WO1992015022A1

WO1992015022A1 - Testing heavy current batteries

Info

Publication number: WO1992015022A1
Application number: PCT/AU1992/000055
Authority: WO
Inventors: Anton Schubert; Dennis Charles Mills; Terence Harvey; George Michael Leigh
Original assignee: Commonwealth Scientific And Industrial Research Organisation
Priority date: 1991-02-13
Filing date: 1992-02-13
Publication date: 1992-09-03

Abstract

A tester for testing a heavy current battery (13) includes a load (17) and a flat battery conditioner (11) for connection to the battery. Measurement circuits (19) provide information relating to the terminal voltage, temperature and charging current at the battery, to a microprocessor (12). A method of testing a battery involves measuring the battery output voltage and determining therefrom if the battery is charged sufficiently to enable a valid heavy current discharge test to be performed and if so the test is performed using said load to draw between about 300 to 500 amps for five seconds. The battery voltages during and sometime after the test are recorded and the microprocessor rates the battery, good, bad, good requiring a recharge and requiring conditioning in order to conduct a valid test. The battery rating (R) is determined using an empirical formula based on known battery data and the parameters measured during said discharge test. Conditioning of the battery involves testing the battery for shorted cells, slow charging rate, irreversible sulphation during conditioning and rating the battery as bad, or requiring an extended charge before retesting as the case may be. If conditioning proceeds to a conclusion the heavy current discharge test is performed and the battery rating (R) is determined.

Description

TITLE: TESTING HEAVY CURRENT BATTERIES

This invention relates to the testing of heavy current batteries such as lead-acid batteries, and more particularly to an improved method and apparatus for testing such batteries. The invention has particular utility in the field of automotive starting, lighting and ignition (SLI) batteries but is not limited to this use.

The most common and most widely accepted method for determining the condition (i.e. good, fair, bad, etc.) of an automotive SLI battery is to subject it to a heavy current load and to measure its voltage during the load period and as it recovers when the load is removed. The load is in the range from 100 Amps to 300 Amps and is intended to be a test of the battery's ability to deliver the currents drawn by a starter motor during the cranking of an automotive engine. In disclosing their test methods, tester manufacturers normally do not reveal details of the decision making process whereby the voltages recorded are used to classify the battery under test as good, fair, recharge and retest, or bad, etc..

There is a fundamental difficulty with the heavy current test in that a battery which is good, but is in a low state of charge, will not be able to deliver the heavy current and so will be wrongly classed as bad. The test method and the intelligence built into the test apparatus determine where the lower limit to the state of charge is, but in the end a cut-off voltage typically in the range 12.1 v - 12.3 V exists at which testing is too unreliable.

It is of importance to understand that batteries which are presented for testing are most often in a flat condition. A battery will not generally be presented for testing unless it has been giving trouble, and frequently it will have been flattened by continuously attempting to start an engine. A large need to test flat batteries exists in the battery reclaim industry, where replaced batteries, in numbers of several hundreds at a time, are to be sorted. Virtually all of these batteries will be in states of charge below 11 volts, with many as low as 8 volts.

Known test apparatus deals with this difficulty by measuring the battery voltage before applying the heavy current and, if the voltage is too low for the decision making method to be reliable, discontinues the test and displays an instruction to recharge the battery and retest. Since the time to charge a battery is at least several hours, having to recharge it before achieving a decision on the battery condition is a considerable inconvenience.

It is also known to subject a battery to a charging cycle as part of a diagnostic test procedure. For example, Great Britain Patent No. 2182155 discloses a method for determining the state of charge of a battery by applying a brief high current boost charge (of less than 1% of the battery's capacity in a time of about 1 second) and instantly scanning the thus boosted battery to establish the voltage level at which the battery is capable of providing a discharge current at its C-rate (current numerically equal to the Amp-hour rate of the battery) . This voltage level allows the degree to which the battery is charged to be estimated by comparison with data accumulated from large numbers of batteries of the type concerned at various known degrees of charge.

