DE9321638U1 - Electronic battery tester with automatic compensation of a low state of charge - Google Patents

Description

Modified registration documents

Branch from EP 93 91 0920.3

Champlin Keith,

uZ: F 1224 GM-DE/D &psgr; "' SIEBERTS STR. 4

Electronic battery tester with automatic compensation of low charge level

HNTF.RGRTTND DF,R FRFTNDTJNG

This invention relates to an electronic measuring or monitoring device for evaluating the ability of a storage battery to supply current to a load. More particularly, it relates to an improved device of the type previously disclosed in U.S. Patent Nos. 3,873,911, 3,909,708, 4,816,768, 4,825,170, 4,881,038 and 4,912,416, all issued to Keith S. Champlin.

Storage batteries are used in many applications that require the storage of electrical energy for later use. They are most commonly used in motor vehicles that use internal combustion engines. In such applications, the energy stored during engine operation by "charging" the battery is later used to power lights, radio, and other electrical equipment when the engine is not running. The heaviest use of a motor vehicle battery is usually by the starter motor. Failure to supply the starter motor with sufficient power to start the vehicle engine, especially in cold weather, is often the first sign of battery degradation. Obviously, a simple measurement that accurately evaluates the battery's ability to provide sufficient cranking power is of considerable value.

Before the introduction of the dynamic conductance test disclosed in the above six U.S. patents,

procedure, the method available in most cases for evaluating the ability of a battery to provide sufficient cranking power was the standard load test. This test subjects a battery to a high direct current of a predetermined value given by the battery's rating. After a predetermined time interval, the battery's voltage under load is considered. The battery is then deemed to "pass" or "fail" the load test depending on whether its voltage γ under load is greater or less than a specified value.

Although the standard load test has been widely used for many years, it has some serious drawbacks. These include:

1. The test draws a large current and therefore requires a device that is heavy and cumbersome.

2. Considerable "sparking" may occur when the test fixture is connected or disconnected under load conditions. Such "sparking" may cause an explosion in the presence of battery gases with the possibility of serious injury to the user.

3. A standard load test leaves the battery in a significantly reduced state of charge and therefore less capable of starting the engine than before the test was performed.

4. The battery terminal voltage will decrease over time while performing the load test. As a result, the test results are generally inaccurate and often depend on the experience of the person performing the test.

5. The load test results are not repeatable because the test itself temporarily polarizes the battery. Such

Test-induced polarization significantly changes the initial conditions of all subsequent tests.

A practical alternative to the standard load test is taught in the six U.S. patents listed above. These documents disclose an electronic device for accurately evaluating a battery's condition using small-signal AC measurements of its dynamic conductance. They teach that the dynamic conductance of a battery is directly proportional to its dynamic power - the maximum

&iacgr;&ogr; power that the battery can deliver to a load. Dynamic conductance is therefore a direct measure of a battery's ability to deliver cranking power. Compared to the load test method for battery estimation, the dynamic conductance test method has many advantages. For example, the dynamic conductance test uses an electronic device that is small and lightweight, draws very little current, produces virtually no sparking on connection or disconnection, does not significantly discharge or polarize the battery, and provides accurate, highly reproducible test results. In fact, millions of battery measurements taken over the years have fully confirmed these teachings and proven the accuracy of this alternative test.

A disadvantage of the dynamic conductance test method, however, was the fact that the test results are somewhat dependent on the state of charge of the battery. Accordingly, the methods and apparatus in the first five of the six U.S. patents cited above generally required that the battery being tested be substantially fully charged. Since many batteries are in fact quite discharged when they are returned for warranty replacement or when they are otherwise deemed defective,

To date, it has often been necessary to recharge a battery before testing it. Such recharging is expensive and time consuming. Clearly, a simple method for performing accurate dynamic conductance tests on batteries "as is" - batteries that may only be partially charged - would be of considerable benefit.

A major advance in solving this problem has been made by the methods and apparatus disclosed in the sixth U.S. patent cited above, US 4,912,416.

Î As is well known to those skilled in the art, the state of charge of a battery is directly related to its open circuit (unloaded) voltage. Utilizing this fact, along with extensive experimental data, an empirical relationship has been established between the state of charge of a battery, reflected by its open circuit voltage, and its relative dynamic conductance, normalized to its fully charged value. *This empirical relationship was first disclosed in U.S. Patent No. 4,912,416. Furthermore, devices disclosed therein utilized this empirical relationship, along with open circuit voltage measurements, to appropriately correct dynamic conductance readings - thus providing battery ratings that are substantially independent of the battery's state of charge.

The measuring device disclosed in U.S. Patent No. 4,912,416 lacked any form of connection between a means for measuring the open circuit voltage and the means for correcting the dynamic conductance readings. Consequently, it required an inconvenient two-step test procedure using an intermediate operator intervention. First, the open circuit voltage of the battery was measured. Then, using the results of the voltage measurement, the user set a variable voltage.

ible attenuator to a suitable setting. Finally, the dynamic conductance was measured. Due to the previously set variable attenuator, the quantitative or qualitative conductance information ultimately displayed to the user corresponded to that of a fully charged battery, even if the battery was actually only partially charged during the test.

The problem of the state of charge was thus in principle solved by the methods and

ββ devices. However, the procedure required was somewhat inconvenient. It was therefore quite obvious that an improved device which provides an automatic state of charge correction - a correction which does not require any intermediate intervention by the user - would be very advantageous. Such an improved electronic battery testing device which provides automatic compensation for a low state of charge is disclosed hereinafter.

The above-referenced U.S. Patent No. 4,825,170 teaches a method and apparatus for automatically selecting an appropriate measuring circuit within a dynamic conductance-based battery tester based on the nominal (not actual) terminal voltage. That is, it teaches an apparatus which, for example, distinguishes between a battery having a voltage rating of 6V and a battery having a voltage rating of 12V and tests that battery in a manner appropriate for its particular rating. Although the selection of the appropriate circuit is automatic, the voltage measuring circuit cannot determine the actual terminal voltage of the battery. It only responds to whether the voltage of the battery is greater or less than a fixed threshold. Thus, unlike the method described herein,

disclosed device, such a device is not capable of implementing a state of charge correction.

