Battery health monitoring for pulsed load power supplies

Abstract

Disclosed herein is a method of detecting a state of health of a battery in a power circuit having a charger and an energy storage device and for a pulsed load having a repeating interval, the method comprising: controlling a charging current of the energy storage device by controlling an operating frequency of the charger to recharge the energy storage device at an optimized charger operating frequency over the repeating interval; at least one of: comparing the optimized charger operating frequency to a threshold frequency, and comparing a rate of change of the optimized charger operating frequency to a threshold rate of change; and indicating an end of life state of the battery in response to at least one of: the optimized charger operating frequency exceeding the threshold frequency, and the optimized charger operating frequency exceeding the threshold rate of change. Other embodiments are disclosed.

Description

Technical Field

[0001] This disclosure relates to a power supply and method for providing energy pulses to a load having a pulsed load distribution. Background Technology

[0002] Various electrical applications require a pulsed energy supply. Non-limiting examples include industrial and medical pumps used to periodically provide rotation or translation of a part of a machine. One such application is medical dosing, used for the intermittent or periodic administration or delivery of fluids such as analgesics or insulin.

[0003] Increasingly, medical dosimetry devices, such as insulin pumps, are being developed to be lightweight and portable, or even placed inside or under the skin of a patient. The power supply for such devices is subject to an increasingly stringent set of constraints; for example, the device cannot be easily accessed inside the patient to replace the battery. Furthermore, space or volume constraints are likely to exist. The intermittent power requirements of such devices may be relatively high, for example, requiring the delivery of energy pulses in relatively short intervals, corresponding to relatively high power delivery levels, and such devices typically include an intermediate energy storage unit capable of providing energy pulses at high power levels and being slowly recharged or "trickle-charged" from energy or power sources that may not be able to provide such high power (e.g., batteries). Battery life is becoming increasingly important for such devices; therefore, including battery health monitoring in such resource-constrained devices can be helpful. Summary of the Invention

[0004] According to a first aspect of this disclosure, a power supply circuit is provided for a pulse load and includes: a battery; an energy storage device connected to an output of the power supply circuit and configured to provide periodic current pulses to the load at repetitive intervals; a charger connected at its input to the battery and at its output to the energy storage device, and configured to provide a charging current to the energy storage device; a charging controller configured to control the charging current by controlling the operating frequency of the charger and to determine an optimized charger operating frequency for recharging the energy storage device within the pulse intervals; and a health monitor configured to perform at least one of the following operations: comparing the optimized charger operating frequency with a threshold frequency, and indicating a battery end-of-life state in response to the optimized charger operating frequency exceeding the threshold frequency; and comparing a rate of change of the optimized charger operating frequency with a threshold rate of change, and indicating a battery end-of-life state in response to the rate of change of the optimized charger operating frequency exceeding the threshold rate of change. By using the operating frequency of the charger as an indicator of the battery's state, the power supply circuit can provide health monitoring at low cost without requiring additional circuitry or components.

[0005] In one or more embodiments, the energy storage device is a capacitor. A capacitor represents a particular, simple form of energy storage device; however, this disclosure is not limited thereto, and secondary batteries, electrochemical batteries, or other forms of energy storage devices may be used, for example.

[0006] In one or more embodiments, the charger is a DC-DC converter, and the charging controller is configured to operate the DC-DC converter in a pulse frequency modulation (PFM) mode. Typically, the PFM mode includes a constant on-time.

[0007] In one or more embodiments, the charging controller is configured to determine the optimized charger operating frequency at the start of each load pulse. This avoids the complexity caused by possible variations in the optimal charging current throughout the recharge interval.

[0008] In one or more embodiments, the power supply circuitry is used for an additional load that is less than one-quarter of the pulsed load during the repetition cycle. In a preferred embodiment, any constant load should be small relative to the pulsed load to avoid obscuring or masking the effect of the pulse on the PFM operating frequency, but this disclosure is not limited thereto.

[0009] In one or more embodiments where the health monitor compares the optimized charger operating frequency with a threshold frequency, the threshold frequency is a predetermined multiple of the lifetime-starting optimized operating frequency. The multiple need not be an integer multiple, but may be, for example, 1.5 times or 4 times the lifetime-starting optimized operating frequency.

[0010] In one or more embodiments where the health monitor compares the rate of change of the optimized charger operating frequency with a threshold rate of change, the threshold rate of change of the frequency is a predetermined multiple of the rate of change of the lifetime starting point of the optimized operating frequency. This multiple need not be an integer multiple and may, for example, be 10 times or 20 times the rate of change of the lifetime starting point. However, the value may vary depending on the application; in some applications, the value may be 100 times or even 1000 times the rate of change of the lifetime starting point. In other embodiments, the threshold rate of change may be an absolute level, rather than depending on the initial rate of change, particularly in applications where the initial rate of change may be very small.

