GB2531302A

GB2531302A - A method and apparatus for charging a battery

Info

Publication number: GB2531302A
Application number: GB1418292.7A
Authority: GB
Inventors: Michael Mcdonagh
Original assignee: ECOTECH ENERGY SOLUTIONS Ltd
Current assignee: ECOTECH ENERGY SOLUTIONS Ltd
Priority date: 2014-10-15
Filing date: 2014-10-15
Publication date: 2016-04-20
Also published as: GB201418292D0

Abstract

A charging protocol is applied to a battery 15 in dependence on the battery response to an applied test current pulse. The test current pulse may be generated by pulse circuitry 15 using a high frequency switch 21 and battery voltage monitored by a feedback loop 22 to a controller 18 that determines a magnitude, shape, duration and number of charging pulses in a pulse array, the charging pulses being separate and distinct from the test pulse. Charging protocols may be selected from predetermined protocols stored in a memory of controller 18 depending on the battery charge state, temperature and electrochemical response indicated by the rate of change of voltage decay in a waveform subsequent to applying the test pulse. A fault in a battery can be detected during charging and a signal representing a fault condition is output depending on a comparison between an off-charge reference voltage for a single cell multiplied by the number of cells and an actual of charge voltage of the entire battery determined in a non-charging condition. A short circuit may be detected by comparing an on-charge voltage of the battery with an on-charge reference voltage of a single cell, whilst charging.

Description

A method and apparatus for charging a battery This invention relates to a method and apparatus for charging a battery, in particular for lead acid batteries.

Batteries are used in almost all spheres of life from toothbrushes to standby power in power stations, with power requirements ranging from less than one watt to more than a gigawatt.

The lead acid battery is the oldest type of rechargeable battery wherein there is a reversible reaction between the electrode plates and the acid when a lead acid battery is charged and discharged.

In the fully-discharged state the lead acid battery has electrodes consisting of chemically similar lead sulphate plates, with a dilute sulphuric acid electrolyte there-between. It is common to charge a battery using a voltage current relationship. Under normal constant current charging, lead dioxide is produced at the anode and the sulphate ions from the lead sulphate plates are rejected to react with the water in the electrolyte solution to form sulphuric acid.

A similar reaction occurs at the cathode except that the process consumes electrons to reduce the lead sulphate on the cathode to a lead metal.

In summary, in the charging process the anode generates sulphuric acid, hydrogen ions and electrons, whereas the cathode consumes the electrons to produce lead and sulphuric acid. During this reaction the sulphate ions are ejected into the bulk electrolyte solution by the plates producing a Gouy-Chapman double layer at the plate surface. This provides a surface charge with a concentration gradient that acts as a barrier to sulphate ions, making it more difficult for them to leave the plates and diffuse into the bulk solution, thereby hindering the rate at which the battery can reach the fully-charged state. This consequently increases the internal resistance of the battery cell, thereby increasing the charging voltage required. This not only increases energy consumption, but also, due to use of a dilute acid, creates unwanted secondary reactions at the plates, whereby water may be broken down into hydrogen and oxygen ions which are released as gases via electrolysis. This gas production removes water from the electrolyte, introducing the requirement for inspection of the electrolyte level and replacement of water at regular intervals, increasing maintenance requirements and the likelihood of early battery failure. These unwanted exothermic reactions occur mainly above 2.4 volts per cell and are a major contribution to the heating experienced, which can also shorten battery life.

In particular lead acid batteries are used in motorised vehicles, for example a fork lift vehicle. In consideration of a fork lift vehicle, such a battery is usually used for a single shift which can last for 8 hrs. Recharging of the battery typically takes 8 to 12 hours using standard chargers with on charge voltages reaching 2.75 volts per cell, which is clearly above the 2.4 volts per cell limit for exothermic reactions causing the above-mentioned unwanted secondary effects.

This standard charging technique has been shown to provide coulombic inefficiencies of up to 20% and the energy efficiency is further reduced due to the higher differential between the charging voltage (2.SV per cell average) and the discharging voltage of around 1.85 volts per cell. Therefore, in total the recharging inefficiency has been found to be approximately 65%. Further, the down-time of fork lift trucks in order to enable charging of the battery is n problematic, with back-up vehicles or further batteries with associated battery changing equipment, being required at an extra cost.

Opportunity charging is a charging technique which attempts to minimise the downtime of, for example, fork lift trucks. The fork lift truck batteries are charged for brief periods during shifts without interruption of the working pattern e.g. during breaks and when the truck is usually idle waiting for a process to be finished, to enable the vehicle to continue working for 24 hours, 7 days a week.

However, the existing methods of opportunity charging and the chargers which are used do not efficiently charge batteries. The result is that very high on-charge voltages are used to maximise the current input to the battery. This in turn gives high levels of water loss and high operating temperatures. The net effect is to reduce the life expectancy of the batteries from four years to less than two. For this reason opportunity charging is not widely accepted in certain parts of the world, e.g. Europe, however, its use is increasing in the US.

Battery chargers with improved energy efficiency are known that utilise a conversion from ac mains to dc output via switch mode thyristor controlled systems. However, these still have an invariant current pattern which can cause overcharging and resultant degradation of the battery life.