SUMMARY OF THE INVENTION

The present invention provides an electronic device for testing an electrochemical cell or battery having a dynamic parameter which is dynamic conductance or dynamic resistance, an open circuit voltage and a state of charge, the electronic device comprising means electrically connected to the cell or battery for measuring the dynamic parameters and means electrically connected to the cell or battery for measuring the open circuit voltage, the electronic device further characterized by:

a correction device connected to the device for measuring the dynamic parameter and directly electrically connected to the device for measuring the open circuit voltage, the correction device serving to respond to the open circuit voltage by adjusting a measured dynamic parameter value depending on the open circuit voltage to obtain a state of charge-corrected dynamic parameter value; and

a device for displaying a test result corresponding to the state of charge corrected parameter value.

Three embodiments of an improved electronic device for testing and monitoring storage batteries which may be only partially charged are disclosed. The first embodiment uses an analog circuit of the type described in U.S. Patent No. 4,912,416 to measure the dynamic conductance of the battery. This dynamic conductance is automatically corrected for the state of charge by a non-linear correction amplifier which

• · 4 ♦ «

electrically connected directly to the device for measuring the open circuit terminal voltage of the battery. The second embodiment also uses an analog circuit for measuring the dynamic conductance. However, in this embodiment, the state of charge correction is performed by a microprocessor which is electrically connected directly to the device for measuring the open circuit terminal voltage of the battery. The third embodiment uses an analog circuit for measuring another dynamic parameter - the

Î dynamic resistance of the battery. A microprocessor, directly electrically connected to the means for measuring the open circuit terminal voltage of the battery, in turn performs a state of charge correction. In the third embodiment, however, the microprocessor can mathematically invert the measured dynamic resistance to obtain the dynamic conductance if a numerical result proportional to the battery performance is desired. If only a qualitative ("pass/fail") result is desired, this last step is not necessary. Two common elements link the three embodiments. These are: 1). The existence of a direct electrical coupling between the means for sensing the open circuit voltage and the means for correcting the measured dynamic parameter; and 2). The ability of the correction means to respond to the coupled open circuit voltage by appropriately correcting the dynamic parameter with respect to the state of charge. In each of the three embodiments, the device determines the dynamic small-signal conductance of the battery to produce either a proportional numerical result displayed in appropriate battery measurement units or a corresponding qualitative assessment of the relative condition of the battery based on the dynamic conductance and the electrical rating. Without additional

-8th-

With manual user intervention, the instrument also measures the battery terminal voltage in a substantially unloaded condition and electrically couples this information directly to a suitable means for correcting the measured dynamic conductance for state of charge. This automatic correction makes the quantitative or qualitative information presented to the user that of a fully charged battery even when the battery may actually be only partially charged. If the battery state of charge is too low to make an accurate assessment, no information is displayed. Instead, an indication is given to the user that the battery must be charged before the test.

The electronic battery tester/monitor with automatic low state of charge compensation can be used to provide either a qualitative or quantitative evaluation of a wide variety of electrochemical energy sources other than lead acid batteries. For example, individual electrochemical cells can be tested or monitored in a manner identical to that used to test or monitor batteries connected in series. Furthermore, by using appropriate numerical correction factors, the invention can be used to test or monitor other electrochemical energy sources such as alkaline, nickel-cadmium, or lithium cells and batteries. The present invention is widely applicable to such testing or monitoring of energy sources because of its simplicity, safety, accuracy, ease of use, and low cost.

DF ₁ R ZKTCHNTTNGFIN

In the drawings:

Fig. 1 shows Thevenin's equivalent circuit of a lead-acid storage battery with its open circuit voltage V ₀ and its series-connected internal resistance R _x .

Fig. 2 is an empirical plot of the normalized dynamic conductance G _x versus the open circuit voltage V ₀ , showing the correlation with measurements taken on four different lead-acid storage batteries having different electrical ratings and built by different manufacturers.

Fig. 3 is a simplified block diagram of an improved electronic battery tester/monitoring device employing automatic low state of charge compensation in accordance with a first embodiment of the present invention.

Fig. 4 is a graphical representation of the state of charge correction factor obtained by determining the reciprocal of the empirical normalized dynamic conductance curve of Fig. 2.

Fig. 5 is a plot of a four-segment piecewise linear approximation of the correction factor curve of Fig. 4 implemented by the correction amplifier circuit disclosed in Fig. 6.

Fig. 6 is a circuit diagram of a DC correction amplifier embodiment which implements the four-segment piecewise linear transfer function shown in Fig. 5.

Fig. 7 is a complete circuit diagram of an improved electronic battery test/monitoring device, with automatic compensation of a low state of charge, which is used to test

••»•-10*- '"

ing/monitoring 12 V car batteries according to a first embodiment of the present invention.

Fig. 8 is a simplified block diagram of a second embodiment of an improved electronic battery test/monitoring device with automatic compensation for a low state of charge.

Fig. 9 is a simplified block diagram of a third embodiment of an improved electronic battery tester/monitor with automatic low state of charge compensation.

DFTAIT.T.TKRTF. RKSClHRFTBTTNC? DFR PREFERRED EXECUTIONJNC4.qFORMFN

Fig. 1 shows Thevenin's equivalent circuit of a lead-acid storage battery. In this equivalent representation, the battery is described by its open circuit voltage V ₀ and its internal resistance R _x connected in series.

As fully disclosed in the first five of the six U.S. patents cited above, a conventional dynamic conductance battery test of a fully charged battery typically neglects the open circuit voltage V ₀ . Instead, the electronic test device directly measures the dynamic conductance G _x = 1/R _x . The test/monitor then provides the operator with either a numerical indication expressed in proportional battery measurement units (such as "cold cranking amps,""amp-hours," or "watts") or, otherwise, a qualitative indication ("Pass/Fail") based on a comparison of the measured G _x value with a corresponding reference value determined by the battery's electrical rating and temperature.

Although the open circuit voltage V ₀ is not normally used in the dynamic conductance test of fully charged batteries,

, it was previously used to determine the state of charge. As is well known to those skilled in the art, the state of charge of a battery is directly related to its (unloaded) open circuit terminal voltage. For example, in lead-acid automotive batteries with a nominal voltage of 12 V, the open circuit voltage is known to vary from about 11.4 V for batteries that are practically completely discharged to about 12.6 V for batteries that are almost fully charged.