[0011] In one or more embodiments, the battery is an alkaline primary battery, also known as an alkaline manganese oxide battery. However, this disclosure is not limited thereto, and in other embodiments, the battery may be another type of dry cell chemical energy storage system, such as, but not limited to, lithium-ion or nickel-cadmium (NiCad) batteries. Therefore, those skilled in the art will understand that the term "battery" will be interpreted broadly as used herein.

[0012] According to a second aspect of this disclosure, a method for detecting the health status of a battery in a power supply circuit having a charger and an energy storage device and for a pulse load having repetitive intervals is provided. The method includes: controlling the charging current of the energy storage device by controlling the operating frequency of the charger to recharge the energy storage device at an optimized charger operating frequency within the repetitive intervals; at least one of the following operations: comparing the optimized charger operating frequency with a threshold frequency, and comparing the rate of change of the optimized charger operating frequency with a threshold rate of change; and indicating a life end state of the battery in response to at least one of the following: the optimized charger operating frequency exceeds the threshold frequency, and the rate of change of the optimized charger operating frequency exceeds the threshold rate of change.

[0013] In one or more embodiments, the repetition interval is a first repetition interval, and at least one of the threshold frequency and the threshold change rate is a first threshold frequency and a first threshold change rate, respectively. The pulsed load further has a second repetition interval, for which at least one of the threshold frequency and the threshold change rate is a second threshold frequency and a second threshold change rate, respectively. In one or more embodiments, the second threshold change rate is equal to the first threshold change rate. In one or more embodiments, the energy storage device is a capacitor. In one or more embodiments, the charger is a DC-DC converter, and controlling the charger includes operating the DC-DC converter in Pulse Frequency Modulation (PFM) mode.

[0014] In one or more embodiments, the PFM mode includes a constant on-time. In one or more such embodiments, the DC-DC converter operates in a discontinuous conduction mode (DCM) with variable idle times. In one or more embodiments, the optimized charger operating frequency is determined at the start of each load pulse. In one or more embodiments, power supply circuitry is used for an additional constant load.

[0015] In one or more embodiments that include comparing the optimized charger operating frequency with a threshold frequency, the threshold frequency may be a predetermined multiple of the lifetime-start optimized operating frequency. In one or more other embodiments that include comparing the rate of change of the optimized charger operating frequency with a threshold rate of change, the threshold rate of change may be a predetermined multiple of the lifetime-start rate of change of the optimized operating frequency. Attached Figure Description

[0016] Please refer to the attached diagram, in which:

[0017] Figure 1 The circuit schematically illustrates a load that requires a pulsed energy supply.

[0018] Figure 2 A power supply circuit according to one or more embodiments of the present disclosure is shown;

[0019] Figure 3 The auxiliary control loop is conceptually illustrated;

[0020] Figure 4 The effect of control is shown when the repetition rate of the load pulse decreases;

[0021] Figure 5 A schematic graph showing the battery's power supply voltage and the charger's PFM frequency over time is shown.

[0022] Figure 6 Schematic graphs of the battery's power supply voltage and the charger's PFM frequency over time are shown under two different operating conditions.

[0023] Figure 7 The effect of the change in charging current across the charging interval is shown for two load pulses with different periods;

[0024] Figure 8 Various signals are shown, illustrating the operation of the power supply circuit during transitions between load conditions;

[0025] Figure 9 The values ​​or magnitudes of the change in charging current between consecutive load pulses relative to the repetition rate were plotted for various control methods; and

[0026] Figure 10 This illustrates the relationship between the periodic decrease of the load pulse and... Figure 4 The signals in the diagram are similar to various other signals.

[0027] It should be noted that the figures are illustrative and not drawn to scale. For clarity and convenience in the figures, the relative dimensions and scales of the parts have been shown in enlarged or reduced form. The same reference numerals are generally used to refer to corresponding or similar features in modified and different embodiments. Detailed Implementation

[0028] Figure 1 A circuit 100 including a load 110 that requires a pulsed energy supply is schematically shown. As shown, the load requires a pulsed supply of energy (including periodic or intermittent energy) 115. The load is an example of a heavy duty cycle system that achieves very low power consumption on average, for example by remaining in an "off" or "sleep" mode for a long time until an event or time signal wakes them up, allowing them to perform operations or activities that require energy pulses for a short period of time and then return to sleep.

[0029] Energy is supplied from a power source (in this case, battery 120) to the load. In the context of small or weak energy or power sources (e.g., small batteries or small energy harvesters with limited output power), it is often necessary to have an energy storage unit, such as an energy storage device or unit 130, in the system. This energy storage unit can accumulate the low power supplied by the energy or power source during long sleep periods and supply the peak power required by the load to perform its activities. As shown, the energy storage unit 130 may be a capacitor or a set of capacitors. A common approach is to charge the capacitors slowly, either directly from the energy or power source or battery 120 or via a charging circuit 140, such as a DC-DC converter, as shown at 135. The details of the circuit may depend on the size of the capacitor, the required stability of the voltage supply to the load, and the required energy efficiency of the entire process. A voltage regulator 150, or other components that individually control the energy from the energy storage device or unit 130 to the load 110, may be provided.