To prevent overcharging of a battery cell and to improve on battery life it is known to apply a test pulse at intervals to determine whether the battery charging process is completed, thereby providing a charging progress monitor.

It is also known for rapid pulse charging methods to reduce charging times and reduce the coulombic input. However, such methods use a current pulse with a profile which in most instances is unlikely to be suitable for the charge state of the battery, since this introduces inefficiencies with higher consumption and may as a result also be detrimental to the lifetime of the battery.

The impedance and internal resistance of a battery will vary according to the state of charge of the battery, which is a consequence of the concentration of lead sulphate in the plates and the concentration of sulphuric acid in the electrolyte.

Embodiments of the present invention are derived from the realisation that there exists the need to provide a charging method and apparatus that delivers an optimum charging pattern to the batteries to which it is applied, based on the varying electrochemical conditions within the battery at all states of the charge, thereby providing a faster, more energy efficient and economic arrangement that minimises the need for battery maintenance and eliminates degradation of the battery due to overcharging. The method and apparatus will also minimise secondary reactions thereby improving the chemical efficiency of the charging process.

Therefore, the present invention in at least some of its embodiments, is intended to address at least some of the above described problems and desires.

According to a first aspect of the invention there is provided a method of charging a battery comprising: applying a charging step, comprising applying a test current pulse to a battery to be charged, and subsequently applying a charging protocol to the battery, wherein the charging protocol is dependent on the response of the battery to the previously applied test current pulse.

A plurality of charging steps may be applied to the battery in a period of time.

The charging protocol relates to the criteria for the next stage in the charging process-for example which current pulse is to be selected, should all the cells be selected, or just a select few cells and whether the frequency of the pulse should be changed. Importantly, the actual electrochemical state of the battery is considered prior to applying charging protocol.

The charging protocol may be a charging pulse or charging pulse array may have a predetermined charging pattern.

For example a charging algorithm will determine the charging protocol.

The test pulse may be separate and distinct from the charging pulse or charging pulse array.

This ensures that the information received regarding the state of charge of the battery is accurate.

The charging pulse may have a minimum frequency of at least 1 Hz.

The charging pulse may have a frequency of between 1 Hz to 50,000 Hz.

The charging pulse may be a current pulse.

The method may further comprise: monitoring the charge state of the battery subsequent to applying the test pulse and prior to determining the charging protocol so as to provide information on the state of charge of the battery.

This enables a suitable charging protocol to be selected that will optimise the characteristics of the battery to be charged, thereby prolonging the efficiency and life-time of the battery. This can have beneficial economic and energy effects to a user.

The information may relate to a voltage response waveform of the battery in response to the previous test current pulse.

This enables the electrochemical characteristics of the battery to be determined.

The information may relate to the electrochemical response of the battery at a plate acid interface located within the battery.

This again provides an accurate measure of the battery which is independent of the battery design. This makes the charger extremely versatile and optimises the charging pattern for a given battery, where-by different batteries will require different optim ised charging patterns.

The information may relate to the rate of change of voltage with respect to time on the waveform provided by an electrochemical reaction within the battery to be charged.

This is considered to be the only true measure of the state of charge of the battery which is independent of battery design.

The information may relate to the part of the waveform attributed to the voltage decay due to diffusion of sulphate ions from a plate surface of the battery into the bulk electrolyte.

The information may be provided by a temperature sensor located in thermal communication with at least one cell of the battery to be charged.

The selected charging protocol may comprise a charging pulse or pulse array that has been predetermined based upon a given charge state of a battery. The charging pulse is a current pulse and the charging pulse array is a current pulse array.

The charge state is the actual electrochemical condition of the battery and not an estimation of the percentage state of charge.

The selected charging protocol may comprise a current pulse or pulse array that has been predetermined based upon the temperature of the battery.

The method may further comprise: terminating the charging protocol; applying a further test current pulse to the battery to be charged; monitoring the charge state of the battery in response to the further test current pulse; selecting a further charging protocol from a series of predetermined charging protocols, depending on the monitored charging state of the battery to be charged; and applying the selected charging protocol to the battery to be charged.

This ensures that the electrochemical state of the battery is considered prior to applying a charging process to the battery so that the overall charging pattern of the battery may be optimised.

The monitoring of the charge state of the battery may occur during a time interval located between a first current charging protocol and successive current charging protocols.

Therefore the test pulse is applied in a charging rest period so that the electrochemical state of the battery may be determined off-charge.

The first and subsequent current charging protocols have different pulse characteristics.

This ensures that on a change of electrochemical charging state as a consequence of a charging pulse, a different more suitable charging pulse can be determined as being appropriate and subsequently supplied to the battery.

The charging pulse characteristics comprise at least one of duration, amplitude, shape or number of pulses in a fixed time period.

These characteristics will be adjusted depending on the voltage waveform information provided subsequent to the application of a test pulse.

In a further embodiment of the invention there is provided an apparatus for charging a battery, comprising: a first current pulse circuitry for generating a test current pulse; a conductor for transferring the current pulse from the current pulse generator to the battery to be charged; a second charge current state circuitry for generating a charging current pulse; a charge state monitor for monitoring and receiving information on the charge state of the battery to be charged in response to the test current pulse applied; and a controller for selecting the charging protocol from a series of predetermined charging protocols dependent on the charging state of the battery to be charged and for communicating with the second charge state circuitry for generating the charging protocol.