The charge and discharge procedure disclosed in

GB 2182155, however, is for the purpose of determining the charge in a battery, i.e., it is a type of battery

"fuel gauge" technique and is not a procedure for evaluating the goodness or badness of a battery.

United States Patent No. 3857087 also discloses a battery test method involving various sequential discharging-charging steps. The objective in this case is the determination of the goodness/badness of lead-acid batteries including those that are low in charge. The discharging and charging sequences are based upon the inventor's theory that the goodness/badness of a battery can be assessed from polarisation effects which accompany the passage of current out of and into the battery. The various test sequences include applying very short duration, controlled voltage, high current charge impulses (for example 100-300 amps for about 55 milliseconds at 15 volts) and heavy current discharging (for about 5 seconds) with rest periods therebetween to allow the battery to recover from ion depletion or ion absorption in the region of the electrodes, i.e., to recover from the polarisation of the electrodes. The polarisation changes are measured in terms of voltage changes and the battery condition is assessed from comparisons with reference values obtained from a large number of tests on batteries in known conditions.

Of central importance with regard to the present invention is that the charging impulses which are utilised in US 3857087 are not for the purpose of enabling a high rate discharge test to be performed, but are for the purpose of causing polarisation characteristics to be revealed in accordance with the inventor's theory.

It is an object of this invention to provide an improved method and apparatus for testing batteries of the kind in question.

The invention therefore provides a method of testing a heavy current battery to determine the condition of the battery, characterised in that, said method involves the steps of measuring the battery output voltage and determining therefrom whether the battery is sufficiently charged to enable a valid heavy current discharge test to be performed and, if sufficiently charged, performing a heavy current discharge test and recording the battery loaded voltage V. after a predetermined time during said test, ceasing said test and recording the battery recovery voltage V_R after a predetermined recovery time, calculating the battery rating R according to the formula:

R = f (V_L, V_R, C, T) where C is the battery's cold current cranking rating (CCA) and, T is the temperature of the battery and categorizing the condition of the battery based on the values of V_R and R.

In another form the invention provides a battery tester for heavy current batteries including charging circuitry for charging said battery, characterised in that, said charging circuitry is capable of delivering a voltage/current/time controlled charge for a short time for conditioning a battery under test, and said tester further comprises a load circuit for applying a load test to said battery, and a microprocessor connected to monitor parameters of said battery under test and for directing charging of said battery by said charging circuitry, said microprocessor containing an algorithm for calculating a battery rating R dependent upon parameters measured during and after a said load test according to the formula: R = f (V_L, v., C, T) where V. is the battery loaded voltage measured at the end of the said load test, V_R is the battery recovery voltage after a predetermined rest period,

C is the battery's cold current cranking rating (CCA) and,

T is the temperature of the battery, said microprocessor being programmed to direct charging of said battery so as to condition said battery for further testing when a predetermined battery rating is calculated and the battery recovery voltage is below a predetermined value. The invention rests upon a discovery concerning the charging of automotive batteries and is directed at enabling the heavy current discharge method of testing to be utilised with batteries in such a low state of charge that they would not otherwise be testable by that method. It has been found that a battery in a low state of charge may be "conditioned" to render it suitable for a heavy current discharge test by the application of a measured amount of charge in the order of 10% of the battery's capacity in a period of time of a few minutes. When tested by the heavy current method the battery will then exhibit voltage changes which are substantially the same as if it had been fully charged prior to the test. A further discovery with regard to conditioning is that the way in which the battery responds to the voltage, current and time sequences yields information which is of value in conjunction with a subsequent high rate discharge test, or may reveal battery states which would cause the battery to fail such a test. With a suitable conditioning sequence, the ability of the tester comprising conditioning and high rate discharge is extended to batteries which have initial voltages as low as 8 volts. Previous testers which utilise discharge and charge sequences have not been based upon, nor have expressed any awareness of, these principles.

In order that the invention may be more readily understood a particular embodiment will now be described with reference to the accompanying drawings wherein:

FIGURE 1 is a simplified circuit block diagram of a heavy current battery tester according to the invention; FIGURE 2 is a more detailed circuit block diagram of the tester shown in FIGURE 1; and FIGURE 3 is a flowchart of the program of a microprocessor of the tester of FIGURES 1 and 2.