[0030] Figure 2 A power supply circuit for providing optimized charging current to energy storage device 130 is shown. A power source or battery 120 supplies current at voltage vbat to a charger or charging circuit 140, which provides periodic charging pulses to energy storage device 130. Various charging devices can be applied: typically, the charging circuit 140 is a DC-DC converter, such as an inductor-based DC-DC converter or a switched capacitor converter. When enabled, the charging circuit provides charging current to the capacitor or energy storage device or unit 130 at a variable voltage vsto. The charging circuit 140 is enabled or disabled based on the state of the output en of comparator 260, which compares the voltage vsto with a reference voltage vref. Therefore, the enable / disable can be described as an internal or main control loop 290. The power supply circuit 200 includes a feedback mechanism 270, as shown, which can be routed from a load 210 or the rest of the system to a controller 250. The feedback mechanism 270 indicates the repetition rate of the pulses of the pulsed load. The controller 250 controls the charging circuit, specifically the charging current Ich supplied by the charging circuit, based on the repetition rate of the pulse load. The controller is configured to change the charging current in response to variations in the repetition rate of the pulse load. Therefore, the power supply circuit implements an auxiliary or external control loop 295 that modifies the charging current according to the system state. Specifically, the controller 250 can adjust the charging circuit 140 to supply a higher charging current in response to an increase in the repetition rate of the load pulses, and a lower charging current in response to a decrease in the repetition rate of the load pulses. According to this disclosure, the controller 250 includes a health monitor 280 configured to identify the end-of-life state of the battery 120, as will be described in more detail below.

[0031] Figure 3 Auxiliary control loop 295 is conceptually shown. Controller 250 provides one or more control signals uich to charging circuit 140, such that charging circuit 140 provides charging current within interval tch when the load draws energy pulses. Depending on the application, and particularly the nature of charging circuit 140, the one or more control signals uich can take various forms. For example, uich can take the form of timing control signals for the high-side and low-side FETs of a half-bridge DC-DC converter. Ideally, and to minimize the charging current, the current should be specified so that the energy storage device or unit 130 is fully replenished just in time to provide subsequent energy pulses drawn by the load. That is, for the load pulse repetition rate tp, in steady state, tp = tch. Any error te between tp and tch is provided back to controller 250 to adjust one or more control signals uich to charging circuit 140. The control loop responds to changes in the average current consumption of the load, particularly changes in the affected repetition rate tp, or changes in energy delivery per pulse, which can be considered as a disturbance Qp to the charging circuit in this control loop.

[0032] Figure 4 The effect of control is shown when the repetition rate of the load pulses decreases. A series of intermittent pulses of the load are shown at 410. In the left part of the figure, at 412, it is shown that the power supply circuit is in steady state and the pulses occur periodically with a frequency of tp0, that is, the pulses have a repetition rate of 1 / tp0. After several pulses, the repetition rate decreases to 1 / tp1, that is, the pulses occur periodically with a frequency of tp1.

[0033] The voltage vsto at the energy storage device or unit 130 is shown at 420. During each load pulse, energy is discharged from the storage unit, and in each case, as long as the energy Qp transferred in each pulse is constant, the voltage of the storage unit drops by the same voltage drop to below the full charge value vref. Since the voltage vsto is less than the reference voltage vref, the charging circuit is enabled, causing the voltage vsto to increase between load pulses, just reaching vref when the next load pulse arrives. Curve 430 depicts the charging enable signal en, which is therefore kept on or high throughout this period; thus, the charging circuit 140 is permanently turned on during this period. The charging circuit provides a charging current Ich0 to the energy storage device or unit 130, as shown by curve 440.

[0034] As shown at 414, the pulse repetition rate can be reduced to 1 / tp1, meaning the pulse occurs periodically at tp1. Since tp1 is greater than tp0 and tch0 equals tp0, the voltage vsto returns to the reference voltage level vref before the next load pulse. Therefore, the charging enable signal en goes low, and the charging circuit 140 is disabled. The voltage vsto at the energy storage device or unit output remains stable as the voltage reference until the next pulse of the pulsed load occurs at the end of the time period tp1, as shown at 416. During this load pulse, the voltage vsto drops. The time te during which the enable signal en is low can then be used as a control parameter to determine an appropriate charging rate at which the output charging current Ich1 from the charging circuit 140 is controlled, such that the voltage vsto will fully or appropriately reach the target level, i.e., the charging level vref, at the end of the subsequent time period tp1. Assuming the control was previously in a steady state (i.e., Ich0 = tp0), the charging current can be calculated as:

[0035]

