WO1991007000A1

WO1991007000A1 - A method and a charger circuit for the charging of alkaline manganese dioxide-zinc rechargeable batteries

Info

Publication number: WO1991007000A1
Application number: PCT/HU1989/000049
Authority: WO
Inventors: Gábor SZÓRÁDY; Gabriella SZÓRÁDYNÉ OSTER; Karl Kordesch; Mihály LANTOS
Original assignee: Szorady Gabor; Szoradyne Oster Gabriella; Karl Kordesch; Lantos Mihaly
Priority date: 1988-04-29
Filing date: 1989-10-26
Publication date: 1991-05-16
Also published as: HU49757A

Abstract

Method for the charging of alkaline manganese dioxide-zinc rechargeable batteries, in which predetermined charging and discharging currents are periodically passed to the battery during alternating sequences of charging and discharging sections, in which the charging sections are substantially longer than the discharging sections, in which in the charging sections a predetermined bias current and steep charging pulses with predetermined width and intensity superimposed thereon are passed to the battery, in which the pulses follow each other with a predetermined time delay therebetween, in each of the discharging sections the battery is loaded with a predetermined bias loading current and steep discharging pulses with predetermined width and intensity superimposed on the bias loading current, in each of the discharging sections the discharging process is stopped after a period of at least about 3, preferably 5 seconds for a predetermined stop period, the voltage of the battery is sensed during the stop periods, the charging and discharging processes are both stopped if the sensed voltage reaches a predetermined maximum value, and the energy of the pulsated charging current is changed as a function of the sensed battery voltage so that the energy decreases with increasing voltage. A charger circuit for carrying out the method comprises a charger and a discharger generator (CHG, DCHG) coupled to the battery, four timers (T1...T4), a window comparator with bistable output properties (CP) which is connected to stop inputs of the generators and enabled by the first timer, the comparator senses the resistance free battery voltage in a time period in the respective discharging sections defined by the fourth timer (T4), the timers are arranged to provide the pulsated charging and discharging current waveforms defined in the method.

Description

A METHOD AND A CHARGER CIRCUIT FOR THE CHARGING OF ALKALINE MANGANESE DIOXIDE - ZINC RECHARGEABLE BATTERIES

The invention relates to a method and a charger circuit for the charging of alkaline manganese dioxide - zinc rechargeable batteries .

The methods for charging batteries vary in many respects, depending on the type of batteries, their constructions and their applications. However, all methods need to detect the ful¬ ly charged state. A good end-of-charge detection method exists with cathode-limited Ni-Cd and Ni-Zπ cells which are charged in a constant current mode and use the eπd-of-charge voltage jump as indicator.

Because of their specific properties rechargeable alkaline Mn0₂-ziπc cells should be charged in a different way than other rechargeable batteries should.

Various properties of such batteries were discussed in the Seventh Australian Electrochemistry Conference, Feb. 14-19, 1988 in the presentation of K. Kordesch, D. Freeman and K. Tomantsch- ger entitled "The Technology of the Rechargeable Alkaline Zπ- Mn0₂ Battery". Regarding the recharging of such batteries the presentation pointed out that a resistance free taper charger seemed to be optimum.. It was added that research and development should be directed to charging methods capable of more efficient elimination of the generation of gases during the recharging process and of decreasing the formation of dendrites at the zinc electrode. The primary object of the present invention is to provide a charging method and a charger which complies with these require¬ ments .

According to the invention a method has been provided for the charging of alkaline manganese dioxide-zinc rechargeable batteries, which comprises the steps of periodically passing predetermined charging and discharging currents to the battery during an alternating sequence of charging and discharging sectioπs, in which the charging sections are substantially longer than the discharging sections, in which in each of the charging sections a predetermined bias current and steep charging pulses with predetermined width and intensity superimposed thereon are passed to the battery,in which the pulses follow each other with a predetermined time delay therebetween, in each of the discharging sections the battery is loaded with a predetermined bias loading current and steep discharging pulses with predetermined width and intensity super- imposed on the bias loading current, in each of the discharging sections the discharging process is stopped after a period of at least about 3, preferably 5 seconds for a predetermined stop period, the voltage of the battery is sensed during the stop periods, the charging and discharging processes are both stopped if the sensed voltage reaches a predetermined maxium value, and the energy of the pulsated charging current is changed as a function of the sensed battery voltage so that the energy decreases with increasing voltage.

