Fast Battery Charging System and Method

NiCad (nickel-cadmium) battery technology has been employed successfully in portable hand-held applications for many years. Photographic equipment, power tools, data terminals, personal radio transceivers, and pagers commonly utilize NiCad batteries as a power source. The charging systems that have been provided with these products have ranged from a simple transformer/rectifier type to rather complex systems to monitor and control the charging function. An increasing need is the ability to charge NiCad batteries quickly. To reliably and efficiently charge NiCad batteries at high rates requires careful control of the charging operation to avoid damage to the cells, particularly under extreme ambient temperature conditions.

The NiCad charge cycle consists of two basic parts: the coulombic portion and the overcharge portion. The coulombic portion of the charge cycle is characterized by the fact that most of the charge that is applied to the battery is stored in the form of electrochemical energy. This portion of the charge cycle is slightly endothermic, consequently high charge currents may be applied during this time without resulting in temperature increase. Most of the available battery capacity is stored during the coulombic portion of the charge cycle. The overcharge portion of the charge cycle is characterized by the fact that most of the applied charge current causes generation of oxygen gas at the positive electrode of the NiCad cell, with only a relatively small amount of charge actually being stored in the cell. The released oxygen chemically recombines with cadmium at the negative electrode of the cell which serves to equalize the internal pressure of the cell. If the overcharge rate is too high, the rate of oxygen recombination may be insufficient to prevent excessive internal pressure and cell venting, which drastically reduces the useful life of the cell.

The most critical factors in determining the maximum allowable charge current that may be safely applied to a NiCad battery are temperature and state of charge. At low temperatures the oxygen recombination rate is significantly reduced which limits the allowable overcharge current that may be applied without venting the cells if they are fully charged. At high temperatures the heat released by the oxygen recombination reaction may cause excessive cell temperature to be experienced leading to premature failure of the plate separator material and subsequent short-circuiting.

If the battery is fully discharged, minimal oxygen generation will occur until the battery nears the fully charged condition. If the battery is nearly fully charged, it will quickly enter the overcharge condition and begin oxygen generation. The difficulty lies in accurate determination of the previous state of charge to avoid damage to the battery.

Additionally, battery chargers sometimes use a constant-current source to charge batteries. If a varying load is placed in parallel with a battery under charge, the current delivered to the battery may not be sufficient to charge it in a specified time. If the load is too great, the battery may actually be discharged. This is particularly true if the constant current source producing the charging current does not have the capability of adequately servicing both battery recharging and varying load, or with a varying load is too unpredictable in magnitude or duration to insure it will not adversely affect battery recharging.

One solution to this problem is to provide a current source which is either variable or without question can produce enough current to handle all situations regarding battery charging and varying load. However, a significant problem with these types of current source is the possibility of excessive heat dissipation which can adversely affect either the recharging process, operation of the device, or both.

Still further, there is room for improvement in recharging batteries of the type being discussed. A variety of factors can interfere with the effective and efficient charging of batteries. An improvement in the flexibility and efficiency of recharging would be desirable in the art.

A basic objective of the present invention is to provide a monitoring and control system which provides for effective fast charging. In order to avoid damage to the battery, the system automatically tests to determine the initial state of charge and selects the charging rate accordingly.

In a preferred implementation, a microprocessor receives measures of battery temperature and battery terminal voltage and selects an optimum charging rate. A unique feature of the preferred system is its ability to provide a safe controlled charge to a NiCad battery without sensing the current flow through the battery directly. This allows the effective battery impedance to be held at a minimum, thereby delivering the maximum available battery energy to the load.

Further features of the preferred system relate to the automatic processing of batteries subject to temperature extremes beyond the range where rapid charging operation is permitted.

An additional objective or feature of the invention is to accommodate a varying load and effective charging of the batteries while connected to a constant current charging source. A further feature is the ability to remove circuitry from the body of the device holding the batteries which may generate heat which is detrimental to the device or batteries.

A still further feature of the present invention relates to a flexible yet very efficient way to recharge batteries so that effective recharging can be accomplished regardless of the nature of the environment, the load on the device, or other factors involved with the device or the charging current source.

The invention will now be described, by way of example and not by way of limitation, with references to the accompanying sheets of drawings; and other objects, features and advantages of the invention will be apparent from this detailed disclosure and from the appended claims.

FIGS. 1-5, 9A, 9B, 10-15, 16A, 16B and 17 and the brief description thereof are incorporated herein by reference to U.S. Ser. No. 876,194, now U.S. Pat. No. 4,709,202 issued Nov. 24, 1987.

FIG. 18 is an electric circuit diagram for illustrating a preferred embodiment of battery charging current control system in accordance with the present invention;

FIG. 19 shows an exemplary current pulse waveform which may correspond with a maximum battery charging rate in a substantially linear operating range for an exemplary control system in accordance with FIG. 18;

FIG. 20A shows selected control pulses which may be generated during control of battery charging current in the control system of FIG. 18, and FIG. 20B shows respective corresponding battery charging current pulses on the same time scale with vertically aligned portions of the waveforms of FIGS. 20A and 20B occurring at the same time;

FIG. 21 is a diagrammatic view of use in explaining the aliased sampling of actual current pulses in the battery charging circuit, and illustrating the case where thirty-two samples form a complete sampling cycle;

FIG. 22 is a block diagram for illustrating exemplary sampling circuitry for association with the V sense input of the processor means of FIG. 18;

FIG. 23 illustrates a battery conditioning system as described at col. 17, lines 51-68 of the incorporated U.S. Pat. No. 4,455,523, and wherein two-way communication may be established between memory means associated with a portable unit comprised of rechargeable battery means, and non-portable central computer controlled conditioning station;

FIG. 24 shows a battery conditioning system wherein a battery identifying circuit element directly controls the set point of a battery charging circuit to determine a battery charging parameter, e.g., battery charging current;

FIG. 25 shows a highly integrated semiconductor device, e.g. for implementing the system of FIGS. 18-22;

FIG. 26 is an electric circuit block diagram showing a battery conditioning system wherein a preferred hand-held terminal unit contains battery parameter sensing means and computer operating means for optimizing battery charging current as supplied by an external circuit (which may correspond with a standardized circuit such as shown in FIG. 24 applicable to a complete family of hand-held terminals);

FIG. 27 shows a new RF terminal unit including charge control and temperature transducer outputs as in FIG. 26, and also incorporating an interface for coupling with a local area network so as to enable batch transmission of data to and from the RF terminal;

FIG. 28 is a block diagram illustration of a preferred fast charging system in accordance with the present disclosure;

FIGS. 29A and 29B show a flow chart for illustrating a preferred fast charging algorithm for use with the microprocessor of FIG. 28;

FIG. 30 is a circuit diagram illustrating a preferred arrangement for the automatic identification of various types of batteries which may be associated with a fast charging system according to FIGS. 28 and 29;

FIG. 31 is a diagram useful in explaining certain steps of the flow chart of FIG. 29;

FIGS. 32A and 32B are a circuit diagram for illustrating an exemplary implementation of the block diagram of FIG. 28;