As shown in FIGURE 1 the tester comprises essentially a heavy current tester 10, a flat battery conditioner 11 and a microprocessor 12 containing an algorithm for classifying the battery 13 according to predetermined empirical formulas. The heavy current tester 10 and conditioner 11 are connected in parallel to the battery 13 as shown and each is connected to provide information to and receive information from the microprocessor 12.

The algorithm contained in the microprocessor 12 classifies batteries according to formulas which have been established on the basis of tests involving a range of sizes of new and used batteries at various states of charge and the algorithm in particular directs the conditioning of a battery to make it testable at voltages heretofore considered too low.

An important advantage of the method and apparatus according to this invention is that in relation to the heavy current testing of batteries in a low state of charge they are not left in a weaker condition after testing than they were in before testing. There is therefore less risk that the operator of the tester will be accused by a customer of having damaged his or her battery by performing the test.

As is evident in FIGURE 2, the tester according to this embodiment has three main stages; viz. a high power stage 14 in which heavy current testing and flat battery conditioning are performed; a control section 15 which directs the testing procedure and computes the battery condition; and a communication section 16 which enables data to be entered by the user and displays instructions and results. In FIGURE 2 the battery 13 to be tested is connected to a load resistor 17 by means of a switch 18. The load resistor 17 and switch 18 together with thermal and overcurrent protection circuitry 22 essentially make up the heavy current tester 10 of Figure 1. The switch 18 may be a semiconductor type or an electromechanical type depending upon cost. Also connected to the battery is the flat battery conditioner 11 which supplies a controlled charge to batteries which are too low in voltage for the load test to be immediately applied. The battery terminal voltage, battery temperature and the current supplied by the conditioner are measured by measurement circuits 19 and for protection purposes temperatures of the load and solid state switch are also measured by circuits 19. The microprocessor 12 receives information from the measurement circuits 19 and from a keypad which is part of display and keypad 20 and supplies control signals to the flat battery conditioner 11 and to the solid state switch 18 and data to a display which is part of the display and keypad 20. The display and the keypad 20 are mounted on the front panel of the instrument and are the means whereby the user is given instructions and information and the results of the tests and is enabled to enter data into the tester. By means of serial interface circuits 21 the tester can output its results to a printer and a computer and can also be controlled from a computer (not shown). The procedure in the use of the tester is as follows.

(a) When the tester is switched on it shows the message "Connect battery" on its display screen 20 to indicate that it is ready for testing. The user then connects the battery to be tested by means of leads provided. If the leads are connected the wrong way around an instruction is given to reverse them. Within one of the clips on the leads is a temperature transducer, e.g. a semiconductor diode, by means of which the temperature of the battery is measured.

(b) By means of the display the user is next asked to select the size of the battery from a set of displayed choices.

(c) By means of the display the user is next asked to press a key labelled "start" to cause the testing program to commence.

The details of a testing program which is contained within the microprocessor 12 are as follows, with reference to the flowchart given in FIGURE 3. (i) Commencing with "Start" box 30 the first actions of the testing program are to measure the battery's open circuit voltage (box 31) and from that measurement to decide whether the battery is flat and needs a conditioning charge or whether it is of a high enough voltage to proceed with a heavy current load test.

If the battery voltage is greater than 12.10 V which corresponds to the battery being charged to about 40% of its full capacity, the heavy current load test (box 32) is immediately applied for a duration of 5 seconds. The load 17 is a 20 milli-ohm (approx) stainless steel resistor and the current which is drawn from the battery is in the range 300 to 500 Amps depending upon the size of the battery. The voltage drop across the switch, the resistance of the leads and wiring and the internal resistance of the battery (which varies in the range from a few milli-ohms to about 10 milli-ohms according to size and condition of the battery) determine the actual current.

(ii) At the end of the 5 seconds the microprocessor records the voltage of the battery and then opens the switch 18 in the load circuit and the heavy current is thereby switched off. After a 30 second rest period the voltage to which the battery has recovered is recorded.