The first and second circuits are different such that the test pulse is provided in the rest state of the charging pulse.

The apparatus may further comprise a microprocessor for selecting a charge protocol from a charging protocol library in dependence on the charging state of the battery.

The charging protocol is determined by a charging algorithm that is stored in the library.

The apparatus may further comprise a charge pulse generator for generating a current pulse having a frequency of at least 1 Hz to 50,000 Hz.

The charge pulse generator may comprise a mains DC power supply and a high frequency switch.

The controller may be communicatively coupled to the charge pulse generator to initiate a current dependent on the selected charging protocol.

The charge state monitor may comprise a temperature sensor.

In an alternative aspect to the invention there is provided a charging station comprising at least one charging bay configured to charge a battery according to the above-mentioned method.

The charging station may comprise the above-mentioned apparatus.

The charging station may comprise a first charging circuit capable of generating a test pulse for determining the charge state of a battery to be charged and a second charging circuit capable of generating the charging pulse and/or an array of charging pulses in response to the determined state of charge of the battery.

The state of charge of the battery refers to the electrochemical state of the battery and not the percentage charge of the battery at a given time.

In a yet further aspect of the invention there is provided an electrically powered vehicle incorporating the above-mentioned apparatus.

This removes the need of a separate charging station.

In a further embodiment of the invention there is provided a method of detecting a fault in a battery during charging, the battery having at least one cell, the method comprising: determining an off-charge reference voltage of a single cell of the battery during a non-charging condition; multiplying the determined off-charge reference voltage by the number of cells in the battery to obtain an off-charge effective voltage for the entire battery; determining an off-charge actual voltage across the entire battery during a non-charging condition; comparing the off-charge effective voltage of the entire battery with the off-charge actual voltage of the entire battery, and outputting a signal representative of a fault condition, the signal being dependent upon the comparison between the off-charge effective battery voltage and the off-charge actual battery voltage.

The signal may be generated when the value of the off-charge effective voltage and the off-charge actual voltage differ by a predetermined threshold.

The signal may be an audio, visual or feedback signal to a processor or a controller.

The method may further comprise determining an on-charge reference voltage of a single cell of a battery during an on-charge condition; determining the on-charge actual voltage of the entire battery during an on-charge condition; comparing the on-charge actual voltage of the battery during an on-charge condition with the on-charge actual voltage of the battery during an on-charge condition; outputting a signal representative of the fault condition of the one or more cells.

When the on-charge actual voltage of the battery is determined to be less than the on-charge effective voltage of the battery, the signal output may be indicative of the presence of a shod-circuit.

Alternatively, when the on-charge actual voltage of the battery is determined to be greater than the on-charge effective voltage of the battery, the signal output may be indicative of the presence of sulphation or a faulty contact.

In a further embodiment of the invention there is provided a method of improving the operational state of a faulty battery, the method comprising, detecting a fault condition using the above-mentioned method, selecting a charging pulse in dependence upon the signal and applying the charging pulse to the battery. The signal may effectively act as a feedback signal.

Whilst the invention has been described above it extends to any inventive combination of the features set out above, or in the following description, drawings or claims. For example, any features described in relation to any one aspect of the invention are understood to be disclosed also in relation to any other aspect of the invention.

The invention will now be described, by way of example only, with reference to the accompanying drawings, in which:-Figure 1 is a graph of a voltage response to an applied current pulse; Figure 2 is a graph of a voltage response to an applied current pulse; indicating the different contributions to the waveform; Figure 3 is a flow diagram of the method of the first aspect of the invention; Figure 4 is a schematic of the apparatus of a second aspect of the invention; and Figure 5 is a schematic of the charging station incorporating the apparatus of Figure 4.

Referring firstly to Figure 1, there is shown the voltage response 1 to a current pulse 2 applied to a lead acid battery. Figure 2, shows that the voltage response can be broken down into different contributions from the various resistive components in the battery. The two essential components are the metallic parts, for example the lead grids in the plate, the metallic inter-cell conductors, the metallic terminals and the contribution from the electrochemical activity of the reactions between the plates and the electrolyte. -In I.3

Without limiting to a particular theory or conjecture, the metallic components conduct electricity directly according to ohms law, however the chemical components which also have a resistance, conduct electricity by ion transfer, which alters with time according to the rate of the chemical reactions and diffusion processes between the electrolyte and the surface of the electrodes.

The chemical components comprise the resistance attributed to the electrolyte and the conduction of the ionic species at the metal/electrolyte interface. The voltage at this region is determined by the concentration of the ionic species produced from the anodic and cathodic reactions at the anode and cathode.

The state of the charge of the battery is a crucial factor in delivering the current efficiently and this is measured using the voltage/time response curve of the battery to specific input pulse signals. The charge pattern varies depending upon the state of charge, so as to make allowance for the specific chemical condition within the battery at each stage of the charging process.

There are four stages of the voltage response to be considered including 1) the resistance from all the metallic components inside the battery 3 which accounts for 50% of the total resistance, 2) the resistance due to the energy requirements to initiate the chemical reactions, the reactions ultimately forming an electrical double layer creating a capacitive effect 4; 3) the diffusion and steady state condition (i.e. maximum voltage) 5 associated with the reaction forming sulphuric acid and 4) the surface concentration causing diffusion of sulphuric acid into the bulk electrolyte 6.

