Battery cell end of life and defect detection system

Abstract

An integrated circuit device includes a controller, an optional current source coupled to the controller, a voltage sampler coupled to the controller, a current detector coupled to the controller, and memory coupled to the controller, where memory includes code segments executable by the controller to: (a) apply a current pulse to a cell; (b) measure a voltage of the cell during the current pulse; (c) calculate an impedance of the cell from the measured voltage; and (d) determine an operational state of the cell from the impedance.

Claims

1. A method for testing a battery cell comprising: (a) applying a current pulse to a battery cell with a battery cell analyzer, wherein the battery is a device under test (DUT) by the battery cell analyzer and is not connected to an operating load; (b) measuring a voltage of the battery cell solely during the current pulse; (c) calculating a calculated impedance of the battery cell from the measured voltage and storing the calculated impedance in a non-transitory computer readable media; (d) determine that the battery cell is defective to the point of failure if the calculated impedance is greater than a failure threshold and, if so, stopping the testing of the battery cell; (e) determine that the battery cell may have a defect to provide a warning if the calculated impedance is less than the failure threshold but greater than a warning threshold; (f) determine an indication of end of life (EOL) of the battery cell when an impedance of the battery cell during the current pulse is greater than an impedance of the battery cell during a preceding pulse that was stored in the non-transitory computer readable media; (g) discharge the cell with the battery cell analyzer for a cell discharge period; and (h) continuously repeat operations (a)-(g) after the cell discharge period until the battery is determined to be defective to the point of failure in operation (d).

2. An integrated circuit device comprising: a controller; a voltage sampler coupled to the controller; a current detector coupled to the controller; and memory coupled to the controller, the memory including code segments executable by the controller to: (a) apply a current pulse to a battery cell, wherein the battery is a device under test (DUT) and is not connected to an operating load; (b) measure a voltage of the battery cell solely during the current pulse; (c) calculate a calculated impedance of the battery cell from the measured voltage and storing the calculated impedance in a non-transitory computer readable media; (d) determining that the battery cell is defective to the point of failure if the calculated impedance is greater than a failure threshold and, if so, stopping the testing of the battery cell; (e) determine that the battery cell may have a defect to provide a warning if the calculated impedance is less than the failure threshold but greater than a warning threshold; (f) determining an indication of end of life (EOL) of the battery cell when an impedance of the battery cell during the current pulse is greater than an impedance of the battery cell during a preceding pulse that was stored in the non-transitory computer readable media; (g) discharging the cell for a cell discharge period; and (h) continuously repeating operations (a)-(g) after the cell discharge period until the battery is determined to be defective in operation (d).

3. An integrated circuit device as recited in claim 2 further comprising a power unit providing power to the controller, the memory, the voltage sampler, and the current detector.

4. An integrated circuit device as recited in claim 3 further comprising a current source coupled to the controller and powered by the power unit.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Several example embodiments will now be described with reference to the drawings, wherein like components are provided with like reference numerals. The example embodiments are intended to illustrate, but not to limit, the invention. The drawings include the following figures:

(2) FIG. 1 illustrates three examples of resistive layer change on active material of battery cells due to discharge reactions;

(3) FIG. 2 includes graphs illustrating the application of test pulses during the lifespan of a battery;

(4) FIG. 3 is a table including cell conditions, impedance patterns, results and causes for several examples of battery cells;

(5) FIG. 4 includes graphs illustrating tests of alkaline cells at C/100 load and different pulse rates;

(6) FIG. 5 is a graph plotting State of Charge (SOC) vs. Pulse Impedance Ratio;

(7) FIG. 6 is a graph plotting State of Charge (SOC) vs. Pulse Impedance Ratio at different temperatures;

(8) FIG. 7 is a graph illustrating a discharge curve of a lithium cell having LiSOCl.sub.2 chemistry;

(9) FIG. 8 is a graph plotting State of Charge (SOC) vs. impedance;

(10) FIG. 9 is a graph illustrating that impedance is still high after hitting cut-off voltage and relaxation;

(11) FIG. 10 is a block diagram of an example battery cell end of life detection system;

(12) FIG. 11 is an illustration of an integrated circuit including an example battery cell end of life detector;

(13) FIG. 12 is flow diagram illustrating an example method for detecting end of life (EOL) and defects of a battery cell;

(14) FIG. 13 is a block diagram of an example battery cell end of life and defect detection system;