Following the recording of the 30-second recovery voltage the microprocessor calculates the battery rating R (box 34) by the formula: R = f (V_L, V_R, C, T) where V_L is the loaded voltage measured at the end of the five second heavy current test,

V_R is the recovery voltage after the 30 second rest period, C is the battery's cold current cranking rating

(CCA) and

T is the temperature of the battery in degrees C. More specifically the rating R is calculated according to the formula: R - V_L + 14.844 - 1.6114V. - 8.888 x 10-³C - 2.378 x 10"²T + 1.913 x 10^'6C² + 1.498 x 10^"5CT + 1.245 x 10^"4T²

The rating R is devised so that it has a value of zero for a perfect battery and is negative for deteriorated batteries. A battery with a rating greater than -0.5 Volt is a good battery; a battery between -1 Volt and -0.5 Volt probably has some deterioration but is still good; a battery between - 1.5 V and -1.0 V is probably a bad battery; and a battery with a rating of less than -1.5 V is definitely a bad battery.

Before accepting the rating however, a further check is carried out. This is because batteries which are tested immediately after an extended charging may be hot and exhibit artificially high recovery voltages and thereby artificially raise the R rating for batteries which have low loaded voltages. The recovery voltage V„ is corrected to a value V_RC and the rating R is accepted only if the corrected recovery voltage exceeds 12.1 Volts. The formula for the corrected recovery voltage is:

V_RC - V_R + 0.080 - 3.19 x lO^T

If the battery has a corrected recovery voltage greater than 12.1 V the R-rating is accepted, that is, the battery is good if R is greater than -1 V and the display will show a message "Good battery" (box 35) and the battery is bad if R is less than -1 V and the display will show a message "Bad battery, replace" (box 36). This is the end of the test in these cases. If the battery has a corrected recovery voltage less than 12.1 V and a rating R greater than -1 V it is considered to be good but in need of charging, and the display will show a message "Good battery, recharge"(box 37). The type of battery distinguished by this process is one which has an acceptable loaded voltage at the end of the 5 second load test, although its corrected recovery voltage is somewhat low. In this case this is also the end of the test.

If the battery has a corrected recovery voltage less than 12.1 V and a rating R less than -1 V it has both low loaded voltage and low recovery voltage but is not necessarily bad because it was just above the 12.1 V when the initial assessment to give a load test was made. It may therefore be flat enough to warrant conditioning in order to properly assess it. In this case the conditioning process (box 38) is commenced and the display shows a message "Conditioning battery, test may take 15 minutes". (iϋ) If the battery open circuit voltage V,,,. measured at (i) is less than 12.1 V, or the conditioning decision is reached as in (ii), the microprocessor switches on the conditioning circuit and controls the delivery of the appropriate amount of charge to make the battery reliably testable in the minimum time. A conditioning time of about 18 minutes is based on a charging current of about 40 Amps (because that is the highest current from an economical transformer in the present embodiment). However a considerably greater charging current could be used and would result in reduced conditioning time. During this conditioning process monitoring of the battery's reaction to the charging current is carried out to decide whether the battery is bad and no further attempt should be made to charge it, such as due to irreversible sulphation or to shorted cells, or that it cannot reach a testable condition in reasonable time, such as 18-20 minutes, and should therefore be given an extended charge. At low temperatures, e.g. -5 C in particular, it may take an excessive time to deliver the 4 to 6 A-h needed to make a flat battery testable.

The conditioning process (box 38) starts with a voltage of about 19 V, which voltage is reduced to 14.5 volts when the current rises to about 10 amps. If the current will not rise to 10 amps within 5 minutes, the battery is deemed to be too sulphated and it is rejected as bad (box 40). Furthermore, if the current rises to 10 amps with the high voltage but fails to rise to 5 amps within the remainder of the 5 minute period from the commencement of the conditioning when the voltage is reduced, it is also deemed to be bad (box 40). If the currents exceed the 10 amp and 5 amp levels in the five minute period, and the initial voltage was greater than 12.1 V, then the battery is considered suitable for conditioning with 6 A-h of charge (box 41) followed by a subsequent high rate discharge test (box 42). If it does not pass the 10 amp/5 amp/12.1 V test, the conditioning is abandoned and the battery is deemed too flat to test and the message "Give extended charge and retest" is displayed and the program moves to box 43. If the battery passes this assessment stage as to the degree to which it can accept charge for the high rate discharge test, the conditioning continues.