In Figure 2, the background current 7 and the point where the current

drops to the background level S is also displayed.

Considering stage 3 the length of time to reach the maximum voltage value, and the amplitude of the maximum voltage, is dependent upon the concentration of the sulphuric acid and the rate of production of the sulphate ions from the discharge electrodes, which are both related to the state of charge of the battery.

In stage 4, the concentration gradient relaxes as the current is switched off and the electrons diffuse into the bulk acid solution. The surface concentration of the acid at the plates then approaches that of the bulk solution.

The more concentrated the acid in the bulk solution, the slower the diffusion rate.

This is also a temperature dependent process, therefore monitoring the temperature of the battery can provide information relating to stage 4 and 6 of the voltage response. Therefore, measuring the temperature of the battery provides essential information for measurement of the charge state of the battery.

With the above in mind, the state of charge of the battery can be accurately measured from the slopes of stage 3 and stage 4.

In the invention as shown in Figure 3 a charging step is applied, whereby the charging step comprises the application of a test current pulse to a battery 9 which is to be charged. The voltage response to the test pulse is determined 10, whereby the voltage response provides information on the charge state of the battery and a charging algorithm is selected from a library storage area in the processor based on the voltage response. Therefore, a charging protocol is selected 11 dependent upon the voltage response of the battery to the previously applied test current pulse. The charge state of the battery is therefore used to select a charging protocol. The charging protocol comprises a charging pulse, or array of charging pulses that have a suitable amplitude, duration or number of pulses (in the case of the array) that would be tailored to the particular state of charge of the battery at that time. This charging pulse is applied to the battery to be charged 12.

The test pulses are distinct separate pulses to the charging pulses] whereby the test pulses last a few seconds and are scheduled for variable time periods of between 1 and 5 minutes in the charging cycle. The test pulse is a short current pulse of fixed frequency and fixed amplitude related to the capacity of the battery, which is designed to invoke an electrochemical response which has clear properties relating to the state of charge. It is important that this response is capable of being reproduced at each state of charge position in the sequence.

At variable programmed intervals, the charging pulse algorithm switches off for a period of several seconds before the test pulse is applied.

Whilst the charging pulses provide chemical conversions in the active material they do not provoke a consistent response from the battery. The test pulse, however, provides this desirable consistent response.

The application of the test pulse and the subsequent monitoring occurs in a rest state applied between a first and second charging protocol or is the first action in a charging cycle, whereby the charging protocol comprises a high frequency current pulse burst that is defined by an algorithm.

The voltage response from the test pulse is adjusted according to a temperature reading at the battery. This adjustment is obtained from a standard table of values for lead acid batteries. Accounting for the temperature dependency ensures that the correct state of charge and corresponding programming algorithm is correctly identified.

The temperature correction not only affects the value of the voltage, but also the speed of ionic diffusion and the reaction rate of the production of lead] lead oxide, sulphate ions and sulphuric acid. Therefore, this affects the slopes of the voltage response at the on/off stage.

By selecting the charging protocol in response to the charge state information of the battery a suitable charging pattern can be applied over a period of time that can, in certain circumstances, reduce the average on-charge voltage by up to 20%. This has the effect of increasing the battery's charge acceptance, allowing faster charging as well as reduced water loss, reduced temperature and lower energy usage of recharge.

The optimisation of the charging pattern is based on the identification of the state of charge of the battery and providing as a consequence a series of unique charging algorithms specific to that state of charge and battery type. To enable this detailed knowledge of the charge state, reactions specific to the state of charge have been determined. This knowledge is used to select the appropriate charge current pulse or array of pulses to be supplied to the battery at each key stage of the charge cycle. The method utilises a charging protocol and feedback monitoring system, providing different pulsed current and voltage profiles at each state of charge of the battery so as to improve the charging process. Ultimately, it is the synchronisation of the rapid current pulses to the battery chemistry at each stage of the charge which provides the maximum efficiency of the process.

As shown in Figure 4, the apparatus 13 comprises a current pulse generator 14 for generating a current pulse. The current is pulsed at a high frequency of between 1 Hz to 1 MHz. The pulses are of a rectangular waveform, the edges of which define the on/off state of the pulse. The amplitude, and frequency of the pulses are dependent upon the state of charge of the battery 15. Further, the duration of the on and off times of the current pulse are not necessarily of equal duration. A conductor 16 in the form of an attached cable is used to transfer the current pulse from the current pulse generator to the battery to be charged.

A microprocessor 17 monitors information on the charge state of the battery 15 in response to the test current pulse applied and a controller 18 selects the charging protocol from a series of charging algorithms stored in a library 19. The library may be stored in memory within the microprocessor 17 or may be stored in a distinct and separate storage means which is accessible by the microprocessor. The charging protocols relate to a current pulse or current pulse array having predefined characteristics, for example amplitude, duration, shape and number of pulses.

