WO1997032384A1

WO1997032384A1 - Method and apparatus for charging a battery

Info

Publication number: WO1997032384A1
Application number: PCT/US1997/002729
Authority: WO
Inventors: Yury M. Podrazhansky; Michael Y. Podrazhansky
Original assignee: Advanced Charger Technology, Inc.
Priority date: 1996-02-27
Filing date: 1997-02-26
Publication date: 1997-09-04
Also published as: AU1967097A

Abstract

A method and apparatus for rapidly charging a battery is described to reduce internal heat generation. In the preferred battery charging method a charge cycle is employed wherein a direct charging current pulse (10.1) is followed by a small wait period (12.1) after which a short discharge pulse (13.1) is applied followed in turn by a first wait period (15.1) during which there is neither a charging nor a discharging of the battery. The first wait period is altered by a first pack of high frequency alternating current pulses (14.1A) at a DC voltage level equivalent to that for an open circuit battery voltage. The first pack of high frequency pulses are selected to enhance a mixing action at the battery electrodes. A second discharge pulse (16.1) is then applied followed by a second wait period (18.1) which is altered by a second pack of high frequency alternating current pulses (14.1B). These latter pulses are selected so that measurement of the battery capacitance can be obtained and an evaluation of the charge of the battery can be derived. The charge cycle can be repeated until a battery parameter attains a desired value indicative of a sufficient battery charge.

Description

Method and Apparatus for Charging a Battery

Prior Applications

This application claims the benefit of the filing date of U.S. Provisional Patent Application entitled Efficient Charging Algorithm, Charge Control And Charging Process Termination filed February 27, 1996 with The United States Patent and Trademark Office and bearing Serial No. 60/012,362 and filed by Yury M. Podraz¬ hansky and Michael Y. Podrazhansky and assigned to the same assignee as for this application.

Field Of The Invention

The present invention generally relates to a method and apparatus for rapidly charging of a battery and more particularly to a method and apparatus for more precisely determining the state of or completion of a battery charging process.

Background Of The Invention

A special techniques for charging a battery with a single discharge pulse is disclosed in US patent No. 4,829,225 to Podrazhansky et al. With US patent No. 5,307,000, which also employs discharge pulses, an improved charging speed is obtained. US Patent No. 3,816,807 to Dale F. Taylor describes a technique for modulating a DC charging current with an AC voltage. The phase change of the modulating AC voltage is sensed with a phase detector and the change in phase sent as a feedback signal to vary the DC power supply. The control is referred to as an impedance controlled battery charger.

US Patent No. 5,329,219, to S. Garret describes a method of charging a battery with a different charge rate. US Patent No. 5,331,268 to J. Pantino et al teaches a control for a trickle charge, which begins when a baseline voltage of the battery during a rapid charge attains a predetermined value. US Patent 5,200,689 to A. Interiano et al describes a charge control and a charge termination for nickel cadmium and nickel hydride batteries. The charge profile is similar to previously described techniques and is topped off with a trickle charge which will cause dendrite and other problems that are also applicable to the technique described in the '268 patent. In US Patent No. 4,878,007 to S. Garbor and N. Sandor steep current pulses are superimposed on both charge and discharge pulses to produce homogeneous electrode surfaces. US Patent No. 4,746,852 to Ray J. Martin teaches a controller that terminates battery charging as a function of a time derivative of the measured battery voltage.

US Patent 4,577,144 to John S. Hodgman and Ferdinand H. Mullersman describes a technique for distinguishing between primary and secondary batteries by monitoring the line ripple voltage of the rectified charging voltage. This ripple voltage is deemed to reflect the low frequency impedance of the battery during charging. A distinction between primary and secondary batteries can be thus be made because, as stated in this patent, the low frequency impedance for a secondary battery is lower than that for a primary battery of the same physical size. US Patent No. 4,740,739 to Leon D. Quammen and James M. Hisle describes the use of very high frequency discharging pulses superimposed upon an unfiltered charging current applied to a battery. US Patent No 3,987,353 to James Adrian Macharg describes a technique for using charging pulses separated by intervals during which a change in the battery voltage is monitored and used to control the magnitude of the charging pulse. US Patent No 3,857,087 to David C. Jones describes a method to test lead acid batteries to separate good from bad batteries by using a charge, wait, discharge, wait technique.

