 204 may be measured at different frequencies of the charging power signal and provided to the circuit controller 210 to determine or estimate a minimum real impedance value of the curve 334 of the battery cell 204. The circuit controller 210 may then control one or more components of the charging signal shaping circuit 206 to generate a series of sinusoidal charging pulses 308, 310 at harmonics of the frequency f _Min 332 corresponding to the minimum real impedance value 330 of the battery cell 204. As described in more detail below, the circuit controller 210 may also perform an iterative process to measure or otherwise determine estimated real impedance values for the current state of the battery cell 204 at various times during a recharging session, and accordingly adjust the pulses 308, 310 of the charging power signal 314 to match the newly estimated frequency f _Min 332. By controlling circuit 200 to generate charging signal 314 having a harmonic frequency of pulses 308, 310 based on the determined or estimated minimum real impedance value, the energy of charging signal 314 can be more efficiently applied to recharging battery cells 204 while minimizing wasted energy due to high impedance at the electrodes caused by the high frequency portion of the charging signal.

One particular embodiment of a circuit for charging battery cells using charge pulse shaping is shown in FIG. 4. Circuit 400 may be controlled by controller 210 to shape the battery cell's recharge signal based on a frequency f _Min corresponding to a minimum impedance value. In one example, controller 210 may be a feedback control system using a voltage or current amplifier. In general, controller 210 may be an analog controller, a digital controller, a microcontroller, or a microprocessor, or a customized integrated circuit such as an application-specific integrated circuit (ASIC). Controller 210 may be configured or programmed to perform one or more of the operations discussed herein to control the performance of shaping circuit 400. Additionally, as described below, circuit 400 may consider the imaginary component of impedance, the conductance component of admittance, the susceptance component of admittance, or any combination thereof. Circuit 400 may include more or fewer components, or may be replaced by other components of equivalent functionality. In some implementations, some components may be replicated in parallel to charge multiple cells in parallel or to provide greater charging capacity for a given cell or configuration of cells. Circuit 400 of Figure 4 is just one example of a power signal shaping circuit that can be controlled to provide a harmonic sine wave charging signal as discussed herein.

Circuit 400 may include a power supply 402 coupled to rail 442 to provide a charging signal to battery cell 404. Power supply 402 may be any type of energy source, such as a DC voltage source, an AC voltage source, or a current source. In some implementations, power supply 402 may be controlled via an input (V _CONT 434) to vary the waveform or pulse magnitude of the energy provided to circuit 400. For example, circuit controller 210 may provide a control signal V _CONT 434 to power supply 402 to turn on the power supply, select the magnitude of the power signal, select between a DC power signal and an AC power signal, etc. In one particular example, power supply 402 may be configured to adjust the magnitude of the charging signal it provides based on the voltage value of the received V _CONT 434 signal.

A filter circuit 406 may be connected to the power supply rail 442 to receive power generated by the power supply 402. The filter circuit 406 may include components that output a charging signal to the battery cell 404 having a portion generally corresponding to the frequency f _Min 322. For example, the output signal from the filter circuit 406 may include a leading edge of a harmonic at or near the frequency f _Min 322 corresponding to the determined minimum real impedance value. In some cases, the components of the filter circuit 406 may be controlled by one or more pulse control signals 416 sent to the filter circuit 406 by the circuit controller 210. In the particular example shown in FIG. 4 , the filter circuit 406 may include a first inductor 410 connected in series between the power supply rail 442 and a first transistor 412. The inductor value of the first inductor 410 affects the shape of the leading edge of the pulse, and therefore the selection of the inductor value may depend, among other things, on the charging characteristics of the battery cell 404. The first transistor 412 may also be connected to a first electrode of the battery cell 404. The first transistor 412 may receive an input signal, such as a pulse control signal 416, to operate as a switching device or component. Generally, the first transistor 412 may be any type of FET transistor or any type of controllable switch for connecting the first inductor 410 to the first electrode 440 of the battery cell 404. For example, the first transistor 412 may be a FET transistor having a drain 412 connected to the first inductor 410, a source connected to the battery cell 404, and a gate receiving the pulse control signal 416. In one embodiment, the pulse control signal 416 may be provided by the circuit controller 210 to control operation of the first transistor 412 as a switch that connects the node 436 to the first electrode of the battery cell 404 when closed and disconnects the connection between the inductor 410 and the battery cell 404 when open. Controlling the first transistor 412 to generate charging pulses is described in more detail below with reference to method 500 of FIG. 5 .

The first inductor 410 may generally operate to prevent a rapid increase in current delivered to the battery cell 404 upon connection to the battery cell 404 via the first transistor 412. More specifically, the first inductor 410 may provide resistance to the rapid conduction of current through the inductor to the battery cell 404 (when the first transistor 412 is conducting). This resistance to the rapid increase in current prevents the leading edge of the pulse of the charging signal provided by the power supply rail 442 from being sharp, thereby suppressing high frequency harmonics that may be generated at the battery cell 404 upon application of a square wave input. When transistor 412 becomes conductive in response to a signal on the pulse control signal input 416, a current or other form of energy flux from the power supply rail 442 may be provided to the battery cell 404 via the first inductor 410 and the first transistor 412 to charge the battery cell 404 while minimizing the effects of high frequency noise. The filter circuit 406 may also optionally include a flyback diode 414 connected in parallel with the first inductor 410. The flyback diode 414 provides a return path for the energy flux provided by the power supply rail 442 when the first transistor switch 412 is open or non-conducting. For example, the first transistor 412 may be controlled by the pulse control signal 416 to stop conducting current from the power supply rail 442 to the battery electrode 440. The current may then be returned to the upper rail 442 via the flyback diode 414. A storage capacitor 432 may also be connected between the upper rail 442 and ground or common, such that the current provided by the power supply rail 442 and returned through the flyback diode 414 is supplied to the storage capacitor 432 via the upper rail 442 during the open period of the first transistor 412. As described in more detail below, upon closing of the first transistor 412 (e.g., on the next pulse of the charging signal), the energy stored in the storage capacitor 432 may be returned to the upper rail 442 and the input of the filter circuit 406, thereby preventing energy loss in the circuit 400 during the period when the first transistor 412 is open, further improving the efficiency of the circuit 400.

While FIG. 4 shows components of a single filter circuit 406, additional filter circuits of the same or similar configuration may be connected in parallel with the filter circuit 406. For example, the filter circuit 406 and any number of additional filter circuits (up to filter circuit N 418) may be connected in parallel in the charging circuit 400. Each filter circuit 406, 418 may be independently controlled by the circuit controller 210 with a respective pulse control signal 406 to remove one or more harmonics from the current provided to charge the battery cell 404. In another example, two or more filter circuits 406 may be controlled by the same pulse control signal 416. One or more of the additional filter circuits 418 may include similar components, with the same or different values. For example, the first inductor of filter circuit N 418 may have a larger inductance value than the first inductor 410 of filter circuit 406. Generally, a larger inductance value for the first inductor 410 will provide greater resistance to the rapid change in the charge pulse, resulting in a sloped leading edge of the charge pulse relative to a smaller inductor value. Thus, by controlling the various filter circuits 406, 418, the circuit controller 210 may shape the leading edge of the energy pulse delivered to the battery cell 404 with different inductance values selected for the first inductor 410.

One or more input shaping circuits 420 may be connected to a first electrode 440 (e.g., an anode or positive terminal) of the battery cell 404 to further modify the pulse of the charging signal provided to the battery cell 404. In particular, the input shaping circuit 420 may include a second inductor 424 connected between the first electrode 440 of the battery cell 404 and a second transistor 422. In one example, the second transistor 422 may be an FET transistor having a drain 444 connected to the second inductor 424, a source 446 connected to ground or common, and a gate receiving a control signal 426. Similar to the first transistor 412, the second transistor 422 may operate as a switch connecting the source 444 to the drain 446, which is connected to a negative rail, ground, or common. The second transistor 422 may be controlled by the input control signal 426. In one embodiment, the shaping input signal 426 may be a high-frequency pulse-width modulated (PWM) signal that alternates between on and off states at a high frequency. In one example, the PWM signal 426 operates at a frequency greater than 100 kHz, although the PWM signal 426 may operate at any frequency. In response to the high-frequency switching PWM signal 426, the second transistor 422 may rapidly alternate between a conducting state (or "on") and a non-conducting state (or "off"). This operation of the second transistor 422 may cause the shaping circuit 420 to extract energy to ground from the charge pulse delivered to the battery cell 404. The extracted current is stored in the second inductor 424, but because the current in an inductor lags the voltage, no current flows to ground while stored in the second inductor 424. However, the off portion of PWM signal 426 may be configured to quickly close transistor 422 such that when current flows out of second inductor 424, transistor 422 turns off and little or no energy signal extracted from the charging pulse is transferred to ground via connection 446. Rather, the extracted energy may be transferred via flyback diode 430 to upper rail 442 and stored in storage capacitor 432 for reuse by charging circuit 400.

By extracting energy from the charging signal, the input shaping circuit 420 may modify the magnitude of the charging pulse to shape or modify the pulse to the battery 404. In particular, the frequency of the PWM signal 426 may be controlled to extract more or less energy from the charging signal. Furthermore, the duty cycle of the PWM signal 426 may be selected or controlled to correspond to the duration of the modified or shaped charging pulse. Thus, the PWM signal 426, optionally provided by the circuit controller 210, may modify the charging signal from the filter circuit 406 to the battery cell 404. Similarly to the filter circuit 406, one or more additional input shaping circuits 428 may be connected in parallel with the input shaping circuit 420. Each input shaping circuit 420, 428 may be independently controlled by the circuit controller 210 via a respective PWM control signal 426. In another example, two or more shaping circuits 420 may be controlled by the same PWM control signal 426. One or more of the additional input shaping circuits 428 may also include similar components with the same or different values. For example, the second input inductor of shaping circuit N 428 may have a larger or smaller inductance value than the second input inductor 424 of filter circuit 420. Pulse control signal 416 and PWM signal 426 applied to filter circuit 406 and/or input shaping circuit 420 may be controlled to shape one or more pulses of the charging signal applied to battery cell 404 to achieve a harmonic charging signal. Additional shaping of the input charging signal may also be controlled by circuit controller 210 to further modify the signal pulse profile, as described in more detail below. Various control signals from circuit controller 210 may also be used to control aspects of the charging signal provided to battery cell 404. For example, the control signals may control the voltage at battery cell 404, the current provided to the battery cell, or the overall energy or power provided to the battery cell. Thus, although discussed herein as controlling or shaping the charging signal to the battery cells, it should be understood that any aspect of the charging signal may be controlled by the circuit controller 210.

The circuit 400 of FIG. 4 may also include an impedance measurement circuit 408 connected to the battery cell 404. Typically, the impedance measurement circuit 408 measures the impedance characteristics seen at the electrodes of the battery cell 404. In one example, the impedance measurement circuit 408 may include a voltage sensor that measures the voltage across the electrodes of the battery cell 404 and a current sensor that measures the current flowing through the battery cell. However, the impedance measurement circuit 408 may include any known or later-developed circuit for measuring the impedance of the battery cell 404. Furthermore, the impedance measurement circuit 408 may be controlled by the circuit controller 210 to measure the cell impedance at various times or intervals. For example, the impedance measurement circuit 408 may be configured to measure the impedance of the battery cell 404 during a test period in which a charging signal is applied to the battery cell 404 over a range of frequencies. These measurements may be obtained and provided to the circuit controller 210 to determine the minimum real impedance of the battery cell 404, as described above with respect to the graph 322 of FIG. 3B.