The pulse generator 14 comprises a mains DC power supply 20 and a high frequency switch 21. A smooth DC voltage is applied to the high frequency switch and to the system controller. The controller is communicatively coupled to the battery source to communicate a signal to initiate a charging current pulse dependent on the selected charging protocol. The circuitry for providing the charging pulse is separate and distinct from the circuitry of the test pulse.

The microprocessor 17 or charge state monitor monitors the voltage waveform response to the applied test current pulse. In particular, the stage 3 and 4 of the waveform is monitored to provide information on the electrochemical process in the cell which relate to its state of charge.

The bursts of high frequency charging current are followed by a short period of rest, with no applied current during which the test pulse can be generated and the condition of the battery can be measured off-charge. Therefore, there is a periodic check of battery conditions wherein the monitoring occurs during an interval located between a first charging current pulse burst (formed of a train of current pulses) and successive current pulse bursts, whereby the first charging pulse bursts and the second charging pulse bursts define the charging pattern.

The first and subsequent current pulse burst can have different pulse characteristics, for example the duration and the amplitude of the pulses can be different.

The microprocessor 17 and controller 18 determine the required time and amplitude for the next charging pulse burst of high frequency current by comparing the measurement from a state of the charge feedback system using a feedback loop 22 which takes information directly from the battery relating to the effect that the previous charge burst or charge protocol had on the battery condition i.e. a test pulse is applied to the battery and the new state of the battery is determined. The controller 18 then determines the required time and amplitude for the next current pulse by comparing the rest period measurements in which the test pulse is applied against a battery-type specific algorithm, in order to provide an optimum charging pattern from a predetermined schedule stored within the microprocessor 17.

Therefore, the charge input is delivered using algorithms and a unique feedback to control delivery of the charging current which is sensitive to the internal cell chemical reactions during the charging cycle. This charge technique removes the ion concentration effects at the surfaces of the plates in the cell, which reduces the charging voltage required, reduces the temperature of the battery and the time taken to charge the battery. This is because reducing the ion concentrations increases the efficiency of the charging process by reducing the charging voltage and thereby avoiding the secondary reactions, so water loss and heat generation will also be minimised.

The controller 18 is programmed and takes information directly from the battery, comparing the battery response to the test pulse to previous responses from the battery in order to calculate its state of charge. The processor 17 has inbuilt memory and computer interfaces, e.g. USB to enable this process.

The average on charge voltage at most stages is reduced to below 2.4 volts per cell. Coulombic efficiencies have been found to be less than 5% due to better efficiency during charging. Combining this with a much lower average on charge voltage can provide a resultant energy saving of between 15 and 45% and considerably shorter charging times when compared with conventional traction battery chargers.

In use, the charger 13 is switched on and a first charging step is applied, whereby a first or initial test current pulse is applied to a battery 15 to be charged. The charge state of the battery in response to the test current pulse of the first charging step is then monitored. A charging protocol from a series of predetermined charging protocols is then selected from a library 19, in dependence of the monitored charging state of the battery 15 to be charged.

The selected charging protocol is then applied to the battery 15 to be charged to complete the first charging step. After a predetermined period of time, the charging protocol is terminated and a second test pulse is applied so as to initiate a second charging step. The charging process is repeated until the battery is sufficiently charged or the charging cycle is terminated by another variable limit.

Information on the charge state of the battery 15 is obtained prior to selecting the charging protocol, thereby forming a charging pattern that provides optimum charging efficiency for the battery. As mentioned above, the information received from the battery relates to the voltage waveform of the battery in response to the charging protocol or test current pulse. Specifically the information relates to stage 3, stage 4 or a combination of the both of the voltage response waveform.

It can also relate to the voltage at a specific point within the second and fourth stages of voltage rise and decay.

Therefore, the information relates to the rate of change of voltage with respect to time on the waveform provided by an electrochemical reaction within the battery to be charged. The information may also relate to, or alternatively relate to the part of the waveform attributed to the voltage decay due to diffusion.

The selected charging protocol comprises a current pulse or array of pulses that has been predetermined based upon a given charge state and state of health of the battery.

Referring to Figure 4, the charger 13, in use is connected across the main battery bank 1 Sto obtain the overall voltage response of the batteries. Also, connection is made to an individual cell/battery or to a group of cells/batteries within the main battery bank. This is used to monitor the difference between individual cells and the total battery voltage, wherein the microprocessor can determine if there are any faulty cells within the battery as well as the condition and health of the overall battery. As a final addition, the feedback may result from a representative temperature sensor within the battery bank in order to obtain temperature compensated values for the measured voltages. It also will make an allowance for the change in the voltage/time slope.

The measurement is analysed by the microprocessor 17 included in the controller. From the voltage response of the battery 15 and individual cells within the battery iSa, 15b,15c,15d, the state of charge of the battery 15 can be determined, along with a mathematical analysis to determine if there are faulty cells, out of balance cells or a high resistance due to internal sulphation.

The microprocessor 17 then decides whether or not the battery 15 should be charged. This ultimately depends on the fault diagnosis. For example, if the measurement establishes that there are damaged or inoperative cells, the battery will not be charged and an alarm will sound. If a high resistance is determined due to sulphation, the charger will operate a recovery pulse algorithm for a specified time then re-measure the voltage response. For cells out of balance an equalisation algorithm can be applied. Ultimately where the microprocessor 17 indicates the start of a charging protocol, the charger will decide what level of charge is required and select the appropriate charging algorithm.