[0036] It should be understood that, based on the previously observed repetition rate, the charging rate is adjusted or changed to match the estimated repetition rate of the pulsed load. For this reason, and for any subsequent decrease in the repetition rate (corresponding to a further increase in the time between load pulses), the voltage at the output of the energy storage device will always reach the reference voltage before the next pulse. Of course, in the case of an increased repetition rate, the voltage will not reach the reference voltage before the next load pulse. The following will refer to... Figure 9 Consider this in more detail. However, for the purposes of this disclosure, those skilled in the art will understand that, under steady-state conditions, the charging current (e.g., Ich0 or Ich1) is constant. The inventors of this invention have recognized this fact and observed that, for a charger or charging circuit 140 that supplies charging current by means of pulse frequency modulation (PFM) charging pulses from, for example, a DC-DC converter, the PFM frequency can indicate the health status of the battery 120 under steady-state pulse loads: in addition to the load repetition rate, it is inversely proportional to the input or battery voltage.

[0037] Reference Figure 5 This will be described. Figure 5 A schematic graph showing the power supply voltage of battery 120 and the PFM frequency of charging circuit or charger 140 over time 510. It should be noted that for... Figure 5 (and for the following) Figure 6 ), time scale ratio Figure 4 The time scale shown is much longer. Specifically, Figure 5 It can depict the entire lifespan of the battery. That is to say, although... Figure 4The time scale in the data can be on the order of milliseconds or microseconds, but Figure 5 The time scale can be on the order of months or years.

[0038] The diagram illustrates the monotonic decrease of the supply voltage 520 of battery 120 over time. Those skilled in the art will understand that the nature of this decrease depends on the specific type and construction of the battery. Typically, the decrease is caused by a combination of increased output impedance and the battery itself becoming more depleted (i.e., its energy storage capacity decreasing). Specifically, for some battery types, the decrease may begin at a high rate, plateau at a lower rate, and then begin to decrease rapidly near the end of the battery's life. For other battery types, the decrease may be relatively linear, as shown, and may exhibit different curves.

[0039] Figure 5 The PFM frequency of the charger or charging circuit 140 is also shown at 530. A PFM frequency is required to supply a constant charging current at any time throughout the battery's life as the battery output voltage decreases (which may decrease linearly as shown at 520). The PFM frequency increases as the supply voltage from the battery decreases. Those skilled in the art will understand that PFM typically corresponds to a discontinuous conduction mode (DCM), where charge pulses are supplied periodically at a pulse frequency. The nature of the pulses may vary depending on the specific operating mode (e.g., peak current control or turn-on time control), but regardless of the specific operating mode, the pulse frequency increases as Vbat decreases, where Vbat is the input supply voltage (in this case, from battery 120). The amplitude of the PFM frequency and its rate of change over time both increase as the battery voltage begins to approach the voltage of the energy storage device. Figure 5 As shown in the diagram. The inventors of this invention have recognized that, at a constant (i.e., fixed) pulsed load, this rate of frequency change can be used to indicate the health status of a battery without requiring a complex circuit system. Specifically, in one or more embodiments, a threshold frequency (in) can be predetermined. Figure 5 (As shown at 540). When the charger's operating frequency exceeds the threshold frequency 540, the controller can provide an end-of-life status indication. Considering that frequency changes are typically non-linear, in other embodiments, the rate of change of the PFM frequency can be used to provide an end-of-life status indication. Such an embodiment... Figure 5 The gradient or rate of change of the PFM frequency is shown at 550.

[0040] Those skilled in the art will understand that, in order to Figure 5 Correct. The pulse load, specifically the product of the energy Qp per pulse and the repetition rate, needs to be constant. Otherwise, the PFM operating frequency 530 will not show a monotonically increasing trend.

[0041] Now turning Figure 6 This shows the same graphs of the power supply voltage of battery 120 and the PFM frequency of charger or charging circuit 140 over time 510 for two different pulse load conditions. The different pulse load conditions may be due to different applications or usage of the power supply; conversely, they may be due to different operating conditions of the same application, for example, due to two different pulse amplitudes (i.e., energy per pulse Qp) of the pulse load, due to two different repetition rates of pulses with the same energy per pulse Qp, or due to a combination of both differences. The drop in battery power supply voltage 520 is related to... Figure 5 Same as shown; the variation of the first PFM operating frequency 530 is the same as... Figure 5 The same as shown, and corresponding to the Figure 5 The same pulse load conditions are shown.