In a preferable embodiment the decreasing step of the charging current is carried out in such a way that the time de¬ lay between the charging pulses is increased as a function of the battery voltage.

In a further embodiment said time delay between the discharging pulses is also increased as a function of the battery voltage.

The lifetime of the batteries will be optimum if the periodical sequence is stopped when the sensed voltage reaches a predetermined first threshold level being substantially between about 1.68V and 1.78V. Automatic charging can be obtained if the periodical sequence is started again when the battery voltage drops to a predetermined second threshold level being substantially between about 1.5V and 1.65V.

In case of several types of alkaline manganese dioxide-zinc batteries the cyle lifetime increases if the intensity of the charging pulses are at most about 3 times as high as the current I,g designating one tenth of the amper-hour capacity of the battery. In such cases it is preferable if the intensity of the charging bias current is at most about half as high as the charging pulses.

For certain sensitive types of batteries it is preferable i the intensity of the discharging pulses is at most about 1.5 times as high as the current I,_Q designating one tenth of the amper-hour capacity of the battery and the discharging bias current is at most one half of the discharging pulses.

According to the invention a charger circuit has also been provided for charging alkaline manganese dioxide- zinc re¬ chargeable batteries, which comprises a controllable charging current generator which in enabled state is capable of providin a first or a second predetermined output charging current de¬ pending on the value of a binary control signal coupled to puls input thereof, a controllable discharging current generator which in enabled state is capable of providing a first or a second predetermined output discharging current in response to the value of a binary control signal, the outputs of the generators are coupled to the battery to be charged, a first timer means coupled to enable and inverted enable inputs of the generators, respectively, for alternatingly enabling and dis¬ abling the generators so that the enable time of the charging generator is at least five times as long as that of the dis¬ charge generator, a second timer means activated by the first timing means providing a timing which corresponds to the time period after which the resistance free battery voltage gets largely independent from the ambient temperture, this timing is shorter than the enabling time of the discharging generator and being about at least 3-5 seconds long, a third timer means coupled to binary control pulse inputs of the generators for generat g periodically the binary signals as pulses substan¬ tially shorter than the enabling time of the dischargir.] generator, and comparator means with bistable output properties, the comparator means is enabled by the output of the second timer means and it has a voltage sensing input coupled to the battery, a pair of reference inputs connected to reference sources defining respective minimum and maximum permitted voltage thresholds, the output of the comparator means is concted to stop inputs of both of the generators for allowing the alternating charging and discharging processes if the battery voltage in the discharge periods following the expiry of the second timing lies between the two reference values.

A preferable embodiment comprises a fourth timer means enabled by the output of the second timer means providing a timing whithin the discharge periods that is at least as long as the time required for the transient phenomena in the battery after a discharging process following thetiming of the second timing means and the output of the fourth timing means is coupled to stop input of the discharging generator to disable the discharging process within the fourth timing to allow thereby a load free state for the battery when the comparator means is in enabled sensing state.

In a preferable embodiment the timing provided by the third timer means depends on the value of the battery voltage so that the periods within which the binary signal corresponds to the generation of larger current is increases with increasing battery voltage.

It is preferable if the timing of the fourth timer means is about 2 seconds.

The invention will now be described in connection with pre¬ ferable embodiments thereof, in which reference will be made to the accompanying drawings. In the drawing:

FIG. 1 shows first preferable current forms;

FIG. 2 shows characteristic waveforms of the circuit arrangements of FIG. 8; FIG. 3 shows the battery voltage during a full charging period with somewhat distorted time scale;

FIG. 4 shows the battery voltage during a period of a charging and discharging sequence in enlarged time scale; FIG. 5 shows second preferable current forms; FIG. 6 shows a typical discharging diagram;

FIG. 7 shows the charging capacity versus charging time diagram for different charging times in different cycles; and FIG. 8 shows the functional block diagram of the charger according to the invention.