The circuit controller 210 may utilize the circuit 400 of FIG. 4 to shape the pulses of the battery cell charging signal based on a frequency corresponding to a minimum impedance value. In particular, FIG. 5 illustrates a method 500 for generating a battery cell charging signal based on a frequency corresponding to a minimum impedance value, according to one embodiment. The operations of method 500 may be performed by the circuit controller 210, which controls various components of the circuit 400, particularly by providing control signals to the power supply 402, the filter circuit 406, and/or the shaping circuit 420. Other circuit designs and components may also be controlled by the circuit controller 210 to perform one or more of the operations of method 500. Thus, although described herein with reference to the circuit 400 of FIG. 4, the operations of method 500 may be performed by any number of hardware components, software programs, or a combination of hardware and software components.

Starting at operation 502, circuit controller 210 may select an initial frequency of a charge pulse used to recharge battery cell 404. For example, a sinusoidal charge pulse may be selected to recharge battery cell 404 to avoid the inefficiencies of a square-wave charge pulse. The initial frequency of the charge pulse may be selected by circuit controller 210. In some cases, the selected frequency may be determined to minimize or reduce the real impedance of battery cell 404 during initial charging of the battery. Initially, circuit controller 210 may not know the real impedance of battery cell 404 because no charge signal is applied to the battery and one or more characteristics (such as the state of charge of the battery cell or other electrochemical components of the battery) may not be known. Therefore, circuit controller 210 may select an initial frequency of the charge pulse to begin providing energy to battery cell 404. In one particular embodiment, circuit controller 210 may determine the initial frequency of the charge pulses based on historical data for battery cell 404, historical data for other battery cells, historical data for circuit controller 210, or other battery recharge data. For example, circuit controller 210 may analyze past recharge sessions for battery cell 404 or other battery cells. Based on this analysis, circuit controller 210 may estimate a frequency f _Min for battery cell 404 at which the real impedance value of the battery cell is at a minimum. As more recharge sessions are analyzed, a best estimate for the initial frequency of the charge pulses may be determined that corresponds to the estimated minimum real impedance value for battery cell 404. The initially selected frequency may not correspond to the actual minimum real impedance value of the state of charge of battery cell 404, but rather may be based on one or more historical real impedance measurements of the battery cell of interest or any other battery cells.

Depending on the selected initial frequency of the charge pulse, the circuit controller 210 may control the pulse control signal input 416 and/or the PWM signal input 426 of the charging circuit 400 to generate harmonic charge pulses for the battery cell 404. In particular, the circuit controller 210 may provide the pulse control signal 416 to drive a first transistor 412 for a first period of time. Driving the first transistor 412 may transmit an energy pulse from the power supply rail 442 to the battery cell 404. The first inductor 410 of the filter circuit 406 may provide resistance to the rapidly increasing pulse (e.g., a square wave pulse) received from the power supply rail 422 and output an angled leading edge (e.g., a leading edge of a sinusoidal pulse) sent to the battery cell 404. The duration of the charge signal pulse may correspond to the first period during which the first transistor 412 is driven to conduct. Furthermore, the magnitude of the pulse may correspond to the magnitude of the signal provided by the power supply 402 (potentially controlled by V _CONT 434) and/or the duration of the pulse signal as controlled by the pulse control signal 416. In particular, the duration that the first transistor 412 is conductive corresponds to the duration of the energy pulse provided to the battery cell 404. In many cases, the circuit controller 210 may repeatedly enable and disable the first transistor 412 to provide a periodically repeating pattern of energy pulses to the battery cell 404.

In addition to the leading edge and pulse duration, the input shaping circuit 420 may control the energy pulse delivered to the battery cell 404 to vary. In particular, the PWM signal 426 provided to the second transistor 422 may rapidly activate and deactivate the transistor, causing the input shaping circuit 420 to extract energy from the pulse and reduce the pulse magnitude at any time during the pulse's duration. The frequency of the PWM signal 426 may further modify the profile by controlling the amount of energy extracted from the energy pulse signal. Precise control of the PWM signal 426 may reduce the pulse magnitude (by extracting energy from the pulse) or increase it (by deactivating the transistor 422 to prevent the input shaping circuit 420 from extracting energy from the pulse) to produce a shaped pulse for charging the battery cell 404.

By controlling inputs to the circuit 400, such as the pulse control signal 416 and/or the PWM signal 426, the circuit controller 210 may generate sinusoidal pulses to charge the battery cells 404 at a selected initial frequency, similar to waveform 314 of FIG. 3A. However, as previously discussed, the minimum real impedance of the battery cells 404 may change during battery charging. For example, the state of charge and temperature of the battery cells 404 may alter the minimum real impedance characteristics. By adjusting the frequency of the pulse charging signal to correspond to the minimum real impedance of the battery cells 404 at the battery's current state, efficiency benefits can be achieved when charging the battery. Therefore, in operation 506, the circuit controller 210 may determine a function of the real impedance values of the battery cells at different frequencies by measuring the impedance of the battery cells at different frequencies. In one embodiment, the circuit controller 210 may apply one or more test signals at different frequencies to the battery cells 404 to determine the charging signal frequency corresponding to the measured minimum real impedance of the battery cells 404. The frequencies of the test signals may be predetermined by the circuit controller 210 to provide a range of test signals to the battery cell 404. For each test signal, a corresponding real impedance value at the battery cell 404 may be determined and/or stored. In addition to using multiple frequencies, a constant current intermittent titration (GITT) technique may be used. Generally, GITT uses the characteristics of a square wave pulse (which is a sum of sinusoidal frequencies across a spectrum) to represent a complex impedance that can be used to determine the impedance of the battery cell 404.

In operation 508, a minimum real impedance value of the measured test impedances may be determined. For example, the circuit controller 210 may select the smallest real impedance value from the received test results as the minimum impedance value. In another example, the circuit controller 210 may analyze the received real impedance values and determine the minimum real impedance value by extrapolating the values. For example, the measurements may indicate that the real impedance value decreases for a series of increasing test frequencies, and then increases for the next series of increasing test frequencies. The circuit controller 210 may determine that the minimum real impedance value of the battery cell 404 corresponds to a frequency between a first set of increasing test frequencies and a second set of increasing test frequencies. In this situation, the circuit controller 210 may estimate the minimum real impedance value of the battery cell 404 between the measurements. In operation 510, the circuit controller 210 may determine the frequency corresponding to the minimum real impedance value determined for the battery cell 404. For example, a graph 334 of the real impedance value 324 of the battery cell 404 versus the frequency 326 of the test signal may be generated, and the minimum real impedance value 330 may be determined from the graph. The frequency corresponding to the minimum real impedance value 330 may also be determined from the graph 334. In general, the determination of the corresponding frequency may utilize any correlation algorithm for determining the frequency of the input signal to the battery cell 404 that results in the minimum real impedance value.

At operation 512, the circuit controller 210 may determine whether the frequency corresponding to the minimum real impedance value of the measured test impedance is different from the previously selected frequency at which the charge pulse is delivered. If the circuit controller 210 determines that the corresponding frequency resulting from application of the test signal to the battery cell 404 is different from the frequency at which the charge pulse is delivered, then at operation 514, the circuit controller 210 may select a corresponding frequency for additional pulses of the charge signal. Furthermore, the circuit controller 210 may return to operation 504 and generate and provide an input signal to the shaping circuit to adjust the frequency of the charge pulses of the battery cell to the determined corresponding frequency. If the corresponding frequency is not different from the frequency at which the charge pulse is delivered, then the circuit controller 210 may maintain the frequency of the additional charge pulses at operation 514 and provide a corresponding control signal to the shaping circuit at operation 504. Thus, the method 500 of FIG. 5 may select a frequency corresponding to the minimum real impedance value of the battery cell for the sinusoidal charge pulse generated to recharge the battery cell 204.

A potential drawback of using a sine wave charging signal is that such a signal may deliver less power to the battery cells for recharging than a square wave charging signal. This potential drawback can be particularly pronounced in fast charging situations, where the goal is to deliver the maximum amount of energy to the battery cells in the shortest amount of time. Graph 602 in FIG. 6 illustrates this potential drawback. In particular, FIG. 6 illustrates a graph 602 of an input voltage value 604 versus time 606, with square wave pulses 612, 614 and sine wave pulses 608, 610 of a battery charging signal superimposed. Generally, the area under each pulse indicates the amount of charge available to the battery for recharging. It should be appreciated that the area under the pulse represents the amount of available charge. As noted above, due to battery and charging characteristics, not all of the energy of the square wave pulses is generally delivered for cell charging. Nevertheless, the difference in the amount of charge delivered by the square wave pulses 612, 614 and the sine wave pulses 608, 610 is illustrated by the hatched areas 616, 618. As shown, the sinusoidal pulses 608, 610 reduce the battery impedance due to the selected harmonic frequency estimation described above, while potentially charging the battery less per pulse than the square wave pulses 612, 614. Therefore, charging based on the minimum impedance frequency may improve charging relative to other systems. However, further refinements and optimizations may be available.

A potential way to deliver a similar amount of charge to the battery at the selected harmonic corresponding to the minimum real impedance value is to increase the magnitude of the charge pulses 608, 610. However, because many battery characteristics impose an upper threshold on the magnitude of the charge signal, simply increasing the magnitude of the sinusoidal pulses may not be beneficial for fast charging the battery cells. For example, many battery electrolytes begin to break down at a specific power level correlated with a voltage threshold, shortening the battery's lifespan due to the irreversibility of such chemical reactions. Furthermore, such electrolyte breakdown may occur during abrupt changes in the recharge power signal applied to the battery's electrodes. The abrupt application of the recharge power signal may also cause breakdown or damage to other battery components. For example, a high-power signal may form one or more permanent channels across the solid electrolyte interphase (SEI) layer of a lithium-ion battery, permanently creating spatial inhomogeneity throughout the anode. Furthermore, the SEI layer may increase in thickness in response to a high-power signal, potentially reducing the battery's efficiency. Additionally, increasing the magnitude of the recharge power signal can cause the battery to heat up faster than it can dissipate heat, potentially damaging the battery and increasing the risk of thermal runaway. Therefore, simply increasing the pulses 608, 610 to provide additional charging can damage the battery while it is being recharged.

An alternative method for increasing charging with the sinusoidal pulses 608, 610 is to maintain the pulse at or near the pulse peak where the sinusoidal pulse normally begins to taper off, while combining harmonics and broadening the peak and/or adjusting the leading edge of the pulse to a target real impedance minimum frequency (and/or target imaginary impedance, as discussed in more detail below). In one example, the methods and circuits discussed herein may be applied to determine a range of frequencies corresponding to one or more minimum real impedance values of a battery cell, and then provide the battery cell with a charging signal that includes harmonics within these identified frequency ranges. For example, FIG. 7A is a graph 702 of a battery cell's measured real impedance value 714 versus the corresponding frequency 706 of a charging signal applied to the battery cell. It will be appreciated that the values may be measured in real time, measured and stored (i.e., not measured in real time), characterized or derived from other information, measured only periodically, the frequency may be set to some initial value and then adjusted in a feedback loop, etc. Of course, other battery cell characteristics, such as imaginary impedance, admittance, and/or susceptance, may also be measured or estimated and used to shape the charge pulse. While the graph shows a maximum frequency 710 and a minimum frequency 708 within the range of allowable minimum impedance values, the minimum impedance frequency is not strictly defined. Graph 702 of FIG. 7A is similar to graph 322 of FIG. 3B, described above, in that it plots the real impedance of the battery cell versus the frequency of the charging signal provided to the battery. However, in this example, rather than determining the frequency f _Min 332 corresponding to the minimum real impedance value 330, a frequency range defined by the minimum frequency f _RMin 708 and the maximum frequency f _RMa 710 may be determined near the minimum real impedance value 712 of the battery based on the range of allowable impedance values for charging the battery cell. A minimum frequency f _RMin 708 and a maximum frequency f _RMa 710 may be selected and included in the generated battery charging signal pulses to widen the pulse profile and increase the charge delivered to the battery cells in each pulse. Based on the range of frequencies with allowable impedance values, including multiple harmonics in the charging pulses of the recharging power signal can increase the charge available from a single harmonic sine wave for recharging the battery cells while maintaining a lower impedance for the battery cells receiving the charging pulse.