FIG. 33 illustrates a plot showing a maximum permissible overcharge rate for fast charge cells as a function of cell temperature, and provides information which may be incorporated in the programming of the system of FIGS. 28-32 for establishing an optimum value of charging current (Ichg) during sustained overcharging;

FIG. 34 illustrates a plot of maximum charge rate for fast charge cells as a function of cell temperature to show exemplary data which may be used for the system programming FIGS. 28-32 for establishing an optimum value of charging current for a battery which has not yet reached the overcharge state;

FIG. 35 shows a plot of measured battery pack temperature as a function of time for a previously fully charged enclosed battery pack where an overcharge current (Ichg) of three hundred milliamperes is applied and the ambient temperature T.sub.A is about fourteen degrees celsius (14.degree. C.) and also illustrates successive approximate slope values for selected successive time intervals; and

FIG. 36 is a plot of measured battery pack temperature as a function of time for the case of an enclosed battery pack which is initially at a much lower temperature than ambient temperature; specifically the battery pack was initially at a temperature of about minus fifteen degrees celsius (-15.degree. C.) while the ambient temperature was about twenty degrees celsius (20.degree.), the battery pack receiving only a small charging current of six milliamperes; FIG. 36 also shows successive approximate slope values by means of straight lines covering successive equal time intervals of 600 seconds;

FIG. 37 shows an improved procedure for carrying out fast charging and maintenance of a nickel-cadmium battery pack, for example in conjunction with a microprocessor system as shown in FIG. 28, the variable P temp, V batt and Atemp of FIG. 28 being represented by PT, PV and AT in FIG. 37;

FIG. 38 is a schematic diagram for illustrating a charge current regulator circuit such as indicated at 28-20 in FIG. 28; and

FIG. 39 shows the voltage to current transfer function for the circuit of FIG. 38.

FIG. 40 is a block diagram of a closed loop battery charging circuit to control the nature of the charging current to increase the efficiency of charging of the battery even under varying load.

FIG. 41 is an electrical schematic showing a charging current control scheme for efficient battery recharging where the recharging current source is located externally of the housing of the battery and associated device.

FIG. 42 is an electrical schematic similar to FIG. 41 showing specifically a circuit combination utilizing a closed loop feed back for a controlled transistor switch to control charging current to the battery.

FIG. 43 is a graph of a pulse width modulated signal illustrative of what could be generated by the circuitry of FIG. 42.

FIG. 44 is an electrical schematic with alternative circuitry to that of FIG. 42 for providing a pulse width modulated recharging current to the battery but where the pulses are modified from a generally square wave to a generally trapezoidal wave.

FIG. 45 is a diagram of a trapezoidal-shaped current pulse illustrative of what could be produced by the circuitry of FIG. 44.

FIG. 46 is a diagram of the voltage pulse generated in conformance with the current pulse of FIG. 45.

The detailed description of FIGS. 1 through 27 is incorporated herein by reference to the specification at col. 4, line 25, through col. 66, line 4, of the incorporated U.S. Pat. No. 4,709,202, by reference to the specification at col. 4, line 15, through col. 11, line 37 of the incorporated U.S. Pat. No. 4,737,702, and by reference to the specification of incorporated U.S. Pat. No. 4,885,523.

A block diagram of the charging system is shown in FIG. 28. A microprocessor 28-10 is preferably of a type that has analog to digital inputs such as 28-11 to 28-13 and digital to analog outputs such as 28-14 for interface to sensor and control-functions. Both the temperature of battery 28-15 and ambient temperature are sensed as indicated at 28-16 and 28-17 so that absolute and relative temperature measurements may be made. The terminal voltage of the battery pack is sensed as indicated at 28-12 so that charge trends may be determined. The charge regulator consists of a voltage controlled current source 28-20 whose output current (Ichg) is controlled by the level of the charge control signal at 28-14 from the microprocessor. A load 28-21 may or may not be connected during charge.

In the microprocessor 28-10, analog to digital (A/D) means and digital to analog (D/A) means are indicated at 28-10A and 28-10B. Preferably these means are integrated with the other components of the microprocessor as part of a monolithic unit or "chip" formed from a unitary substrate of semiconductor material.

With a charging system as shown in FIG. 28, an algorithm has been developed for fast charging of NiCad batteries as shown in FIG. 29 (sheets one and two). The charging function is initiated as represented by "start" at 29-1 e.g. by placing the battery 28-15 in the charger. The temperature sensor 28-16 is preferably in a housing 28-22 which together with battery 28-15 forms the battery pack 28-25. The sensor 28-16 is preferably of the type whose output is proportional to absolute temperature e.g. at the rate of ten millivolts per degree Kelvin (10 mv/.degree.K.). The microprocessor 28-10 tests for insertion of the battery in the charger by reading the temperature Ptemp as indicated at 29-2, and checking to determine if Ptemp shows a temperature greater than -100.degree. C.; see decision block 29-3. The decision at block 29-3 will be affirmative only if a battery pack has been inserted to provide a non-zero voltage on the Ptemp signal line 28-11.

Following determination of the presence of a battery pack 28-25 in the charger, the pack type must be identified as represented at 29-4 and 29-5 to allow for cells With different charge characteristics. In the case of an invalid reading of battery pack identity, the program may branch to an error subroutine as indicated at 29-5A. The identification of the type of battery inserted into the charger is a significant step in the battery processing operation since battery cells of specialized types may offer significantly higher capacity than ordinary NiCad cells, but they may require charging at lower maximum rates. Other cells may allow high charging rates at extreme temperatures. Future technology developments may offer new cell types with unusual charging parameters that may be accommodated by applying an appropriate charging algorithm. As shown in FIG. 30, a proposed method for identification of the pack type is to connect a shunt voltage regulator 30-10 to the battery pack 30-27 represented in FIG. 30. The shunt regulator may consist of a simple zener diode or it may be implemented with an active regulator e.g. as indicated in FIG. 30, depending on the number of different pack types that must be identified. Upon determination of the pack type a suitable one of a set of parameter tables may be selected that contains the appropriate values for charging the specified cell type, as shown at 29-6.

As indicated in FIG. 28, there may be a load 28-21 placed on the battery that requires current. Consequently, current supplied by a charger is shared by the load and the battery as shown in FIG. 31. If the load current Iload is larger than the charge current Ichg, the battery will provide the difference, resulting in further discharge of the battery rather than charging. To compensate for this effect, the algorithm senses the terminal voltage (Vh) of the battery (step 29-7, FIG. 29) and applies increasing charge current to battery in small increments (step 29-8) until the terminal voltage trend is positive (steps 29-9 to 29-11) meaning the battery is accepting charge rather than providing current to a load (see block 29-12).

While the absolute terminal voltage of a NiCad battery is a poor indicator of its condition, its trend is a good indicator of charging versus discharging if it is measured over a short enough time that the pack temperature remains relatively constant. Once the battery voltage trend is determined to be positive, the level of current required by the load (Iterm of block 29-12 corresponding to Iload, FIG. 31) is known, and may be added to the desired net battery current level (Ibatt, FIG. 31) to select the actual charge current (Ichg, FIG. 31).