FIG. 1 shows one period of a periodic charging current form applied according to the invention to rechargeable alkaline man¬ ganese dioxide-zinc cells. Each of these periods consists of a charging and a subsequent discharging section. In the example the charging section was about 60s long and it comprises a constant current bias I. and a plurality of steep current pulses superimposed on the bias. In the example the narrow pulses were adjusted to have a predetermined constant width of 200 ms. The delay time t , between subsequent pulses forms an adjustable pa¬ rameter, it depends on the value of the resistance free battery voltage U_Rf measured in each period in predetermined moments.

The maximum charging current Ic.hm is adjusted to be 2.4 times as high as the tenth capacity of the battery. In the examples C type alkaline manganese dioxide-zinc batteries were examined which had a nominal ca ^rpacity ^J of 2.5 Ah,' therefore I„ch_mm ^was 600 mA• The bias current I. was adjusted to be equal to the half of the peak current i.e. to 300 mA.

At the end of each charging section a discharging section of about 10s was started. The discharging section consists mainly of two parts each being about 5 s long. In the first part a dis- charging bias current I ,, of about 0.2.I,_Q = 50mA was used, and discharging pulses of 200 ms were superimposed thereon. The peak current I . of the discharging pulses is adjusted to 0.8.1, _Q i.e. to 200 A. The time delay t. between the pulses is equal to that defined during the charging section. When the time delay t , is varied as ^"a function of the resistance free sampled voltage ^URfree' ^{the cnaπ}9^e affects the repetition period of the pulses both in the charging and discharging sections.

In the second part of each discharging section the battery was not loaded at all. At the starting moment of this part (in 1 θr 2 ms following the disruption of the loading current from the battery) the battery voltage was measured and this value consti¬ tuted the resistance free battery voltage. The disruption of the curreπt eliminates the voltage drop which would otherwise occur in loaded state due to the series resistance of the battery.

In FIG. 1 a 0.5 s disruption of the current is shown in the starting portion of the first part of the discharge section. This was used additionally for determining the resistance free battery voltage just after the charging process.

FIG. 3 shows the changes of the delay time t , between the superimposed pulses as a function of the measured resistance free voltag^ae UrR,f-ree. The delay ^J time t a. keep ^s the starting^a extreme values until the cell voltage is below 0.7 V (FIG.3 section A), whereafter the period time starts to increase (FIG. 3, sections B and C). The initially higher charge rate gradu¬ ally decreases and in average it supplies an energy which is close to the charging energy of conventional resistance free chargers kept ideal for alkaline manganese dioxide-zinc batteries.

In FIG. 3 the time scale is distorted as regards the pulse times. The battery voltage is at maximum during the pulse peaks in the charging periods and at minimum at load peaks during discharge periods. The two dashed line corresponds to the envelope of these two extreme voltage values. The dot-dash line between them corresponds to the resistance free voltag^σe U_DRi„ree

The diagram illustrates how the rate of the charging pulses decreases as the voltage increase. The rate of the discharge pulses also decreases with time. The resistance free sampled battery voltage gradually increases as the charging process goes on, and this increase is sufficiently definite that the charging process can be finished if this voltage reaches a maximum limit value which should lie about 1.72 -1.75 V. When the charging process has stopped, the unloaded battery voltage starts decreasing. The charger according to the inven¬ tion starts operating again if the battery voltage drops below lower limit value which is preferably about 1.55-1.6 V. The charging process is now shorter and the increase in the sampled voltage is faster compared to the charging of a discharged bat¬ tery. The charger stops operating if the voltage reaches the up per limit. This process goes on repeatedly, however, the charg- iπg periods follow each other with higher and higher delays and their durations decrease in discrete steps.