(15) FIG. 14 is a block diagram of an example battery cell and end of life detection system with an external load; and

(16) FIG. 15 is a flow diagram illustrating example methods for operating the battery cell and end of life detection system of FIG. 14.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

(17) Embodiments, set forth herein by way of example and not limitations, relate to methods and apparatus for detecting end of life and health of batteries through measurements of internal cell impedance. Disclosed embodiments may be used in the electrochemical and electronic industries to analyze the state of health of battery cells and provide warnings of approaching end of life (EOL) and other internal problems of primary and secondary batteries. In certain example embodiments, a pulse impedance method is used to determine state of health and EOL of primary batteries. Advantageously, certain example embodiments consume <0.01% of battery energy in the detection process and do not require a reference model.

(18) In FIG. 2, a series of related graphs 28 illustrate the voltage responses to short discharge pulses that are applied during cell discharge of various cells. Graph 28A illustrates the test pulses. These could be artificial discharge pulses or natural pulses associated with higher loads. The current in test pulses should be higher than the normal load of the cell. Once the pulses are applied, the associated voltage drop (which is directly related to the impedance of the cell) can be assessed to determine the state of health of the cell and its end of life. In the following examples, test pulses of the same current are applied at different states of charge.

(19) In FIG. 2, example #1, as illustrated by graph 28B, the voltage response, and therefore the impedance, is high at the beginning, which might signal that the battery is defective. Later on, the impedance decreases, which corresponds to smaller voltage drops during pulses. That means that the battery has recovered and is beginning to operate normally. Closer to the lower state of charge the impedance again starts growing, which signals that the cell is approaching end of life (EOL).

(20) In FIG. 2, example #2, as illustrated by graph 28C, the initial impedance is low, corresponding to insignificant passivation layer in the cell, which could mean fresher battery than in example #1. The cell's impedance again increases towards the end of its lifetime.

(21) In FIG. 2, example #3, as illustrated by graph 28D, the cell's impedance is very high from the beginning, and doesn't recover over time. This means that the cell is either defective, or has previously been discharged. In either case, this battery cell is not operable.

(22) In FIG. 3, a table 30 includes cell conditions, impedance patterns, results and causes for various cell examples. R0* represents expected cell impedance. R1 represents measured impedance associated with the first pulse, R2 represents impedance associated with the second pulse, etc.

(23) In FIG. 4, a pair of graphs 32 include a first graph 32A and a second graph 32B which illustrate the running of the tests on the alkaline cells at a C/100 load and different pulse rates. In this non-limiting example, graph 32A illustrates C rate pulse of C/20, and graph 32B illustrate a C rate pulse of C/5. It is noted that, in both cases, the impedance increases as the cell approaches the end of life.

(24) In FIG. 5, a graph 34 illustrates voltage drops associated with the pulses. It can be seen that the pulse impedance ratio (R(current)/R (100% SOC)) increases significantly towards the end of life of the cell.

(25) As seen in FIG. 6, a graph 36 illustrates that if the temperature is increased, the voltage drop also increases. For example, the curve 38 was plotted at 23° C. and the curve 40 was plotted at 30° C. Nonetheless, the pulse impedance ratio stays approximately the same, independently of the temperature, as can be seen in graph 36.

(26) Is should be noted that it is easier to predict the end of life of alkaline cells, since they have non-flat voltage discharge curves, as opposed to, for example, lithium battery cells. Fuel gauges containing coulomb counter and/or voltage sensing devices can determine the states of charge of alkaline cells, but are ineffective for use with cells with flat curves, such as lithium cells with a LiSOCl.sub.2 chemistry.

(27) In FIG. 7, a graph 42 illustrates a discharge curve 44 of a lithium (Li) cell with LiSOCl.sub.2 chemistry with current pulses to determine impedance. As can be seen, the voltage (and therefore the impedance) stays constant for an extended period of time, until it collapses at the end of life (EOL) of the cell. It should be noted, however, that the cell's impedance starts growing well before that, signaling an impending EOL the cell, providing an early EOL warning. FIG. 8 illustrates with a graph 44 that when impedance ratio starts increasing, the cell is approaching EOL.

(28) It should be noted that the lithium cell that used for the example tests described above had been stored on a shelf for about 3 years, implying that a passivation layer may have built up over time. With additional reference to graph 46 of FIG. 9, this is confirmed by observing the decreasing impedance of the cell at the beginning of the discharge cycle. It should also be noted that once the cell is discharged and relaxed for some time, the current pulse confirms that the impedance level is still high.