× Fig. 2 illustrates the observed relationship between normalized dynamic conductance and open circuit voltage applicable to a large class of automotive lead-acid storage batteries. This information was originally disclosed in U.S. Patent No. 4,912,416. Fig. 2 illustrates an empirical graph of the relative dynamic conductance normalized to the fully charged value, G _x (V ₀ )/G _x (12.6) , plotted as a function of open circuit voltage V ₀ . The solid curve plotted in Fig. 2 is described by a second order polynomial equation having coefficients fitted to best fit the experimental data. The appropriately fitted polynomial equation is

Equation 1:

=-{78.1963} +{12.3939 }F ₀ - {0.4848 }F ₀ ²
)

Fig. 2 also discloses standardized experimental points representing actual measurements obtained from four different batteries with different electrical ratings and built by different manufacturers. The batteries XMO, XMl and XM3 are six-cell batteries with nominal voltages of 12 volts. The

••"-12-

Battery XM2 is a three-cell, 6-volt battery. The open circuit voltage readings of battery XM2 were multiplied by a factor of 2 to plot XM2 data points on the same graph as the other three batteries. It can be seen that the measurements obtained from all batteries agree quite well with the empirical relationship described by Equation 1. The fact that the same empirical relationship shows a strong correlation with experimental data obtained from both 6-volt and 12-volt batteries shows that the empirical state of charge correction disclosed in Figure 2 is quite universal, and is in fact a fundamental property of a single cell.

Referring now to Fig. 3, there is disclosed a simplified block diagram of a first embodiment of an improved electronic battery tester/monitor with automatic low state of charge compensation. Except for specific details relating to the automatic low state of charge compensation circuitry, the explanation of the operation of the block diagram of Fig. 3 is identical to that of the corresponding block diagram designated Fig. 1 in U.S. Patent No. 4,816,768.

Accordingly, signals representing the signal at the output 10 of the high gain amplifier cascade 12 are fed back to an input 20 of the high gain amplifier cascade 12 via two feedback paths; an internal feedback path 14 and an external feedback path 16. The internal feedback path 14 includes a low pass filter (LPF) 18 and feeds a signal directly back to the input 20 of the high gain amplifier cascade 12. The purpose of the internal feedback path 14 and the low pass filter 18 is to generate a strong direct current (DC)

Negative feedback but relatively weak alternating voltage (AC) feedback to define the operating point of the high gain amplifier cascade 12 and to ensure its DC stability without appreciably reducing its AC voltage gain. The external feedback path 16 includes a resistor network 22 and feeds a signal current back to the battery 24 under test. The summing circuit 26 combines the resulting signal voltage 28 thereby generated across the battery 24 with a periodic square wave signal voltage 30.

In the embodiment disclosed in Fig. 3, the periodic square wave signal voltage 30 is generated by the operation of an oscillator 32, a chopper switch 34 and a correction amplifier (Cor Amp) 36. The oscillation frequency of the oscillator may be, for example, 100 Hz. The voltage applied to the input 38 of the DC correction amplifier 36 is the DC terminal voltage of the battery 24. Due to the fact that the electronic device shown in Fig. 3 draws only a very small load current from the battery, this terminal voltage is essentially the open circuit (unloaded) terminal voltage V ₀ . The output signal 40 of the DC correction amplifier 36 is a DC voltage derived from V ₀ with a voltage level that is inversely related to V ₀ and thus inversely related to the state of charge of the battery 24. This derived DC voltage 40 is repeatedly interrupted by the chopper switch 34, whose control input 42 is activated by the output signal of the oscillator 32. The interrupted DC voltage thus comprises a periodic square wave signal voltage 30 with a voltage amplitude that is inversely related to V ₀ and thus inversely related to the state of charge of the battery 24. This signal voltage 30 is applied to the summation circuit 26 together with the signal developed across the battery 24.

voltage 28. The resulting composite signal voltage 4 4 at the output of the summation circuit 2 6 is then capacitively coupled to the input 20 of the high-gain amplifier cascade 12 by means of a capacitive coupling network 46.

As already fully explained in U.S. Patent No. 4,816,768, the voltage at the output 10 of the high gain amplifier cascade 12 comprises a constant DC bias component together with an AC signal component,

σ which is proportional to the dynamic conductance G _x of the battery 24 under test as well as to the level of the square wave signal voltage 30. The constant DC bias component is ignored while the variable AC signal component is detected and accurately converted to a DC signal voltage by a synchronous detector 48 which is synchronized to the oscillator 32 by means of a synchronization path 50.

The DC signal voltage at the output 52 of the synchronous detector 48 passes through the adjustable network 54 to the input of the DC coupled operational amplifier 56. The feedback path 58 of the operational amplifier 56 includes a DC milliammeter 60. Accordingly, the reading of the DC milliammeter 60 is proportional to the DC signal voltage level at the output 52 of the synchronous detector 48 and hence to the dynamic conductance G _x of the battery 24. In addition, the proportionality constant with respect to the milliammeter reading G _x is determined by the value assumed by the adjustable resistor network 54 and the level of the signal voltage at 30 - and hence by the state of charge of the battery which is represented by its unloaded DC terminal voltage V ₀ .

By using a suitably set resistance value in the resistor network 54 and then by potassium

·

3 emulates the direct reading battery tester disclosed in U.S. Patent No. 3,873,911. In addition, as fully taught in U.S. Patent No. 4,816,768, the resistance of the resistor network 54 which drives the reading of the DC milliammeter 60 to a specific set value is directly proportional to the dynamic conductance of the battery 24. Thus, by calibrating the resistor network 54 in conventional battery power units and then assigning "pass" and "fail" ranges on the scale of the milliammeter, the embodiment disclosed in FIG. 3 also emulates the direct reading battery tester disclosed in U.S. Patent No.

3,909,708. Furthermore, one can use a switch to select either a fixed value network 54 or an adjustable value network 54 and can arrange both a numerical scale and "good-bad" ranges on the display of the milliammeter. One can therefore realize both a direct display battery tester and a "good-bad" battery tester with only one device.