Firstly, it should be mentioned that if the battery is low in voltage because it has a shorted cell the voltage will fail to rise above 12 V and furthermore, it is dangerous to continue to put a large current into a battery which has a shorted cell. Therefore if the battery voltage fails to rise above 12.0 V within 2 seconds of the application of the high current conditioning process (see box 48), it is suspected that the battery has a shorted cell and the test is ended with message such a "Bad battery, shorted cell" and the battery is rejected at box 40.

The actual current which flows from the charging circuit depends upon the battery internal resistance and if this is relatively high, such as 20 milli-ohms, the current is reduced and the time to deliver the charge of say 6 Amp-hours to make the battery testable is increased. The microprocessor monitors the current and predicts whether the necessary charge can be delivered within a reasonable time, such as 15 minutes (box 49). If it is not possible to deliver the charge in the short time the conditioning is discontinued and the display shows a message such as "Give extended charge and retest" and the program moves to box 44. If the necessary charge such as 6 Amp-hours can be given within the time limit the conditioning continues. At the end of the conditioning the heavy current load test as described above (box 42) is immediately applied, i.e. 5 seconds of heavy current load followed by a rest period of 30 seconds. The load and recovery voltages are recorded and other empirical formulae, to suit the conditioned battery, are used to determine ratings (box 45) . As before these formulas give a value of 0 for a perfect battery and negative values for a deteriorated battery with the division between good and bad batteries occurring for R = -1 Volt. Messages such as "Good battery, recharge" when the program moves to box 46, "Bad battery, replace" when the program moves to box 47 are then displayed according to the degree to which the value of R is less than zero.

It should be evident from the description hereinabove that the tester of the present invention incorporates an intelligent, microprocessor controlled, conditioning process which enables batteries in a low state of charge, generally called flat batteries, to be tested by the heavy current load method in a relatively short space of time. The tester conditions a battery in a low state of charge to make it testable by the heavy current load method by giving it a charge of about 10% of its capacity

(for example about 4 to 6 Ampere-hours) in a short period (for example 5 to 18 minutes). The tester applies a heavy current discharge for a few seconds

(for example 5 seconds) and measures the loaded voltage and then allows a rest period and measures the recovery voltage. It applies the heavy current discharge method but does not further weaken a battery which is already in a low charge condition. The tester incorporates a temperature sensor in a battery clip and automatically makes corrections for battery temperature. Also it incorporates a state of charge correction using the recovery voltage and the corrected recovery voltage. The tester further incorporates a battery size (CCA) correction and is also able to reject batteries which have shorted cells so that they are not mistaken for batteries in a low state of charge. The tester incorporates an algorithm which uses formulas developed from hundreds of tests upon automotive batteries of various sizes, states of health ranging from new to unserviceable, temperatures ranging from -18°C to 50°C and states of charge ranging from 20% to fully charged. Thus, the tester increases the convenience and usefulness of the heavy current load test.

Claims

WE CLAIM:

1. A method of testing a heavy current battery to determine the condition of the battery, characterised in that, said method involves the steps of measuring the battery output voltage and determining therefrom whether the battery is sufficiently charged to enable a valid heavy current discharge test to be performed and, if sufficiently charged, performing a heavy current discharge test and recording the battery loaded voltage V_L after a predetermined time during said test, ceasing said test and recording the battery recovery voltage V_R after a predetermined recovery time, calculating the battery rating R according to the formula: R = f (V_L, V_R, C, T) where C is the battery's cold current cranking rating (CCA) and,

T is the temperature of the battery and categorizing the condition of the battery based on the values of V_R and R.

