Battery pack and charging method for battery pack

Abstract

The invention relates to a battery pack (100) having a predefined number (n) of individual battery cells (Z 1,…,n ) connected to one another and a monitoring unit (110) for the individual battery cells (Z 1,…,n ), the monitoring unit being in the form of an SCM circuit (112) in particular, wherein the monitoring unit (110) has a passive input circuit (120), and the monitoring unit (110) is configured to identify different electrical faults (F 1,…,n ) in the passive input circuit (120) by means of a time measurement unit (132) assigned to the monitoring unit (110), in particular based on the measurement of the rise time (T An1,…,n ) of a voltage (U 1,…,4 ).

Description

Technical Field

[0001] The invention relates to a battery pack having a predetermined number of individual battery cells connected to one another and a monitoring unit for the individual battery cells, in particular in the form of an SCM circuit (“single cell monitoring” circuit), wherein the monitoring unit has a passive input circuit. Background Art

[0002] Such a battery pack is known from the prior art, which has a predetermined number of interconnected individual battery cells and a monitoring unit. Here, the monitoring unit is constructed according to the SCM circuit form for individual battery cells and has a passive input circuit.

[0003] In such battery packs, a plurality of battery cells are usually electrically connected in series and / or in parallel to each other in order to achieve higher electrical power. In order to achieve a particularly high energy density, lithium-ion batteries or lithium-polymer batteries are used in particular. In these battery technologies, each cell must operate within a certain operating parameter range defined by voltage, current and temperature. In order to ensure that the operating parameter range is always met for each individual battery cell of the battery pack, the voltage of the individual battery cells is continuously monitored, for example, by means of the so-called "single cell monitoring" (SCM) method. The SCM method ensures that, first of all, during the charging process, no battery cell operates above its permissible end-of-charge voltage or target voltage. For the SCM method, it is necessary to further transmit the electrical signal of the connection point between two battery cells connected in series to a monitoring unit.

[0004] The SCM method can be implemented, for example, with the aid of so-called SCM circuits, such as Texas Instruments. The integrated circuit of BQ771800, or the combination of a microcontroller with a so-called passive front end or input circuit. The electrical connection between the battery cell connection point and the SCM circuit is usually checked by means of the so-called "open wire detection". In this context, the switches are briefly activated in sequence, which at least briefly bridge the voltage measurement input of the SCM circuit. As a result, the respective assigned filter capacitors are briefly discharged in a limited manner. Subsequently, the relevant switches are opened again. After a short time, the voltage at the voltage measurement input of the SCM circuit that is briefly bridged is checked again. If there is an "open wire", the capacitor has not yet been charged again, which can be detected by means of the SCM circuit. In order for the SCM circuit to reliably limit the end-of-charge voltage even when a single electrical fault occurs in the input circuit, it is also necessary to design the relevant circuit parts as single fault safety (einzelfehlersicher). However, this requires a high-cost redundant design of the relevant circuit parts. The redundancy of the input circuit of the SCM circuit is also problematic because the electrical signals of the individual branches of the battery cells are required in the sense of special insulation protection, that is, in the event of a fault, they may lead to high currents and / or high voltages, which may lead to thermal secondary faults. Furthermore, electrical faults may occur in the input circuits of the SCM circuits which cannot be detected by previously known "open wire detection". Summary of the invention

[0005] The invention relates to a battery pack having a predetermined number of individual battery cells connected to one another and a monitoring unit for the individual battery cells, in particular in the form of an SCM circuit, wherein the monitoring unit has a passive input circuit. The monitoring unit is designed to detect various electrical errors in the passive input circuit by means of a time measuring unit assigned to the monitoring unit, in particular based on the measurement of the rise time of the voltage.

[0006] The invention thus enables the passive input circuit of the monitoring unit to be continuously checked for different electrical faults. A redundant design of the passive input circuit, which is otherwise required for safety reasons, can be omitted. All functionality of the monitoring unit is preferably implemented by means of an integrated digital microcontroller.

[0007] Preferably, the monitoring unit has at least one voltage meter and an electronic switch for detecting the voltage. Thus, each individual battery cell of the battery pack can be monitored individually. A separate voltage meter, in particular an analog / digital converter, can be provided for each battery cell in the monitoring unit or microcontroller. Alternatively, the individual voltages can also be measured in a time-staggered manner using only one voltage meter and a multi-channel analog multiplexer connected upstream thereof.

[0008] The passive input circuit preferably has a plurality of resistors and capacitors, thereby making it possible to generate a voltage rise curve or a change over time.