7B is a signal diagram 722 of a battery cell charge pulse including multiple frequencies corresponding to a highest frequency f _RMax 710 and a lowest frequency f _RMin 708 based on the real impedance value of the battery cell, according to one embodiment. Signal diagram 722 shows input voltage 724 versus time 726, above which a highest voltage threshold 730 may be exceeded, potentially damaging the battery. In particular, charge pulse 728 of diagram 722 may be generated based on the range of frequencies shown in graph 702 of FIG. 7A. For example, charge pulse 728 of FIG. 7B may include a range of harmonics that exist between the lowest frequency f _RMin 708 and the highest frequency f _RMax 710. In one example, the minimum frequency f _RMin 708 and the maximum frequency f _RMax 710 may be based on a range around the minimum real impedance value 712, such that the frequency f _Min 711 corresponding to the minimum real impedance value 712 determined for the battery cell may fall between the minimum frequency f _RMin 708 and the maximum frequency f _RMax 710. For each selected harmonic frequency in the charge pulse 728, a corresponding magnitude may be determined based on the corresponding real impedance value of the battery at that frequency, resulting in some non-uniformity in the charge pulse. However, any selected magnitude may not exceed an upper voltage or power threshold 730 that could potentially damage the battery cell during recharging or cause thermal runaway of the battery. Expanding the charge pulse by including the range of frequencies corresponding to the minimum real impedance value 712 allows for a larger amount of charge to be applied to recharge the battery while maintaining a low battery impedance. In this manner, using a multi-charge, low-impedance charge signal to recharge the battery cell may improve efficiency compared to a square wave recharge signal.

FIG. 8 is a flowchart illustrating one embodiment of a method for generating a charging signal for a battery cell based on different frequency ranges corresponding to the battery's maximum and minimum real impedance values. As discussed above, similar methods may be performed to generate a charging signal for a battery cell based on other battery cell factors, such as imaginary impedance, admittance, and/or susceptance values. Similar to method 500 of FIG. 5, the operations of method 800 of FIG. 8 may be performed by circuit controller 210, which controls various components of circuit 400 of FIG. 4, particularly by providing control signals to power supply 402, filter circuit 406, and/or input shaping circuit 420. Other circuit designs and components may also be controlled by circuit controller 210 to perform one or more of the operations of method 500. Thus, although described herein with respect to circuit 400 of FIG. 4, the operations of method 500 may be performed by any number of hardware components, software programs, or combinations of hardware and software components.

Starting with operation 802, circuit controller 210 may determine a minimum real impedance value for the battery cells. Determining the minimum real impedance value may be similar to the above in that circuit controller 210 may measure or receive the impedance of the battery at different frequencies of the charging signal. The minimum real impedance value may also be determined by a loop process or a process driven by circuit controller 210. For example, circuit controller 210 may charge the battery of the circuit at different frequencies (e.g., a range of frequencies) and measure the impedance of battery cells 204 until a minimum impedance value for battery cells 204 is found. Such measurements may be taken during active charging of the battery cells, or may be performed, stored in memory, and searched for. For some batteries, the impedance measurements versus charging signal frequency may resemble graph 702 of FIG. 7A. Similar to graph 702, circuit controller 210 may determine a minimum real impedance value 712 for the battery cells based on multiple impedance measurements. Additionally, the impedance measurement process may acquire and store impedance values at different frequencies (e.g., impedance measurements may be acquired at frequencies above and below the lowest frequency f _Min 711).

In operation 804, the circuit controller 210 may select an upper real impedance value 720 of the corresponding range of allowable impedance values. In particular, the circuit controller 210 may determine or be provided with an allowable impedance value 716 for the battery cell based on application of the charging signal. The allowable impedance value 716 is illustrated and described as a single allowable impedance value that is above the minimum impedance value and occurs at frequencies lower and higher than the frequency f _Min 711 at which the minimum impedance occurs. It is recognized that the allowable impedance value 716 need not be the same for frequencies higher or lower than the minimum impedance. Furthermore, the allowable impedance 716 may change with the progress of charging, changes in cell temperature, may be based on the charging current level, etc. The allowable impedance value 716 may be greater than the determined minimum impedance value 712. For example, the circuit controller 210 may determine or be provided with the impedance value 716 as the allowable impedance value for the charging signal. Generally, the allowable impedance value 716 may be any impedance of the battery cell during recharging. However, a smaller allowable impedance value 716 may be selected or determined to limit the overall impedance of the battery cell when a charging signal is applied. Furthermore, the upper impedance value 720 of the range may be an impedance value that occurs at a different frequency or combination of frequencies than the frequency at which the minimum frequency f _Min 711 occurs. In many cases, there will be a range of frequencies above the minimum impedance 712 and below the allowable impedance 716, above and below the frequency f _Min 711 at which the minimum impedance occurs. For example, the allowable impedance in this range may occur at a frequency f _RMax 710 that is higher than the frequency at which the minimum impedance occurs. Thus, the circuit controller 210 may be configured to determine or select the upper impedance value 720 for the allowable range by tracing the impedance value plot 714 from the minimum impedance value 712 to the right (or toward higher frequencies) until the allowable impedance value 716 is reached. However, in other embodiments, the upper impedance value 720 of the range may be a predetermined difference from the minimum impedance value 712 (programmatically calculated from a minimum value that takes into account other factors such as battery charge, temperature, etc.) For example, the upper impedance value 720 of the range may be determined as twice the minimum impedance value 712 or some other multiple of the minimum impedance value.

Although shown as a smooth curve in FIG. 7A , the shape of the impedance plot 714 may include various artifacts, such as noise at different frequencies. For example, the plotted impedance 714 may be generated at various signal magnitudes, so that it may include dips at higher frequencies, particularly as the harmonics grow. Thus, the plot 714 may be the sum of several different plots, each associated with a different increment of harmonic power. In this situation, the frequency f _Min 711 corresponding to the minimum impedance 712 may remain relatively constant as the harmonics grow, up to a certain value at which the impedance value begins to rise rapidly.

Furthermore, due to parasitic capacitive and inductive losses, the physical orientation of the cells in the pack (e.g., parallel or series connection) can affect the shape of the impedance curve. For example, at certain frequency bands, energy may jump short distances in the air from one cell to another, effectively bypassing cells in the battery pack structure and further inhibiting or enhancing current flow at that point. The measured impedance at these frequencies may result in dips in the impedance curve or areas where the impedance appears low due to the omission of cells in the pack, and local minimum impedance values may be determined, particularly for some higher-frequency harmonics. However, charging the battery cell or pack at these higher frequencies may not improve the efficiency of battery cell charging for the reasons explained above. Therefore, determining the frequency f _Min 711 corresponding to the minimum impedance 712 may include excluding dips in the impedance values or relatively noisy bands at high frequencies due to parasitic losses in the battery pack. Such high frequency rejection may be achieved through selection of inductor value 410 (or filter circuits 406, 418) or by including an additional high frequency filter in the path of the charging signal in circuit 400. In one embodiment, controller 210 may compare multiple battery cell or pack parameters, such as real and imaginary impedance and admittance, to identify regions that contain local minimum impedance values while being high frequency and therefore should be rejected. Additionally, controller 210 may determine the range of frequencies associated with the detected minimum impedance values, since impedance dips due to parasitic losses in the battery pack may be associated with a small frequency range.

The controller 210 may also use the impedance curve plot 714 obtained by a pack experiencing energy jumps between cells to characterize or identify the pack's configuration. For example, a first battery pack configuration with cells connected in series may have a different impedance plot than a second battery pack configuration with cells connected in parallel. Detectable differences between packs with different cell counts or orientations may also be used. To this end, the controller 210 may obtain impedance plots of the battery pack (as well as plots of other battery pack elements, such as conductance and/or susceptance) and compare the obtained plots to a database of impedance plots. The database of impedance plots may correlate each plot with a particular battery pack configuration or battery cell type, allowing the controller 210 to determine or estimate the battery pack configuration or cell type being charged by comparing the obtained impedance plot with the stored plot. The controller 210 may then further adjust or shape the charge pulse based on the estimated battery pack configuration.

Regardless of how the range of upper impedance values 720 is determined, circuit controller 210 may determine a corresponding frequency f _RMax 710 for the upper impedance value 720 in operation 806. As previously mentioned, the impedance at the electrodes of a battery cell may vary based on the frequency of the charging signal applied to the electrodes. As such, frequency f _RMax 710 may correspond to a selected upper impedance value 720 in the acceptable range. Circuit controller 210 may determine a frequency f _RMax 710 that corresponds to the selected upper impedance value 720.

Also, in operation 808, the circuit controller 210 may select a lower impedance value 718 of a corresponding range of allowable impedance values based on the minimum impedance value 716 obtained for the battery. Like the upper impedance value 720 of the range, the lower impedance value 718 may be selected or determined based on the allowable impedance value 716 and may be a frequency f _RMin 708 that is lower than the frequency f _Min 711 at which the minimum impedance value 712 occurs. In other words, the circuit controller 210 may be configured to determine or select the lower impedance value 718 for the range of allowable impedance values by tracing the impedance value plot curve 714 from the frequency f _Min 711 at which the minimum impedance value 712 occurs to the left (or toward lower frequencies) until the allowable impedance value 716 is reached. Thus, the upper impedance value 720 and the lower impedance value 718 may sometimes be equal (e.g., the allowable impedance value 716 of the range) but occur at different frequencies of the charging signal (e.g., above and below the frequency f _Min 711 of the minimum impedance). In another embodiment, the lower impedance value 718 of the range of impedance values may be a specified difference from the minimum impedance value 712, as may the upper impedance value 720 of the range. Regardless of how the upper impedance value 720 is determined, the circuit controller 210 may determine a corresponding frequency f _RMin 708 of the lower impedance value in operation 810. Generally, the corresponding frequency f _RMin 708 is a lower frequency than the corresponding frequency f _Min 711 of the minimum impedance value 712. In some examples, the allowable range or set of harmonics for generating the charging pulse may be based on a range of frequencies included between the range of frequencies f _RMax 710 and the range of frequencies f _RMin 708, including the frequency f _Min 711.

In yet other embodiments, circuit controller 210 may not determine one or both of upper impedance value 720 or lower impedance value 718. Rather, circuit controller 210 may select (e.g., perform a table lookup, etc.) the frequency f _RMax 710 and frequency f _RMin 708 for a range of impedance values. In some cases, one or both of the upper and lower frequency values may be based on a minimum impedance frequency f _Min 711, which may be measured based on past modeling, extrapolation from past measurements, etc., or retrieved from memory. By selecting frequency f _RMax 710 and/or frequency f _RMin 708 based on the minimum impedance frequency f _Min 711, etc., circuit controller 210 may control the frequency range or bandwidth of the charging signal. Furthermore, the frequency range may be selected so that corresponding impedance values within the frequency range remain below tolerance threshold(s) 716 for charging the battery cell based on the measured impedance value of the battery cell or historical measurements of the battery cell or other battery cells.