Summary Of Invention

With a technique and apparatus in accordance with the invention a battery can be rapidly and safely charged even when the battery is frozen by reliably determining the battery's state of charge and as a result determine when the charging process needs to be either reduced or terminated.

These advantages are achieved in accordance with one embodiment of the invention by providing a microprocessor controlled battery charger wherein the processor provides a charge pulse followed by a discharge pulse and a first wait period during which there is neither a charging nor a discharging of the battery. The first wait period is altered by a first pack of pulses alternating about the open circuit voltage of the battery and at a frequency selected to enhance a mixing of the electrolyte. During selected charge cycles the wait period is altered with a second set of alternating frequency pulses at a frequency selected to enable a measurement of the impedance of the battery. The impedance measurement is used to determine the capacitance of the battery and this value is then used to ascertain whether the charging process should be continued or terminated. In another embodiment of the invention the wait period is in turn followed by a second discharge pulse. After the second discharge pulse there is a second wait period during which there is neither a charging nor a discharging of the battery and this second wait period is altered by a second pack of high frequency pulses alternating about the open circuit voltage of the battery and having a frequency sufficiently high to enable a determination of the capacitance of the battery. The measurement of the capacitance is then used to determine whether the battery is sufficiently charged.

With these battery charging techniques a frozen battery can be thawed much more quickly and enables the microprocessor to determine the state of the charge of the battery at any given time more correctly. By applying different pulse frequencies during wait periods a dramatic reduction of the concentration of an over potential at the electrodes and a more uniform battery plate surface structure are obtained with a reduction in the overall charging time for the battery while preserving a high plate porosity level and reduce corrosion of the collector grid.

The high frequency pack of pulses can be in the form of alternating current (AC) sine waves. These high frequency pulses modify the mass transport process inside the battery by reducing the concentration of a diffusion layer wherein the concentration of ions fluctuates with time near the electrode surface. The reduction of polarization concentration and over potential depends not only on the magnitude of the AC current pulses but also on their frequency. In addition, the AC pack of pulses allows a measurement of metallic conversion during charging and enables a determination of the battery capacitance, and thus battery capacity, as well as when to terminate battery charging. The use of high frequency packs of pulses increase the precision of the measurement of the internal impedance and the determination of the battery status. It is, therefore, an object of the invention to provide a technique and apparatus for rapidly charging a battery. It is a further object of the invention to provide a method and apparatus for determining when to terminate the charging of a battery.

These and other advantages and objects of the invention can be understood from the following description of several embodiments as shown in the drawings.

Brief Description Of The Drawings

Figure 1 is a plot of charge current cycles in accordance with the invention for use in the charging of a battery;

Figure 1 A is an enlarged plot a charge cycle used in the charging cycles of Figure 1 ;

Figure 1B is an enlarged view of a charge cycle shown in Figure 1 for the charging of a battery in accordance with the invention;

Figure 2 is a plot of an alternate charge current cycle in accordance with the invention for use in charging a battery;

Figure 3 is a plot of still another charge current cycle in accordance with the invention for use in charging a battery; Figure 3A is an enlarged view of another charge cycle for use in a battery charging process in accordance with the invention;

Figure 4 is a block diagram of an apparatus in accordance with the invention for use in charging a battery;

Figure 5 is a general flow chart for use in a charging process in accordance with the invention for applying a battery charging technique of this invention; and

Figure 6 is a flow chart of steps used in a microcontroller to implement a charging of a battery in accordance with the invention.

Detailed Description Of The Drawings

With reference to Figures 1 , 1A, and 1B charge current cycles 10 are shown and are used during the charging of a battery. The charge current cycle 10 need not be used continuously but can be employed on a regular interval basis, say once every minute or several minutes, depending upon the type and size of battery being charged. Other charge cycles such as shown in US Patent 5,307,000 to Podraz- hansky can then be used in between the charge cycles 10. Hence, the description of US Patent 5,307,000 is incorporated herein by reference thereto subject to such modifications as described herein for the instant invention. The use herein of decimals after numerals identifies specific items with the numerals after the decimal points whereas the use of the numeral on the left side alone denotes the same item in a general manner.

In the charge cycles 10, 11 are charge pulses; 12 ar general or technical rest periods; 13 are first discharge or depolarization reverse current pulses; 14 are alternating current pulses or sine waves selected to enhance a mixing of the battery electrolyte; 15 are first wait or rest periods; 16 are second discharge pulses; 17 are second high frequency packs of alternating current pulses or sine waves; 18 are second wait or rest periods; and 19 are following charge current pulses.