[0042] However, this diagram also shows the PFM operating frequency F2 corresponding to the second pulse load condition. PFM The second variation of 532 over time. In the example shown, the pulses of the pulsed load are more frequent, or have greater amplitude, or both; in any case, the combination of frequency and amplitude results in an increase in the average load drawn from the energy storage device. Therefore, the PFM frequency F2 PFM Usually higher Figure 5 The first set of operating conditions is shown. Due to F2 PFM Usually higher than F PFM Therefore, it has a longer battery life than F. PFM The condition reaches and exceeds threshold 540 earlier. A separate threshold 542 can be predetermined for these conditions to correspond to the same end-of-life state of the battery. Conversely, as... Figure 6 As shown, the rate of change of the PFM frequency can be the same regardless of the load amplitude (and therefore the amplitude of the PFM frequency at any particular stage of battery life). Therefore, in some embodiments, a gradient-dependent threshold (i.e., comparing the rate of change of the PFM frequency to a threshold rate of change) may be appropriate. This is in Figure 6 As shown in the figure, it can be seen that the PFM frequency F2 under the second operating load condition is [value missing] at any specific stage of the battery's lifespan. PFM The gradient 552 corresponds to the PFM frequency F under the first operating load condition. PFM The gradient is 550. Therefore, gradient 552 is the same as the step 550 shown in the lifetime end state. Therefore, in this case, a single threshold can be used regardless of the load conditions of the pulse load.

[0043] Those skilled in the art will understand that, although in Figure 6Two different pulsed load operating conditions are shown, but the principle can be extended to other load operating conditions, as well as thresholds for any one or both of the amplitude and rate of change of the PFM frequency.

[0044] The thresholds can be predetermined and stored in the health monitor 280 in any suitable form. For example, but not limited to, in a digital implementation, thresholds for several known operating conditions or operating modes of the same application can be provided as a lookup table.

[0045] Figure 7 The diagram illustrates the effect of variations in the output voltage vout when the energy storage device is charged between load pulses. Specifically, the diagram shows load pulses for two operating modes at 710, with the first mode on the left and right sides of the diagram having a repetition rate tp0, and the second node in the center of the diagram having a repetition rate tp1. The charger's output voltage vout is shown at 720. This voltage drops during the load pulse and then rises back to the voltage of a fully charged and ready storage device during the interval between pulses, while the energy storage device is recharged. In other words, the output vout equals vsto. The charger's operating frequency is shown at 740. As mentioned above, the frequency is different for different operating modes or operating conditions of the pulse load. The corresponding charging current required by the charger is shown at 730. Although a constant current can be supplied as shown at 732, in practice, the charger's output may drop during the charging interval due to variations in the output voltage shown at 720, variations in the input battery voltage, and the charger's control operations during this period, as shown at 734. However, even if the charging current does vary within the charging interval, the PFM charging frequency can be determined only once for each charging interval at the beginning of the interval. If this measurement is taken at the beginning of a charging interval, it will continue to increase throughout the battery's lifespan, as mentioned above. Figure 5 and Figure 6 As described.

[0046] As described above, the method relies on the load being constant throughout the application's lifespan, or at least constant for a sufficient period of time, allowing for the setting and implementation of appropriate thresholds for end-of-life comparisons. In the latter case, the load may change, and the control mechanism then needs to adjust the charging current Ich to match the new load. References will now be made to... Figure 8 and Figure 9 This describes one way of implementing this.

[0047] Figure 8 Various signals are illustrated, demonstrating the operation of a power supply circuit according to one or more embodiments of this disclosure. Specifically, the diagrams illustrate operation during the periodic increase and decrease of the load pulse tp. Similar to... Figure 5A series of intermittent pulses of the load are shown at 810. The voltage vsto at the energy storage device or unit 130 is shown at 820, and curve 830 depicts the charging enable signal EN. For reference purposes, the variation of vsto over time according to a conventional system is shown at 822, and the charging enable signal EN of a conventional system is shown at 832. The charging circuit providing the charging current Ich to the energy storage device or unit 130 is shown at 840.

[0048] The first two time intervals shown at 812 between t0 and t1 and at 814 between t1 and t2 respectively correspond to Figure 5 The time period shown is [not specified]. Compared to 822a in a conventional system, the shallower recharge gradient 824 according to the embodiments of this disclosure results in a slower recharge rate, i.e., a lower charging current, which is due to the reduced repetition rate within time period 814 relative to 812. It should be noted that the relatively lower battery current 844 drawn during this time period may be beneficial compared to the charging current from a conventional system as in 842a.

[0049] An increase in the repetition rate of the load pulses (relative to the repetition rate in time period 814) is shown at 816. That is, the interval between load pulses at t2 and t3 (or t3 and t5) is shorter than the interval between load pulses during time period 814. For a conventional power source, as shown at 822b, the energy storage device or unit 130 recharges at the same rate as after any other pulse. And therefore, the current 842b from the battery or other power source 120 is the same as in the previous interval, as shown at 842a, for example. However, the controller sets the recharge current 826 based on the previous periodicity tp1 (which is longer than the current interval tp2). Therefore, at the end of the interval, when the next load pulse occurs, the energy storage device or unit 130 has not yet been fully recharged to the reference voltage vref. Therefore, the next energy pulse drawn from the energy storage device or unit 130 reduces the voltage vsto to below the previous minimum voltage, as shown at 826b. To compensate for this, the controller modifies uich to increase the charging current, as shown at 846. It should be understood that in this situation, because the energy storage device or cell 130 is not fully recharged during the final interval between load pulses, the increase in charging current must not merely be to make the current proportional to the repetition rate in order to fully recharge the energy storage device or cell 130. Therefore, the voltage vsto increases back to vref more quickly, and once the voltage reaches vref, comparator 260 turns off the enable signal en to stop charging, as shown at t4. Those skilled in the art will understand that during this interval, the battery must supply a higher power level than in the case of a fixed battery current.