2. A method according to claim 1, characterised in that, categorizing the condition of the battery includes categorizing the battery as good, bad, good but requiring charging and, insufficiently charged for valid assessment, and wherein said method further includes, in the case of a battery categorized as insufficiently charged for valid heavy current test or valid assessment, the step of conditioning the battery by applying a voltage/current/time controlled charge to the battery sufficient to render the battery capable of delivering, for a short period of time, a current in the order of the fully charged current capability of the battery, thereafter performing a said heavy current discharge test to determine the said battery rating R and categorising the condition of the battery as good or bad based on the value of R.

3. A method according to claim 2, characterised in that, the recovery voltage V_R is corrected to account for varying temperatures of the battery before categorizing the condition of the battery.

4. A method according to claim 3, characterized in that, said method further includes monitoring said battery during said conditioning step to determine the reaction to charging current and if the battery charging current does not meet predetermined criteria within predetermined times, classifying said battery as bad or requiring an extended charge before retesting.

5. A method according to claim 4, characterized in that, said monitoring during said conditioning step includes monitoring the battery voltage and if it fails to rise above the rated battery voltage within a very short period of time, categorising said battery as bad.

6. A method according to claim 5, characterised in that, said monitoring during said conditioning step further includes predicting whether the battery will accept a predetermined amount of charge within a predetermined time limit and if not categorizing said battery as requiring an extended charge before re¬ testing.

7. A method according to claim 6, wherein said battery is a 12 V battery characterised in that, said voltage/current/time controlled charge commences with a voltage of about 19 V, which, when the current rises to about 10 amps, is reduced to about 14.5 V, and said predetermined amount of charge is 4 to 6 ampere-hours and said predetermined time limit is between 5 and 18 minutes.

8. A battery tester for heavy current batteries including charging circuitry for charging said battery, characterised in that, said charging circuitry is capable of delivering a voltage/current/time controlled charge for a short time for conditioning a battery under test, and said tester further comprises a load circuit for applying a load test to said battery, and a microprocessor connected to monitor parameters of said battery under test and for directing charging of said battery by said charging circuitry, said microprocessor containing an algorithm for calculating a battery rating R dependent upon parameters measured during and after a said load test according to the formula:

R = f (V_L, V_R, C, T) where V_L is the battery loaded voltage measured at the end of the said load test,

V_R is the battery recovery voltage after a predetermined rest period,

C is the battery's cold current cranking rating (CCA) and,

T is the temperature of the battery, said microprocessor being programmed to direct charging of said battery so as to condition said battery for further testing when a predetermined battery rating is calculated and the battery recovery voltage is below a predetermined value.

9. A battery tester according to claim 8, characterised in that, said algorithm adjusts said recovery voltage V_R to a corrected value V_RC to account for varying temperatures of the battery, before categorising the condition of the battery.

10. A battery tester according to claim 9, characterised in that, said load circuit comprises a resistor of about 20 milli-ohms capable of drawing about 300 to 500 amps from said battery, a switch controlled by said microprocessor and thermal and overcurrent protection circuitry.

11. A battery tester according to claim 10, characterised in that, said charging circuitry is adapted to provide about 4 - 6 Amp-hours of charge in about 5 to 18 minutes.

12. A battery tester according to claim 11, characterised in that, said charging circuitry is adapted to provide a voltage of about 19 volts to said battery at the start of a conditioning process, which voltage is reduced to about 14.5 volts when the current to said battery rises to about 10 amps and said algorithm causes rejection of said battery if the current will not rise to said about 10 amps within a period of about 5 minutes from the commencement of conditioning or if it fails, after reaching said about 10 amps, to reach a current of about 5 amps within the remainder of the period with the reduced voltage.

13. A battery tester according to claim 12, characterised in that, it includes a measurement circuit connected to measure the battery terminal voltage, battery temperature and current supplied by said charging circuitry to the battery and provide information to said microprocessor relating thereto.

14. A battery tester according to claim 13, characterised in that, it includes a temperature transducer within a terminal clip of a battery lead for the purpose of providing data relating to the temperature of said battery.