In operation 812, the circuit controller 210 may determine magnitude values corresponding to multiple frequencies within the range of frequencies defined by frequency f _RMax 710 and frequency f _RMin 708. In one embodiment, the magnitudes corresponding to the frequencies within the range may be proportional to the impedance measured or estimated at that frequency. For example, the magnitude determined to be included in the charge pulse at frequency f _RMax 710 may be proportional to the real impedance value 720 at that frequency. Similarly, the magnitude determined to be included in the charge pulse at frequency f _RMin 711 may be proportional to the real impedance value 712 at that frequency. Thus, each frequency within the range may have an associated magnitude corresponding to the impedance value 714 at that frequency. Note, however, that the impedance of each harmonic is not necessarily independent of the magnitudes of other harmonics in the waveform.

In operation 814, the circuit controller 210 may control the pulse control signal and PWM signal of the charging circuit 400 to generate shaped charge pulses for the battery cells 404. As described above, the circuit 400 of FIG. 4 may be utilized to generate pulses of a charge signal to the battery cells 404 during charging. In particular, the filter circuit 406 and/or the input shaping circuit 420 may be controlled to shape the power from the upper rail 442 into a charge pulse train that includes one or more frequencies or harmonics corresponding to the determined frequency range. In one example, the filter circuit 406 may be controlled to generate a leading edge that corresponds to a sinusoidal signal at the frequency f _RMax 710 or the frequency f _RMin 708. Furthermore, the duration of the pulse control signal 416 determines the range of harmonics of the charge pulse, with a longer duration of the pulse control signal 416 corresponding to a wider charge pulse (or wider bandwidth of the charge pulse). Input shaping circuit 420 may also vary the magnitude of the charge pulse at particular events or harmonics of the signal under control of PWM signal 426. Thus, by providing one or more inputs to circuit 400, circuit controller 210 may shape the charge pulse to include multiple harmonics based on a determined range of frequencies defined by frequency f _RMax 710 and frequency f _RMin 708. According to method 800 of FIG. 8, circuit controller 210 may generate a series of shaped charge pulses that deliver an optimal amount of charge to battery 404 while maintaining or reducing the impedance at the battery cell electrodes.

The determined frequency range and the generated charging signal based on this frequency range may be used in accordance with method 500 of FIG. 5. In particular, circuit controller 210 may generate a charging signal from a range of frequencies based on a first set of measured impedance values to initiate charging of the battery cells. A second set of measured impedance values may be obtained during a battery cell recharging session using the iterative process discussed with respect to FIG. 5. A second frequency range may then be determined based on the second measured impedance values, and the charging signal may be adjusted accordingly. In this manner, an iterative process (including recalculating the range of frequencies or harmonics included in the charging signal) may be performed to adjust or modify the pulses of the charging signal during battery cell recharging based on additional measurements of the battery cell impedance values.

FIG. 9A illustrates a signal diagram 902 of a shaped charging pulse train 902 generated by a battery charging circuit, according to one embodiment. In one example, the circuit 400 may generate shaped pulses 914, 916 based on the controller 210. The signal diagram 902 shows the input voltage 904 or input current versus time 906 for a current control hardware circuit for pulses 914, 916 of the charging signal. As can be seen, each pulse 914, 916 is asymmetric, with the leading edge 912 shaped differently relative to the trailing edge 910. In one example, the pulses 914, 916 may be defined by a harmonic or combination of harmonics associated with a minimum impedance value seen at the electrodes of the battery cell. In particular, the charging signal pulses 914, 916 may include a leading edge 912 corresponding to a selected frequency associated with the minimum impedance value of the battery cell. For example, the shape of the leading edge 912 of the pulse 914 may correspond to the harmonic f _Min 332 identified by the circuit controller 210 as the frequency at which the battery cell experiences a minimum real impedance value. In one example, the shape of the leading edge 912 may be based on the leading edge of the corresponding sine wave at the frequency of the minimum impedance. In another example, the shape of the leading edge 912 of the pulse 914 may correspond to the harmonic f _RMax 710 or the harmonic f _RMin 708. The identification of the minimum impedance frequency may be based on, among other things, measurement(s), battery characteristics, alone or in combination. Regardless of the selected frequency, the leading edge 912 of the pulse 914 may be shaped to be the same as the leading edge of a portion of the harmonic sine wave charging signal that minimizes or reduces the impedance seen by the battery cell for more efficient application of the recharging power signal.

To generate the leading edge 912 of the pulse 908 of the selected harmonic, the circuit controller 210 may control one or more of the filter circuits 406 described above. For example, the shape of the leading edge 912 of the pulse 908 may be correlated to the inductance value of the first inductor 410. In particular, the first inductor 410 provides resistance to the rapid conduction of current such that the current through the inductor rises slowly and increases over time. The resistance to current flow through the inductor is determined by the inductance value of the first inductor 410. Thus, to shape the leading edge 912 of the charging signal pulse 914, the circuit controller 210 may drive the first transistor 412 (via the pulse control signal 416) to cause current to begin flowing through the inductor 410 and reach the battery cell 404. This current flow may begin slowly and increase over time, and because the voltage of the charging signal is related to the current of the charging signal, the voltage may track the current to form the leading edge 912 of the pulse 914 as shown in FIG. 9A . In general, the rate of increase of current through the first inductor 410 may be based on the inductance value of the inductor, which may impart the shape of the leading edge 912 to the pulses 914, 916 of the charging signal. Accordingly, the harmonic of the leading edge 912 may correspond to the inductance value of the first inductor 410. To apply a target harmonic to the leading edge 912, the circuit controller 210 may select from multiple filter circuits 406, 418 or the first inductor to create a slope to the leading edge 912 that corresponds to the determined harmonic of the minimum real impedance. Furthermore, the resistance of the first inductor 410 to a rapid increase in current prevents the leading edge of the charging signal pulse from becoming too sharp, suppressing high frequency harmonics that may be generated at the battery cell 404 when a square wave input is applied.

By driving the first transistor 412 with the pulse control signal 416, the circuit controller 210 may generate a leading edge 912 of a pulse 914 of a selected harmonic as current flows through the first transistor 412. Later in the pulse 914, the magnitude of the pulse may reach the upper voltage or floating voltage of the power rail 442, corresponding to the constant voltage 908 at the top of the pulse 914. The duration of the pulse 908 may be controlled by the circuit controller 210 by maintaining the first transistor 412 in a conductive state so that power is supplied to the battery cell 404 through the first inductor 410 and the first transistor 412. In this manner, the pulse control signal 416 can control the duration or width of the pulse 914 of the charging signal.

In some cases, the circuit 400 may be controlled to include a sharp trailing edge 910 of the pulse 914. The circuit controller 210 may generate the sharp trailing edge 910 of the pulse by deactivating the first transistor 412 to isolate the battery cell 404 from the power supply rail 442. In particular, the circuit controller 210 may cause the first transistor 412 to stop conducting by deactivating the pulse control signal 416. As described above, when the first transistor 412 is non-conductive, the current flowing through the first inductor 410 may be returned to the power supply rail 442 through the flyback diode 414. Controlling the first transistor 412 in this manner may result in the sharp trailing edge 910 of the pulse 914. Furthermore, although the sharp trailing edge 910 may typically correspond to a large harmonic content, such harmonics are unlikely to increase impedances that would damage the battery cells 404 because, after the sharp trailing edge 910, the current and voltage magnitudes approach or equal zero (or, in the case of voltage, zero overpotential) across the battery 404. As described in more detail below with reference to FIG. 12 , the separation between large harmonics and damaging impedances is maintained so long as the voltage magnitude temporarily drops below the battery's floating voltage (e.g., the battery voltage when not receiving a charging current) to reduce the time it takes for the charging current to reach zero. In this manner, the filter circuit 406 may be controlled to produce a shaped charge pulse 418 that includes a harmonic sinusoidal leading edge 912 corresponding to a minimum impedance value of the battery cells 404, a duration at the upper magnitude 908, and a sharp trailing edge 910 that delivers sufficient charge to the battery cells 404 while maintaining a low impedance at the battery's electrodes.

In general, the circuit 400 can be controlled to generate or shape pulses of the charging signal into any desired shape. For example, FIG. 9B is a signal diagram 922 of a second shaped charging pulse train 924, 932 generated by the battery charging circuit 400, according to one embodiment. In this example, the leading edge 928 of each pulse 926, 932 may be similar to the leading edge 912 described above with respect to FIG. 9A. In particular, the leading edge 912 of the charging pulses 924, 932 may be generated by controlling one or more of the filter circuits 406 described above. However, in this example, rather than a flat pulse voltage level 908 for the duration of the pulse after the shaped rising edge 928, the circuit controller 210 may further shape the pulse 924 by controlling one or more of the input shaping circuits 420, 428 of the charging circuit 400. In the illustrated example, the portion 926 of the pulse 924 following the leading edge 928 may include a non-uniform voltage (or current) drop to a sharp trailing edge 930. While the decreasing level (or slope) 926 is shown linearly, this is not required, and the pulse 924 may be shaped to include many forms. In one embodiment, the circuit controller 210 may provide a PWM signal 426 to the second transistor 422 of the input shaping circuit 420. As discussed above, the PWM signal 426 may be a high-frequency switching signal that alternates the second transistor 422 between a conducting (or "on") state and a non-conducting (or "off") state. The rapid alternating action of the second transistor 422 may cause current from the pulse 924 to flow through the second inductor 424. This current extraction from the pulse 924 removes current, resulting in the downward slope 926. Generally, the duty cycle of the PWM signal 426 may control the amount of current drawn from the pulse 926, and this duty cycle may be configured by the circuit controller 210 to generate the slope 926 of the pulse 924. Additionally, as explained above, the off portion of PWM signal 426 may rapidly close transistor 422 so that little or no energy signal extracted from the charging pulse is transmitted to ground via connection 446. Rather, the extracted energy may be transmitted via flyback diode 430 to upper rail 442 and stored in reservoir capacitor 432 for reuse by charging circuit 400.

At the end of the duration of the charge pulse 924, the circuit 400 may be further controlled to define a sharp trailing edge 930 as described above with respect to FIG. 9A. In particular, the circuit controller 210 may generate the sharp trailing edge 910 of the pulse by deactivating the first transistor 412 to isolate the battery cell 404 from the power rail 442. In particular, the circuit controller 210 may deactivate the pulse control signal 416 to cause the first transistor 412 to cease conduction. In yet another example, the trailing edge of the pulse 924 may be further shaped by bleed-off current at the trailing edge 930 by driving the input shaping circuit 420 with the PWM signal 426. Of course, the charge pulses 924, 932 shown in FIG. 9B are merely examples of shaped charge signals that may be generated by controlling the charge circuit 400. In particular, the circuit controller 210 may control the filter circuit 406 and/or the input shaping circuit 420 to generate charge pulses of various shapes as desired. Thus, other shapes of the charging signal, such as those shown in Figures 3A, 7B, and/or 9A, may be generated by circuit 400.