[0050] Once the next load pulse occurs, the controller can set the charging current to be proportional to the previous repetition rate (that is, inversely proportional to the time interval between t3 and t5) since the energy storage device or unit 130 is fully recharged. Therefore, as shown at 848, this current level is lower than level 846 due to the need to recover lost energy at t3; however, it is higher than current level 844 because the new repetition rate 816 is higher than the repetition rate in time period 814.

[0051] It should be understood that various alternative implementation schemes can be implemented based on the above adaptation concept. First, consider the case where the repetition rate of the load pulse has decreased relative to the previous repetition rate (which is typically, but not necessarily, steady-state) to such that tch < tp. Without limitations, at least three alternative methods can be used to adjust the recharge current. References will now be made. Figure 9 The right-hand side discusses various alternatives. Figure 9 A graph is plotted showing the value or magnitude of the change in charging current ΔI (i.e., ΔI = Ich1 - Ich0) between consecutive load pulses relative to the change in repetition rate te for various control methods according to embodiments of the present disclosure.

[0052] The first method “(a)” shown at 910 is as described above: the auxiliary loop can be configured to use, for example Figure 2 The external clock clk shown is used, or an internal time reference is used to measure the time between the load pulse (tp) and the charging time (tch). As mentioned above, the time difference te can be calculated because te = tp - tch, and the charging current is modified proportionally to the difference, as indicated by equation (1):

[0053] .

[0054] Or equivalently:

[0055] .

[0056] In the second method "(b)" shown at 920, the charging current can be reduced by a step amount in response to te being higher or lower than a predetermined threshold level. Therefore, for example, if the value of te is within te... x and te y Between these points, the charging current can change by a first amount ΔI. x If the value of te is in te y and te z Between these points, the charging current can change by a second quantity ΔI. y And so on.

[0057] In the third method "(c)" shown at 930, the charging current can be simply reduced by a fixed amount based on the sign of te (in other words, whether te is negative). This method is particularly simple and easy to implement because it does not require remeasuring any time amplitude, but only requires observing whether the charging process is complete before triggering the next load pulse. However, the solution may result in "bounces," or cyclic oscillations limited to around the optimal value, which can be detrimental in some applications and may lead to longer settling times for the auxiliary loop compared to other embodiments, resulting in a longer suboptimal charging interval for the energy storage device or unit 130.

[0058] The repetition rate of the load pulse is now increased relative to the previous repetition rate (which can typically, but is not necessarily, in a steady state) to make tch > tp. This change is slightly more complex because the optimal charging rate for the previous repetition rate cannot fully charge the energy storage device or cell 130. Therefore, since the charging process is not complete, the charging time tch required to fully recharge the energy storage device is unknown. Similarly, various methods can be implemented according to this disclosure, some of which will refer to... Figure 9 left side and Figure 10 To describe in more detail.

[0059] Figure 10 This illustrates the effect caused by the periodic decrease in tp of the load pulse. Figure 6 The signals are similar to those in the diagram. A series of intermittent pulses of the load are shown at 1010, initially periodic with tp0 within time period 1012, then changing to a decreasing periodicity of tp1, corresponding to an increased repetition rate. The voltage vsto at the energy storage device or unit 130 is shown at 1020, and a charging circuit providing a charging current of Ich to the energy storage device or unit 130 is shown at 1040.

[0060] During the first time interval, the charging current is optimized for the periodic tp0 of the load pulse, such that the voltage vsto on the energy storage device or unit 130 reaches the reference value vref just when the load pulse 1002 arrives. The controller continues to recharge the energy storage device or unit 130 at the same rate during the next interval; however, since this interval tp1 until the next load pulse 1004 is shorter than the previous interval, vsto does not fully recover to vref, but is lower by a value ΔV, as shown. Various alternative control methods according to one or more embodiments can now be implemented. In one method (d) shown at 940, the load pulse is detected while the charging circuit is still enabled. Maximum charging current Ich maxis selected by the controller such that the voltage vsto across the energy storage device or unit 130 increases at the fastest rate to ensure that vsto reaches the reference level vref before the next subsequent load pulse. As Figure 9 shows, the increase in charging current is Ich max - Ich0. And as Figure 10 shows, current 1042a causes a relatively rapid rise of vsto, as shown at 1022a, such that it reaches the reference voltage vref at time t1, where t1<t3 (t3 is the start of the next subsequent load pulse 1006). For the subsequent pulse 1008, the period tp1 is now known, so assuming subsequent pulses occur at steady-state periodicity, the current charging level 1042b can be set such that vsto recovers to vref exactly at time t5 when the next load pulse 1008 occurs. To achieve this, the maximum charging current Ich max should be significantly larger than the maximum value required by the load at the highest expected repetition rate. Therefore, during the period from t0 to t1, a maximum or relatively high current is drawn from the power source or battery 120. However, this will only last for a single period, so any associated degradation of the battery can be expected to be relatively minor.