For each emulation, the amplitude of the detected signal at the output 52 of the synchronous detector 48 is directly proportional to the amplitude of the square wave signal 30 at the output of the chopper switch 34. Thus, both the level of the numerical quantity displayed during the direct reading operation and the relative qualitative rating provided during the "Pass/Fail" operation are affected by the state of charge of the battery represented by its unloaded terminal voltage V ₀ . In order for the information displayed to be unaffected by the battery state of charge, it is necessary that V _out , the DC output voltage at 4 0 on the cor-

—16 —

rectification amplifier 36 is proportional to the reciprocal of G _x (V ₀ ). Under these circumstances, V _ou t can be written as:

Equation 2:

V _out (V ₀ ) = V ₀ Ut (12, 6) x F(V ₀ )

where:
Equation 3:

G _X (U,6) G _x (V ₀ )

is a suitable state of charge correction factor fed into the correction amplifier 36. A transformation of equation 2 leads to:

Equation 4:

F(V ₀ ) =

V _out (12.6)

which shows that F(V ₀ ) can be considered simply as the DC output voltage of amplifier 38 normalized to the DC output voltage obtained with a fully charged battery; ie, for a battery where V ₀ = 12.6 V.

In addition to producing a DC output signal 40, the correction amplifier 36 also provides a "chopper lock" output signal 62 and an LED output signal 64. These two additional output signals are activated when the battery terminal voltage V ₀ and thus its state of charge is too low to perform an accurate dynamic conductance test. Under these special conditions, the chopper switch 34 is locked so that no qualitative or quantitative information is displayed to the user. Instead of

LED 66 will light up to indicate to the user that the battery must be recharged before it can be tested.

Fig. 4 is a graphical plot of the state of charge correction factor F(V ₀ ) obtained by taking the reciprocal of the empirical curve of [G _x (V ₀ )/G _x (12,6)] shown in Fig. 2. A four-segment piecewise approximation for this empirical curve is disclosed in Fig. 5. The parameters specifying the four breakpoints of this piecewise linear approximation are listed in Table 1.

TARF.TTE T
PARAMETERS OF EACH STfTPKWKTSE T.TNEARF.N APPROXIMATION

V ₀ (Volts) Correction factor F Breakpoint A 12.60 1.00 B 12.15 1.21 C 11.80 1.78 D 11.60 2.91

The piecewise linear input/output relationship of Fig. 5 is implemented by the transfer function of the embodiment of a DC correction amplifier circuit disclosed in Fig. 6. Referring to Fig. 6, the DC correction amplifier contained in block 36 comprises a circuit of four operational amplifiers 70, 92, 74 and 76 together with a comparator 78. The circuit input line 38 connects to the positive terminal of the battery 24 under test, while the negative battery terminal is grounded. Due to the fact that very little current is drawn from the battery 24, the

Circuit input voltage at 38, Vi _n , measured with respect to ground, is substantially equal to the open circuit terminal voltage V ₀ .

Within the DC correction amplifier circuit disclosed in Fig. 6, a constant reference voltage V _R is generated by means of a voltage reference diode 80, which receives its operating current through a series resistor 82. The reference voltage V _R can be, for example, 2.5 V. The V _R is also operated on a voltage divider chain consisting of resistors 84, 86, 88 and 90. Accordingly, the

β voltage levels V _R applied to the non-inverting inputs of the operational amplifiers 70, 72, while increasingly smaller portions of the reference voltage V _R are applied to the non-inverting inputs of the operational amplifiers 74, 60 and to the inverting input of the comparator 78.

In addition to these fixed voltage levels, a variable voltage V _x , proportional to battery voltage V ₀ , is derived from V _in by means of voltage divider resistors 92 and 94. This variable voltage is applied directly to the non-inverting input of comparator 78 and to the inverting inputs of operational amplifiers 72, 74 and 76 through resistors 98, 102 and 106.

The outputs of the four operational amplifiers 70, 72, 74 and 76 are connected to a common bus V _out via four diodes 108, 110, 112 and 114. Due to the operation of the four diodes, only one of the operational amplifiers is active at any given time - the amplifier with the largest (highest positive) output voltage. Only this operational amplifier is connected to the output bus and thus controls the output voltage V _out . The other three operational amplifiers, those which have smaller output voltages, are thereby disconnected from the output bus.

• m ?? * φ · *·* 9 ·

by having reverse biased diodes in their output circuits.

The operational amplifier 70 has its inverting input connected directly to the output bus, and is therefore configured as a voltage follower amplifying with a gain of 1 the reference voltage of V _R . The other three operational amplifiers 72, 74 and 76 use feedback resistors and are configured as inverting amplifiers; each amplifying the variable voltage V _x ; and each producing a negative incremental voltage gain given by the corresponding resistance ratio {R(96)/R(98)}, {R(100)/R(102)}, {R(104)/R(106)}, respectively.

The circuit of Fig. 6 operates as follows: For V _in > V _A = 12.6 V, the output voltage of op-amp 70 is greater than the output voltages of the other three op-amps. Consequently, the output bus is controlled by voltage follower 70 with a gain of unity so that V ₀ Ut = V _R This region of constant output voltage is represented by segment 1 in the piecewise linear transfer function shown in Fig. 5.

When V _in has dropped to V _A = 12.6 V, V _x has become sufficiently less than V _R so that the output voltage of the inverting amplifier 72 is equal to that of the amplifier 70. Thus, for Vi _n < V _A, the diode 108 is reverse biased while the diode 110 is forward biased, and the amplifier 72 controls the output bus. Due to the gain of the inverting amplifier 72, a further decrease in Vi _n causes V _ou t to rise with an incremental gain or "slope" of
{R(96)/R(98)} . This range of amplification, which lasts until V _in is equal to V ₃ , is represented by segment 2 in Fig. 5.

&bull; » * * · ■· k4f&diams;

'-2(3-

equal to V _B is represented by segment 2 in Fig. 5.

When V _in has dropped to V _B , V _x has dropped sufficiently that the output voltage of inverting amplifier 74 exceeds that of amplifier 72. Diode 110 is therefore reverse biased while diode 112 is forward biased and amplifier 74 now controls the output bus. Due to the gain of inverting amplifier 74, a further decrease in Vi _n γ causes V _ou t to rise ^with the larger incremental gain or "slope" of -{R(100)/R(102)}. This gain region continues until Vi _n is equal to V _c and is represented by segment 3 in Fig. 5.