While the above discussion focuses on real impedance values at the battery's electrodes, the reactance, or imaginary part, of the impedance at the battery's electrodes may also be considered when shaping the charging signal. Other factors, such as admittance and/or susceptance values, may also be considered. In particular, FIG. 10A is a signal diagram illustrating a sinusoidal voltage signal 1004 used to generate a charging current 1006 that recharges a battery cell. Generally, the charging current 1006 measured at the battery cell may have the same shape as the applied voltage signal 1004. However, due to the battery's impedance, the charging current 1006 applied to the battery may be considered to be smaller and delayed in time relative to the voltage signal 1004. The qualitative difference in magnitude between the voltage signal 1004 and the current 1006 at the battery is the measurement of the real impedance _ZR 1008, expressed as _ZR = (dV/dI) or (ΔV/ΔI). One or more of the methods and circuits described above take this real component into account when shaping the pulses of the charging signal to recharge the battery. The time delay between the voltage signal 1004 and the application of current 1006 at the battery is shown as _ZI 1010 and is due to the reactive, or imaginary, component of the battery impedance. Like the real component of the impedance, the reactive portion of the impedance 1010 also causes inefficiencies in the application of the charging signal to the battery during a charging session. For example, the duration of a charging waveform is typically measured from the start of the charging voltage or current recharging the battery until the voltage slowly returns to zero overpotential (the voltage at the terminals matches the floating voltage of the battery) and no charging current flows to the battery (zero amperes). However, a charging system that ignores the reactive portion of the battery cell impedance may assume that the voltage to the battery and the resulting charging current waveform start and stop at the same time. However, when the reactive portion of the impedance is considered, a capacitive or inductive time delay occurs between the voltage and current waveforms at the battery cell. This delay between the voltage and current of the charging signal indicates a longer charging period per pulse. This can lead to a lower average current over the entire charging period of the pulse, potentially increasing the inefficiency of the charging pulse at the battery cell. Depending on the reactance level, the reactive component can also divert energy toward heat generation rather than stored chemical energy within the battery. Reactance can also be problematic, generating heat in the conductive paths (such as cables, wires, and wiring board traces) and the cell itself. High reactance can also lead to non-uniform electrochemical activity across the area of the electrode, resulting in significant ohmic drop across the current collectors, electroactive materials, and other components within the battery cell.

To address potential inefficiencies in the application of charge pulses to the battery cells, the system may generate a charge signal whose pulses correspond to a determined or estimated reactive component of the battery cell's impedance. In particular, the shape and overall duration of the pulses of the charge signal for recharging the battery cells may be tailored to correspond to the real component of the impedance as well as the imaginary component of the impedance. For example, refer now to FIG. 10B , which illustrates a graph 1022 of various components of the battery's impedance 1024 versus the frequency 1026 of the charge signal applied to the battery. In particular, graph 1022 includes a plot of real impedance values 1028, a plot of imaginary impedance values 1032, and a plot of a calculated reference impedance value 1030. The method discussed herein determines a frequency f _Zr 1034 corresponding to the minimum real impedance value, which may be used to generate a charge signal whose pulses include harmonics at or within a range of frequencies above and/or below that frequency. However, as shown in graph 1022, the frequency f _Zr 1034 corresponding to the minimum real impedance value may be associated with a relatively high imaginary impedance value 1032 at the battery electrodes. Therefore, simply considering the real impedance does not account for the imaginary impedance and its effect on charging efficiency, and therefore may not provide an optimal charging solution. Therefore, some implementations of the circuits and methods described herein may optimize the frequency to define the pulse shape and the duration of the overall charging signal at which such pulses are applied by taking into account, to varying degrees, both the imaginary and real impedances, such as by understanding the frequency of both the imaginary and real components of the impedance at the battery cell. Still other implementations may use admittance and/or susceptance values calculated from the measured real and/or imaginary impedance at the battery cell.

In one example, the circuit controller 210 may calculate or obtain a combination of real and imaginary impedance values to select the frequency and harmonics at which to generate the pulse of the charging signal. Such combinations may include reference calculations of the real and imaginary impedance values. A plot of the impedance reference values 1030 is shown in graph 1022 of FIG. 10B . Other combinations of both components of the battery's impedance may also be calculated or determined by the circuit controller 210 and used to shape the pulse of the charging signal. For example, one or both of the real and imaginary impedance values may be weighted unequal (e.g., 20% weighting of the real impedance value and 80% weighting of the imaginary impedance value) or proportionally, and may be used to determine various components of the pulse of the charging signal, such as the leading edge or width of the charging pulse. Similarly, the circuit controller 210 may determine a minimum impedance reference value and corresponding frequency (shown in graph 1022 as frequency f _ZMod 1036). As can be seen in graph 1022, by generating charge pulses having harmonics of frequency f _ZMod 1036, a higher real impedance is induced into the battery than at other frequencies (especially compared to f _Zr ), while the imaginary impedance component can be minimized or reduced. Thus, by considering both components of impedance at the battery cell (real impedance 1028 and imaginary impedance 1032), a more efficient charge signal can be generated. Considering both components of impedance at the battery cell can be particularly useful in multi-cell systems where connections between cells add impedance.

In some cases, circuit controller 210 may select a frequency for the charging signal that is different from the frequency f _Zr 1034 corresponding to the minimum real impedance value and the frequency f _ZMod 1036 corresponding to the minimum reference impedance calculation. Rather, circuit controller 210 may determine the harmonics of the charging signal by balancing the real and imaginary impedance values, such that the frequency selected for the charging signal can be between frequency f _Zr 1034 and frequency f _ZMod 1036.

In one particular implementation, separate portions of the pulse of the charging signal may be shaped by the circuit controller 210 based on two or more impedance measurements. For example, FIG. 11 is a signal diagram of shaped pulses 1108 of a battery cell charging signal 1102 corresponding to two or more frequencies generated by a battery recharging circuit, according to one embodiment. Similar to the power signal pulse described above with reference to FIG. 9, the pulse 1108 may include a leading edge 1110 configured as a harmonic corresponding to a minimum real impedance value. For example, the shape of the leading edge 1110 of the pulse 1108 may correspond to harmonic f _Zr 1034. However, a second portion 1112 of the pulse 1108 may include a harmonic based on another frequency different from frequency f _Zr 1034. For example, the leading edge 1110 and the second portion 1112 may collectively include a first harmonic f _ZMod 1036 corresponding to the minimum reference impedance calculation 1030. The application of the harmonic f _ZMod 1036 corresponding to the minimum reference impedance calculation may determine the duration of the second portion 1112 of the pulse 1108 to reduce the imaginary impedance at the battery's electrodes due to application of the recharge power signal. Determining and applying the harmonic based on not only the real impedance component of the battery but also the imaginary impedance component allows for a more efficient recharge power signal to be used to charge the battery cells.

Additional aspects of the charge signal pulse may be controlled by circuit 400. In particular, controlling the trailing edge of the charge signal pulse can provide efficiency benefits when charging a battery cell. FIGS. 12A and 12B are plots of applied/measured voltage 1208 across a battery cell and measured charging current 1210 of a battery cell versus time 1206, according to one embodiment. As noted above, the charge signal may include a sharp trailing edge that removes the charge signal 1212 to the battery cell. However, as can be seen in the plot of FIG. 12A , even when the voltage applied to the battery is set to zero, the current I does not immediately drop to zero, but rather has some delay before reaching zero. However, the time between pulses may be set so that the next pulse does not begin until the current reaches zero (the cell is depolarized). Thus, in one example, circuit 400 may be controlled to wait until the current of battery cell 404 reaches zero before beginning the next pulse of the charge signal, to prevent potential damage to the battery cell or inefficient charging due to starting polarization of the cell before it is fully depolarized. Because charging can only occur in pulses, shortening or minimizing the time between pulses shortens the overall charging time, all else being equal. In the case of voltage-controlled circuit 400, the current component 1210 of the charging signal may lag behind the voltage component 1208. More specifically, as shown in FIG. 12A , it may take some time for the battery current 1210 to return to zero after the battery voltage 1208 is removed. This delay in the battery current returning to zero may further reduce the efficiency of the charging pulse. Therefore, in some embodiments, the voltage 1208 of the charging signal may be controlled to be below a transition voltage corresponding to zero current, represented as line 1206 in plot 1222 of FIG. 12 , as shown in plot 1222 of FIG. 12B . Generally, the transition voltage 1206 is the voltage of the charging signal at which current flow to the battery reverses and may resemble the floating voltage of a battery cell. In particular, allowing the voltage 1208 to fall below the transition voltage 1206 for a period of time (shown as period T _T 1216) following the trailing edge 1212 of the pulse may allow the current 1210 to reach zero amperes more quickly than a pulse without the drop. The duration T _T 1216 during which the voltage 1208 of the voltage controlled charging circuit 400 is controlled below the transition voltage corresponding to zero current may be determined or set by the circuit controller 210 to minimize the time it takes for the current 1210 of the battery cell 404 to return to zero amperes. In one example, the voltage drop may be controlled to not fall below the minimum recommended cell voltage of the battery cell to prevent degradation of the battery cell electrodes. Also, the magnitude of the voltage drop may be controlled to be a percentage of the magnitude of the charging pulse relative to the transition voltage. Furthermore, the voltage return to the transition voltage may be controlled at a rate that maintains the current at zero amperes as long as the charge in the battery cell remains balanced. If the current 1210 returns to zero amperes for a specified rest period, another charge pulse 1202 may be applied to the battery cell 404. In this manner, by reducing the time it takes for the current 1210 in the battery cell 404 to return to zero, the rate at which charge pulses can be applied to charge the battery cell can be increased.

While generally described above as a power control circuit, it should be understood that the charging circuit 400 may be voltage-controlled or current-controlled, and that each may be utilized in different situations. Both approaches are similarly controlled by measuring the voltage drop across the battery cells 404 and measuring the current through a current-sense resistor connected in series with the battery cells 404. The primary differences between the control schemes are based on whether the current-sensing hardware (e.g., the current-sense resistor) is external or internal to the power supply circuit (e.g., the power amplifier of the power supply circuit 402) and whether the voltage drop across the battery cells 404 or the current-sense resistor is processed first. In the case of a voltage-controlled power supply, an initial voltage measurement may be made across the battery cells 404, while the corresponding voltage drop across the external current-sense resistor may then be measured, and the current in the battery cells 404 may be calculated, such as by utilizing Ohm's Law. This allows for precise control of the voltage of the charging signal, while the current calculation involves measuring the voltage across the battery cells 404 first and then calculating the current in the battery cells.

The voltage-controlled charging circuitry may optionally be controlled to provide a charging signal having components as shown in FIG. 12. In particular, the voltage of the charging signal 1202 may be controlled to provide a flat voltage for the remainder of the pulse after the sinusoidal leading edge 1214 as described above. A voltage-controlled charging signal may provide the benefits of the charging pulse as described above. The voltage control circuitry 400 may also provide a trailing edge 1212 that includes a portion 1216 where the voltage is below a transition voltage corresponding to zero current in the battery cell 404. Also as shown in FIG. 12, the current 1210 in the battery cell 404 lags the control voltage 1208, indicating that the calculation of the current follows the control of the voltage 1208. Control of the voltage signal 1208 may cause the circuitry 1210 to return to zero amperes before an additional charging pulse is similarly provided to the battery cell 404. Another advantage of the voltage control circuit 400 is that its precise control prevents the thermodynamic thresholds of the battery cells 404 from being exceeded, preventing breakdown characteristics of the battery cells 404, such as maintaining the voltage below the voltage at which the electrolyte of the battery cells 404 begins to break down.