(29) In FIG. 10, a block diagram of a battery cell end of life (EOL) and defect detection system 48 includes a battery cell analyzer 50, a device under test (DUT) battery cell 52, and a power source 54. The battery cell analyzer 50 is, in a non-limiting example, provided as an integrated circuit (IC). However, the battery cell analyzer 50 may be implemented with discrete components, or combinations of discrete components and/or ICs. Furthermore, the battery cell analyzer 50 can be combined with the power source 54 and/or the battery cell 52, as will be appreciated by those of skill in the art.

(30) Example battery cell analyzer 50 includes a controller 56, a memory 58, a power unit 60, a current source 62, a voltage sampler 64 and a current detector 66. Controller 56 is preferably a digital processor capable of executing program code segments to implement various computer-implemented processes. A positive terminal of cell 52 is coupled to current source 62 and to voltage sampler 64, while a negative terminal of cell 52 is coupled to the current detector 66. The power source 54 can be any suitable power source, including a battery charger, an battery (e.g. a battery pack), a capacitor (which may be charged by the cell 52 or otherwise), etc.

(31) In FIG. 11, an integrated circuit (IC) 50′, set forth by way of example and not limitation, has four control/data input pins {S0, S1, S2, S3}, a BAT+ pin, a BAT− pin, five output pins indicating the condition of the battery (e.g. OK, Defect, Failure, Recovered, EOL), a power input pin V+ and a ground pin GND. A device under test (DUT) battery 52′ comprising two cells is coupled to the Bat+ and Bat− pins. The control/data input pins can be used, by way of non-limiting example, to select a current range, a cell impedance range, a test period, etc.

(32) It should be noted that a number of protocols (a/k/a processes and/or methods) can be implemented for testing battery cells based upon the principals described herein. Some of these protocols, set forth by way of example, can be implemented by the battery cell analyzer 50. For example, code segments, executable by controller (e.g. digital processor) 56, can be stored in memory 58 to implement various example protocols. However, it will be appreciated by those of skill in the art that combinations, subsets and extensions of these protocols will also provide useful results in various situations, applications and environments.

(33) FIG. 12 is a flow diagram illustrating an, example process (a/k/a method and protocol) 70 for detecting end of life (EOL) and defects in a battery cell. The process 70 begins at 72 and, in a step 74 counter variable “i” is set to zero, and a vector array element IMP(0) is arbitrarily set to a failure threshold T.sub.F. Next, in an operation 76, the variable i is incremented. Next, in operations 78 and 80, a test current pulse is applied to the cell and the voltage of the cell is measured during the test current pulse. In an operation 82, the impedance is calculated from the measured voltage and is stored in the vector array element IMP(i). A decision operation 84 determines if IMP(i)>T.sub.F and, if so, an operation 86 determines that the battery cell is defective. If not, then a decision operation 88 determines if IMP(i)>T.sub.W<T.sub.F and, if so, an operation 90 warns that there is a possible defect in the cell. If not, a decision operation 92 determines if IMP(i)>IMP(i−1), i.e. if the impedance detected with the current, pulse is greater than the impedance detected with the previous pulse and, if so, provides an early EOL warning in an operation 94. Next, an operation 94 provides for the discharge of the cell, and an operation 98 determines if the cell discharge period is over. If so, process control returns to operation 76 to apply the next test current puke.

(34) In another example protocol, the following operational steps are performed: 1. Initial current pulse applied; 2. Measurement of cell voltage during current load; 3. Calculation of impedance and comparison to expected value; 4. If it higher than a threshold report a warning; 5. Apply a second pulse after a certain discharge time; 6. Assess impedance if impedance lowered or is increasing; 7. Report a cell failure if impedance remains higher than a threshold; 8. If impedance was low or recovered over the initial testing period apply current pluses at certain time of discharge; 9. If impedance increases report a warning; and 10. If impedance increases to an certain threshold report a EOL warning.

(35) In FIG. 13, a block diagram of a battery cell end of life (EOL) and defect detection system 100, set forth by way of example and not limitation, includes a battery cell analyzer 102 and a device under test (DUT) battery cell 104. The battery cell analyzer 102 includes a processor 106, a memory 107 coupled to the processor 106, an impedance-based health and EOL detection unit 108, a capacity-based health and EOL detection unit 110, and an open circuit voltage (OCV) based health and EOL detection unit 112. The processor 106 executes code segments stored in computer-readable media, such as memory 107, to process and combine impedance, charge and OCV data from detection units 108, 110, and 112, respectively to output, by way of non-limiting examples, include health and end of life output values and/or matrix, a state of charge (SOC) assessment, a failure mode of the cell, etc.