Typical NiCad cell specifications call for charging in a temperature range of 0.degree. to 40.degree. C. Many of the products that utilize NiCad batteries may operate in environments with temperatures that range from -30.degree. C. to 60.degree. C. Consequently, it is possible that a battery pack may be placed in a charger immediately after being removed from either of these temperature extremes. If the pack temperature is greater than 40.degree. C., (see decision step 29-13), the pack must be "cooled" to no more than 40.degree. C. before charging may proceed. This is accomplished (as shown by step 29-14) by applying a charge current Ichg that equals the terminal load Iterm so that no net charge current is received by the battery and it may be by the ambient environment. If the battery pack is cold, it must be warmed to a temperature above 0.degree. C. This is carried out by steps 29-15 to 29-19. By applying a safe (low) charge current per the charge table of steps 29-17 and 29-19 (and FIG. 33), the pack my be warmed by the ambient environment of the charger.

Although charging may begin when the battery temperature exceeds 0.degree. C. according to the battery charging specifications, additional information is needed to determine the state of charge of the battery. The clearest method to determine whether a battery is fully charged is to detect the presence of the overcharge condition. In overcharge, the oxygen recombination reaction is highly exothermic which results in rapid heating of the battery. By applying twice the permissible substained overcharge rate as at steps 29-20 to 29-23 and monitoring cell temperature, it is possible to reliably determine that the overcharge condition has been reached. Unfortunately, when a cold pack is placed in a warm environment, there is a resultant temperature rise due to ambient warming that can actually occur at a rate faster than the heating due to the supply of a high value of overcharge current. Consequently, the only reliable means of detecting heating due to overcharge current is first insure that the battery temperature is not substantially less than the ambient temperature (as determined by step 29-18). Once the battery is warmed to ambient temperature, the overcharge condition can be quickly detected by means of steps 29-20 to 29-23 since any further substantial increase in temperature can be attributed to internal heat being evolved by the battery. If the pack has been in a hot environment, the cooling (steps 29-13 and 29-14) will bring its temperature down to no more than 40.degree. C., which is above the ambient temperature of the charger. Overcharge induced heating will cause the pack temperature to begin to increase again as shown by FIG. 35. According to the described algorithm, the charge current applied to the battery for overcharge detection (step 29-20) is double the standard overcharge table value of steps 29-17, 29-19 and 29-28 (and of FIG. 33) to improve the ability to detect a temperature increase. Since the test time is relatively short, little gas pressure increase and potential for cell venting is involved.

Once it has been determined that the battery is not in the overcharge condition (at decision block 29-23), it is a relatively simple matter to apply the appropriate charge value from the fast charge parameter table (as at step 29-24, FIG. 29). The fast charge table value may correspond to that indicated in FIG. 34 and is a function of temperature so that a temperature regulation capability is implemented for reducing the current applied at elevated temperatures. During the fast charge operation, battery temperature increase is closely monitored (steps 29-25 to 29-27) to determine when overcharge has been reached, so that the fast charge cycle may be terminated (as represented by branch line 29-27a) and a controlled temperature overcharge cycle may be initiated as represented by step 29-28 and FIG. 33 to "top-off" the battery for maximum capacity. After the overcharge cycle is complete (after step 29-29), a trickle charge current is applied per step 29-30 to maintain the full battery capacity and offset the effects of self-discharge normally seen when a battery rests in an idle condition.

FIG. 32 shows a preferred embodiment of the described fast charging system utilizing a microprocessor system with a programmed algorithm for fast charging of battery packs. Other embodiments involving control circuits contained within a data terminal or other utilization device may employ identical algorithms without departing from the concepts described.

FIG. 32 represents an implementation of FIG. 28, and corresponding reference numerals have been applied in FIG. 32 so as to facilitate correlation therewith. The major components of. FIG. 32 may comprise commercially available parts which are identified as follows:

microprocessor chip 32-10 of microprocessor system 28-10, type SC83C552

voltage regulator 32-11, type LP2951AC

amplifiers 28-12 and 28-13 of charge regulator 28-20, type LT1013 transistor 32, type 1RF9Z30

temperature sensor 28-17, type LM335

The programming for microprocessor element 32-10 of FIG. 32 may correspond with that represented in FIGS. 29, 33 and 34, as described with reference to these figures and the circuits of FIGS. 28, 30 and 31. By way of example, terminals 32-21, 32-22, 32-23 and 32-24, FIG. 32, may be connected with terminals 30-21, 30-22, 30-23 and-30-24 respectively in FIG. 30. Temperature sensor 30-26, FIG. 30, which is connected between terminals 30-22 and 30-23, may correspond with sensor 28-16 and may be mounted in intimate heat transfer relation with battery 28-15 and within the housing of the battery pack 28-25 as represented in FIG. 28. Resistor R1, FIG. 30, has a respective one of a set of values so as to provide a voltage level between terminals 30-22 and 30-24 selected so as to identify the particular type of battery pack 30-27 with which it is associated.

Terminals 30-21A, 30-22A, 30-23A and 30-24A may be connected with a utilization circuit to supply energy thereto during portable operation. It will be noted that the battery pack 30-27 can be associated with the circuitry of FIGS. 16A and 16B hereof (corresponding to FIGS. 16A and 16B of incorporated U.S. Pat. No. 4,709,202), terminals 30-21A and 30-22A having a quick-release connection with terminals JP-1, JP-2, FIG. 16B, and terminals 30-23A having a quick-release coupling with terminal J7-3, FIG. 16B. Terminal 30-24A can be used by the portable device to identify the battery pack, where the portable device provides a circuit such as associated with terminal 32-24, FIG. 32, leading to an analog to digital input such as 32-25, FIG. 32.

FIG. 35 illustrates by a plot 35-10 the increase in temperature as a function of time of an enclosed battery pack such as 27-10B, FIG. 27, 28-25, FIG. 28 or 30-27, FIG. 30, due to an overcharge current of 300 milliamperes, where the battery means 27-27, 28-15 is initially fully charged and is at a battery temperature of about minus eight degrees celsius, the ambient temperature being about fourteen degrees celsius. The slopes between successive points 35-1, 35-2, 35-3 and 35-4, are represented by straight line segments 35-11, 35-12 and 35-13, with respective slope values of 0.54 degrees celsius per minute, 0.36 degrees celsius per minute and 0.21 degrees celsius per minute.

FIG. 36 for the sake of comparison shows by a curve 36-10 the rate of warming of such a battery pack due to an ambient temperature which is substantially higher than battery temperature. Specifically FIG. 36 shows the case where initial battery temperature is about minus fifteen degrees celsius and ambient temperature is about twenty degrees celsius. Straight line segments 36-11, 36-12, 36-13, 36-14 show approximate slope values of 0.98 degrees celsius per minute, 0.5 degrees celsius per minute, 0.33 degrees celsius per minute and 0.24 degrees celsius per minute. The relatively high slope values indicate that the differential between a high ambient temperature and a low battery temperature must be taken into account when using steps 29-20 to 29-23 to determine whether a battery is in the overcharge range.

FIGS. 33 and 34 represent in effect a series of tables of charge rate versus temperature since the ordinate values are in units of charge rate (e.g. current Ibatt in milliamperes divided by capacity C in milliampere-hours). The following TABLES I and II give values of overcharge and fast charge corresponding to FIGS. 33 and 34 for successive temperatures in increments of two degrees celsius, and give corresponding current values milliamperes for two different values of battery capacity C, namely C equals 800 milliampere-hours and C equals 1200 milliampere hours.