The circuits and methods discussed herein may also be implemented using a current-controlled power supply. If the power supply of circuit 400 is current-controlled, a pre-calibrated sense resistor in the power supply circuit may provide the initial measurement, as the current through this resistor may be determined by the current through the battery cell 404. Thus, by precisely knowing the charge current, the charge current to the battery cell 404 can be precisely controlled without knowing the voltage drop across the battery cell. In this embodiment, the current to the battery cell 404 (as measured at the current sense resistor) may be inherently known (by the pre-calibrated voltage across the sense resistor), while the voltage of the battery cell 404 is measured as a result of this applied current. Figure 13 is a plot of the measured current 1314 and voltage 1310 across a current sense resistor at a battery cell versus time 1306 in response to a charge signal 1304 applied to the battery cell, according to one embodiment. As shown in plot 1302, the current to the battery cell 404 may be controlled to produce a similar pulse as described above, with a sinusoidal leading edge 1314, possibly corresponding to a minimum impedance value of the battery cell 404, followed by a constant current. Current control circuit 400 may also provide a trailing edge 1312 with a portion 1316 where the current is below zero amperes, corresponding to a stable transition voltage of the battery cell 404. Also shown in FIG. 13, the voltage response 1310 of the battery cell 404 lags behind the control current 1308, indicating that voltage acts as a feedback response rather than the primary control factor.

In applications where simple components are available or where the process is constrained by the existing power hardware of the device being charged, current control may be the default mechanism. Alternatively, in implementations where both the controller response time and the battery transient response are fast, voltage control and current control methods may behave similarly. However, as frequency increases and/or battery reactance reaches high levels, the behavior of these two methods may diverge, and practical control considerations may be necessary.

The above-described embodiments involve measuring or obtaining the impedance (real and/or imaginary) of the battery cell 204 to determine the frequency components of at least a portion of the pulse of the charging signal. The impedance value of the battery cell 204 may be obtained in a variety of ways. In one embodiment, the impedance of the battery cell 204 may be measured or estimated in real time as the charging pulse is applied to the battery cell. For example, the magnitude and time components of the voltage and current waveforms of the charging signal at the battery cell 204 may be measured and/or estimated. The difference between the measured magnitude and time components of the voltage and current waveforms may be used to determine or estimate the real, imaginary, or approximate impedance of the battery cell 204. For example, the real and imaginary impedance values may be determined from the leading edge of the charging pulse because the leading edge is comprised of a single known harmonic, and the magnitude difference of the voltage and current waveforms may be obtained at consistent minimum and maximum values of the edge. Similarly, the impedance components may be approximated from measurements of the magnitude of the voltage and current waveforms at the trailing edge of the charging pulse. In still other embodiments, various measurements of the voltage and current waveforms of the charging signal may be adjusted based on weights applied to the measurements. Generally, the impedance of the battery cell 204 may be determined or estimated by determining or measuring multiple components of the voltage and current waveforms of the charging signal. In other embodiments, hundreds or thousands of measurements of the voltage and current waveforms may be acquired and analyzed by a digital processing system. Generally, improved waveform fidelity and/or more measurements may allow for a more accurate analysis of the impedance of the waveform applied to the battery cell 204, thereby more fully determining the harmonic content of the charging signal where impedance minima occur or other aspects of the waveform's effect on the battery cell 204, and thus determining the pulse shape of the charging signal.

FIG. 14 is a block diagram illustrating an example of a computer device or computer system 1400 that can be used to implement embodiments of the disclosed network. In particular, the computer device of FIG. 14 is one embodiment of a circuit controller 210 that performs one or more of the operations described above. The computer system (system) includes one or more processors 1402-1406. The processors 1402-1406 may include one or more internal level caches (not shown) and a bus controller or bus interface unit that directs interaction with a processor bus 1412. The processor bus 1412, also known as a host bus or front-side bus, may be used to couple the processors 1402-1406 to a system interface 1414. The system interface 1414 is connected to the processor bus 1412 and may interface other components of the system 1400 with the processor bus 1412. For example, system interface 1414 may include a memory controller 1418 that interfaces main memory 1416 with processor bus 1412. Main memory 1416 typically includes one or more memory cards and control circuitry (not shown). System interface 1414 may also include an I/O interface 1420 that interfaces one or more input/output (I/O) bridges or I/O devices with processor bus 1412. As shown, one or more I/O controllers and/or I/O devices, such as I/O controller 1428 and I/O device 1430, may be connected to I/O bus 1426.

I/O devices 1430 may also include input devices (not shown), such as an alphanumeric input device including alphanumeric and other keys for communicating information and/or command selections to processors 1402-1406. Another type of user input device may include a cursor control, such as a mouse, trackball, or cursor direction keys, for communicating directional information and command selections to processors 1402-1406 and controlling cursor movement on a display device.

System 1400 may include a computer-readable device, such as a dynamic storage device (referred to as main memory 1416) or random access memory (RAM), coupled to processor bus 1412 for storing information and instructions executed by processors 1402-1406. Main memory 1416 may also be used for storing temporary variables or other intermediate information during execution of instructions by processors 1402-1406. System 1400 may also include read-only memory (ROM) and/or other static storage devices coupled to processor bus 1412 for storing static information and instructions for processors 1402-1406. The system illustrated in FIG. 14 is but one possible example of a computer system that may be employed or configured in accordance with aspects of the present disclosure.

According to one embodiment, the techniques described above may be performed by computer system 1400 in response to processor 1404 executing one or more sequences of one or more instructions contained in main memory 1416. These instructions may be read into main memory 1416 from another machine-readable medium, such as a storage device. Execution of the sequences of instructions contained in main memory 1416 may cause processors 1402-1406 to perform the process steps described herein. In alternative embodiments, circuitry may be used in place of or in combination with software instructions. Thus, embodiments of the present disclosure may include both hardware and software components.

Machine-readable media includes any mechanism for storing or transmitting information (e.g., software, processing application) in a form readable by a machine (e.g., a computer). Such media may be in the form of non-volatile and volatile media, but is not limited to this. Non-volatile media includes optical or magnetic disks. Volatile media includes dynamic memory, such as main memory 1416. Common forms of machine-readable media include, but are not limited to, magnetic storage media (e.g., floppy diskettes), optical storage media (e.g., CD-ROMs), magneto-optical storage media, read-only memory (ROM), random access memory (RAM), erasable programmable memory (e.g., EPROM and EEPROM), flash memory, or other types of media suitable for storing electronic instructions.

It may be desirable for battery-powered electronic systems to be operable while charging. Thus, for example, it would be advantageous for battery-powered tools to be operable while charging. Similarly, electronic systems may be configured to operate in various states while charging. For example, a cell phone, tablet, laptop computer, etc. may be fully operational while charging, or may operate in various low-power modes while charging, or may have some limited functionality while charging. According to aspects of the present disclosure, a power converter, such as a step-down converter or a step-up converter, may operate in synchronization or coordination with circuitry controlling the charging waveform or energy flux at the electrodes of an electrochemical device (e.g., a battery cell). The charging waveform may include frequency components and/or one or more harmonics associated with minimum or low impedance (including its real and/or imaginary components, or any combination thereof) of the electrochemical device being charged. The system may be controlled to match the charging signal, including the power signal, to the load so as not to disrupt the shape or configuration of the charging waveform. Because the charging signal is intentionally controlled, it is advantageous not to change its form or configuration. In particular, the system may control the power signal so as not to disrupt the harmonically shaped leading edge of the charge pulse. Thus, for example, the harmonic-defined leading edge of the charge waveform is maintained (e.g., not distorted) while supplying power to any load. In another example, the system adjusts the operation of a power converter to shape the charge signal and/or to function in conjunction with or instead of a recycling function. Also, the discharge (power signal) from the battery may be adjusted in frequency/harmonic content based on a discharge impedance, which may be the same or different from the charge impedance used to adjust the frequency/harmonic content of the charge waveform. In either case, the discharge signal components may be adjusted.

FIG. 15 is a circuit diagram illustrating one possible example of a circuit topology 1500 that generates a shaped waveform based on the impedance (or other measurement, such as susceptance) of a cell being charged. Because this system includes components introduced with respect to FIG. 4, like reference numerals represent like components in FIGS. 4 and 15. Generally speaking, this circuit includes filter circuits 406, 418 that can independently or cooperatively shape a waveform (e.g., the leading edge of a charge pulse) based on one or more harmonics and their respective effects on the impedance. As noted above, each filter circuit section may include a shaping inductor 410. The filter circuit sections may include inductors of the same or different values. In contrast to the circuit shown in FIG. 4, the circuit of FIG. 15 includes a power converter 1502 coupled between the electrochemical cell 404 and a load 1504. In one example, the power converter is a step-down converter 1506. Generally speaking, a step-down converter reduces the voltage of a power source to the voltage required by the load. In another example, the power converter is a boost converter 1508. Generally speaking, a boost converter increases the voltage of a power source to the voltage required by a load. In another example, both boost and buck converters may be provided in parallel and operated depending on the operating conditions or type of one or more loads. The buck and boost operation may also be adjusted to maintain a voltage output between a maximum and minimum battery voltage. As discussed in more detail below, it is also possible to include one or more parallel buck and/or boost circuits that allow for alternative output pulse control.

There are various possible examples of buck and boost circuit topologies. FIG. 16 shows an example of a charging circuit employing a buck converter coupled between an electrochemical cell and a load. The circuit includes a filter circuit 406, as described above with reference to FIG. 4, controlled by a control signal labeled "PULSE" from a controller (e.g., controller 210) to a filter transistor. The circuit further includes a buck converter 1600. The buck converter is coupled to a battery 404. The buck converter includes a transistor coupled to the battery and controlled by a control signal "BUCK" generated by the controller. FIG. 17 shows an example of a charging circuit employing a boost converter 1700 coupled between an electrochemical cell and a load. Similar to FIG. 16, the circuit includes a filter circuit 406, as described above with reference to FIG. 4, controlled by a control signal labeled "PULSE" from a controller (e.g., controller 210) to a filter transistor. The boost converter is coupled to the battery 404. The buck converter comprises a transistor coupled to the battery and controlled by a control signal "boost" generated by the controller. Other features of the circuits shown in Figures 4 and 15 may also be included in one or both of the circuits shown in Figures 16 and 17. Additionally, other buck or boost topologies may be employed.

Figure 18 shows an example of the control of the various circuits of Figures 15-17. These controls and charge pulses relate to the circuit of Figure 15. However, these concepts are also applicable to circuits with fewer components, such as those shown in Figures 16 and 17, or more complex circuits. Figure 18A shows the voltage (top) and current (bottom) components of a regulated charge pulse. As with the other pulses shown herein, this circuit can be controlled to shape the leading edge to match the frequency and/or harmonics (including their respective real and/or imaginary components) associated with the relatively low or lowest impedance of the cell being charged. Other measures, such as admittance or its susceptance and conductance components, may also be used in conjunction with the impedance used in the above example. As used herein, the term impedance may include its reciprocal, admittance. As noted above, impedance may change over time based on the electrochemical cell's state of charge, temperature, age, and/or number of cycles, etc. Thus, the waveform can similarly be programmatically modified or dynamically modified based on feedback and impedance measurements. In one example, shaping can be performed by driving different combinations of filtering circuits to employ different combinations of inductors that shape the leading edge of the charging signal. It is also possible to characterize the impedance of the cell at various harmonics based on state of charge, temperature, age, etc., and programmatically change the combination of driving the filter circuits to modify the charging waveform based on any such characterization, alone or in combination, rather than on actual measurements of the impedance.