The first discharge pulses 13 reduce electrolyte at a thin layer between the active electrode mass and the current collector grid, not shown, at the electrochemi¬ cal conversion area. Normally this layer moves because the active material is converted and the conversion process starts from material close to the conductive collector grid and moves inside towards the opposite plate or electrode.

The discharge pulses 13 and 16 are created by applying a load or reverse voltage to the battery. The discharge pulses have a significantly shorter duration than the duration of the charge pulses 11. The discharge pulses preferably have a magnitude that is from about 2 to 10 times larger than the magnitude of the direct current charging pulses 11, but with a much smaller duration than the duration of the direct current charging pulses 11. The discharge pulses serve to reduce polarization concentration right after the charge pulses 11.

The first wait periods 15 allow additional mixing with a fresh electrolyte from outside of the reaction area. During first wait periods 15 the battery's open circuit voltage (OCV) can be measured. The second discharge pulses 16 help to reduce polarization concentration on the electrochemically active double layer of the battery and also reduces over voltage potential.

The second wait periods 18 also help with electrolyte exchange within the reaction area. The second (or sometimes more than one) rest periods allow a complete cessation of the charging process and thus a measurement of internal impedance during a time ions are in a neutral position which normally is not achieved within a battery. During the second wait period, after discharge pulse 16 the measurement of the OCV is much closer to the real OCV.

The packs 14 of AC pulses serve to mix a fresh electrolyte from outside the electrode reaction area with the electrolyte within the reaction area and thus reduce the density of the over potential due to a concentration of polarization within the double layer of the battery. The effect from the high frequency pack of pulses is to reduce the internal resistance of the battery. The forced mixing of the electrolyte enables a reduction in the wait periods between charge pulses so that longer wait periods for measurements can be used with little impact on the overall charging time.

The pulses 14 persists for a duration that is less than charging pulses 11. If the number of frequency pulses 14 is too high then the battery tends to heat up. If the number of high frequency pulses is too few then there is not a sufficient mixing action near the electrodes. Generally a duration sufficient to obtain from about three to about five sine waves is sufficient to achieve the desired mixing action. For a frequency of about 200 Hz a duration for the packs 14 of up to twenty five ms is sufficient. The frequency should not be too high lest the electrolyte mixing does not occur and, therefore, a maximum frequency is generally about 500 Hz. The wait periods 15 and 18 may be of equal duration or different and usually are of a longer duration than discharge pulses 13 and 16. Typically the duration of the discharge pulses 13 and 15 are of the order of a few milliseconds.

From time to time a second type of high frequency pulses 17 is used during a charge cycle such as 10.2 to alter a wait period. The pack of pulses 17 has a higher frequency so as to enable its use to measure the internal impedance of the battery. Specifically the frequency and duration of pulses 17 are selected so that the capacitance of the battery can be measured. The accuracy of this measurement is enhanced by virtue of the use of the discharge pulses 13 and 15 and the mixing action from the prior pack of high frequency pulses 14.

The frequency of the pulses 17 is selected so that high frequency current signals can be physically delivered to the battery and a measurement of its capacitance can be made. This may, therefore, preferably be higher in frequency than packs 14 and in the range from about 100 to about 1M Hz. The capacitance measurement is obtained by first measuring the internal resistance of the battery using conventional techniques. The internal resistance is obtained by monitoring the drop in the constant DC charge voltage during charging and dividing this by the corresponding current passed through the battery. Preferably the charge voltage drop and current are measured towards the end of the charge pulse 11 and the internal resistance R, is obtained by dividing the current measurement into the measured voltage drop.

The measurement of the impedance, Z, during the second pack of high frequency pulses 17 is then used to derive its reactive component X, and from this the capacitance C of the battery using the well known relationship: Z= R + 1/τπC = R + 1/2/ιfC

The capacitance is measured during different charge cycles and changes in its value monitored. When the capacitance continues to show a change in value this is interpreted as an indication that the battery is still accepting a charge. When the capacitance measurement fails to indicate a change from the last measurement, or some other previous measurement, a determination is made that the battery is fully charged and battery charging is terminated.

Since battery charging takes a relatively long time it is not necessary, as illustrated in Figure 1, to measure the capacitance during each charge cycle. The capacitance measurement can be made once every minute or at such other interval as will assure a timely termination of the charging process without adding significantly to the overall duration of the charging process.