[0009] According to an advantageous embodiment, electrical faults in passive input circuits can be detected by at least briefly bridging one of the voltage meters of the monitoring unit by means of an assigned switch, respectively by absolute and / or relative evaluation of the voltage rise time. Fault detection is therefore based on an evaluation of the rise time, rather than on the detection of absolute voltage values, so that fault detection is independent of individual characteristics of the battery pack, such as aging effects, the charge state of individual battery cells and currently flowing charge or discharge currents.

[0010] Preferably, the passive input circuit has a plurality of additional resistors, each of which is connected in series with one of the resistors, wherein the number of additional resistors is equal to the predefined number of battery cells. Electrical faults can thus be detected by evaluating the falling voltage curve. In addition, mutually compensating faults can thus also be detected, which occur, for example, when the fault causes an increase in the resistance value and the fault causes a corresponding decrease in the capacitance of the associated capacitor, so that the product of the resistance value and the capacitance remains essentially unchanged externally.

[0011] Preferably, the additional resistor is dimensioned to be smaller than or equal to the resistor. This makes it possible to improve the detection of electrical faults. Fault detection becomes particularly advantageous if the resistance value of the additional resistor is between 20% and 80% of the resistance value of the resistor.

[0012] In one embodiment, it is provided that the monitoring unit is designed as an integrated circuit, so that it can be integrated into existing battery pack designs in a particularly space-saving manner.

[0013] Furthermore, the invention relates to a method for detecting electrical faults in a passive input circuit of a monitoring unit, in particular in the form of an SCM circuit, for a predetermined number of individual battery cells of a battery pack, in particular the above-described battery pack, comprising the following steps:

[0014] a) Start charging of the battery pack,

[0015] b) measure voltage,

[0016] c) at least briefly bridging the voltage measuring device of the monitoring unit by means of an associated switch,

[0017] d) measuring the rise time until the currently measured voltage has reached a fraction (Bruchteil) of the voltage (0.1 to 0.9),

[0018] e) repeating steps b) to d) for each of a predetermined number of battery cells in the battery pack,

[0019] f) performing an absolute and / or relative evaluation of the rise time based on a predefined tolerance range, and

[0020] g) If at least one rise time is above the tolerance range and a fault has therefore been detected in the passive input circuit of the monitoring unit, the charging operation of the battery pack is stopped, otherwise the charging operation is continued and steps b) to g) are repeated cyclically.

[0021] The method enables continuous monitoring of the input circuit for possible electrical faults. The cycle time for a complete method run may be, for example, at most one second.

[0022] If the first voltmeter is not connected to the negative pole of the battery pack via a resistor, the rise time is preferably corrected by a time correction factor when the predefined number is equal to one. Preferably, the rise time is corrected by the time correction factor in units whose input filters differ from the input filters of the other units. This makes it possible to numerically simply compensate for electrical asymmetries of the passive input circuit of the monitoring unit caused by the lack of a resistor. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] The invention is further explained in the following description based on the exemplary embodiments shown in the drawings. The drawings show:

[0024] Figure 1 A battery pack according to the invention is shown, which has a monitoring unit and an input circuit for monitoring individual battery cells;

[0025] Figure 2 Display for identifying by means of a monitoring unit Figure 1 A schematic diagram of an algorithm for detecting an electrical fault in an input circuit of a monitoring unit of a battery pack;

[0026] Figure 3 A schematic diagram showing an alternative algorithm for identifying electrical faults;

[0027] Figure 4 A further embodiment of a battery pack according to the invention is shown. DETAILED DESCRIPTION

[0028] Figure 1 A battery pack 100 is shown, which here has only a number n of battery cells Z. 1,…,5 , n is exemplarily equal to five (n=5), the battery cells are only exemplarily connected in series. 1,…,5Can be connected in series or in parallel for voltage and current matching. 1,…,5 There is a (central) tap A between 1,...,4 The battery pack 100 has a negative pole M or ground branch and a positive pole P, wherein the negative pole M is identical to the negative pole of the battery cell Z1 and the positive pole P is identical to the positive pole of the battery cell Z5. Any (not shown here) electrical load can be connected to the negative pole M or ground branch and to the positive pole P by means of two electrical connecting lines 102 , 104 .

[0029] Furthermore, the battery pack 100 comprises a device for permanently monitoring each individual battery cell Z of the battery pack 100. 1,…,5 The electronic monitoring unit 110 is implemented in particular in the form of an SCM circuit 112, wherein the monitoring unit 110 is connected upstream of the passive input circuit 120. A high-performance digital microcontroller 130 and a time measurement unit 132 are preferably also integrated in the monitoring unit 110. The monitoring unit 110 is preferably implemented as an integrated monolithic circuit 114.