In any event, in systems that may require the application of some power to the load during charging, this power may be applied in a manner that does not disrupt the shape and frequency/harmonic characteristics and/or content of the charging waveform and helps avoid applying waveforms associated with suboptimal impedance or that affect the control of the charging waveform. However, as will be appreciated from examples discussed in more detail below, in some cases the charging pulse may be shaped by driving a buck or boost circuit in some combination with a filter circuit. In any event, in one example, to avoid disrupting the leading edge shape and/or control of the waveform shape or content, the operation of the charge controller may be interleaved with the operation of the buck or boost converter so that the buck or boost is not "on" during at least a portion of the charging pulse. In one example, the power converter is turned on only after the charging pulse is turned off. In another example, the power converter may be turned on only after the leading edge transitions into the second "body" portion of the pulse following the shaped leading edge while the charging pulse is on. In another example, the power converter is turned off when the charging pulse is on. In another example, the power converter is turned off for some time before the charge pulse is turned on.

FIG. 18B illustrates control pulses that can be applied to various components of the circuit of FIG. 15 to drive the power converter to form and deliver the charge pulse of FIG. 18A and deliver power to a load. More specifically, FIG. 18B illustrates three different control signal pulses associated with forming and delivering the regulated charge signal pulse. These pulses can be executed in sequence, defined at any frequency or duty cycle, and delivered in a variety of different configurations depending on the shape of the desired charge pulse. This example definition is merely illustrative of the various concepts discussed herein and should not be construed as limiting. A first pulse, labeled "soft pulse," is applied to switch 412 and is the pulse control signal 416 for switch 412. A second pulse, labeled "hard pulse," is applied to a switch in filter circuit N 418. Depending on the desired shape of the charge pulse, one or more second pulses may be applied to one or more of the N filter circuits. Furthermore, if the first "soft pulse" is sufficient to shape the charge pulse, the second pulse can be omitted. A third "recycle" pulse is then applied to switch 422 as a recycle signal. The combination of the first and second pulses shapes the leading edge of the pulse. Depending on the inductance value of any given filter circuit and the desired harmonic characteristics of the leading edge, the filter circuits may be driven in a variety of possible combinations, with the first and second sequences discussed herein being merely exemplary. Similarly, various possible implementations may employ one or more filter circuits with the same or different inductor values, and various control schemes may be applied to the various filter circuits to control the leading edge of the charge pulse, define other characteristics of the charge pulse, or generally define the charge signal, whether pulsed or not. Additionally, to provide the desired inductance value according to the desired charge pulse shape, various filter circuits may be driven synchronously in parallel, such that the parallel combination of inductors achieves the inductance value when driven by any combination of filter circuits. Various possible values may also be achieved by directly connecting inductors in series or parallel within the filter circuits.

Finally, buck/boost pulses are applied to either the buck or boost circuitry, depending on whether the circuit includes a buck or boost branch and whether buck or boost functionality is desired regardless of the load's operating mode. As noted above, in some implementations, providing either a buck or boost power converter may be sufficient, while other implementations may include both buck and boost types. These exemplary control pulses are examples of discrete pulses that are part of such a pulse train (e.g., a pulse-width modulated (PWM) signal), typically applied at high frequency as part of a charge train to generate a charge train for charging an electrochemical device. It will be appreciated that in accordance with the present disclosure, various possible charging and/or discharging functions as discussed herein can be achieved by using control signals (which may be PWM signals) for control of filter circuits (e.g., soft or hard), recycle functions, and boost and buck circuits (e.g., PWM "buck" or "boost" control signals for each transistor in buck circuit 1600 or boost circuit 1700) discretely and synchronously in various possible combinations.

Referring to Figures 18A, 18B, and 15, it can be seen that when the first rising edge of the soft pulse occurs at time T0, turning on switch 412, current begins to flow through electrochemical device 404 and a voltage rises at load terminal node 440. At time T1, the rising edge of the hard pulse follows the soft pulse, while the soft pulse is still high (and circuit 406 is still operating). At time T1, current from filter circuit N 418 begins to flow to the load, in combination with the current from circuit 406 through switch 412. Thus, the charge pulse (shape of the leading edge) is governed by the combination of filter circuit 406 and circuit N 418.

In this example, the first pulse is labeled a "soft" pulse because a larger inductor results in a slower rise time for the current, and therefore driving the circuit with a larger inductor results in a slower rise time for the leading edge of the pulse. In this example, the second pulse is labeled a "hard" pulse because a smaller inductor results in a faster rise time for the current, and therefore driving the circuit with a smaller inductor results in a faster rise time for the leading edge of the pulse. In the illustrated example, two filter circuits are combined side-by-side to shape the leading edge of the charge pulse beginning at time T0. Additional combinations may also be employed to shape the leading edge to mimic a sinusoidal rising edge (e.g., additional filter circuits and/or fine-tuning of the filter circuit switches could smooth the leading edge and shape it similar to the first half of a sinusoidal pulse). Different inductor values could also be applied to the various circuits N 418, and the shape of the leading edge of the pulse could be defined by adjusting the controls between any possible combinations.

At V2, while the soft and hard pulses are still at a high level, the current flow through circuit N 418 is at a maximum when the voltage at terminal node 440 reaches a maximum, essentially the rail voltage minus the voltage drop across the switches in filter circuits 406 and 418. Because the amount of current that can be drawn by the battery load is determined by the terminal voltage and tends to decrease over time at a given voltage, the charge current to the battery decreases over the times marked V2 and V3 while the terminal voltage remains relatively constant.

At time T3, both the hard and soft pulses drop to zero, stopping the charging current from both circuit 406 and circuit N 418. At this point, a recycle pulse may be applied to switch 422 to activate the recycle portion of the circuit. As discussed above, activating the recycle pulse allows the current to quickly return to zero by directing charge at the terminal node to storage capacitor 432.

Additionally or alternatively, a power converter (which may include a step-down circuit and/or a step-up circuit) may be turned on to supply source energy to load 1504. As introduced above, it may be desirable to charge a battery while simultaneously powering a load (e.g., a power tool, a cell phone, a vehicle function, etc.). Also, as introduced above, in some cases, step-up may be required to operate the load, and in other cases step-down may be required to operate the load.

As shown in FIG. 18B, the boost/buck pulses that drive the boost switch or buck switch, respectively, occur even while the charge pulse is not activated. In this example, if either the buck or boost circuit is active during the charge pulse, the battery powers the buck or boost circuit, and thus the load. In some examples, if the power converter is not operating, a recycle function may be used to return the charge pulse voltage at the terminals to zero as quickly as possible after the charge pulse is turned off at the time associated with voltage V3. In one example, application of the recycle pulse drives recycle switch 426. If a power converter function is present and operating, the power converter may act as an alternative to or coordinate with the recycle pulse.

While the buck or boost circuit is shown operating with the charge pulse inactive, it is also possible for it to operate with the charge pulse active to further shape the pulse. However, in one example, such operation occurs after the rising edge, or at least after the initial portion of the rising edge, so as not to distort the shape of the rising edge. The buck or boost operation in this example may also replace the function of shaping circuit 428. Similarly, the buck or boost may also act as a recycle function to quickly return the charge pulse to zero, rather than recycling initial energy rather than drawing, or in conjunction with, energy from the battery to power the load. It should be recognized that, in conjunction with various uses of the buck or boost circuit for coordination with the shaping and regulated charge waveform function in cases of dynamic charging, one or more capacitors may be used in the buck or boost branch to maintain a stable voltage at the load.

In addition to controlling power delivery to a load through a power converter, aspects of the present disclosure also include controlling the power converter to shape the output pulse delivered from the electrochemical device to the load. Such pulse shaping may be performed in conjunction with or independent of charging. Thus, output pulse shaping may be performed by a step-down or step-up circuit, either alone or in various possible combinations, separate from the charging function. In one example, shaping the output pulse from the electrochemical device to the load may provide benefits similar to those achieved by harmonic shaping or conditioning of the input charging waveform, such as shaping at least the leading edge of the charging pulse to the electrochemical device. In one example, the shape of the output waveform may be associated with low or minimal impedance power delivery from the battery. In some cases, the output impedance may be expected to be the same or nearly the same as the input impedance under the same conditions of the electrochemical device (e.g., battery state of charge, temperature, age, etc.). In other cases, the output impedance may be measured or characterized differently from the input impedance under different conditions, and these different measurements or characterizations may be used to select optimal output frequency characteristics (which may be harmonics). Impedance measurement circuitry 408 may be used to measure the output impedance from the load at different frequencies, similar to that described above for measuring the input impedance to the battery 404. In any case, in various examples, the output waveform (e.g., shaped pulse) from the battery to the load may be shaped, in particular examples, by shaping the leading edge of the output pulse to a particular shape corresponding to the frequency and/or by harmonic shaping. The optimal harmonic or frequency characteristics are associated with values representing current flow to or from the electrochemical device, depending on whether charging or discharging (power delivery from the electrochemical device) is discussed.

The optimal frequency or harmonic can be associated with whatever minimizes the input or output impedance to the electrochemical device. However, in any given situation, this may not be the absolute minimum impedance because the system may select a value close to the minimum or iteratively select a value up to the minimum. In other situations, the nature of feedback loops and dynamic systems may cause the system to select a range of values around or associated with the minimum. For example, even in a characterized system, the battery may not be fully characterized for all state-of-charge, age, temperature, or other conditions, and reasonable extrapolations and assumptions from the characterization may be made in selecting harmonic components or frequencies to define certain portions of the charge or discharge waveform (e.g., to shape the leading edge of a discharge or charge pulse). Thus, the use of "optimal" in the context of values such as impedance, harmonics (frequencies), or other measures discussed herein that represent current flow to or from an electrochemical device does not necessarily imply that the minimum impedance value is known, or that the system knows the harmonics or frequencies that provide that minimum. Also, as described elsewhere herein, other measures may be used, such as power, admittance, or its susceptance and conductance components. In the case of admittance, the optimum value may be associated with a harmonic that provides the maximum admittance or a value within some range of the maximum admittance during charging or discharging.

In one example, the leading edge of a pulse from a battery may be shaped by controlling a switch in a buck or boost circuit. For example, the switch (e.g., a transistor) in the buck circuit of FIG. 16 may be controlled by applying a variable duty cycle or variable period pulse train to the buck input. Alternatively, the boost switch in the boost circuit of FIG. 17 may be controlled by applying a variable duty cycle or variable period pulse train to the boost input. To harmonically shape the leading edge of a pulse from an electrochemical device to have a sinusoidal shape at a certain frequency, the system may control the duty cycle or period of the PWM signal driving the buck or boost switch at the leading edge, and then maintain the duty cycle or period for the remainder of the pulse.

The harmonics, leading edges, etc. of the charge or discharge signal may also be adjusted to match the impedance (or other value) of the electrochemical device to optimize the combination of charge or discharge interaction and impact on the electrochemical device. For example, the system may operate to balance the battery's charge rate with cycle life (e.g., the number of charge and/or discharge cycles required before the battery capacity drops below a certain threshold, e.g., 75% (25% capacity loss)). In some cases, the system may determine the harmonics for the highest charge rate, but applying signals to achieve that charge rate may not be optimal for cycle life. Thus, the system may apply a lower charge rate than possible, but applying such a lower charge rate may affect the impedance, which may change the harmonic content of the charge signal. In other cases, the system may apply harmonically adjusted charge or discharge pulses by controlling a combination of duty cycle, frequency (e.g., of the charge pulse), and/or overall frequency (e.g., a combination of charge and rest) to achieve a balance between various possible real-time battery characteristics, such as charge rate, and/or longer-term battery characteristics, such as cycle life. For example, a relatively large charge or discharge current reduces the cell's impedance, which generally benefits the charge or discharge rate, while a high charge or discharge rate, like battery charging and discharging, will have some impact on cycle life, even when harmonically optimized with the complex impedance feedback discussed herein. Duty cycle has a significant effect on peak current. On the other hand, for a fixed current RMS, a frequency with the lowest impedance may result in a lower charge rate but benefit cycle life. Thus, the system may charge or discharge to optimize the balance between different factors. In other words, aspects of the present disclosure may operate to improve charge or discharge rates relative to conventional technology, and such improvements may be performed while considering other desirable outcomes, such as optimizing cycle life under the above conditions. In some such cases, the charge or discharge rate may remain improved relative to conventional systems, but may be operated at a level lower than the maximum to balance other factors.