When V _in has dropped to V _c , V _x has dropped sufficiently that the output voltage of inverting amplifier 76 exceeds that of amplifier 74. Diode 112 is therefore reverse biased while diode 114 is forward biased. Thus, amplifier 7 6 now controls the output bus. Due to the gain of inverting amplifier 76, a further drop in V _1n causes V _ou t to rise with the even larger incremental gain or "slope" of -{R(104)/R(106)}. This region of greatest gain is represented by segment 4 in Fig. 5.

Finally, for V _less than V _D, the desired voltage V _x will be less than the divided reference voltage present at the junction of resistors 88 and 90. Under these special conditions, the non-inverting input of comparator 78 is at a lower potential than the inverting input, causing the output of comparator 78 to be in a "low" state. Consequently, LED 66 will illuminate to provide the user with a

&bull;-2&Idigr;-

to give an indication that the battery must be recharged before it can be tested. In addition, the output line 62 is in a "low" state, thus disabling the chopper switch 34 and preventing any qualitative or quantitative dynamic conductance information from being output to the user.

Fig. 7 discloses a complete circuit diagram of a first embodiment of an improved electronic battery tester/monitor with automatic charge level compensation configured for testing/monitoring 12 volt batteries in accordance with the present invention. Operational amplifiers 120, 122, 124 and 126 represent four elements of an integrated circuit IC 1 with four operational amplifiers. Bilateral analog switches 34 and 28 represent two elements of an integrated circuit IC 2 with four bilateral CMOS switches. The operational amplifiers 70, 72, 74 and 76 represent four elements of an integrated circuit IC 3 with four operational amplifiers. The comparator 78 represents one element of an integrated circuit IC 4 with four comparators. All four integrated circuits IC 1, IC 2, IC 3 and IC 4 are supplied with energy by means of common supply lines V ⁺ 130 and V" 132, which are connected to the battery 24 to be tested via battery contacts 134 and 136, respectively.

The high-gain amplifier cascade 12 of Fig. 3 comprises an operational amplifier 120 and an emitter-follower connected npn transistor 138. The resistor 140 supplies a DC bias voltage to the non-inverting (+) input of the operational amplifier 120 from voltage divider resistors 142 and 144, which are connected to the battery 24 via battery contacts 146 and 148. The output voltage of the high-gain amplifier cascade 12 is fed via the

Feedback resistor 22 of the external path. An internal feedback path consisting of resistors 150 and 152 passes the DC voltage at the common connection between the emitter of npn transistor 138 and resistor 22 to the inverting (-) input of operational amplifier 120. Resistors 150 and 152 together with capacitor 154 form low-pass filter 18 of Fig. 3.

The AC signal voltage developed across the battery 24 is detected at the battery contacts 146 and 148 and in series

σ is added to an input signal voltage component developed across the viewing resistor 156. The resulting composite AC signal voltage is coupled to the differential input of the operational amplifier 120 by means of a capacitive coupling network consisting of capacitors 158 and 160. A feedback current proportional to the voltage developed across the resistor 22 passes through the battery 24 by means of the conductors 162 and 164 of the external feedback path, together with the battery contacts 134 and 136.

A square wave input signal voltage is developed across the viewing resistor 156 and shaped by the action of the oscillator 32, the chopper switch 34 and the correction amplifier 36. The oscillator 32, which produces a 100 Hz square wave synchronization signal, is a conventional astable multivibrator consisting of the operational amplifier 122 together with resistors 168, 170, 172, 174 and capacitor 176. The synchronization output of the oscillator 32 is applied to the control input of the chopper switch 34 through a resistor 178.

Consequently, the chopper switch 34 switches off and on periodically at a 100 Hz rate. The signal terminals of the chopper switch 118 connect the DC signal output 40 of the

·

·

·

·

correction amplifier 36 to the input line of the trimmer potentiometer 180 used for an initial calibration adjustment. The signal voltage across the trimmer potentiometer 180 therefore comprises a 100 Hz square wave with an amplitude proportional to the DC output voltage of the correction amplifier 36. A signal current proportional to the signal output voltage of trimmer potentiometer 180 passes through injection resistor 184 and is injected into viewing resistor 156, thereby producing a 100 Hz signal voltage across viewing resistor 156. Due to the action of correction amplifier 36 described with reference to Fig. 4, the signal voltage " across viewing resistor 156 includes an automatic correction for the state of charge of the battery under test. However, if the battery state of charge is too low for accurate battery evaluation, output lines 62 and 64 are at logic low states. These output lines, in turn, pull the input of chopper switch 34 low and pull the cathode of LED 66 low. As a result, chopper switch 34 is disabled so that no AC signal is injected into viewing resistor 156, and LED 66 is illuminated to indicate to the user that the battery must be recharged before a dynamic conductance test can be performed.

The analog switch 128, together with the operational amplifier 124, which is connected as an integrator, forms the synchronous detector 48 of Fig. 3. A resistor 194 and shunt capacitor 126 form a low-pass filter which biases the non-inverting input of the operational amplifier 124 to the voltage level of the DC bias component developed across the feedback resistor 22. A voltage derived from the total voltage at the common junction between the resistor 22 and the transistor 138

Signal current passes through resistor 198 and analog switch 128 to the inverting input of operational amplifier 124. This signal current is interrupted periodically at the oscillator frequency by the control signal of analog switch 128 which is connected to the synchronization output of oscillator 32. Resistor 200 provides DC negative feedback to operational amplifier 124. Integration capacitor 202 serves to smooth the detected voltage signal output by operational amplifier 124.

The non-inverting input of operational amplifier 126 is biased to the DC level of the non-inverting input of operational amplifier 124, while the inverting input of operational amplifier 126 is connected to an SPDT selector switch 206. Accordingly, a DC current proportional to the detected signal voltage at the output of operational amplifier 124 passes through milliammeter 60 to the output of operational amplifier 160 via one of two paths selected by selector switch 206. With switch 206 in position 1, the instrument current passes through fixed resistor 208. Under these conditions, the disclosed invention emulates a direct reading battery tester which provides a quantitative output displayed in battery measurement units proportional to the dynamic conductance of battery 24. With switch 206 in position 2, the instrument current passes through a fixed resistor 210 and a variable resistor 212. Under these conditions, the disclosed invention emulates a qualitative "pass/fail" battery tester with a manually adjusted battery rating scale that is linearly dependent on the setting of variable resistor 212 and a rating offset determined by the value of fixed resistor 210.