[0030] The passive input circuit 120 preferably has a plurality of resistors R 1,…,n and capacitor C 1,…,n The input circuit 120 of the monitoring unit 110 comprises five resistors R corresponding to the number n being equal to 5 (n=5). 1,…,5 , where the resistor R 1,…,4 Located in individual battery cells Z 1,…,5 Tap A 1,...,4 The resistor R5 is directly connected to the positive electrode P or the positive electrode of the battery cell Z5. In addition, the input circuit includes five capacitors C 1,…,5 , which are connected in parallel to two adjacent input terminals E of the monitoring unit 110. 1,…,6 It should be noted that the monitoring unit 110 may also have more or less than five resistors R 1,…,5 and / or capacitor C 1,…,5 Thus, instead of one resistor, such as resistor R1, multiple resistors connected in series or in parallel may be arranged. Figure 2 The resistance R shown in 1,…,5 It is preferably constructed from two resistors arranged in series. Figure 2 The capacitor C shown in 1,…,5 Preferably, the two capacitors are designed to be arranged in series, so that if one of the two capacitors is mechanically damaged, no short circuit occurs.

[0031] The monitoring unit 110 preferably has a device for detecting the voltage U 1,…,n At least one voltage measurer V1,…,n Here, at every two adjacent input terminals E of the monitoring unit 110 1,2 、E 2,3 、E 3,4 、E 4,5 and E 5,6 A voltage measuring device V is connected between the 1,…,5 , respectively by means of an electronic switch S which can be controlled by a microcontroller 1,…,5 The voltage measurement device V in the monitoring unit 110 can be connected across or short-circuited. 1,…,5 For example, it can be implemented by electronic analog / digital converters (ADC's) of microcontroller 130. Instead of the five voltage measuring devices V shown here only by way of example, 1,…,5 , or only one voltage measuring device can be provided, in the case of an electronic switch S 1,…,5 In the case of a voltage measuring device, the inputs E can be connected sequentially in time by means of an analog multiplexer (not shown). 1,…,6 With the aid of five voltage measuring devices V, which are only used as examples here, 1,…,5 , if necessary, the (input) voltage U can be detected simultaneously with the aid of the monitoring unit 110 1,…,5 .

[0032] The monitoring unit 110 is designed to monitor, in particular, the (input) voltage U selected by the algorithm. 1,…,n The corresponding rise time T An1,…,5 Measurement and evaluation of the 1,…,5 At least briefly short-circuit or connect across the voltage measuring device V 1,…,5 After one of the above, different electrical faults F, cited here as an example, are detected in the passive input circuit 120 by means of the time measurement unit 132. 1,…,4 Preferably, the voltage U is measured continuously in a cycle with a period time of less than one second. 1,…,n , and the associated rise time T is detected by the monitoring unit 110 accordingly An 1,…,5 , and are evaluated absolutely and / or relatively based on a tolerance range.

[0033] With an additional series resistor R F1 A first fault F1 of the form (which may be caused by a bad weld, for example) can verifiably increase the level of at least two input terminals E of the monitoring unit 110. 4,5 Rise time T An 4,5 . With the parasitic parallel resistance R F2 A second fault F2 of the form (which may be caused, for example, by contamination of a circuit board of the input circuit 120 (not shown)) verifiably increases the rise time T at least at the input E1 of the monitoring unit 110. An1. With the resistor R connected in parallel with the capacitor C3 F3 A third fault F3 of the form (which may be caused, for example, by a local insulation fault in the dielectric of the capacitor C3) causes at least a rise time T of the input E3 of the monitoring unit 110 to An3 The measurable change of capacitance C F5 A fourth fault F4 of the form φ , which may occur, for example, due to dielectric damage of capacitor C5 , verifiably affects at least the rise time T φ at input E5 of monitoring unit 110 . An5 .