With reference to Figures 2 and 3 charge cycles 10.2 and 10.3 are illustrated, which are similar to charge cycle 10.1. However, in charge cycle 10.2 in Figure 2, the AC pulse packs 14, 17 do not alter the wait periods 15, 18 at their starts, but are injected to do so at the ends of these wait periods. In the charge cycle 10.3 in Figure 3 the first pulse pack 14 is located at the beginning of wait period 15 while the second pulse pack is timed to occur near the end of wait period 18. 20 points to a small technical wait period as it may be necessary.

In the charge cycle 10.2 of Figure 2 the high frequency AC pack of pulses 17 enable a more precise measurement of the internal impedance of the battery. The mixing role of the AC pulse pack 14 is diminished but the impedance and metal conversion measurements during the second pulse pack 17 is improved.

In Figure 3A a charge cycle 40 is used wherein a single discharge pulse 13 is employed following a wait period 12. The discharge pulse 13 is followed by a pulses 14, a wait period 15 and at the letter's end a high frequency pack of pulses 17 to measure the battery's impedance. The pulse pack 17 need not occur every cycle 310 as previously explained. The high frequency pack of pulses 14 preferably occurs with each charge cycle 310.

In Figure 4 a block diagram of a charger 40 in accordance with the invention is shown. 21 is a microcontroller or microprocessor; 22 is a circuit for measuring the battery voltage; 23 is a circuit for measuring the charging current; 24 is a control circuit for control over the charge currents; 25 is a discharge control circuit; 26 is a transistor or other semiconductor device for controlling the charge current to the battery; 27 is a discharge current transistor; 28 is a the battery to be charged and which can be a lead acid cell, a nickel battery or a metal hydride cell. 29 is a shunt resistor used to measure the charging and discharging currents; 30 is a display which can show capacity or battery voltage; 31 is a power supply using to drive the microcontroller and supply electrical energy to charge battery 28.

When a battery 28 is installed into the charger 40 the microcontroller 21 , via circuitry 22 senses the battery voltage. If the voltage is normal for the battery 28 the controller 21 starts to charge the battery via circuitry 24 and transistor switch 26. If the voltage on battery 28 is lower than expected, the controller 21 will not charge the battery 28 and send information to display 30 with an explanation of the problem. If the voltage on the inserted battery 28 is normal the processor 21 begins a testing procedure. The testing procedure consists of measuring the battery's internal resistance, so processor 21 , via circuitry 24, sends a signal to transistor switch 25 to enable it to apply a charging pulse 11 as shown in Figure 1.

Figure 5 illustrates a simplified flow chart 50 for operating the microcontroller 21 and implement the invention. Other flow techniques can be implemented and what is shown is for illustration only. Figure 6 shows a flow chart 70 for determining the capacitance of a battery in accordance with the invention and comparing the measurement to determine when to terminate the charging process.

In Figure 6 at 72 a regular charge cycle such as 10.1 in Figure 1 is begun and a timer is set at 74 to ascertain at 76 with a test when a capacitance measurement is to be made. Then if a capacitance measurement is to be made at 80 the value for capacitance from a previous measurement is stored as C_p,,^. and at 82 a charge pulse 11 is started. At some time towards the end of the charge pulse 11 the internal resistance of the battery is measured by sensing at 84 the current l_dc through the battery and at 86 the voltage V_dc across the battery. These parameters are determined with the apparatus as described with reference to Figure 4. The current through the battery is determined from a measurement of the voltage across the known resistance 29. The internal resistance is determined at 88 from the voltage and current values using R-Wl⁰ .

The several events as described with reference to Figures 1 and 1A are then applied at steps 90-96. At 98 a pack of high frequency current pulses 17 with which the impedance of the battery can be measured is applied to the battery. While these are applied, the ac current, l^ac, and ac voltage, V", are measured at 100 in a similar manner as the dc equivalents. The battery impedance Z is then determined at 102 from the relationship Z=V_ac l_ac and the measured capacitance C at 104 from the relationship as previously set forth. This value for C is stored at 106 and a test is made at 108 whether the stored value for C is greater than the value C_prβvl0US. If not, a return is made to step 72 to continue the charging process.