TABLES I and II may be stored in machine readable form in the memory of microprocessor system 28-10 or 32-10, e.g. in first and second read only memory segments. Thus if step 29-4 identified an 800 milliampere-hour capacity fast charge nickel-cadmium battery means, the microprocessor would access the first memory segment corresponding to TABLE I for steps such as 29-17, 29-19, 29-20, 29-24 and 29-28, while if step 29-4 showed a 1200 milliampere-hour capacity fast charge nickel-cadmium battery means the second memory segment corresponding to TABLE II would be addressed.

If for example, the battery temperature (Ptemp) in step 29-16 were greater than nineteen degrees celsius but less than or equal to twenty-one degrees celsius, the overcharge value read from memory segment I would be 240 milliamperes (0.300 units in FIG. 33 times 800 milliampere-hours, the battery capacity C, equals 240 milliamperes). Thus according to step 29-17 and step 29-19, an overcharge current of 240 milliamperes (plus any needed load current) would be supplied by regulator 28-20 until temperature sensor 28-16 showed that battery temperature exceeded ambient temperature (Atemp, 28-13, FIG. 28).

If ambient temperature were thirty degrees celsius and the battery temperature were in the range from thirty-one to thirty-three degrees celsius, a current of 480 amperes would be applied according to step 29-20, but for a limited duration (e.g. about ten minutes per step 29-21) such as to avoid substantial detriment to the useful life of the battery means.

FIG. 36 illustrates warming of the battery pack as a function of time with the battery pack initially at a temperature of about minus fifteen degrees celsius. From FIG. 33, it can be seen that maximum permissible overcharge current corresponds to about 0.08 units. For a battery capacity of 800 milliampere hours, this would correspond to an overcharge current value of greater than sixty milliamperes, while FIG. 36 shows the warming rate with an ambient temperature of about twenty degrees celsius and a relatively negligible value of charging current (i.e. Ichg equals six milliamperes). It will be noted that the warming rate in FIG. 36 in the first 600 seconds is 0.98 degrees celsius per minute which considerably exceeds the warming rate produced by a current of 300 milliamperes in FIG. 35.

For representing an embodiment such as that of FIG. 30, a microprocessor system such as indicated at 28-10 in FIG. 28 would be shown with a fourth input to A/D means 28-10A corresponding with input 32-25, FIG. 32. For such an embodiment each type of battery means such as the one with 800 milliampere-hour capacity and temperature characteristics as shown in Table I, and the one with 1200 ampere-hour capacity and characteristics according to Table II would have a respective distinct value of R1, FIG. 30, and a respective different shunt voltage level so as to enable the microprocessor system 28-10 to reliably identify each of numerous types of battery means pursuant to step 29-4. The microprocessor system 28-10 or 32-10 may store a set of parameter tables such as Tables I and II in machine readable form with each table of such set having an address associated with the corresponding shunt voltage level. In this way the appropriate stored table can be interrogated by the microprocessor in accordance with a given battery temperature reading so as to obtain appropriate current values for steps 29-8, 29-17, 29-19, 29-20, 29-24 and 29-28.

The battery identification means 26-36 or 30-10 would distinguish the presence or absence of an internal current regulator 26-28 as well as identifying the various battery types requiring different charging and overcharge treatment.

Other stored machine readable tables of computer system 28-10 or 32-10 may include acceptable maximum overcharge rates as represented in FIG. 6 and have charge rates, e.g., as described at Col. 9, line 26 to Col. 10, line 32 of the incorporated U.S. Pat. No. 4,455,523. Such stored tables would insure that the charging system of FIG. 28 or FIG. 32 would be compatible with a battery means such as shown in FIG. 5 or in FIGS. 9A and 9B. For example, the stored table for the battery means of FIGS. 9A and 9B could take account of internal heating within the internal regulator 173 of the battery pack and insure that the current to the battery (20, FIG. 5) and to the battery load did not exceed the power dissipation capacity of the internal regulator network (173, FIG. 9A).

The current regulator 28-20 may be controlled to provide a voltage VCHG at the line CHG in FIG. 9A of approximately seven volts which would result in minimum power dissipation in the interior regulator network (173, FIG. 9A). The presence of an internal current regulator within a hand-held terminal unit is indicated at 26-28, FIG. 26, and charging current control circuit 26-22 could conform with the embodiments of FIGS. 28-36 in the selection of charging and overcharge current values while tending toward minimum power dissipation in the internal regulator network (173, FIG. 9A) or in the internal regulator 26-28, FIG. 26.

In a different embodiment, each battery pack could have an internal digitally stored identification code digitally stored in the battery pack and accessible to an external microprocessor system such as 28-10 or 32-10 as in the embodiment of e.g. FIG. 23 (via contacts 23-51A), FIG. 25 (via data output 25-51), FIG. 26 (via components 26-36, 26-37, 26-20 and 26-23), or FIG. 27 (via D to A component 27-37 or LAN interface 27-39).

Instead of bringing battery temperature up to ambient temperature as in steps 29-15 to 29-19, it would be conceivable to establish ambient temperature to match battery temperature, and then proceed with a test for overcharge condition as in steps 29-20 to 29-23. Similarly before steps 29-24 to 29-27, where the battery is initially at a low temperature, it would be conceivable to control ambient temperature so as to bring battery temperature up to zero degrees celsius or six degrees celsius by control of ambient temperature alone, or in combination with a suitable charging current. In this way, a relatively high charge rate according to FIG. 34 would be suitable, e.g., at least 0.6C, and a maximum overcharge rate according to FIG. 33 would quickly be appropriate for the overcharge cycle of steps 29-28 and 29-29.

The stored charge rate information can take the form of end points such as 33-1, 33-2; 33-2, 33-3; 33-3, 33-4, FIG. 33, for successive substantial straight segments such as 33-11, 33-12 and 33-13, so that the microprocessor could interpolate a precise charge rate multiplier for any measured battery temperature. Thus, if segment 33-12 had end points at -20.degree. C., 0.060 units and at -10.degree., 0.100, a battery temperature of -19.degree. might be computed to correspond to 0.064 by linear interpolation. Of course of course the points given in Tables I and II could be similarly interpolated to obtain intervening more precise overcharge and fast charge values.

With respect to steps 29-7 to 29-12, ean internal microprocessor such as in FIG. 5 or FIGS. 9A, 9B may determine battery load current and communicate the same to an external microprocessor such as 28-10 as shown in FIGS. 23, 25, 26 or 27, for example, FIG. 28 may represent the association of a non-portable battery conditioning station including components 28-10, 28-17 and 28-20 with a hand-held terminal unit containing a quickly removable battery pack 28-25 comprised of a nickel-cadmium rechargeable battery 28-15 and a battery temperature sensor 28-16 within housing 28-22. The hand-held terminal unit may provide load means 28-21, which may comprise a dynamic random access memory and other circuitry which is to be continuously energized during a charging operation.