FIG. 4 shows an enlarged view of the battery voltage during a period consisting of a charging and a discharging section as drawn by a multi-channel voltage recorder. During the charging section respective voltage peaks correspond to the current peaks of Fig. 1 and the lower voltage values correspond to the biasing charging current between the pulses. In opposite sense similar changes in the voltage can be seen in the discharging section, however, the envelope curves of the deep voltage peaks and of the bias values are exponentially decreasing when the charging section has finished. It has been experienced that the form of this decreasing curve depends on the temperature and this depen¬ dency is higher at the starting moments of the discharging sec- tion whereafter it rapidly decreases. Our measurements have de¬ monstrated that about 5 seconds after the beginning of the dis¬ charging section the battery voltage is practically independent from the ambient temperature. This explains the choice of such a long waiting period before the sensing of the resistance free voltage. In any given temperature the resistance free voltage sampled immediately after the end of the charging periods and the voltage sampled 5 seconds after such moments vary in the same way during the whole charging process. It worth mentioning that the battery voltage measured during the existence of the bias load varies similarly during the whole charging process, so that the end of charging control and the control of the pulse parameters can be based on such a voltage as well if sampled about five seconds following the end of the respective charging sections. Such samples possess a certain error which depends on the internal resistance of the battery. The charging of batte¬ ries with differing internal resistance based on such samples will not be as uniform as in case of a control based on a vol¬ tage measured by the resistance free technique, since the mea¬ sured voltage depends on the actual loading current and on the non-uniform internal resistance. In several applications, howe¬ ver, the inaccuracy caused by this phenomenon lies well within a permissible tolerance range. In addition to the example with the charging current forms shown in FIG. 1 tests with an other current form were also car¬ ried out. The current versus time diagram thereof is shown in FIG. 5. In this example the delay time t . between the pulses superimposed on the bias current was varied with changing re¬ sistance free voltage. In this current form both the duration and the amplitude of the charging pulses were constant. The pulse width was 400 ms, while the pulse amplitude I _hm was 600 mA which corresponds to 2.4 -. The bias current I, was 60 mA i.e. 0.241, _Q. In the discharging section the bias load I ., was 50 mA i.e. 0.2.1,,, and the amplitude of the loading pulses was 300 mA i.e. 1.2.1,,, and the width thereof was also 400 ms. Obviously, the charging of the battery with such current forms is a more intensive process than with the current forms of FIG. 1. The active loading time lasted for about five seconds in each discharge section followed by another 5 seconds of rest without load just as in case of FIG.l. The voltage of the battery was sensed just as in the previous example.

With the current form shown in FIG. 5 two sets of C size manganese dioxide-zinc rechargeable batteries were charged. In each sets eight batteries were connected in series. Respective channels of a multi-channel recorder were connected to each of the batteries. The charging was carried out until the resistance free voltage reached a value of 1.72 V/cell. In the first cycle the voltage reached this threshold level in 7,25 hours. After a rest of about 2 hours the sets were discharged by a constant current of 190 mA. The discharge took place until the battery voltage dropped to 0.7 V. Actually, due to the series connection of 8 batteries the discharge process was disconnected when the voltage dropped to 8 . 0.7 =5.6 V. The discharge time was 8.54 hours and the average capacity was 1.62 Ah. During this measure¬ ment it has been noticed that the batteries were not uniform. A typical discharging diagram is shown in FIG.6. In a second charging cycle the charging time increased to 7,5 hours and the discharging time dropped to 8,5 hours which corresponds to a ca¬ pacity of 1.61 Ah. In a third cycle the charging time was 7.8 h. In that case the batteries were discharged by a load of 60 ohms coupled to the respective sets. With constant loading resistance the discharge time was 10.1 h. In that case the capacity was not calculated.

The temperature of each of the batteries was monitored dur- ing the charging process but no change was sensed. In contrast to this fact the same batteries were previously discharged and charged for a cycle with a conventional taper charger that al¬ lowed a constant voltage of 1.72 V on the batteries, in which the maximum charging current was limited to about 1 A, and in the first hours of the charging process an average increase of 3°C was sensed compared to the ambient temperature. The taper charged was left on the batteries for about 28 hours whereafter the batteries were discharged and an average capacity of 1.84 Ah was measured. The tests with the current forms shown in FIG. 5 were fi¬ nished at this stage. Now the current form shown in FIG. 1 was applied to the two sets and a charging process took place which was the fourth cycle of he batteries with the pulse charging. The threshold for the end of charging was set to V-_{> fτeβ}--- - H ^v- The charging process lasted through 7.5 hours, whereafter a rest of 2 hours and a discharging with 190 mA constant load was made. The discharge time was 8.4 h and the capacity was 1.596 Ah.