The state of charge measurement value must be received and stored by the microprocessor 17. The frequency of the pulse of the charge test signal during a specified charge period is fixed for the entire charging process.

An identification tag associated with the current algorithm enables controllable selection of a suitable charge protocol, whereby the controller is capable of selecting a tag, controlling the generator and switching the charge pulse circuit to provide the charging protocol associated with the newly selected algorithm depending on the measured charging voltage information. Each charging pulse must be of such a duration that the secondary reactions (gas evolution and heat evolution) do not have time to be significant. These reactions are slower than the conversion of lead sulphate to lead and lead dioxide. The apparatus of the invention is designed to input the energy with minimum resistance and therefore maximum efficiency. The processor implements a cut off when a charge threshold is reached corresponding to a maximum charge time, to ensure that thermal runaway or overcharging is avoided. It will also restrict the current flow or switch off the charging process if the temperature exceeds a safe charging temperature to protect the battery and the electronic equipment that is implemented e.g. the charge pulse circuitry. The threshold temperature is 60 degrees Celsius. Li

Figure 5 shows a charging station 23 comprising at least one charging bay (not shown) configured to charge a battery according to the above-mentioned method and using the above-mentioned charging apparatus may be used that is adapted to either receive the battery from the vehicle or has a current transferring conductor 24 attachable to the battery located within the fork lift or other vehicle or device utilising the battery. The charging station 23 comprises a micro-controller 25 and an oscilloscope 26. A data logger 27 and a pulsing unit 28 are also provided as well as an output unit 29 for generating the charging pulse.

Alternatively, the pulsing unit 25 and test pulse units (not shown) can be fitted onto the electrically powered vehicle enabling the vehicle to be charged from a simple power source anywhere on the premises. Or, as a further alternative, the entire charger may be mounted on the vehicle. In the case the charger is to be used for opportunity charging, the charger can be scaled down with a lower current output made to operate within the range of 30-80% the state of charge of the battery.

Use of an initial test pulse does not only enable measurement of the state of charge (SOC) with temperature correction, but also measures the state of health of the battery pack. It can achieve this mathematically from the rest voltage and the applied voltage response. Also, by using single cell measurements or multi-cell measurements, it can compare rest voltages with on charge voltages to determine i) if the battery is discharged and sulphated (i.e. zero/low voltage off charge i.e. at rest, and high voltages on-charge) and/or ii) whether it contains cells which are short circuited (i.e. giving 0 volts off charge i.e. at rest and 0 volts on-charge) and/or Hi) whether there are batteries which have over-discharged cells. There is the possibility of having a mixture of these defects which would give similar mathematical outcomes i.e. one or more cells of a battery may have any of, or a multiple of these fault conditions. For this reason a recovery programme is used to identify any health problems of the battery within a prescribed period of time from commencement of charging (e.g. minutes of charging). The recovery programme then determines whether the battery can be recovered or not. For example, subsequent recovery pulses would show if a sulphated cell is present, since the off-charge and on-charge voltages would improve, however the voltage would remain at zero for a short circuited cell, therefore no improvement would be observed. Therefore, it is necessary to provide recovery pulses over a series of time to determine which fault is occurring and how many cells are affected by the particular fault.

In use, to determine whether there is a fault in the battery (whereby the battery may comprise a single cell or multiple cells) a reference voltage of a single cell is determined when the battery is in the non-charging state (i.e. the rest state). The reference voltage is the voltage of the single cell that is used as a reference for the subsequent stages of the method.

A calculation is then performed whereby the reference voltage is multiplied by the number of cells in the battery to obtain an effective voltage for the entire battery. This calculation assumes that each cell in the battery is of the same type and voltage, the type being a lead acid battery and a typical reference voltage being 2.0 V at rest and 2.3V on charge Effective voltage is what the voltage value of the battery is expected to be in dependence upon the reference voltage value. For example, the effective voltage for a 12 cell battery, considering a reference voltage of 2.OV off charge and 2.3 V on charge is 24V off charge and 27.6 V on charge.

The actual i.e. measured voltage across the entire battery (i.e. all present cells are considered) is then determined during a non-charging state of the battery so as to provide off charge the actual voltage value.

The effective voltage of the entire battery off charge is then compared with the actual voltage of the entire battery off charge, and an output signal is provided.

Instead of only considering a single cell in a multiple celled battery, the reference voltage of multiple cells may be determined, whereby the effective voltage calculation becomes more complicated i.e. the reference voltage value is divided by the number of multiple cells multiplied by the total number of cells in the battery.

Regardless of whether the reference voltage is determined for a single cell or multiple cells, the signal may be generated when the value of the effective voltage and the actual voltage differ by a predetermined value.

Whilst monitoring in the off-charge state provides information that a fault condition is present in the battery information on the voltage of the battery in the charge state is required to clearly determine what the fault condition actually is and the number of cells affected. Therefore, in order to establish the type of fault the voltage of the battery is determined in an on-charge state.

The voltage of the entire battery is measured on charge and compared with the voltage of the reference cell on charge. It is essential to do the off-charge comparison of the reference cell and total battery voltage prior to the on-charge comparison so as to establish the ground condition of the battery before applying the charge to create the on charge state. The on charge state will not provide the required information without the initial off charge condition reading.