However, if the value of C did not change appreciably, i.e. no more than a maximum amount, then the test result is interpreted as indicative of a need to terminate the charging process at 110 and a return is made at 112. Note that, depending upon the type of battery involved the change in capacitance indicative of a need to continue the charging process can be negative or positive. When the change is less than some maximum amount or zero the result is indicative of a need to terminate the charging process. This is explained below.

During pulse charging the current flowing across the electrode /electrolyte interface to be composed at a time averaged DC component l_DC , and a fluctuating alternating current (AC) component, l_AC .

l(t) - l_DC + l_AC (t) (1)

For boundary Condition

C = C_β at t=0

C = C„ at y = oo

-D ac/ay - l(t)/nF at y= 0 (2)

Where is C is concentration, t is a time y is vertical distance from electrode surface and the subscript oo denotes the bulk properties. This also split the concentration C into a time invariant steady-state component C₈, and a fluctuation component C_f

C(y,t) - C_β(y) + C_f (y.t) (3)

where C is the concentration, t is a time y is vertical distance from electrode surface and the subscript oo denotes the bulk properties.

Substituting equation (3) into equation (1) - (2) we will obtains two set of differential equations and associated boundary conditions, one for the steady state concentration C, and a second one for fluctuation concentration .C,

The equations for the steady state concentration are

~~C = C_ at y = oo(4a)

-D d~^"Ctøy = l_oc / at y = 0 (4b)

The solution above equation give a surface concentration component ^"~C sur in the form

^"C sur / C„ = 1 - l_DC / l_llm (5)

Where I _lim is the limiting current density for a given convective condition.

The thickness of the steady state Nernst diffusion layer depend from concentration gradient at the electrode surface.

The equation for the fluctuation concentration are: ac/at +~v • ^"~v c = D v² c (6a)

C_ = 0 at t = 0(6b)

C s o at y = oo(6c)

-D aC/ay = l_AC (t)/nF at y= 0(6d)

Normally the concentration fluctuation occur within thin region of the steady - state Nernst diffusion layer where the contribution of convective flux term ~~ V • ~~V C« relatively small. To increase fluctuation of concentration by inserting high frequency AC pulses within Nernst diffusion layer will reduce overall concentration, prevent over potential and reduce neat by reducing internal concentration resistance.

The nickel metal hydride battery ever more getting into the market and up to this point neither has very satisfactory method of a termination charge cycle. The present termination by Δ V(ee) and by maximum temperature did not prove a satisfactory result, since battery heats up and it is no way to define minus V(ee) correctly Δ and, charge termination generally accomplished by maximum tempera- ture, that destroy a battery. The new method aτ²/3t (where is T is temperature and t is time), when temperature will increase in short period effective only if temperature of a battery below 30°C. If temperature of a battery higher, the change did not have sharp curve and termination became very difficult or impossible.

For Lead Acid battery the problems are worse. Even lead acid technology is 150 years old. The accurate termination of a charge cycle did not exist yet. The entire world has number of problems with life (corrosion) and performance of lead acid batteries and it is maintenance. The incorrect methods of charge termination, abuse method of energy transformation make life of lead acid battery insufficient, and increase expense for a battery maintenance. The pulses are Using different frequency during time then high frequency pack of pulses is allow also measure metallic conversion from PbSO₄ to Pb on plates.

For Lithium Ions type of battery or it is modification, inaccurate methods of energy transfer demand costly electronics circuitry inside the battery pack to protect life of the battery and keep a battery operation safety. Limited possibilities of measurement battery parameters during a charge cycle suppress feasibility for rapid charging lithium ion battery.

For Nickel Cadmium type battery incorrect energy transfer technique and termination make battery life very short and in addition using trickle charge increase Nicd battery memory problem. The most creation of memory occur during trickle charge. The creation of crystals with open circuit potential below 1.2 volts per cell did not allow return capacity back during discharge cycle, because minimal voltage that allow to discharge battery 1 volt per cell and crystals that create memory has open circuit potential 0.8 volt per ceil. The AC high frequency pack of pulses preventing grow such crystals with low potential, that is automatically extended life of NiCd battery.

For Nickel Metal Hydrate type battery incorrect energy transfer algorithm, Production more Hydrogen and Oxygen, that battery can absorb. Normally NiMH type battery did not have Ions transport problem and termination can be done by using 3T/3t, where is T is temperature on a battery and t is a time. The NiMH battery takes charge with heat production as exothermic reaction. The potassium hydroxide electrode also polarize at high current and prevent NiMH battery from overpotential during the charge cycle will extend life of a battery. Having thus described several embodiments for practicing the invention its advantages can be appreciated. Variations from the embodiments can be implement¬ ed by one skilled in the art without departing from the scope of the invention as determined by the following claims.