In order to make the capacity test more uniform, those bat¬ teries were selected from the two sets which behaved similarly, and a new set of eight substantially uniform batteries was as¬ sembled. With this new set a new (fifth) cycle was started. In this cycle the charging was finished after 7,5 hours. The charg¬ er was kept in on state for further 2.5 hours during which the charging switched on automatically when the voltage dropped to 1.6 V for two short periods as shown in FIG. 3. The following discharge with the 190 mA load lasted 9.5 hours and the capacity was 1.8 Ah. In the next (sixth) cycle the charging stopped at 7.5 hours and in this moment the charger was switched off and after a rest a discharge period was started. The discharge time was 7.5 h and the capacity was 1.425 Ah. In the following (se¬ venth) cycle the threshold level was increased to 1.74 V and the charging period lasted 10 hours when this new threshold was reached. The following discharge period lasted 9.58 hours and the capacity was 1.82 Ah. In the eighth cycle the threshold le¬ vel was increased to 1.76V and 11.5 hours were required for the sampled voltage to reach this threshold. After a 2 hours rest the discharge period lasted 10.26 hours.

The results of the tests with identical charging current forms but with different charging time are diagrammatically il¬ lustrated in FIG. 7, which shows relative capacity versus asso¬ ciated charging time curves. In this figure 100% corresponds to the 2.5 Ah nominal capacity. The 6th cycle with the 7.5 h charg¬ ing time corresponds to point 3, while the 7th and 8th cycles to points G and H respectively.

According to measurements carried out with a resistance free taper charger on similar batteries it was obtained that in the first cycle the 50% capacity can be reached within 5.5 to 8 hours of charging. These tolerance ranges are indicated by points A and B. Similarly, the ranges for the two-third capacity were between 9 and 13 hours that correspond to points C and D. The area between the dashed curves connecting the points A-C and B-D designate the charging time associated with the charging of the battery with the percentage of the nominal capacity shown in the vertical axis. It is also experienced that such times are increasing with increased number of cycles. The point E and F correspond to points B and D but interpreted for the 20th cycle. Our measurements were carried out a least in the 7th to 9th cycles (a starting cycle was made by the taper charger) and the points 3 G C are well above the expected area even for the firs cycle in which the highest values can be obtained. FIG. 7 indi¬ cates that the charging with the current forms according to the invention provide a high charging efficiency.

For providing lighter charging conditions to the batteries the threshold level was re-adjusted to 1.72 V and the delay tim t . between the charging pulses was increased by about 40% (FIG. 2 corresponds to these increased delay times). The charging and discharging times as well as the associated capacity values for the next three (9th to 11th) cycles are summarized in Table 1. cycle 9th 10th 11th

charging time (h) 16.45 16 . 25 16 . 66 discharging time (h) 8.55 8 . 46 8 . 63 capacity (Ah) 1.625 1 . 607 1 . 64

From the above examples one can see how the charging time varies with the different waveforms. It can well be appreciated that the optimum current forms can and should be adjusted in ac- cordance with the requirements of the user and the properties of the actual batteries to be charged. It is believed that the value of the biasing current during the charging sections should be about as high as 100 mA preferably 150 mA at least during the first few hours of the charging process, since the active state of the zinc electrode can thus be reached more readily. The steep pulses superimposed on the bias assist in pumping more energy in the battery and the steep transients contribute slow¬ ing down the formation of dendrites which increases cycle life. The use of the resistance free voltage samples largely indepen- dent from the ambient temperature and from the internal resist¬ ance of the individual batteries for establishing the end of charge periods and for the adjustment of the pulse parameters result in a largely uniform charging. The automatic start-stop operation prevents the batteries from gas forming if the charger is left in a switched on state and the repeated cycles preserve (sometimes increase) the storage capacity without causing any harm to the batteries or consuming much energy.