In use, in the condition that the off charge voltage is less than the reference cell voltage multiplied by the number of cells by a predetermined value recognised by the microprocessor, then a fault condition is recorded. A test pulse is then generated to the battery and the on charge voltage of the reference cell and the battery are recorded. If the actual on charge voltage is less than the effective voltage (reference cell multiplied by the number of cells on charge) then a short circuit in one or more cells is assumed and the charger will go through a sequence to verify this and set off an audio visual alarm. If the actual on charge voltage is greater than the effective voltage then it is assumed that there is a high resistance which indicates that the battery or some cells within the battery are sulphated, or alternatively the high resistance may be due to a faulty contact.

When the actual on charge voltage is greater than the effective voltage a recovery programme of charging is initiated. Again an audio visual signal is generated with a fault condition registered on a charger display panel.

In the event that the reference cell may be faulty then this will be apparent as an out of balance battery and the recovery programme will also be initiated.

Therefore, improving the operational state of a faulty battery may be provided by, once the fault type has been determined as described above, selecting a recovery program in dependence upon the signal and applying the recovery pulse to the battery. The recovery pulse may be a series of pulses defined by a recovery program or algorithm. Therefore feedback of the signal to the processor or controller is required to enable selection of the appropriate recovery program.

The charging station further provides an option for completely recharging a faulty battery and then making a diagnosis and recommendation for recovery or repair.

The method of charging a battery using the apparatus of Figure 4 and charging station of Figure 5 obtains maximum efficiency since the charging profile varies according to the state of charge of the battery, which depends on the particular chemical state of the battery. The method of charging the battery, therefore recognises the state of charge from a particular battery response in various states of charge and utilises a set of charging profiles which vary according to the state of charge of the battery.

When using the apparatus of the invention, beneficially, the water loss and gassing due to overcharge and inefficient side reactions are negligible when using the charging technique of Figure 3 and apparatus of Figure 4. The temperature rise usually associated with charging is minimised and reduced to below industry norms. The temperature is controlled by varying the charging algorithms in accordance with the battery temperature and recovery algorithms are available to ensure that cell imbalance and acid stratification are eliminated.

There is a large market for lead acid batteries in the material handling industry e.g. electric fork lifts used in warehouses and logistics sectors. The cost savings from reduced energy use, improved vehicle utilisation and reduced capital outlay, due to the operational improvements from faster charging makes a huge impact to this market. However this also benefits the mobility sector that includes mobility wheelchairs and scooters and leisure vehicles, for example golf carts, caravans and lawnmowers. In these applications, faster and more efficient charging not only provides cost benefits, but would also allow increased flexibility and utilisation, as well as the possibility for increasing the operating range of these vehicles.

Another advantage of the charge method and apparatus is the environmental benefit; referring back to the use of the charger for batteries used in fork-lifts, the proposed 15% reduction in energy usage will have measurable impact on the carbon footprint of the operator. Use of the low energy rapid recharging method will also reduce the time on charge to enable opportunity charging. This has the major benefit of removing the need for several battery changes per day which would reduce the numbers of batteries and trucks and associated charging sets needed to cope with the workload for typical warehouse and logistics operations.

This will also be considered as an environmental benefit when the production wastes, emissions and end of life recycling methods are taken into account. Further the battery can be connected to the charger when the battery is only partly discharged without being a detriment to the life cycle of the battery.

Various modifications to the principles described above would suggest themselves to the skilled person. For example, an RF switch may be applied to provide the desired current pulse and the batteries may be used in a golf cart or other leisure or mobility vehicles.

Whilst the reference voltage for a single cell may be determined, it is also possible to determine the off-charge reference voltage for multiple cells in the battery. The reference voltage per cell is then obtained by dividing the off charge reference voltage value for multiple cells by the number of cells forming the multiple cell to provide an off-charge effective voltage for an individual cell (i.e. single cell). The single cell reference value is then multiplied by the total number of cells in the battery to obtain the effective voltage in the off-charge state.