(36) Example battery cell analyzer 102 is advantageous in that it can be used to determine overall health and end of life of both primary and secondary batteries. By combining the impedance based methodologies described above with charge-based and/or open-circuit based methodologies, the results of two or all three methodologies can be processed and/or combined to provide fast, accurate determinations of the overall health and EOL of the cell.

(37) In this non-limiting example, the output of the battery cell analyzer 102 is as follows:
Output=function(Input.sub.Impedance,Input.sub.Charge,Input.sub.OCV)
For example, the function simply sums the inputs such that Output=Input.sub.Impedance+Input.sub.Charge+Input.sub.OCV. Examples of the use of the Input and Output signals is as follows: If Output >Threshold, a warning of an aged battery is made for both primary and secondary cells; A function of Output over time is used for EOL prediction of a secondary cell; A function over time of Inputs and Output is used for failure mode detection such as Li loss, loss of active material, surface impedance build up, etc. A function over time of Inputs and Outputs is used to control charge and discharge modes for mitigation of failure modes.

(38) In FIG. 14, a block diagram of an example battery cell end of life and defect detection system 114 includes a battery cell analyzer 116, a device under test (DUT) battery cell 118, an external load 120, and a power source 121. The battery cell analyzer 116 is, in this non-limiting example, provided as an integrated circuit (IC). However, the battery cell analyzer 116 may be implemented with discrete components, or combinations of discrete components and/or ICs. Furthermore, battery cell analyzer 116 can be combined with the battery cell 118, the external load 120, and/or the power source 121, as will be appreciated by those of skill in the art.

(39) Example battery cell analyzer 116 includes a controller (“processor”) 122, a memory 124, a voltage sampler 126, a current detector 128, and a power unit 130. Controller 122 is preferably a digital processor capable of executing program code segments to implement various computer-implemented processes. A positive terminal of cell 118 is coupled to voltage sampler 126 and a negative terminal of cell 118 is coupled to voltage sampler 126 and current detector 128. External load 120 is coupled between the positive terminal of cell 118 and the current detector 128. The power source 121 can be any suitable power source, including a battery charger, a battery (e.g. a battery pack), a capacitor (which may be charged by the cell 118 or otherwise), etc.

(40) FIG. 15 is a flow diagram illustrating an example process (a/k/a “method” or “protocol”) 132 for early warning of end of life (EOL) and defects in a battery cell. The process 132 may be implemented in code segments stored, for example, in memory 124 and executed on controller 112 of battery cell analyzer 116. The process 132 begins at 134 and, in an operation 136, the process detects and measures a change in load current (di/dt). Next, in an operation 138, a measurement and associated change of cell voltage is determined based upon the change in load current. In an operation 140, the impedance is calculated and/or a change in voltage is used to compare to an expected value (“threshold value”) with one of three results. If it is less than or equal to a threshold, process control is returned to operation 136. If it is higher than the threshold, but has not been proximately proceeded by one or more other higher values, a warning is made in an operation 142 and process control also returns to operation 136. If it is repeatedly higher than the threshold value, e.g. is higher than the threshold a number of consecutive times, a report of cell failure or and end-of-life (EOL) value is provided in an operation 144 and process 132 ends at 146.

(41) As another example, the process 132 can be used to calculate the resistivity (e.g. resistive impedance) of the cell for an EOL prediction from multiple measurements of cell voltage and current taken over time. It is noted that resistivity of the battery cell tends to increase with an approach of its end of life. The change of resistivity can be obtained by filtering the changing voltage and current measurements of the load, and the EOL can be predicted by an analysis of the change in resistivity over time.

(42) Although various embodiments have been described using specific terms and devices, such description is for illustrative purposes only. The words used are words of description rather than of limitation. It is to be understood that changes and variations may be made by those of ordinary skill in the art without departing from the spirit or the scope of various inventions supported by the written disclosure and the drawings. In addition, it should be understood that aspects of various other embodiments may be interchanged either in whole or in part. It is therefore intended that the claims be interpreted in accordance with the true spirit and scope of the invention without limitation or estoppel.