As in the embodiment of FIG. 30, the hand-held terminal units which are to be associated with components 28-10, 28-17 and 28-20 may include coupling means such as 30-21, 30-21A, 30-23 and 30-24 which are automatically engaged with cooperating coupling means of the charging station when the hand-held unit is bodily inserted into a receptacle of the charging station. The coupling means 30-21 and 30-21A in FIG. 30 would be represented in FIG. 28 by a line (+BATT) from component 28-20 corresponding to line 28-26, and a further line (+CHG) leading to a network (representing components 30-28 and 30-29) in turn connected with battery 28-15 and line 28-11.

An exemplary charging station adaptable for a hand-held unit including battery pack 30-27 of FIG. 30 is shown in greater detail in FIG. 27.

Where FIGS. 28-36 are applied to a system as represented in FIG. 27, components 28-10, 28-17 and 28-20 would be part of charger station 27-22. Line 28-26 would lead to a charging station contact engageable with external battery pack contact 27-11. Input line 28-12 would be connected via a further set of mating contacts with internal battery pack contact 27-61. Input line 28-11 would connect with contact 27-17. Alternatively, charger station 27-22 would have a LAN interface corresponding to 27-39 and would receive digital information as to battery terminal voltage for example via amplifier 27-35A, an A to D converter of terminal 27-10A, LAN interface 27-39 and LAN data coupling means 27-19, 27-21. The charging station would then charge the battery packs such as 27-10B of terminals such as 27-10A in accordance with the embodiments of FIGS. 28-36. In place of amplifier 27-37, (representing components 26-36, 26-37), an identifying shunt voltage regulator 30-10 would be part of each battery pack 27-10B.

As a further embodiment, the charger station 27-22 could comprise the components of FIG. 32, the line 28-12 being coupled with a battery such as 27-27 via mating terminals 32-21 and 27-11 and through forward biased diode 27-D1, for example. In this embodiment terminal 27-13 would mate with terminal 32-24, and battery pack 27-10B would conform with battery pack 30-27 for example by including a respective identifying shunt regulator 30-10, and e.g., nickel-cadmium batteries with respective characteristics as shown by FIGS. 33 to 36 and Tables I and II.

Operation of the specific exemplary embodiment as presented in FIG. 29 may be summarized as follows.

As represented by steps 29-2 and 29-3, the presence of a battery pack 28-25, FIG. 28, or 30-27, FIG. 30, may be sensed by means of the input 28-11, FIG. 28 or FIG. 32, from battery pack temperature sensor 28-16, FIG. 28, or 30-26, FIG. 30. A non-zero voltage input level on line 28-11 may signal the presence of a battery pack 28-25 coupled with components 28-10 and 28-20. The physical connections may be analogous to those of FIG. 26, for example, an exemplary arrangement of terminals for a battery pack 30-27 being shown in FIG. 30.

Referring to FIG. 31, the load current may be automatically sensed by means of steps 29-7 to 29-11 since battery voltage as measured at 28-12, FIG. 28 or FIG. 32, will not increase until a current Ichg, FIG. 30, in excess of battery load current Iload is applied to line 28-26, FIG. 28.

Where the battery pack has a given upper temperature limit which must be observed to avoid detriment to battery life, the battery may be automatically allowed to cool to a suitable temperature (e.g. 40.degree.) if it is introduced into the charger at an unacceptably high temperature. This is represented by steps 29-13 and 29-14 which may be automatically performed by microprocessor 28-10 or 32-10 according to battery temperature (Ptemp) as sensed at input 28-11, FIG. 28 or FIG. 32.

As explained in reference to FIG. 36, in order to detect the battery overcharge condition, the microprocessor 28-10 or 32-10 automatically performs steps 29-15 to 29-19 to assure that battery temperature as measured at input 28-11 is not substantially lower than ambient temperature as sensed at input 28-13. Once battery temperature is at least essentially equal to ambient temperature, steps 29-20 to 29-23 are effective to automatically determine whether the battery is to receive a fast charge according to steps 29-24 to 29-27, and e.g. FIG. 34, or whether the microprocessor 28-10 or 32-10 is to govern the supply of charging current at 28-26 according to steps 29-28 and 29-29 and e.g. FIG. 33.

From the foregoing, it will be understood that steps 29-20 to 29-23 are effective where the battery system can be made to exhibit a temperature characteristic which rises as a function of overcharge current over a given time interval generally as illustrated in FIG. 32. To accomplish this the charging system may operate automatically as in steps 29-15 to 29-19 to insure that the battery means has a state such that its temperature will not increase at a substantial rate due to a higher ambient temperature (e.g. as in FIG. 36). In particular, the state of the battery means may be automatically assured to be such that it will exhibit a substantially greater increase in battery temperature in response to a given selected charge rate when the battery is in overcharge condition that when it is not in such a overcharge condition.

As represented by step 29-20, the current automatically applied to the battery means exceeds battery load current by a substantial overcharge magnitude, e.g. twice the overcharge value obtained from FIG. 33, but the application is of limited duration (e.g. ten minutes per step 29-20) such as to void substantial detriment to the useful life of the battery.

According to steps 29-22 and 29-23, the microprocessor system automatically determines whether any increase in battery temperature due to step 29-20 is of a magnitude (e.g. two degrees celsius or greater) which is distinctive of the overcharge condition of the battery means.

From the foregoing TABLE I, it will be understood if battery temperature at step 29-22 has reached thirty degrees celsius, step 29-24 would result in an initial relatively high battery Charging current (Ibatt, FIG. 31) of about 1280 milliamperes if the temperature increases at step 29-23 was not greater than two degrees celsius, while if the increase at step 29-23 were found to be greater than two degrees celsius, step 29-28 would result in supply of a relatively lower battery charging current of about 240 milliamperes.

Where the relatively high battery charging current is applied, battery temperature is measured at regular intervals (e.g. at about one minute intervals per step 29-25) to assure that such high charge rate is terminated sufficiently quickly after overcharge condition is detected so as to avoid any substantial detriment to the useful life of the battery means.

The overcharge relatively lower charge rate is terminated after an overcharge interval so as to insure optimum charging of the battery means without detriment to its useful life.

Of course the charge rate or overcharge rate may be readjusted higher or lower according to FIGS. 33 and 34 at any desired time intervals, e.g. at each step 29-24 in charging mode, and by inserting steps such as 29-25 and 29-26 between steps 29-28 and 29-29 so that overcharge current would be re-selected at suitable intervals such as one-minute intervals.

Referring to the plots of maximum acceptable overcharge rate in FIGS. 6 and 33, it will be observed that there is a minimum temperature for each battery type below which overcharge current is not applied by the microprocessor system 28-10 or 32-10. In FIG. 6, the lower temperature extreme is shown as about zero degrees Fahrenheit (about minus eighteen degrees celsius). At about zero degrees Fahrenheit, the low overcharge rate is less than about capacity divided by fifty. In FIG. 33, the low temperature extreme is about minus thirty degrees celsius where the overcharge current of about 0.04 units corresponds to an overcharge rate of about capacity divided by twenty-five.