FIG. 8 shows the block diagram of a circuit arrangement by which the specific current forms required for the charging pro- cess according to the invention can be generated. For facilitat¬ ing the understanding the exa plary timing data shown in FIG. 1 will be used for the explanations.

The circuit uses three timers Tl, T2 and T3. Timer 1 is capable of generating a pulse train with 60s on-time and 10s off-time. These pulses determine the alternating charging and discharging sections so that the output of timer Tl is coupled directly to enable input of a charging current generator CHG and to an inverted enable input EN of a discharging current generator DCHG. The current generators CHG and DCHG have respective stop input st which, when being activated, disrupt their outputs connected both to a terminal of the battery to be charged. The discharge generator DCHG provides a first constant loading current for the battery when being enabled and a second constant load when it receives a second enable condition at in¬ put tp. The charging current generator CHG operates in a similar way i.e. it has also an input tp. The first current of the charging generator CHG generated during the disabled state of the input tp is related to the second current of the same generator occurring in the enabled state of the input tp. The first current is e.g. the half of the second one. The second timer T2 receives the output of the first timer Tl and generates a pulse which has a low logical state and it is 5s long. This pulse starts with the leading edge of the output pulse of the first timer Tl which separates the charging and discharging sections. The output pulse of the second timer T2 is coupled to enable input EN of a comparator CP realized by a window comparator and to enable input EN of the fourth timer T4. The comparator CP has bistable properties i.e. it keeps an output state reached by crossing a first threshold level in a direction until the other threshold is crossed in the other direction.

In the first 5 seconds of every discharging section the output of the second timer T2 is at logical zero (low) state (FIG. 2 curve T2). In 5 seconds the output of the second timer T2 goes high which enables the comparator CP and starts the timing of the fourth timer T4 (FIG. 2, curves CP and T4). In th timing defined by the fourth timer T4 which is typically 2 seconds, the discharging generator DCHG is disabled by receivin a stop pulse via its stop input st and the comparator CP determines whether the battery voltage lies within the window defined by two reference voltages refl and ref2 being e.g. 1.72 and 1.6 V, respectively (FIG. 2 curves T4, U_RFREE and DCHG). This can be considered as the resistance free voltage of the battery. In the enable time of the comparator CP the battery voltage sensed by the comparator CP takes maximum when there is no loading c -rent i. e. when the fourth timer T4 is on state and stops the discharging current. This period is shown by the hatched area in the U_Rp_Rpp curve of FIG. 2.

The third timer T3 generates 200 ms wide pulses with a repe¬ tition time between them corresponding to the value of the voltage of the battery. The output of the third timer T3 cont¬ rols the second enable inputs tp of both of the generators CHG and DCHG.

When a battery is connected to the battery terminals and the voltage thereof is below 1.6 V, the comparator CP changes its output state and disconnects the stop signal from the stop inputs st of the generators, whereby the alternating sequence of charging and discharging pulses will start to exist, thus the battery will be charged. If the sensed voltage exceeds the 1.72 V threshold level (which can occur in the resistance free sens¬ ing period during the timing made by the fourth timer T4), the comparator CP turns over and stops both generators CHG and DCHG by controlling their stop inputs st.

The circuit arrangement can thus provide all conditions re¬ quired for the generation of the current forms shown in FIGs 1 and 5.

Claims

Cl aims

1. Method for the charging of alkaline manganese dioxide- zinc rechargeable batteries, comprising the steps of periodical¬ ly passing predetermined charging and discharging currents to said battery during an alternating sequence of charging and discharging sections, in which said charging sections being sub¬ stantially longer than said discharging sections, characterized in that in each of said charging sections passing a predetermin¬ ed bias current and steep charging pulses with predetermined width and intensity superimposed on said bias current, in which said pulses follow each other with a predetermined time delay therebetween, in each of said discharging sections loading said battery with a predetermined bias loading current and steep discharging pulses with predetermined width and intensity super¬ imposed on said bias loading current, in each of said dis- charging sections stopping said discharging process after a period of at least about 3, preferably 5 seconds for a predeter¬ mined stop period, sensing the voltage of the battery during said stop periods, stopping the charging and discharging pro¬ cesses if said sensed voltage reaches a predetermined maxium value, and changing the energy of said pulsated charging current as a function of said sensed battery voltage so that the energy decreases with increasing voltage.