[0061] in Figure 9 an alternative method "(e)" is shown at 950. In this case, if the energy storage device or unit 130 is not fully charged before the load is triggered, the charging current is simply increased by an amount Ich each cycle k , wherein said amount Ich k is fixed (as shown) or depends linearly or non-linearly on ΔV. In the case of fixed increments, several load pulse intervals may be required to fully recharge the energy storage device or unit 130, and at the start of each interval, the charging current increases from as shown. This may result in a longer stabilization time to reach the optimal charging current value, but will generally avoid the need to use the maximum charging current.

[0062] in Figure 9 another alternative method "(f)" is shown at 960. In these cases, the optimal charging current Ich to reach vref at the end of time interval tp1 can be determined by using additional information opt , as Figure 10 shown at 1044a. For example, in the first case, if the capacitance value Csto of the energy storage device or unit 130 and the voltage difference (ΔV) between the target recharge level vref and the actual vsto when a new load event is triggered are known, Ich1 can be calculated according to the following formula:

[0063] .

[0064] It should be understood that this method requires a relatively accurate ΔV measurement, and Csto needs to be known to or be computable by the controller. The latter can be achieved, for example, through or possibly using an indirect measurement of the slope based on vsto.

[0065] In the second case, the new optimal charging current Ich1 can be calculated by referring to the previous charging current Ich0 using the following formula.

[0066] .

[0067] It should be understood that this method requires storing the previous interval tp0. In contrast, the optimal charging current can result in the shortest settling time, which has potential advantages.

[0068] Those skilled in the art will understand that the second of these two cases is very similar to Figure 4 The periodic increase between load pulses shown is relative to or opposite to case "(a)". However, when tch < tp (case "(a)"), both can be said to have a time to end the charging process, while in the case of tch > tp, the charging time is unknown because the charging process is truncated before it is completed by the load pulse (or at least its measurement is therefore severely distorted). Therefore, in "(f)", tp0 (the time between two previous load events) is used when tch is not known, although this is based on the assumption that the loop is well stable (tp0 equals or is very close to tch0), which may not always be the case.

[0069] In yet another example, “(g)” 970, by measuring tp1 and storing tp0, Ich0 can be increased based on the difference tp1-tp0 relative to different threshold locations using different constant values. Similarly, as in Figure 9 In other left-handed methods, te is actually tp1-tp0, because the time tch for fully recharging the energy storage device is unknown.

[0070] As already mentioned and as will be familiar to those skilled in the art of process control loops, it may be appropriate to increase the charging current slightly above the optimum value and / or to apply hysteresis around the point te=0, in order to ensure system stability without rebound oscillations. Furthermore, again as mentioned above, in the case where tp<tch, the energy storage device or unit 130 may not be completely recharged between load pulses, and therefore the worst-case drop in vsto may be greater than that in conventional systems. This should be taken into account when designing the overall system: the energy storage device or unit 130 should be capable of providing more than one load pulse without requiring recharging. It should also be further understood that for embodiments where the controller and hence the auxiliary loop can be pre-notified of expected changes in the repetition rate of load pulses, the controller can use this information to reduce transients in this auxiliary loop.

[0071] Although the present disclosure has been illustrated by using rectangular pulses for pulsed loads, those skilled in the art will understand that the present disclosure is not limited thereto. For example, without limitation, load pulses may be triangular, trapezoidal, or have a more complex time profile.

[0072] The schematic illustrations of embodiments described herein are intended to provide a general understanding of the structure of various embodiments, and are not intended to serve as a comprehensive description of all elements and features of the apparatus and systems that may utilize the structures described herein. Many other embodiments will be apparent to those skilled in the art upon review of the above description. Other embodiments may be utilized and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of the present disclosure. The drawings are merely representative and may not be drawn to scale. Certain proportions of the drawings may be exaggerated, while other proportions may be minimized. Accordingly, the specification and drawings are to be regarded as illustrative rather than restrictive.

[0073] Although specific embodiments have been illustrated and described herein, it should be understood that any arrangement that is intended or adapted to achieve the same or similar purpose may be substituted for the specific embodiments shown. The present disclosure is intended to cover any and all adaptations or variations of the various embodiments. The present disclosure expressly contemplates combinations of the above-described embodiments and other embodiments not specifically described herein.