[0034] According to the method, preferably firstly the voltage measuring device V is short-circuited at least briefly. 1,…,5 The voltage U selected by the algorithm is directly determined at the first time t1. 1,…,5 (t1). At least briefly short-circuit the selected voltage measuring device V 1,…,5 Afterwards, it is preferred to continue to measure the relevant current voltage U 1,…,5 (t2), until it has reached the relevant original voltage U 1,…,5 (t1) is a fraction (between 0.1 and 0.9) of the voltage U, thus satisfying the termination condition. 1,…,5 The rise time T An1,…,n is obtained from the difference between the two times t2 and t1. In this case, the second time t2 is obtained from the cessation of the following time: starting from the first time t1 until the above-mentioned end condition is reached. The first and second times t1, t2 are preferably determined by means of a time measurement unit 132 which is also controlled by the microcontroller 130. The smaller the fraction used as the end condition, the faster the fault detection is. In the case of medium-sized fractions, the accuracy is increased, while in the case of large fractions, the sensitivity is lower. Therefore, preferably, the relevant voltage U is carried out with a fraction of 0.5 as the end condition. 1,…,5 (t2) is measured. Then, the voltage U 1,…,5 The rise time T obtained in this way An1,…,5 Based on the predefined tolerance ranges in the monitoring unit 110, four exemplary electrical faults F in the passive input circuit 120 can be detected. 1,…,4 .

[0035] An advantage of the monitoring unit 110 of the battery pack 100 is that the associated rise time T An1,…,5 On the one hand, this allows for a tolerance range for the evaluation and for the individual battery cells Z of the battery pack 100 to be determined. 1,…,5The corresponding charge state of the device is not affected, on the other hand, there is a commercially available microcontroller 130 in which, for example, an analog / digital converter or a voltage measuring device V 1,…,5 threshold, and only the rise time T An1,…,5 , until a predetermined threshold is reached. Due to this situation, more resources of the microcontroller 130 can be used for other tasks. An1 , a correction factor may also be calculated and taken into account if necessary: ​​if (as shown here) input E1 is not connected via a (compensation) resistor R in the ground branch but directly to negative pole M of battery pack 10 .

[0036] Figure 2 The detection by monitoring unit 110 is shown Figure 1 In step 202, the charging process of the battery pack 100 is started. In step 204, different battery cells and different channels of the input filter are selected in each pass. Next, in step 206, the voltage U is measured at the first time t1. 1,…,n (t1). In addition, in step 208, a switch S is selected and at least briefly closed. 1,…,5 In the subsequent step 210, the relevant switch S is turned on again. 1,…,5 , and start the time measurement unit or timer in the monitoring device 110. Thereafter, in step 212, the current voltage U 1,…,n (t2). In the subsequent query step 214, a short test is performed: the current voltage U 1,…,n (t2) Is it zero volts? If not, then the charging operation is immediately stopped in step 230 via path 280. Otherwise, in step 216, the relevant voltage U is continuously measured. 1,…,n (t2), until the voltage has reached the original voltage U 1,…,5 (t1) is a fraction (preferably 0.1 to 0.9). In the subsequent query step 218, a short check is performed: the voltage U 1,…,n (t2) whether the voltage U continues to rise. If not, the charging operation is immediately stopped in step 230 via path 282. However, if the voltage U 1,…,n (t2) If the time measurement unit continues to rise, the time measurement unit is stopped, and the relevant rise time T can be calculated from the difference between the second time t2 and the first time t1 when the time measurement is stopped. An1,…,n For a number n equal to 1 (n=1), it may also be necessary to correct the time t with a numerical correction factor An1 .

[0037] In query step 222, a test is again performed to determine whether the number n of inputs has been traversed, i.e., whether all inputs have been checked by means of the battery cells. If this is not the case, a jump is made back to step 204 via (jump back) path 284. Otherwise, the process proceeds via path 286 to step 224, in which all measured rise times T An1,…,n An absolute evaluation is performed as to whether these rise times are within a predefined tolerance range. In a further query step 226, a check is made as to whether all tolerance ranges are observed. If not, the charging operation is continued in step 228, otherwise the charging operation is immediately stopped in step 230 via a further (jump) path 288. In principle, the rise time T An1,…,n The tolerance range in the absolute evaluation of φ must be designed to be relatively large in order to adequately take into account all deviations arising from aging processes of the battery cells and active and passive electronic components.

[0038] Figure 3 Another alternative algorithm for identifying electrical faults is shown. Preferably, Figure 2 Steps 202 to 230 of the algorithm 200 are identical to steps 202 to 230 and paths 280 to 288 of the algorithm 300 described herein, so that for the sake of brevity and clarity of description, reference may be made to the description of the algorithm 300 described herein at this point. Figure 2 An implementation method is carried out within the scope of the description.

[0039] As Figure 2 The only difference of algorithm 200 is that Figure 3 The algorithm 300 additionally implements the rise time T An1,…,n A relative evaluation or assessment is performed in a program part 302 , which is additionally referenced and enclosed in a dashed line and includes steps 304 , 306 and a query step 308 .