Claims

CLAIMS1. A method of charging a battery comprising: applying a charging step, comprising applying a test current pulse to a battery to be charged, and subsequently applying a charging protocol to the battery, wherein the charging protocol is dependent on the response of the battery to the previously applied test current pulse.
2. A method according to claim 1, wherein a plurality of charging steps are applied to the battery in a period of time.
3. A method according to claim 1 or claim 2, wherein the charging protocol is a charging pulse or charging pulse array having a predetermined charging pattern.
4. A method according to claim 3, wherein the test pulse is separate and distinct from the charging pulse or charging pulse array.
5. A method according to claim 3 or claim 4, wherein the charging pulse has a minimum frequency of at least 1 Hz.
6. A method according to any of claims 3 to 5, wherein the charging pulse has a frequency of between 1 Hz to 50,000 Hz.
7. A method of charging a battery according to any preceding claim further comprising: monitoring the charge state of the battery subsequent to applying the test pulse and prior to determining the charging protocol so as to provide information on the state of charge of the battery.
8. A method according to claim 7, wherein the information relates to the electrochemical response of the battery at a plate acid interface located within the battery.
9. A method according to claim 7, wherein the information relates to a voltage response waveform of the battery in response to the previous test current pulse.
10.A method according to claim 9, wherein the information relates to the rate of change of voltage with respect to time on the waveform provided by an electrochemical reaction within the battery to be charged.
11.A method according to claim 9 or 10, wherein the information relates to the part of the waveform attributed to the voltage decay due to diffusion of sulphate ions from a plate surface of the battery into the bulk electrolyte.
12.A method according to claim 11, wherein the information is provided by a temperature sensor located in thermal communication with at least one cell of the battery to be charged.
13.A method according to any preceding claim, wherein the selected charging protocol comprises a charging current pulse or pulse array that has been predetermined based upon a given charge state of a battery.
14.A method according to any preceding claim, wherein the selected charging protocol comprises a current pulse or pulse array that has been predetermined based upon the temperature of the battery.
15. A method according to any preceding claim, further comprising: terminating the charging protocol; applying a further test current pulse to the battery to be charged; monitoring the charge state of the battery in response to the further test current pulse; selecting a further charging protocol from a series of predetermined charging protocols, depending on the monitored charging state of the battery to be charged; and applying the selected charging protocol to the battery to be charged.
16.A method according to claim 15, wherein the monitoring of the charge state of the battery occurs during a time interval located between a first current charging protocol and successive current charging protocols.
17.A method according to claim 15 or claim 16, wherein the first and subsequent current charging protocols have different pulse characteristics.
18.A method according to claim 17, wherein the pulse characteristics comprise at least one of duration, amplitude, shape or number of pulses.
19.An apparatus for charging a battery, comprising: a first current pulse circuitry for generating a test current pulse; a conductor for transferring the current pulse from the current pulse generator to the battery to be charged; a second charge current state circuitry for generating a charging current pulse; a charge state monitor for monitoring and receiving information on the charge state of the battery to be charged in response to the test current pulse applied; nfl 3.3 and a controller for selecting the charging protocol from a series of predetermined charging protocols dependent on the charging state of the battery to be charged and for communicating with the second charge state circuitry for generating the charging protocol.
20. An apparatus according to claim 19, further comprising a microprocessor for selecting a charge protocol from a charging protocol library in dependence on the charging state of the battery.
21.An apparatus according to claim 19 or 20, further comprising a charge pulse generator for generating a current pulse having a frequency of at least 1 Hz to 50,000 Hz.
22. An apparatus according to claim 21, wherein the charge pulse generator comprises a mains DC power supply and a high frequency switch.
23.An apparatus according to any of claims 21 or 22, wherein the controller is communicatively coupled to the charge pulse generator to initiate a current dependent on the selected charging protocol.
24.An apparatus according to claims 19 to 23, wherein the charge state monitor comprises a temperature sensor.
25.A charging station comprising at least one charging bay configured to charge a battery according to a method of any of claims 1 to 18.
26. A charging station comprising the apparatus of any of claims 19 to 24.
27.A charging station comprising a first charging circuit capable of generating a test pulse for determining the charge state of a battery to be charged and a second charging circuit capable of generating the charging pulse and/or array of charging pulses in response to the determined charging state of the battery.
28.An electrically powered vehicle incorporating the apparatus according to claims 19 to 24.
29.A method of detecting a fault in a battery during charging, the battery having at least one cell, the method comprising: determining an off-charge reference voltage of a single cell of the battery during a non-charging condition; multiplying the determined off-charge reference voltage by the number of cells in the battery to obtain an off-charge effective voltage for the entire battery; determining an off-charge actual voltage across the entire battery during a non-charging condition; comparing the off-charge effective voltage of the entire battery with the off-charge actual voltage of the entire battery, and outputting a signal representative of a fault condition, the signal being dependent upon the comparison between the off-charge effective battery voltage and the off-charge actual battery voltage.
30.A method according to claim 29, wherein the signal is generated when the value of the off-charge effective voltage and the off-charge actual voltage differ by a predetermined threshold.
31. A method according to any of claims 29 to 31, wherein the signal is an audio, visual or feedback signal to a processor or a controller.
32. A method according to any of claims 29 to 31, the method further comprising determining an on-charge reference voltage of a single cell of a battery during an on-charge condition; determining the on-charge actual voltage of the entire battery during an on-charge condition; comparing the on-charge actual voltage of the battery during an on-charge condition with the on-charge actual voltage of the battery during an on-charge condition; outputting a signal representative of the fault condition of the one or more cells.
33.A method according to claim 32, wherein when the on-charge actual voltage of the battery is determined to be less than the on-charge effective voltage of the battery, the signal output is indicative of the presence of a short-circuit.
34.A method according to claim 32 or 33, wherein when the on-charge actual voltage of the battery is determined to be greater than the on-charge effective voltage of the battery, the signal output is indicative of the presence of sulphation or a faulty contact.
35. A method of improving the operational state of a faulty battery, the method comprising, detecting a fault condition according to claim 29 or 33, selecting a charging pulse in dependence upon the signal and applying the charging pulse to the battery.
36.An apparatus for charging a battery with reference to the accompanying text and drawings.
37.A method of charging a battery with reference to the accompanying text and drawings.
38.A charging station with reference to the accompanying text and drawings.