The embodiment of the improved battery test/monitor with automatic low charge compensation shown in Fig. 7 operates as follows: The operator simply connects the device to the battery to be tested and selects one of the two positions of the selector switch 206. When position 1 is selected, the milliammeter 60 displays the quantitative condition of the battery in appropriate battery measurement units, with the displayed quantitative results automatically

&iacgr;&ogr; are adjusted to correspond to those of a fully charged battery. When switch 206 is in position 2 and variable resistor 212 is adjusted to correspond to the battery rating, ammeter 60 indicates the qualitative ("pass/fail") condition of the battery. Again, the displayed results are automatically adjusted to correspond to those of a fully charged battery. In either selection position, if the battery charge level is too low for an accurate evaluation, no information is displayed to the user. Instead, an LED is illuminated to indicate to the user that the battery must be charged before testing.

Table II contains a list of components and values for the first embodiment of an improved electronic battery tester/monitor with automatic low state of charge compensation disclosed in Figure 7.

-26-

TABLE II

COMPONENT TYPE &Tgr;&Tgr;&Ngr;&Ggr;&Pgr; ffiVALUE. FTTR DTK SCT-TAT ₁ TTJNfG OF FTfi. REFERENCE NUMBER COMPONENT

Semiconductor components

120, 122, 124, 126 ICl - LM324N

34, 128 IC2 - CD4066B

70, 72, 74, 76 IC3 - LM324N

78 IC4 - LM339

80 IC5 - LM336-2.5

138 TIP31C Power transistor

108, 110, 112, 114 1N4148 diodes

66 T-I 3/4 LED

Resistances - Ohm (^-W unless otherwise stated)

22 22Ω- - 5 Watt

82 4.7K

84 5.36K

86 6.19K

88 6.04K

90 200K

92 2.25K

94 576

96, 100, 104 &Igr;,&Ogr;&Ogr;&Mgr;

98 174K

102 49.9K

106 13.7K

116, 142, 144 l,0K

156, 210 100

208 470

212 500 Potentiometer

180 1OK trim potentiometer

184 200 198 470K 140, 194, 47K 178, IOOK 172 168, 170 150K 174 152, 270K 150, 1M

Capacitors -

176 158, 160, 1 96 0, 022 154, 0, 47 202 instrument 1, 0

202 1 mA DC milliammeter

Switch
206 SPDT

Fig. 8 discloses a simplified block diagram of a second embodiment of an improved electronic battery tester/monitor with automatic low state of charge compensation. This embodiment eliminates the correction amplifier 36 and chopper switch 34 of the first embodiment disclosed in the block diagram of Fig. 3. Instead, it includes a microprocessor generally represented by block 220 of Fig. 8.

The microprocessor block 220 contains all the usual elements that make up a microprocessor system, such as the necessary logic elements, a clock oscillator, a random access memory, a firmware program contained in a read-only memory and, of course, the processor itself. In the embodiment of Fig. 8, the appropriate correction for the low charge state is carried out by the microprocessor.

·

-28-

processor block 220 controlled by the firmware program contained therein.

A description of the operation of most of the elements of Fig. 8 parallels the description of the operation of the embodiment disclosed in Fig. 3. Signals representing the signal at the output 10 of the high gain amplifier cascade 12 are fed back to the input 20 of the high gain amplifier cascade 12 by means of two feedback paths; an internal feedback path 14 and a γ feedback path 16. The internal feedback path 14 includes a low pass filter (LPF) 18 and feeds a signal voltage directly back to the input 20 of the high gain amplifier cascade 12. The external feedback path 16 includes a resistor network 22 and feeds a signal current back to the battery 24 under test. The summing circuit 26 combines the resulting signal voltage 28 thereby developed across the battery with a periodic square wave signal voltage 30.

In the embodiment of Fig. 8, the signal voltage 30 simply comprises the constant output signal of the oscillator 32. The oscillation frequency of the oscillator 32 may be, for example, 100 Hz. This periodic signal is presented to the summing circuit 26 together with the signal voltage 28 generated across the battery 24. The resulting composite signal voltage 4 4 at the output of the summing circuit 26 is then capacitively coupled to the input 20 of the high gain amplifier cascade 12 by means of a capacitive coupling network 46. Consequently, the voltage at the output 10 of the high gain amplifier cascade 12 comprises a constant DC bias component together with an AC signal component which is proportional to the dynamic conductance G _x of the battery 24 under test. The constant DC

. . .· ii i.ii *► *

Bias component is ignored while the variable AC signal component is detected and accurately converted into a DC signal voltage by the synchronous detector 48, which is synchronized to the oscillator device via the synchronization path 50.

The DC signal level at the output 52 of the synchronous detector 50 is proportional to the dynamic conductance G _x of the battery. This analog voltage is converted into a corresponding digital representation of G _x by an analog/digital (A/D) converter

222 and then input to the microprocessor block 220 via an input terminal 224. In addition, the unloaded battery voltage V ₀ is connected to the input of an analog-to-digital converter 226 via the DC path 38. A corresponding digital representation of V ₀ at the output of the A/D converter 226 is thereby input to the microprocessor block 220 via the input terminal 228.

Through programmed algorithmic techniques known to those skilled in the art, the microprocessor firmware program uses the digital representation of V ₀ to correct the digital representation of G _x for the state of charge of the battery. This can be done, for example, by entering V ₀ into a "look-up table" whose output is the appropriate correction factor F, and then multiplying G _x by the resulting factor F to obtain the corrected conductance G _x (12,6). Alternatively, the correct value of G _x (12,6) could be calculated directly by numerically calculating the reciprocal of the empirical relationship disclosed in Equation 1.

To emulate a quantitative electronic battery tester, a numerical value proportional to G _x (12.6) is output and displayed on a digital display such as

236 connected to an output terminal 234 or printed by a printer such as 240 connected to the microprocessor 220 via an output terminal 238. In addition, whenever V ₀ is less than a predetermined minimum value, the firmware program suppresses the numerical display and instead provides an indication to the user that the battery must be charged before testing. This special information may be displayed, for example, by the digital display 236, printed by a printer 240, or provided to the user via an LED 76 connected to the microprocessor 220 via an output terminal 242.