FIG. 19 illustrates one method for generating a harmonically shaped (e.g., sinusoidal) leading edge of the output signal from the battery (which may be a shaped pulse train). That is, the control pulse width can be varied during the shaping portion of the charge pulse. For example, as shown in FIGS. 19A and 19D (highlighting the area where the duty cycle of the buck/boost transistor control sequence in FIG. 19A varies), the control pulse width can be varied from a relatively short (almost off) pulse width to a relatively long (almost on) pulse width, resulting in a relatively slow-rising voltage/current followed by a relatively faster-rising leading edge over the same period for each discrete pulse, mimicking the shape of the sinusoidal leading edge of a discharge pulse (from a battery). The duty cycle can be uniformly increasing or non-uniformly controlled to achieve a variety of possible shapes. Alternatively, the same pulse width (percentage) can be used for each discrete pulse, while the duration can be varied from pulse to pulse.

In either case, the control sequence shown in FIG. 19A, or its equivalent, may be applied during each discrete step-down or step-up pulse shown in FIG. 19B. In one example, the variable duty cycle or control sequence of FIG. 19A produces a train of output pulses from an electrochemical device having a harmonic-shaped leading edge, as shown in FIG. 19C. The duty cycle or period is controlled to create a leading edge that the system determines (or characterizes) to match the optimal output impedance of the cell at any frequency. The length of application of the duty cycle or period control may also be controlled to shape the output pulse. In the example of FIG. 19C, the duty cycle is controlled to create an output pulse during the shaped leading edge. Then, during the main portion of the pulse, the duty cycle remains constant for the remainder of the pulse width.

PWM control of the buck or boost circuit will likely cause the output current from the battery to the load to gradually increase in a somewhat "step" fashion. These steps may be smoothed out by filtering at the output of the electrochemical device. This may be integrated into the power converter, or may precede the power converter.

Control of the duty cycle or duration of the control pulse may also be applied to shaping the charge pulse. Such duty cycle control may be performed alone or in combination with the methods described above, whereby filter circuits and combinations of filter circuits (e.g., filter circuits 406 and 418) are selected to apply a certain frequency profile based, at least in part, on the inductance value of each filter circuit's inductor 410 and its effect on leading edge shaping. Referring again to FIGS. 19A and 19B, the initial variable duty cycle control signal of FIG. 19A may be applied as pulse control signal 416 to switch 412, as shown in the dashed box for the "soft" pulse in FIG. 19B. As discussed above and shown in FIG. 19B, so-called hard pulses may also be used in combination to shape the leading edge of the charge pulse, for example, as shown in FIG. 18A. Duty cycle control provides additional control beyond the selection of various combinations of filter circuits, allowing for finer tuning of the leading edge when used.

Returning to the discussion of power converter functionality, one or more parallel step-up or step-down circuits may also be employed, as shown in FIGS. 20A/20C and 20B/20D. In either case, the addition of one or more parallel step-down or step-up topologies may provide, among other benefits, opportunities for optimizing efficiency relative to a single power converter design, provide alternative paths for the power converter, and reduce the size of components in each parallel path, potentially reducing thermal losses and improving switching efficiency. In the illustrated example, the step-up or step-down inductors in the parallel circuits are not the same, and a smaller inductor value in one circuit of each pair potentially provides higher efficiency than a parallel circuit with a relatively larger inductor. The inductors may be the same in one or both cases, and in various examples, additional parallel step-down or step-up circuits may be employed. In one example, two or more parallel power conversion circuits may operate in parallel (e.g., step-down or step-up), each feeding a load. In another example, a varying duty cycle may be used in each parallel power converter circuit, as shown, to shape the leading edge of the output pulse, e.g., to match harmonics to provide optimal output impedance, or to vary the pulse duration for the same purpose. In yet another example, one of the circuits may initially operate with a duty cycle that shapes the leading edge of the pulse, particularly to reduce the initial supply current, and if a higher and/or more stable output current is desired, one or more additional parallel circuits may be operated to provide current not available in a single power converter. In some cases, it may be desirable to carefully control both the shape and amount of output current from the electrochemical device, providing additional parallel power converters, either alone or in combination with pulse-shaping control, for control flexibility.

Various embodiments of the present disclosure have been discussed in detail above. While specific implementations have been discussed, it should be understood that this is done for illustrative purposes only. Those skilled in the art will recognize that other components and configurations may be used without departing from the spirit and scope of the present disclosure. Accordingly, the above description and drawings are illustrative and should not be construed as limiting in any way. Numerous specific details are set forth to provide a thorough understanding of the present disclosure. However, in certain instances, well-known or conventional details are not set forth in order to avoid obscuring the description. References to one embodiment or an embodiment in this disclosure may refer to the same embodiment or any embodiment, and such references refer to at least one of these embodiments.

A reference to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic described with respect to that embodiment is included in at least one embodiment of the present disclosure. The appearances of the phrase "one embodiment" in various places throughout this specification do not necessarily all refer to the same embodiment, or to separate or alternative embodiments mutually exclusive of other embodiments. Furthermore, various features have been described that may be exhibited by some embodiments and not by other embodiments.

The terms used herein generally have their ordinary meanings in the art, within the context of this disclosure and in the specific context in which each term is used. Additionally, for any one or more of the terms discussed herein, alternative expressions and synonyms may be used, and no special meaning is intended whether or not a term is specifically described herein. In some cases, synonyms for particular terms are provided. The listing of one or more synonyms does not exclude the use of other synonyms. The use of examples anywhere in this specification, including examples of any term discussed herein, is illustrative only and is not intended to further limit the scope and meaning of the disclosure or any exemplary term. Similarly, the disclosure is not limited to the various embodiments provided herein.

Without intending to limit the scope of the present disclosure, examples of instruments, devices, methods, and their respective associated results according to embodiments of the present disclosure are provided. Note that titles or subtitles may be used in each example for the convenience of the reader, but these are not intended to limit the scope of the present disclosure in any way. Unless otherwise specified, technical and scientific terms used herein have the meanings commonly understood by those of ordinary skill in the art to which the present disclosure pertains. In the case of conflict, the present specification, including definitions, will control.

Embodiments of the present disclosure include various steps described herein. These steps may be performed by hardware components or embodied in machine-executable instructions, such that a general-purpose or special-purpose processor programmed with the instructions performs the steps. Alternatively, the steps may be performed by a combination of hardware, software, and/or firmware.

Various modifications and additions can be made to the described exemplary embodiments without departing from the scope of the present invention. For example, although the embodiments described herein refer to particular features, the scope of the present invention also includes embodiments having different combinations of features and embodiments that do not include all of the described features. Accordingly, the scope of the present invention is intended to embrace all such alternatives, modifications, and variations, together with all equivalents thereof.

Claims (19)

a charging signal forming circuit;
a controller in operative communication with the charge signal shaping circuit to control the charge signal shaping circuit to define a charge signal for the electrochemical device based on harmonics associated with a value representative of current flow to the electrochemical device;
a power converter operatively coupled to the electrochemical device and the controller to provide power to a load;
Equipped with
the controller controls the power converter to generate a discharge waveform from the electrochemical device based on harmonics associated with a value representing a current flow from the electrochemical device.
Charging system. 2. The charging system of claim 1, wherein the charge signal comprises a series of regulated charge pulses and the discharge waveform comprises a series of regulated discharge pulses, and the controller controls the charge signal shaping circuit and the power converter to interleave the series of regulated discharge pulses with the series of regulated charge pulses. 3. The charging system of claim 2 , wherein the regulated discharge pulse immediately follows the regulated charge pulse. 3. The charging system of claim 2 , wherein the regulated discharge pulse is activated in a body portion of the regulated charge pulse, the body portion following a shaped leading edge of the regulated charge pulse. The charging system of claim 1, wherein the power converter includes at least one of a first step-down converter or a first step-up converter. 6. The charging system of claim 5 , wherein the power converter further includes at least one of a second buck converter in parallel with the first buck converter or a second boost converter in parallel with the first boost converter. The charging system of claim 1, wherein the power converter includes a switch that controls the power converter, the switch receiving a control signal having a pulse width that varies from approximately an off pulse width to approximately an on pulse width to shape the edges of an output pulse from the load. The charging signal shaping circuit
a first molded inductor having a first inductance value in electrical communication with the power rail;
a first switching device in electrical communication with the first molded inductor and configured to connect the electrochemical device to the power rail through the first molded inductor;
The charging system of claim 1 , comprising: The charging signal shaping circuit
a second molded inductor having a second inductance value in electrical communication with the power rail;
a second switching device in electrical communication with the second molded inductor and configured to connect the electrochemical device to the power rail through the second molded inductor;
The charging system of claim 8 , comprising: 10. The charging system of claim 9, wherein the controller defines the charging signal for the electrochemical device based on the harmonics associated with the value representing the current flow to the electrochemical device by sending a first control signal to the first switching device and a second control signal to the second switching device. 11. The charging system of claim 10, wherein the first inductance value is greater than the second inductance value, and the first control signal turns on the first switching device before the second control signal turns on the second switching device, thereby shaping the leading edge of the charging signal based on the harmonic associated with the value representing the current flow to the electrochemical device. 11. The charging system of claim 10, wherein the first inductance value is the same as the second inductance value, and the first control signal shapes the leading edge of the charging signal based on the harmonic associated with the value representing the current flow to the electrochemical device by turning on the first switching device before the second control signal turns on the second switching device. The charging system of claim 1, wherein the controller drives the power converter when the charging signal is off. The charging system of claim 1, wherein the value is associated with a minimum impedance when a charging signal is applied to the electrochemical device, and a harmonic component associated with the minimum impedance is applied to the charging signal. 9. The charging system of claim 8, wherein the controller provides a control signal to the first switching device to generate a harmonically shaped leading edge of a pulse of the charging signal, the harmonically shaped leading edge being associated with a minimum impedance value of the electrochemical device . 16. The charging system of claim 15 , wherein the control signal has a varying duty cycle or a varying period to produce the harmonic shaped leading edge. The charging system of claim 1 , wherein the value is at least one of an impedance or an admittance. a controller that obtains a value representative of a current flow to an electrochemical device, the value being associated with a harmonic content of the current flow;
a charge signal shaping circuit that defines a charge signal for the electrochemical device based on the harmonic components;
a power converter operatively coupled to the electrochemical device to provide power to a load;
Equipped with
The power converter, wherein the controller delivers control signals to the power converter to generate a harmonically regulated output of the power converter based on an optimal impedance value of the electrochemical device during discharge. A power delivery method comprising generating a discharge pulse from an electrochemical device, the discharge pulse having a harmonic-shaped edge defined based on harmonics associated with the output impedance of the electrochemical device.