Claims

We Claim

1. A method for charging a battery, comprising at least one charge cycle having the steps of: applying a charge pulse to said battery, said charge pulse having a pulse width not substantially less than 20 milliseconds, after the charge pulse applying a first discharge pulse to said battery to produce a discharge current from the battery, after said discharge pulse applying a first wait period during which said battery is neither substantially charged nor substantially discharged; altering said first wait period by applying a first pack of high frequency current pulses to said battery with said first pack of pulses having a frequency selected to enhance mixing of the electrolyte.

2. The battery charging method as claimed in claim 1 and further including the steps of: altering said first wait period by applying a second pack of high frequency current pulses to said battery with the frequency of said second pack of pulses being selected to enable a measurement of the impedance of the battery; and measuring the impedance of said battery while said second pack of pulses are being applied to the battery.

3. The battery charging method as claimed in claim 1 and further including the steps of: after said first wait period applying at least a second discharge pulse to said battery, after said second discharge pulse applying a second wait period during which said battery is neither substantially charged nor substantially discharged; altering said second wait period by applying a second pack of high frequency pulses to said battery with said second pack of pulses having at least one alternating current cycle; and repeating said charge cycle until a selected battery parameter achieves a desired value.

4. The method as claimed in claim 1 wherein said step of altering said first wait period comprises applying said first pack of high frequency pulses at a time that is near the start of said first wait period.

5. The method as claimed in claim 1 wherein said step of altering said first wait period comprises applying said first pack of high frequency pulses at a time which is towards the end of said first wait period.

6. The method as claimed in claim 3 wherein said step of altering said second wait period comprises applying said second pack of high frequency pulses towards the start of said second wait period.

7. The method as claimed in claim 3 wherein said step of altering said second wait period comprises applying said second pack of high frequency pulses towards the end of said second wait period.

8. The method as claimed in claim 1 wherein said step of altering the wait period comprises applying said first pack of high frequency alternating current pulses at a DC potential that is substantially at the open circuit voltage level of the battery during said wait period.

9. The method as claimed in claim 1 and further comprising the step of: measuring a voltage of the battery after said charge pulse and after said pack of pulses to determine whether charging of the battery needs to be reduced.

10. The method as claimed in claim 9 wherein said step of determining the charge of the battery comprises the steps of: measuring the open circuit voltage of the battery after at least several packs of high frequency pulses have been applied; and comparing sequential ones of said measured open circuit voltages to generate a signal indicative of whether a reduction in charge current is required.

11. The method as claimed in claim 2 wherein the frequency of said second pack of pulses is in the range from about 100 Hz to about 1M Hz.

12. The method as claimed in claim 11 and further comprising the steps of: measuring the internal resistance of the battery and producing a resistance signal indicative thereof; deriving from said impedance measurement and with said internal resistance measurement of the battery an indication of the capacitance of the battery.

13. An apparatus for charging a battery comprising: means for applying charge cycles to the battery wherein the charge cycles are formed of at least one charge pulse and one discharge pulse and a wait period following the discharge pulse during which wait period there is neither a charging nor a discharging of the battery; and means for altering the wait period with a series of alternating current pulses selected to enhance a mixing of the electrolyte.

14. The apparatus for charging a battery as claimed in claim 13 wherein said means for altering said wait period further includes means for altering said wait period with a series of high frequency pulses selected to enable a measurement of the internal impedance of the battery; and means for determining the capacitance of the battery.

15. The apparatus as claimed in claim 14 and further including: means for determining whether the battery capacitance has changed or remained substantially the same; means for terminating the battery charging cycles when the battery capacitance has remained substantially the same.

16. The apparatus as claimed in claim 13 wherein said means for applying charge cycles comprises means for generating several discharge cycles during a charge cycle and following a charge pulse and separated by wait periods; and means for altering said wait periods with mixing current pulses.

17. The apparatus as claimed in claim 16 and further comprising means for altering wait periods with alternating high frequency current pulses of a frequency selected to enable the determination of the internal impedance of the battery; and means for measuring the internal impedance of the battery during said latter high frequency pulses.