[0040] In step 304 following query step 226, firstly the previously determined minimum and maximum rise times T An1,…,n The difference between the two is used to calculate the rise time T An1,…,n Then, in step 306, according to the rise time T An1,…,n The actual relative evaluation is performed by comparing the maximum extension with a predetermined tolerance range or a predefined tolerance interval, preferably stored in the monitoring unit 110. In the query step 308, a test is performed to determine whether the maximum extension lies within the tolerance range. If so, the charging operation is continued in step 228. Otherwise, the charging operation is immediately terminated again in step 230 via another (jump) path 290.

[0041] In principle, it is possible to use a predefined tolerance range to determine T An1,…,n However, the relative evaluation does not allow for systematic errors, such as incorrect installation of all capacitors with the same but incorrect capacitance value. Figure 2 The advantages of the relative tolerance evaluation become apparent only in combination with the absolute tolerance evaluation of the input circuit 120. Furthermore, the tolerance range can be measured narrower in the case of the relative evaluation than in the case of the absolute evaluation. This is because it can be assumed that the same components within the input circuit 120 also always experience the same degree of aging.

[0042] Figure 4 Another battery pack 400 is shown, which is Figure 1 The battery pack 100 is different in that it has an additional resistor R Z1,…,5 Improved input circuit 410, so for the sake of brevity of description, the description of other active and passive electronic components of battery pack 400 can refer to Figure 1 Preferably, the passive input circuit 410 has a plurality of additional resistors R Z1,…,n , the additional resistors are respectively 1,…,n One is connected in series, where the additional resistor R Z1,…,n The number of battery cells Z is preferably equal to 1,…,n The pre-set quantity.

[0043] exist Figure 4 In the improved input circuit 410 of FIG. 4 , the number n of the battery pack 400 which is again given by way of example is equal to five (n=5) battery cells Z 1,…,5 , five additional resistors R are set accordingly illustratively Z1,…,5 , they are respectively connected with the resistor R 1,…,5 connected in series, where they are directly located at the input E of the monitoring unit 110 2,...,6 In front of one of the input terminals. The five additional resistors R Z1,…,5 are determined to be less than or at most equal to the other resistors R 1,…,5 Preferably, the additional resistor R Z1,…,5 The resistance value is resistance R 1,…,5 The additional resistor R Z1,…,5 The resistance value is resistor R 1,…,5 The additional resistor R Z1,…,5 The number of battery cells Z preferably corresponds to 1,…,5 The number n and the resistance R 1,…,5 The number n therefore corresponds here to the number 5. However, it should be noted that the additional resistor R Z1,…,5The number of battery cells Z may not be equal to 1,…,5 and / or resistor R 1,…,5 The number of additional resistors R Z1,…,5 A current decay is caused in the input circuit 410 of the monitoring unit 110 .

[0044] Instead of the five electronic switches S in the monitoring unit 110 1,…,5 Other active or passive electronic components may also be provided, which achieve: instead of causing a defined short circuit, draw a specific current or connect a defined resistance. These components may be part of the SCM circuit 112, or alternatively, may be integrated into the passive input circuit 410, such as Figure 4 As exemplarily described in .

[0045] For the evaluation (input) voltage U given here 1,…,5 In the case of a fault detection in the input circuit 410 of the battery pack 400, the electronic switch S is determined by the algorithm. 1,…,5 At the beginning of the time interval or at the relevant switch S 1,…,5 The closing process measures the current (input) voltage U immediately after the first time t1 takes effect. 1,…,5 (t1). After the time interval has expired or at the second moment t2, the relevant switch S is opened again. 1,…,5 , and immediately thereafter measure the associated (input) voltage U 1,…,5 (t2). Next, preferably, U is calculated by means of the microcontroller 130 of the monitoring unit 110. 1,…,5 (t2) / U 1,…,5 (t1). "Directly" because, in other cases, the capacitor C 1,…,5 will be assigned to its battery cell Z 1,…,5 Recharging, which would result in a measurement distortion of the falling curve. The detection of the two times t1, t2 can be performed, for example, by means of a time measurement unit 132 of the monitoring unit 110 controlled by the microcontroller 130. In this case, the difference between the second time t2 and the first time t1 is equal to the determined time interval.