To emulate a qualitative electronic ("pass/fail") battery tester, the battery rating is first input to the microprocessor 220 through an input device such as a shaft encoder 230 connected to the microprocessor 220 through an input terminal 232. A dial associated with the shaft encoder is calibrated in battery rating units such as cold cranking amps or amp-hours. Through programmed algorithmic techniques well known to those skilled in the art, the microprocessor firmware program then instructs the microprocessor block 220 to compare the state-of-charge corrected dynamic conductance with a nominal value appropriate for the entered battery rating and output the resulting pass/fail information to the user. This qualitative output information may be displayed, for example, by a digital display 236, printed by a printer 240, or provided to the user via an LED 246 connected to the microprocessor 220 via an output terminal 244. Again, if V ₀ is less than a predetermined minimum value

suppresses the information displayed and informs the user that the battery must be charged before the test. This special information may be displayed, for example, by the digital display 236, printed by the printer 240, or conveyed to the user by the LED 66.

Fig. 9 discloses a simplified block diagram of a third embodiment of an improved electronic battery tester/monitor with automatic low state of charge compensation. Like the embodiment disclosed in Fig. 8, this third embodiment includes a microprocessor block 220 for implementing the appropriate low state of charge correction. It differs from the embodiment of Fig. 8 in that the hardware presents a digital representation of the battery's dynamic resistance R _x to the microprocessor 220, which then uses a firmware program to calculate the reciprocal of the dynamic conductance G _x = 1/R x and to perform a battery low state of charge correction.

The hardware disclosed in Fig. 9 functions as follows: An oscillator 32 generates a periodic square wave signal 42 which is output directly to the current amplifier 244. The oscillation frequency of the oscillator 32 may be, for example, 100 Hz. The output of the current amplifier 244, a periodic signal current 246, then passes through the battery 24. Due to the fact that the output resistance of the current amplifier 244 is substantially greater than the dynamic resistance R _x of the battery, the amplitude of the signal current 246 is practically independent of R _x . Consequently, the resulting AC signal voltage 248 appearing across the battery terminals is directly proportional to the dynamic resistance R _x of the battery. The capacitive coupling

-its* ti · · * ··

-32-

network 250 couples the AC signal voltage 248 to an input 252 of a voltage amplifier 254. This coupling network suppresses the DC terminal voltage of the battery, but allows the AC signal voltage to be amplified by the voltage amplifier 254. The output voltage 256 of the voltage amplifier 254 represents the input voltage to the synchronous detector 48, which is synchronized to the oscillator 32 via the synchronization path 50. As a result, a DC signal voltage 52 is produced at the output of the synchronous detector 48 that is directly proportional to the dynamic resistance R _x of the battery.

The analog voltage 52 is converted to a corresponding digital representation of R _x by an analog-to-digital converter 222 and then input to the microprocessor block 220 via input terminal 224. In addition, the battery open circuit voltage V ₀ is connected to the input of the analog-to-digital converter 226 via a DC path 38. A corresponding digital representation of V ₀ at the output of the A/D converter 226 is then input to the microprocessor block 220 via input terminal 228.

Through programmed algorithmic techniques well known to those skilled in the art, the microprocessor firmware program instructs the microprocessor block 220 to invert the digital representation of R _x to obtain a digital representation of the dynamic conductance G _x of the battery. It then uses the digital representation of V ₀ to correct the digital representation of G _x for the battery state of charge. This can be done, for example, by entering V ₀ into a "look-up table" whose output is the appropriate correction factor F, and then multiplying G _x by the resulting factor F to obtain the corrected conductance

G _x (12,6); or by directly calculating G _x (12,6) from the reciprocal of the empirical relationship disclosed in Equation 1. Alternatively, a corrected value of R _x , R _x (12,6), can be calculated first and then algorithmically inverted to obtain G _x (12,6).

To emulate a quantitative electronic battery tester, a numerical result proportional to G _x (12.6) is output and shown on a digital display, such as

The numerical display may be displayed on a digital display such as 236 connected to an output terminal 234 or printed on a printer such as 240 connected to the microprocessor 220 via an output terminal 238. In addition, whenever V ₀ is less than a predetermined minimum value, the firmware program suppresses the numerical display and instead generates an indication to the user that the battery must be charged before testing. This special information may be displayed, for example, by the digital display 236, printed by the printer 240, or provided to the user via an LED 66 connected to the microprocessor 220 via the output terminal 242.

To emulate a qualitative electronic ("Pass/Fail") battery tester, the battery rating is first input to the microprocessor 220 through an input device, such as a shaft encoder 230 connected to the microprocessor 220 through an input terminal 232. A dial associated with the shaft encoder is calibrated in battery rating units, such as cold cranking amps or amp-hours. Through programmed algorithmic techniques well known to those skilled in the art, the microprocessor's firmware program then instructs the battery to be rated at the correct voltage.

microprocessor, the microprocessor block 220 is to compare the calculated quantity G _x (12.6) with a reference quantity corresponding to the inputted battery rating to determine whether the battery is good or bad. If the calculated quantity is greater than the reference quantity, the battery is good. Otherwise, it is bad. Alternatively, a comparison can be made between the calculated quantity R _x (12.6) and a reference quantity corresponding to the inputted battery rating to determine whether the battery is good.

γ or bad. If the calculated value is less than the reference value, the battery is good. Otherwise, it is bad. In either case, this qualitative output information is displayed by the digital display 236, printed by the printer 240, or brought to the user by the LED 246 which is connected to the microprocessor 220 by an output terminal 244. Again, if V ₀ is less than a predetermined minimum value, the display of the information is suppressed and the user is informed that the battery must be charged before the test. This particular information may, for example, be displayed by the digital display 236, printed by the printer 240, or brought to the user by the LED 66.

Although three specific ways of carrying out the invention have been described, it should be understood that many modifications and variations may be made without departing from what is considered to be the scope and spirit of the invention. For example, the invention may consist of a single instrument housed in a housing which is temporarily connected to the battery for testing the battery in situ. Alternatively, this device may consist of a monitoring device which is substantially permanently connected to a battery for continuous monitoring of the battery condition at a remote

-35-

location. In this latter case, the device may be separated into two parts - one part connected to the battery and located at the battery location, while the other part containing the remote display is located at the remote location. The division between the two parts may be made arbitrarily. However, I submit that all such divisions, modifications and variations are within the scope of the invention disclosed herein and are therefore intended to be covered by the appended claims.