Above the lower limit temperature, there is a range of temperatures where the upper overcharge rate exceeds the lower overcharge rate by a factor of at least about four. For example, in FIG. 6, the acceptable overcharge rate at a relatively high temperature of about one hundred and ten degrees Fahrenheit is close to capacity divided by five (0.2C), while the acceptable overcharge rate at the low temperature extreme of about zero degrees Fahrenheit is about capacity divided by fifty (0.02C) a ratio of overcharge rates of ten to one. According to FIG. 33, the microprocessor system 28-110 or 32-10 may supply values of overcharge at about five degrees celsius of about 0.30 units (C/3.3) while at a low temperature extreme of about minus thirty degrees celsius, the acceptable overcharge rate to be supplied by the microprocessor system is about 0.04 units (C/25), a ratio of about seven to one. Between the temperature values of the temperature range of FIG. 6, the permissible overcharge rate progressively increases with successively higher temperature values such as zero degrees, fifteen degrees, thirty-five degrees, fifty-five degrees, seventy-five degrees and ninety-five degrees (Fahrenheit). Similarly in FIG. 33, between temperatures of minus thirty degrees celsius and about five degrees celsius, the permissible overcharge rate progressively increases for successively increasing temperature values (such as -20.degree. C. -10.degree. C. and 0.degree. C.).

Referring to FIGS. 28, 30 and 31, it will be understood that the embodiments of FIGS. 28 to 36 avoid series resistance means of substantial ohmic value such as shown at 131, FIG. 9A, 18-26, FIG. 18, 24-30, FIG. 24, 25-26, FIG. 25, for sensing battery current. Instead charging current source 28-20, FIG. 28, may be automatically operated to supply desired current values in an open loop manner. An automatic sequence such as steps 29-7 to 29-11 may be used to measure load current if this would be a fluctuating and possibly significant amount for a given hand-held terminal unit and would not be reported to the charging station by the hand-held unit. As shown by FIGS. 28, 30 and 31, the battery 28-15 has external terminals e.g. as at 30-21A, 30-22A, FIG. 30, with external circuit means connecting such terminals with the battery, such external circuit means having essentially negligible ohmic resistance such that the battery means supplies load current to a load via the external terminals with minimized ohmic loss at the battery side of said external terminals.

As portable hand-held data and radio terminals continue to be used more widely in certain demanding applications, the for fast charging of the terminal batteries becomes more significant. The increased use of high powered scanner attachments and peripherals as well as other connected devices often causes the terminal battery capacity to be taxed to the point where only a portion of the intended period of usage may be served with the stored charge available from a single battery pack. Consequently, it has become increasingly necessary to provide multiple packs which may be exchanged in such a way that a depleted pack may be replaced by a fresh one with minimal downtime. When a depleted pack is removed, it should be fully recharged in a least the amount of time that a fresh pack is able to operate the terminal. With a recharging capability of this type, it is then possible for virtually perpetual operation to be provided with as few as two battery packs per terminal.

A similar but further complicated application involves the utilization of the described data terminals on a vehicle such as an industrial fork lift truck. In this type of application, the terminal may receive power for operation from the vehicle the majority of the time. Often, however, it may be necessary for the terminal to be physically removed from the vehicle and operated in a fully portable mode for potentially extended periods of time. For this reason, it is imperative that the terminal batteries be maintained in their fully charged or "topped off" state at all times.

The above stated objectives for a battery charging system have traditionally been extremely difficult to achieve. FIGS. 37, 38 and 39 show an embodiment that addresses both aspects of fast charging and maintenance in a novel and unique way.

As described with reference to FIGS. 28 through 36, the characteristics of the popular NiCad battery technology are such that the rates at which charging may be accomplished are a strong function of temperature and state of charge. If the cell is in a discharged condition, the rate at which charge may be applied is relatively high, regardless of the cell temperature. If the cell is in a charged condition, the rate at which charge may be applied to the cell is determined by the temperature of the cell. At the limits of cell temperature, excessive charge current may cause permanent damage resulting in premature failure of the cell. Consequently, for fast charging to be accomplished safely, the temperature and state of charge of a battery must be determined.

Battery temperature (herein designated PT) may be measured directly by the use of a pack temperature sensor 28-16 thermally coupled to the battery pack 28-25. State of charge of a NiCad battery type is more difficult to determine. In general, the most reliable indication that a NiCad battery is fully charged is the release of heat while under charge. This only occurs when the battery is in the overcharge condition in which most or all of the current supplied to the battery causes evolution of oxygen gas at the positive electrode. When oxygen chemically recombines with cadmium at the negative electrode, heat is released. No other condition of operation of a NiCad battery causes appreciable heat to be generated.

In general, the process of converting charge current to stored charge in a NiCad battery is a slightly endothermic chemical reaction, that is, heat is removed from the environment of the battery and it gets slightly cooler. Consequently, it is possible to apply quite high rates of charge to the battery if it is not in the overcharge condition. Once the overcharge condition is reached, the slightly endothermic charge reaction is overwhelmed by the highly exothermic overcharge/oxygen recombination reaction. The rate of applied charge must then be quickly reduced to prevent excessive heating and battery damage.

As described in reference to FIGS. 28-36, a microcomputer 28-10 with the ability to measure the temperature of a battery and control the applied charge to that battery may be employed to effect charging at the maximum safe (non-damaging) rate and may also terminate the charge function to prevent damage to the battery when it is fully charged. The present embodiment also employs a microcomputer to measure battery temperature and control applied charge as indicated in FIG. 28, however, the process that is used to determine the appropriate charge current is quite substantially different, and significantly modifies and improves the performance of the charging system.

A flowchart of the procedure that accomplishes the described charging characteristics is shown in FIG. 37. It should be noted that the charging method described may be applied to either a terminal or utilization device with the circuits as shown in FIG. 28 or to a standalone pack charger with one or more sets of the same circuit. In the case of the pack charger configuration, the block 28-21 labeled "load" would not be present. In the terminal configuration the operating power required by the terminal itself would represent a load that requires current to be delivered by the charge circuit or battery.

In FIG. 37, the initial decision block 37-1, .vertline.AT-PT.vertline.>10.degree. C., provides two pieces of information based on the ambient temperature value, AT, from sensor 28-17, and battery pack temperature PT. The first information (when the temperature difference is not greater than ten degrees celsius) is that the temperature sensors 28-16 and 28-17 are in at least approximately agreement which provides confidence that they are functioning properly. If the temperature difference is relatively great, it is possible that the battery pack and the charger are at significantly different temperatures, in which case they need to stabilize to an acceptable level before further procedure steps may be taken. If this condition is detected, a constant current of sixty milliamperes (60 ma.) is selected as indicated at 37-2 to provide a safe low maintenance current that may minimize further discharge of the battery if it is already in a relatively depleted state.

If the initial temperature difference is not excessive, the absolute temperature of the battery pack is examined at steps 37-3 and 37-4. The temperature range allowed for charging is between 10.degree. C. and 36.degree. C. If the battery temperature is not within this range, the battery must be allowed to cool or warm as the case may be for the charging process to continue. It may be assumed (or specified) that the ambient temperature environment of the charger itself is between these limits, so that the battery temperature will stabilize after some time to an acceptable level. During this temperature stabilization time, it is preferable that no charge current be supplied to the battery, though it may be necessary for current to be supplied to a load, as in the case of a terminal which receives it operating power from the battery or charge while charging is in progress. Since the load current is generally not known, a mechanism must be provided to adjust the current provided by the charger to accurately match the load current of the terminal. The means by which this is accomplished is as follows:

1. Examine the battery pack terminal voltage designated PV as indicated at block 37-5.

2. Select an initial charge current of sixty milliamperes (60 ma.) as indicated at 37-6.

3. Examine the pack temperature PT at 37-7 and 37-8 to determine if it has stabilized within the desired limits. If so, return to the main charging process.