[0046] Corresponding to Figures 1 to 3 In the embodiment described in the context of FIG. 1 , the ratio U of the two voltages can again be evaluated absolutely and / or relatively based on a suitable predetermined tolerance range. 1,…,5 (t2) / U 1,…,5 (t1), and consider using it to detect four electrical faults F in the input circuit 410 of the battery pack 400 1,…,4A particular advantage of the battery pack 400 and the evaluation of the drop curve is that two faults that compensate for each other can also be detected, for example if 1 / R 1,…,5 and C 1,…,5 The values ​​of the pairs are changed by the same degree so that the product of R and C remains unchanged.

Claims

1. A battery pack (100) having a predetermined number (n) of interconnected individual battery cells (Z 1,…,n ) and for monitoring the individual battery cells (Z 1,…,n ) of a monitoring unit (110), wherein: The monitoring unit (110) has at least one voltage measuring device (V 1,…,n ) and the respective assigned electronic switches (S 1,…,n ) and having a passive input circuit (120), wherein the at least one voltage measuring device is used to detect the single battery cell (Z 1,…,n ) voltage (U 1,…,n ), the passive input circuit (120) has a plurality of resistors (R 1,…,n ) and capacitor (C 1,…,n ), the monitoring unit (110) being configured to, by means of a time measuring unit (132) assigned to the monitoring unit (110), determine the time of the individual battery cells (Z 1,…,n ) voltage (U 1,…,n ) rise time (T An1,…,n ) to identify different electrical faults (F 1,…,4 ), characterized in that by means of an allocated electronic switch (S 1,…,n ) at least briefly across the voltage measuring device (V 1,…,n ) in each of which a voltage measuring device is connected through the voltage (U 1,…,n ) rise time (T An1,…,5 ) to identify the electrical fault (F) in the passive input circuit (120) by absolute and / or relative analysis 1,…,4 ).

2. The battery pack according to claim 1, characterized in that: The passive input circuit (410) has a plurality of additional resistors (R Z1,…,n ), the additional resistors are respectively connected to the resistor (R 1,…,n ) is connected in series, wherein the additional resistor (R Z1,…,n ) is equal to the number of battery cells (Z 1,…,n ) of a predetermined number (n).

3. The battery pack according to claim 2, characterized in that: The additional resistor (R Z1,…,n ) is determined to be less than or equal to the resistance (R 1,…,n ).

4. The battery pack according to claim 1, characterized in that: The monitoring unit (110) is a monitoring unit in the form of a single cell monitoring circuit (112).

5. The battery pack according to any one of claims 1 to 4, characterized in that: The monitoring unit (110) is implemented as an integrated circuit (114).

6. A method for detecting an electrical fault (F) in a passive input circuit (120) of a monitoring unit (110) 1,…,4 ), the monitoring unit being used for monitoring a predetermined number (n) of individual battery cells (Z) of a battery pack (100) 1,…,n ), the battery pack is a battery pack (100) according to any one of claims 1 to 5, wherein: The passive input circuit (120) has a plurality of resistors (R 1,…,n ) and capacitor (C 1,…,n ), characterized in that the method comprises the following steps: a) starting the charging operation of the storage battery pack (110); b) At the first moment (t1), the battery cell (Z 1,…,n ) voltage (U 1,…,n (t1)); c) With the aid of an assigned electronic switch (S 1,…,n ) at least briefly across the voltage measuring device (V 1,…,n ); d) Measure the voltage (U 1,…,n (t1)) rise time (T An1,…,n ), until the voltage (U 1,…,n (t2)) has reached the voltage (U 1,…,n (t1)) is a fraction between 0.1 and 0.9; e) a predetermined number (n) of battery cells (Z) of the battery pack (100) 1,…,n ) repeats steps b) to d) for each battery cell in; f) Based on a predefined tolerance range, the individual battery cells (Z 1,…,n ) voltage (U 1,…,n ) rise time (T An1,…,n ) performs absolute analysis processing and / or relative analysis processing; g) If at least one rise time (T An1,…,n ) is outside the tolerance range and therefore a fault (F) has been detected in the passive input circuit (120) of the monitoring unit (110) 1,…,4 ), the charging operation of the battery pack (100) is stopped; otherwise, the charging operation is continued and steps b) to g) are repeated cyclically.

7. The method according to claim 6, characterized in that The monitoring unit (110) is a monitoring unit in the form of a single cell monitoring circuit (112).

8. The method according to claim 6 or 7, characterized in that: If the voltmeter is not connected to the negative pole (M) of the battery pack (100, 400) via the first resistor (R), the rise time is corrected by the time correction factor when the predetermined number (n) is equal to one.

Images