[0074] For example, one or more features or aspects from one or more embodiments may be combined with one or more features or aspects from one or more other embodiments. In one or more embodiments, a positively referenced feature may also be negatively referenced and excluded from the embodiments, whether or not it is replaced by another structural and / or functional feature. The steps or functions described with respect to embodiments of this disclosure may be performed in any order. The steps or functions described with respect to embodiments of this disclosure may be performed individually, in combination with other steps or functions of this disclosure, or according to other embodiments or according to other steps not described in this disclosure. Additionally, more or fewer of all the features described with respect to embodiments may be used.

[0075] Fewer than all the steps or functions described with respect to the exemplary process or method may be performed in one or more exemplary embodiments. Furthermore, unless explicitly stated otherwise, the use of numerical terms such as first, second, third, etc., to describe devices, components, steps, or functions is not intended to describe a sequence or function. Unless explicitly stated otherwise, the use of the terms first, second, third, etc., is generally for distinguishing devices, components, steps, or functions. Additionally, one or more devices or components described with respect to exemplary embodiments may facilitate one or more functions, wherein said facilitation (e.g., facilitating access or facilitating connection establishment) may include fewer than each step required to perform the function, or may include all the steps required to perform the function.

[0076] In providing this abstract, it should be understood that the abstract is not intended to interpret or limit the scope or meaning of the claims. Furthermore, in the foregoing detailed description, it can be seen that various features are grouped together in a single embodiment for the purpose of simplification. The approach of this disclosure should not be construed as reflecting an intention that the claimed embodiments require more features than are expressly recited in each claim. Rather, as reflected in the following claims, the subject matter of the invention lies in fewer than all features of a single disclosed embodiment. Therefore, the following claims are hereby incorporated into the embodiments, wherein each claim stands alone as a separately claimed subject matter.

Claims

1. A power supply circuit, characterized in that, For pulse load (210) and includes: Battery (120); An energy storage device (130) is connected to the output of the power supply circuit and is configured to provide periodic current pulses to the load at repeated intervals; A charger (140) is connected to the battery at its input and to the energy storage device at its output, and is configured to provide charging current to the energy storage device; A charging controller (250) is configured to control the charging current by controlling the operating frequency of the charger, and to determine an optimized charger operating frequency for recharging the energy storage device within the pulse interval; and A health monitor (280) is configured to perform at least one of the following operations: comparing the optimized charger operating frequency with a threshold frequency, and indicating the end-of-life state of the battery in response to the optimized charger operating frequency exceeding the threshold frequency; and comparing the rate of change of the optimized charger operating frequency with a threshold rate of change, and indicating the end-of-life state of the battery in response to the rate of change of the optimized charger operating frequency exceeding the threshold rate of change.

2. The power supply circuit according to claim 1, characterized in that, The energy storage device is a capacitor.

3. The power supply circuit according to claim 1 or 2, characterized in that, The charger is a DC-DC converter, and the charging controller is configured to operate the DC-DC converter in Pulse Frequency Modulation (PFM) mode.

4. The power supply circuit according to claim 3, characterized in that, The PFM mode includes a constant on-time.

5. The power supply circuit according to any one of the preceding claims, characterized in that, The charging controller is configured to determine the optimized charger operating frequency at the start of each load pulse.

6. The power supply circuit according to any one of the preceding claims, characterized in that, The power supply circuit is used for an additional load, which is less than one-quarter of the pulse load during the repetition cycle.

7. The power supply circuit according to any one of the preceding claims, characterized in that, The health monitor compares the optimized charger operating frequency with a threshold frequency, wherein the threshold frequency is a predetermined multiple of the optimized operating frequency at the start of the lifespan.

8. The power supply circuit according to any one of the preceding claims, characterized in that, The health monitor compares the rate of change of the optimized charger operating frequency with a threshold rate of change, wherein the rate of change of the threshold frequency is a predetermined multiple of the rate of change of the lifetime starting point of the optimized charger operating frequency.

9. A method for detecting the health status of a battery in a power supply circuit, characterized in that, The power supply circuit has a charger and an energy storage device and is used for a pulse load with repetitive intervals; the method includes: The charging current of the energy storage device is controlled by controlling the operating frequency of the charger, so as to recharge the energy storage device at an optimized charger operating frequency within the repeating interval. At least one of the following operations: comparing the optimized charger operating frequency with a threshold frequency, and comparing the rate of change of the optimized charger operating frequency with a threshold rate of change; and The battery's end-of-life state is indicated in response to at least one of the following: the optimized charger operating frequency exceeds the threshold frequency, and the rate of change of the optimized charger operating frequency exceeds the threshold rate of change.

10. The method according to claim 9, characterized in that, The repetition interval is a first repetition interval, and at least one of the threshold frequency and the threshold change rate is a first threshold frequency and a first threshold change rate, respectively. The pulse load also has a second repetition interval, and for the second repetition interval, at least one of the threshold frequency and the threshold change rate is a second threshold frequency and a second threshold change rate, respectively.

Images