4. Examine the present terminal voltage PV at 37-9 and 37-10.

5. If the battery terminal voltage has increased, decrease the charge current by twenty milliamperes (20 ma.) as indicated at 37-11.

6. If the battery terminal voltage has decreased, increase the charge current by twenty milliamperes (20 ma.) as indicated at 37-12.

7. If no terminal voltage change is detected, leave the charge current unchanged and return to step 3 above. This method serves to provide an adaptive current that will prevent the battery from being further depleted while its temperature stabilizes to an allowable level.

After the battery temperature has stabilized to an allowable level, it is then possible to begin charging at high rates of charge. As described with reference to FIGS. 28-36, a stored table containing values of currents that may be safely applied to a battery of a known capacity at a given temperature is used to determine the charge current, this being indicated at 37-13. While the table values for fast charge current will not cause stress or damage to a battery when it is discharged and efficiently receiving charge, in general, these charge currents are high enough to cause permanent damage to the battery if not terminated properly. The indication that the battery is nearing full charge is based on detection of the overcharge condition, which is the only condition of a NiCad cell that releases significant heat. In flowchart block of FIG. 37, the condition for decision block 37-14, PT<AT+10.degree. C., provides the test for overcharge detection. In essence, the test for overcharge is to detect that the battery is becoming warmer than the ambient environment, in this case by an amount of ten degrees celsius (10.degree. C.). When this amount of heating is detected indicating that the battery has reached the overcharge condition in an appreciable amount, the fast charge function is terminated.

Upon completion of fast charge, a maintenance charge function is initiated which continues to monitor the battery temperature rise above the ambient environment (steps 37-15) and maintains an applied overcharge current at a level that regulates that battery temperature rise. The overcharge temperature rise is held to eight degrees celsius (8.degree. C.) as shown by steps 37-16, 37-17, and 37-18; this being a safe sustainable level that may be maintained indefinitely without appreciable cumulative damage to the battery. The temperature regulation process is implemented by selecting between a low charge current of sixty milliamperes (60 ma.) and the higher overcharge current table value depending on the measured temperature rise. By maintaining the battery n a state of continuous safe overcharge, it is possible to hold the battery in its maximum state of charge at all times, thereby ensuring that the user has the full battery capacity available whenever needed. If the battery temperature falls below the ambient temperature as determined at step 37-15, the fast charge state will be re-entered which will apply the maximum safe charge current for the measured temperature. It should be noted that this situation might occur if a very warm battery pack is placed in a pack charger at nominal temperature. Initially if the temperature difference is greater than 10.degree. C. the pack will be charged at a fixed current of 60 ma until the temperature difference is reduced. If the pack temperature is less than 36.degree. C. at this time, its temperature difference may still be very close to 10.degree. C. which might allow the process to advance to the final maintenance state of the charge system. As the pack cools further due to ambient cooling and the endothermic charging reaction, its temperature may go below the ambient temperature (step 37-15), at which point the fast charge state would be re-entered.

In the maintenance mode, the current required for operation of a terminal is provided by the fact that the charge current (step 37-17 or 37-18) will exceed the terminal operating current by an amount necessary to maintain the temperature rise of the battery. Consequently, this charging system provides broad flexibility for fast charging of NiCad batteries in utilization devices with widely varying current demands.

A useful feature of this charging method is that it is not critical that the charging voltage source be able to provide the maximum current specified by the controlling microcomputer, for reliable charging to be accomplished. For example, if the selected value of charge current for a certain battery pack is 1500 ma., but the voltage source has a current capacity of only 600 ma., the fast charge state of the procedure would be maintained in exactly the same way except it would take correspondingly longer for the overcharge state to be reached. This feature of the charging method is particularly useful in configurations where multiple battery packs may be charged in a single unit and it is necessary to place constraints on the unit power supply for economic or size reasons. It is a relatively simple matter to externally limit the maximum delivered current so that the actual charge current is less than the value selected by the controlling microcomputer.

FIG. 38 shows a schematic diagram of a charge current regulator circuit which has the capability of delivering a constant current output to a battery in proportion to an input control voltage. In addition, this circuit has a maximum delivered current limit that may be set by a resistor selection in power supply constrained applications.

The CHARGE CONTROL input 38-10 is intended to be driven by a digital to analog (D/A) converter output of a microcomputer based utilization device such as a data terminal. The CHARGE CONTROL input develops a control voltage at pin 3 of differential amplifier 38-U1. The output pin 1 of 38-U1 drives 38-Q3 which establishes a current through 38-R8 that develops a voltage at 38-UI, pin 4 equal to the voltage at 38-U1, pin 3. Since the current gain h.sub.fe of 38-Q3 is relatively high (about 200) the collector current of 38-Q3 is nearly equal to the emitter current, resulting in an equal current through both 38-R5 and 38-R8. Since these resistors are of equal magnitude, the input voltage at CHARGE CONTROL 38-10 appears across 38-R5 referenced to the +12 volt supply voltage. The amplifier at 38-U1A pins 8, 9 and 10 is a differential configuration operating in a negative feedback mode. With a voltage developed across 38-R5, the voltage at pin 9 of 38-U1A will be decreased, which increases the voltage at the output pin 8. This increased voltage drives current into 38-Q2 which increases the drive current to 38-Q1 establishing a current through current sense resistor 38-R6. When the voltage drop across 38-R6 equals the voltage across 38-R5, the amplifier output will stabilize, holding the output current constant. With a sense resistor value of one ohm at 38-R6, the voltage to current conversion factor is one ampere per volt (1 amp./volt). If the CHARGE CONTROL input is left unconnected, the 1.25 volt voltage reference 38-CR1 and resistors 38-R2 and 38-R4 establish an open circuit voltage of 0.120 volts which establishes a "default" output current of 120 ma. This condition may be useful in cases where a utilization device is either unintelligent or its battery is completely depleted in which case its processor is unable to operate and the battery must be brought up to at least minimal capacity for the processor to function.

The circuit block consisting of the amplifier at 38-U1B pins 5, 6 and 7 is a clamp circuit that limits the maximum voltage that may be applied to 38-U1, pin 3. By limiting the input voltage, the maximum available charge current may then be limited to some selected value dependent on the selection of 38-R15 and 38-R16. With values of twenty-one kilohms for 38-R15 and ten kilohms for 38-R16, a voltage of 0.40 volts is applied to the clamp circuit input. If the input voltage driven on CHARGE CONTROL is less than 0.40 volts, the output pin 7 of 38-U1B remains low which biases 38-Q4 off. If the CHARGE CONTROL input voltage reaches or exceeds 0.40 volts, 38-Q4 is turned on sufficiently to maintain a voltage of exactly 0.40 volts at 38-U1B pin 5 which prevents the input voltage to the control amplifier from exceeding this voltage. The voltage to current transfer function of the system is shown in FIG. 39. It should be noted that the clamp voltage and maximum available current may be modified by selecting different values for 38-R15 and 38-R16 or the voltage reference 38-CR1. A maximum available current of 1.25 amps may be implemented by deleting 38-R16 in which case the full reference voltage appears at the clamp circuit input.