2. The method as claimed in claim 1, characterized in that said decreasing step of the charging current is carried out by increasing said time delay between the charging pulses as a function of said battery voltage.

3. The method as claimed in claim 1, characterized in that said time delay between the discharging pulses is increased as a function of said battery voltage.

4. The method as claimed in claim 1, characterized in that said periodical sequence is stopped when said sensed voltage reaches a predetermined first threshold level being substantial¬ ly between about 1.68V and 1.78V.

5. The method as claimed in claim 4, characterized in that said periodical sequence is started again when the battery voltage drops to a predetermined second threshold level being substantially between about 1.5V and 1.65V.

6. The method as claimed in claim 1, characterized in that the intensity of said charging pulses being at most about 3 times as high as the current I,_n designating one tenth of the amper-hour capacity of said battery.

7. The method as claimed in claim 6, characterized in that the intensity of said charging bias current being at most about half as high as said charging pulses.

8. The method as claimed in claim 1, characterized in that the intensity of said discharging pulses being at most about 1.5 times as high as the current I,« designating one tenth of the amper-hour capacity of said battery and said discharging bias current being at most one half of said discharging pulses.

9. Charger circuit for charging alkaline manganese dioxide- zinc rechargeable batteries, characterized in that said circuit comprises a controllable charging current generator (CHG) which in enabled state is capable of providing a first or a second predetermined output charging current depending on the value of a binary control signal coupled to pulse input (tp) thereof, a controllable discharging current generator (DCHG) which in enabled state is capable of providing a first or a second predetermined output discharging current in response to the value of a binary control signal, the outputs of said generators (CHG and DCHG) are coupled to the battery to be charged, a first timer means (Tl) coupled to enable and inverted enable inputs (EN, EN ) of the generators, respectively, for alterπatingly enabling and disabling said generators so that the enable time of the charging generator (CHG) is at least five times as long as that of said discharge generator (DCHG), a second timer means (T2) activated by said first timing means (Tl) providing a timing which corresponds to the time period after which the resistance free battery voltage gets largely independent from the ambient temperture, this timing is shorter than the enabling time of the discharging generator (DCHG) and being about at least 3-5 seconds long, a third timer means (T3) coupled to binary control pulse inputs (tp) of said generators (CHG and DCHG) for generating periodically said binary signals as pulses substan¬ tially shorter than the enabling time of said discharging generator (DCHG), and comparator means (CP) with bistable output properties, said comparator means is enabled by the output of said second timer means (T2) and has a voltage sensing input coupled to said battery, a pair of reference inputs connected to reference sources defining respective minimum and maximum permitted voltage thresholds, the output of the comparator means is concted to stop inputs (st) of both of said generators (CHG and DCHG) for allowing the alternating charging and discharging processes if the battery voltage in the discharge periods following the expiry of the second timing lies between said two reference values.

10. The charger as claimed in claim 9, characterized by comprising a fourth timer means (T4) enabled by the output of said second timer means (T2) providing a timing whithin said discharge periods that is at least as long as the time required for the transient phenomena in said battery after a discharging process following the timing of said second timing means (T2) and the output of the fourth timing means (T4) is coupled to stop input (st) of said discharging generator (DCHG) to disable the discharging process within said fourth timing to allow thereby load free state for the battery when said comparator means (CP) is in enabled sensing state.

11. The charger as claimed in claims 9 or 10, characterized in that said timing provided by said third timer means (T3) depends on the value of said battery voltage so that the period within which said binary signal corresponds to the generation o larger current is increases with increasing battery voltage.

12. The charger as claimed in claim 11, wherein the timing of said fourth timer means (T4) is about 2 seconds.