Although the present invention has been described with reference to preferred embodiments, those skilled in the art will recognize that changes in form and detail may be made therein without departing from the spirit and scope of the invention.

Claims (18)

1. Electronic device for testing an electrochemical cell or battery (24) which has a dynamic parameter, which is either a dynamic conductance or a dynamic resistance, an open circuit voltage and a state of charge, the electronic device further comprising devices which are electrically connected to the cell or battery (16, 28, 246, 248) for measuring this dynamic parameter (10, 12, 14, 16, 18, 20, 22, 26, 28, 30, 32, 34, 42, 44, 46, 48, 50, 52, 222, 224, 244, 246, 248, 250, 252, 254, 256), and further devices which are connected to the cell or battery for measuring the open circuit voltage (38), the electronic Device is further characterized by:
Correction means (36, 220) connected to the means for measuring the dynamic parameter (40, 62, 226, 228) and directly electrically connected to the means for measuring the open circuit voltage (38, 226, 228), the correction means being responsive to obtain a dynamic parameter value corrected for state of charge by adjusting a detected dynamic parameter value corresponding to the open circuit voltage to said open circuit voltage; and
Means (56, 58, 60, 64, 66, 234, 236, 238, 240, 242, 244, 246) for displaying a test result in accordance with the state of charge corrected dynamic parameter value. 2. Electronic device according to claim 1, wherein the dynamic parameter is the dynamic conductance, and the correction device comprises a non-linear amplifier device (36), which in turn has an input voltage (38) and an output voltage (40), the input voltage being proportional to the open circuit voltage and the output voltage being proportional to the ratio of the dynamic conductance corrected with respect to the state of charge to the detected dynamic conductance. 3. Electronic device according to claim 1 or 2, wherein the electronic device comprises a microprocessor device (220), the dynamic parameter is the dynamic conductance (G _x - Fig. 8), the correction device comprises a firmware correction program which is implemented by means of the microprocessor device, and wherein the digital representations of both the open circuit voltage (228) and the dynamic conductance (224 - Fig. 8) are input to the microprocessor device (220) and algorithmically combined by means of the firmware correction program to obtain the corrected dynamic conductance. 4. Electronic device according to claim 1, wherein the electronic device comprises a microprocessor device (220), the dynamic parameter is the dynamic resistance (R _x - Fig. 9), the correction device comprises a firmware correction program which is implemented by means of the microprocessor device, and wherein the digital representations of both the open circuit voltage (228) and the dynamic resistance (224 - Fig. 9) are input to the microprocessor device (220) and algorithmically combined by means of the firmware correction program to obtain the state of charge corrected dynamic resistance value. 5. Electronic device according to one of claims 1 to 4, wherein the test result comprises a qualitative result (60, 236, 240, 246) determined by comparing the corrected dynamic parameter with a reference value. 6. An electronic device according to claim 5, wherein the electrochemical cell or battery has an electrical power, the electronic device has means for inputting (54, 230, 232) the value of that power, and the reference value is determined by the determined value of the electrical power which is thus input. 7. Electronic device according to claim 2 or 3, wherein the test result (at 60, 236, 240) comprises numbers proportional to the dynamic conductance corrected for the state of charge. 8. Electronic device according to claim 1, wherein the microprocessor means (220 - Fig. 9) mathematically inverts the state-of-charge corrected dynamic resistance value to obtain a state-of-charge corrected dynamic conductance, and the test result (at 236, 240 - Fig. 9) comprises numbers proportional to the state-of-charge corrected dynamic conductance. 9. An electronic device for monitoring or testing an electrochemical cell or battery having a dynamic conductance and an open circuit voltage, comprising:
a device operatively connected to the cell or battery for determining the dynamic conductance;
a device operatively connected to the cell or battery for determining the open circuit voltage;
means for automatically correcting the dynamic conductance in an inverse correspondence to the open circuit voltage to obtain a corrected dynamic conductance; and
a device for displaying test results according to the corrected dynamic conductance. 10. Electronic device according to claim 9, wherein the test results represent numerical values proportional to the corrected dynamic conductance. 11. The electronic device of claim 9, wherein the test results represent qualitative results according to the corrected dynamic conductance with respect to a reference conductance. 12. Electronic device for monitoring or testing an electrochemical cell or battery having a dynamic conductance, an open circuit voltage and a rated electrical power, comprising:
a device operatively connected to the cell or battery for determining the dynamic conductance;
a device operatively connected to the cell or battery for determining the open circuit voltage;
means for automatically correcting the dynamic conductance in an inverse correspondence to the open circuit voltage to obtain a corrected dynamic conductance;
a device for comparing the corrected dynamic conductance with a reference conductance to obtain a qualitative result;
a device for displaying the qualitative result. 13. An electronic device according to claim 12, wherein the means for automatically correcting the dynamic conductance includes a correction amplifier means having an input voltage and an output voltage, the input voltage being proportional to the open circuit voltage and the output voltage being proportional to the ratio of the corrected dynamic conductance to the dynamic conductance. 14. An electronic device according to claim 12, wherein the means for automatically correcting comprises microprocessor means, and wherein digital representations of both the open circuit voltage and the dynamic conductance are input to the microprocessor means and algorithmically combined to obtain the corrected dynamic conductance. 15. Electronic device for testing or monitoring an electrochemical cell or battery having a dynamic conductance, an open circuit voltage and a rated electrical power, comprising:
a device operatively connected to the cell or battery for providing a signal proportional to the dynamic conductance;
a device operatively connected to the cell or battery for measuring the open circuit voltage;
a device for automatically adjusting the level of the signal according to the measured open circuit voltage and
means for indicating a rating of the cell or battery in response to the level of the signal. 16. An electronic device according to claim 15, wherein the means for automatically adjusting the level of the signal according to the measured open circuit voltage adjusts the level in an inverse correspondence to the value of the open circuit voltage with respect to a fully charged value. 17. Electronic device according to claim 16, wherein the means for automatically adjusting the level of the signal according to the measured open circuit voltage comprises a correction amplifier means having an input voltage and an output voltage, the input voltage being derived from the measured open circuit voltage, and the output voltage being proportional to the level of the signal. 18. The electronic device of claim 17, wherein the correction amplifier means comprises a reference voltage means, a plurality of operational amplifier means, a plurality of diode means, and a plurality of resistor means connected together to produce a piecewise linear output voltage/input voltage transfer function.