The microprocessor system means 28-10 or 32-10 operates automatically to apply substantially maximum charging current to the battery means consistent with avoiding substantial detriment to the useful life of the battery means e.g. as represented in FIG. 34.

FIGS. 40-46 illustrate further optional features of the present invention. As previously explained, there is room for improvement in the art regarding flexibility and efficiency of the recharging process. As illustrated previously in FIG. 31, the conventional way of recharging is to simply supply constant charging current to the battery. If, like in FIG. 31, a varying load is connected in parallel with the battery, this may affect not only the effectiveness and efficiency of recharging, but may even cause loss of recharging capabilities; or worse, discharge of the battery.

It has been found that one way to provide flexibility for recharging and increase efficiency of recharging is to pulse the recharging current. The pulse width can be modulated according to one or more controls to adjust the net charging current going to the battery.

In FIG. 40, a circuit is illustrated which feeds back information to the current source regarding the amount of recharging current the battery is receiving. The current source can therefore alter the nature of the charging current (for example into a pulse width modulated waveform) which can in turn be varied or manipulated to provide a net charge to the battery.

By referring back to FIG. 31, it can be seen that a constant current source provides constant current I.sub.chg (28-26). I.sub.chg is divided into battery current I.sub.batt, directed to the battery (28-15), and load current I.sub.load, supplied to the load 28-21. Total current I.sub.chg equal to the sum of I.sub.batt plus I.sub.load. If load current I.sub.load exceeds constant current I.sub.chg, current is drawn from the battery 28-15.

FIG. 40 shows current source 40-10, load 40-12, and battery 40-14 similar to FIG. 31. Additionally, however, a sensor device 40-16 is connected in series to battery 40-14 to sense the amount of charging current I.sub.C that is given to battery 40-14 by the current I.sub.B. Sensor 40-16 in turn provides a signal to integrator 40-18 which in turn communicates with current source 40-10. The signal from integrator 40-18 tells charging current source 40-10 the amount of current I.sub.B received by battery 40-14 over a given period of time. Charging current source 40-10 can include some sort of control to vary charging current I.sub.C to insure effective battery recharging even in light of a varying load which draws the varying load current I.sub.L.

The configuration of FIG. 40 therefore will allow flexible, efficient control of charging current I.sub.C to in turn allow flexible and efficient charging of battery 40-14. The pulsing of the current provides a net charge over time to the battery. The magnitude of the net charge can be adjusted of fine-tuned by varying the pulse width or duty cycle of the pulses.

FIG. 41 shows a slightly different configuration from FIG. 40. Dashed line 41-10 indicates schematically the wall of an enclosed housing containing the components of a device including battery 41-12. In this configuration, a controlled switch designated by reference number 41-14 pulses the charging current I.sub.C to battery 41-12 (current I.sub.B) and to other parts of the circuit (current I.sub.L for current load). The advantages of a pulsed charging current have been previously mentioned.

In FIG. 41 an additional feature is the placement of the source of charging current 41-16 externally of the housing of the device. Charging current source 41-16 would be connectable to the circuitry inside the housing by a plug-in 41-18 or other suitable connection existing on wall 41-10 of the device. FIG. 41 also illustrates an electrical power source 41-20 which sends electrical power to charging current source 41-16. Charging current I.sub.C would basically be a DC value of constant magnitude. Control switch 41-14 would then produce a pulsed output. By placing the source of charging current 41-10 outside of the housing device, any heat dissipated from such a component would be removed from presenting any problems to the circuitry inside the housing. This embodiment therefore provides the advantages of controlling the nature of the form of the charging current to the battery and load, as well as moving a heat dissipating component outside of the housing. This would further allow the current source 41-15 to produce a higher level of charging current I.sub.C then would be possible if source 41-16 were positioned inside the housing. This current could then be controlled by switch 41-14 to provide an adequate net charge to the battery without substantial danger of the varying load affecting sufficient charging current to the battery.

FIG. 42 illustrates the basic configuration of FIG. 41, with a specific feedback circuit and specific controlled switch. In this embodiment, constant current I.sub.C is pulse-width modulated by transistor 42-10. Transistor 42-10 is a low on-resistance, high power field effect transistor providing very low power dissipation. Blocking diode 42-20 prevents current flow from battery 42-30 through transistor 42-10. I.sub.c is divided between battery 42-30 and load 42-80. Sensing register 42-40 is a low resistance device which produces a voltage corresponding to current flow through the battery 42-30. Voltage from sensing resistor 42-40 is conditioned by amplifier 42-50. The output voltage of amplifier 42-50 is presented to tho input of integrator 42-60. Integrator 42-60 integrates the voltage input presented to it, thereby integrating the charge which has flowed into the battery 42-30. The output voltage of integrator 42-60, corresponding to the net charge in a given integration time interval delivered to the buttery 42-30, is presented to pulse width modulator 42-70. As the output voltage of integrator 42-60 increases, corresponding to increased net charge delivered to the battery 42-30 per given integration time interval, the width the output pulse of pulse-width modulator narrows. As the duty cycle of pulse width modulator 42-70 decreases, transistor 42-10 reduces the average value of current delivered to battery 42-30 and load 42-80.

FIG. 43 depicts the type of generally square wave, pulse modulated charging current I.sub.C that is possible with circuits such an shown in any of FIGS. 40-42. In particular, the circuit FIG. 42 could produce this sort of signal which would have a maximum magnitude well above that needed to effectively net charge the battery. However, by pulse width modulating I.sub.C the net charge can be dynamically controlled to provide just enough charging current I.sub.B in addition to accommodating a varying load; due to varying conditions throughout the whole circuit. The circuit of FIG. 42 also has the added benefit of taking the current source outside the housing wall to eliminate any heat dissipation problems.

FIG. 44 illustrates a still further alternative for the invention. Similar to FIGS. 41 and 42, electrical power source 44-10 and constant current source 44-12 are located outside the housing wall 44-14 for the device. The major difference between the configuration of FIG. 44 and that of FIG. 42 is that it can produce a generally trapezoidal shaped pulse such shown in FIG. 45, instead of generally rectangular or square wave pulse of FIG. 43. The benefit of such a pulse-shape is to reduce both conducted and emitted transients delivered to the load circuits.

FIG. 44 shows the use of a "miller integrator" with a transistor switch (see dashed line 44-16) to produce the trapezoidal shape. In this configuration, the miller integrator receives the signal from the pulse width modulator (see FIG. 42) and produces the trapezoidal shape pulse width modulated current signal illustrated in FIG. 45. FIG. 46 shows the voltage rendition of the pulse width modulated current wave of FIG. 45.

It will be apparent that features of the various embodiments described or incorporated herein may be combined, and that various of the features may be utilized independently of others, and that many further modifications and variations may be effected without departing from the scope of the teachings and concepts of the present disclosure.