METHOD FOR DETECTING FAULT ARCING DURING THE CHARGING OF ELECTRICAL BATTERY SYSTEMS

Description

[0001] The invention relates to a method for detecting arc faults during the charging of electrical battery systems. A lead-acid battery is preferably used as the battery system. The invention further relates to a method for producing electrical battery systems and a shutdown device.

[0002] Lead-acid batteries are commonly used as car batteries. These typically consist of six or twelve 2-volt battery cells connected in series. Each cell has two lead electrodes immersed in dilute sulfuric acid, which acts as the electrolyte. When discharged, a layer of lead(II) sulfate is deposited on the two electrodes. When charged, one of the two electrodes, the positive electrode, has a layer of lead(IV) oxide, while the negative electrode consists primarily of porous lead. To prevent short circuits between the electrodes, which are usually arranged relatively closely together, a separator, such as a perforated, corrugated polyvinyl chloride (PVC), is placed between them.

[0003] After mechanical production, the lead-acid batteries must be charged, as they are mostly used as starter batteries in the automotive sector, and the vehicle must be ready for use immediately after the starter battery has been replaced. For this to happen, an electrical voltage must be applied to the lead-acid batteries for a specific period of time during production, thus charging them. In order to produce a large number of lead-acid batteries in a relatively short period of time, they are usually connected in series and contacted with an electrical voltage source. This requires that the lead-acid batteries be electrically contacted with one another. This is usually done by loosely placing connecting cables, relatively simple clamps or the like onto the individual connections of the lead-acid batteries.However, it's possible that the connection between one of the cables and the associated connector could become loose, for example, due to dirt or improper installation. Due to the applied electrical voltage, an arc can form between these components, through which the electrical current continues to flow. In other words, the charging process is not interrupted.

[0004] When an electrical voltage is applied, the sulfuric acid electrolyte dissociates, and at least some hydrogen is formed, which mixes with oxygen to create oxyhydrogen gas. This is usually contained within the casing of the lead-acid battery. However, it is also possible that a relatively large amount of oxyhydrogen gas is formed due to manufacturing tolerances. In this case, to prevent the lead-acid battery casing from bursting, the excess pressure causes closures in the casing to open, allowing the oxyhydrogen gas to escape. In this case, it is possible for such a cloud of oxyhydrogen gas to move into the area of ​​an arc. The spark generated by the arc, in particular the plasma, and the relatively high temperature of this plasma, can ignite the cloud of oxyhydrogen gas.This may result in damage to the lead-acid battery manufacturing facility and the lead-acid batteries located there, as well as injury to any persons present.

[0005] A charging system for batteries is known from US 2017 / 0126036 A1. WO 2017 / 079041 A1 also shows a battery system that can be used to identify charging errors.

[0006] DE 10 2012 020 718 A1 discloses a method for detecting an arc in a motor vehicle with a multi-voltage electrical system. The motor vehicle has an energy storage device configured as a multi-voltage battery, which provides two different electrical voltage sources.

[0007] EP 2 916 455 A1 discloses a device for detecting a serial arc in an electrical system comprising a DC voltage source. The DC voltage source is provided, for example, by several photovoltaic modules, electric accumulators, wind turbines, or by generators driven by internal combustion engines.

[0008] The invention is based on the object of improving the charging of electrical battery systems, in particular reducing the risk of explosion and / or increasing reliability.

[0009] With regard to a method for detecting arc faults during the charging of electrical battery systems, this object is achieved according to the invention by the features of claim 1, with regard to a method for producing electrical battery systems by the features of claim 9, and with regard to a shutdown device by the features of claim 11. Advantageous further developments and refinements are the subject of the respective subclaims.

[0010] The method is used to detect arc faults, specifically the formation of a plasma between two electrically charged parts, whereby a current flow between the two electrically charged parts is created or maintained by the plasma. For the formation of such arc faults ("ARC"), an electrical voltage greater than a minimum voltage is required. This minimum voltage depends, for example, on contact materials, current strength, and / or air temperature. In particular, the minimum voltage is essentially equal to 15 volts.

[0011] The method is used to detect arc faults during the charging of electrical battery systems. The battery systems are electrically connected in series to form a string. The string itself is fed by a DC-DC converter. This results in an energy flow from the DC-DC converter to the string. Consequently, the string forms at least partially, preferably completely, the load, and the DC-DC converter forms at least partially, preferably completely, the source. In other words, an electrical DC voltage is applied to the string by means of the DC-DC converter, which voltage is expediently essentially constant. For example, several strings, each with an associated control system, are connected in parallel to one another and are in particular structurally identical to one another.The string has at least two battery systems, preferably between 5 battery systems and 50 battery systems, in particular between 10 battery systems and 30 battery systems, and for example 15 battery systems.

[0012] In particular, each battery system has an electrolyte, and / or each battery system is electrochemical. The electrolyte of each battery system is in particular aqueous and / or in an aqueous solution. Expediently, each battery system is a battery system with an aqueous electrolyte. For example, each battery system has a number of electrochemical cells that are suitably interconnected. Suitably, the battery systems are structurally identical. The battery systems are, for example, accumulators and are thus rechargeable. In particular, the battery systems are open systems. In other words, when the battery system is charged, a certain chemical substance escapes, for example, or is absorbed, for example, from the ambient air, such as in particular air, oxygen, or hydrogen. Particularly preferably, the battery system is a lead-acid accumulator, such as a so-called starter battery for a motor vehicle.For example, the battery system, in particular the lead-acid battery, has an electrical voltage of 12 volts or 24 volts when charged (nominal voltage values). To charge the electrical battery system, a charging profile is suitably used, which includes section-by-section current and / or voltage regulation. In particular, an electrical voltage (charging voltage) of up to 2.4 volts is applied per cell of the battery system, for example, up to 13.8 volts or 14.8 volts per battery system. Due to the series connection, it is therefore necessary for the DC-DC converter to provide a multiple of this charging voltage, with the multiple corresponding to the number of battery systems per string.

[0013] For example, the battery systems are charged by regulating the voltage applied to the string or by regulating the current flowing through the string. For example, the current flowing through the string is initially regulated. Once the battery systems are charged to a certain level, the system switches to regulating the voltage applied to the battery systems and thus to the string.

[0014] The method provides that in one work step a first value corresponding to an electrical voltage applied to the string is created. For example, the electrical voltage applied to the string is used directly as the first value. Alternatively, the first value is created based on the electrical voltage. For example, the first value is calculated based on the electrical voltage or determined based on a characteristic map. However, at least the first value is dependent on the applied electrical voltage. In a further work step, which takes place, for example, before or after the first work step, a second value is created which corresponds to the electrical current flowing through the string. Suitably, the first value and the second value are created essentially simultaneously.Preferably, the first value is based on the electrical voltage present when the electrical current flows, which is used to determine the second value. For example, the electrical current is used directly as the second value, or the second value is calculated based on the electrical current or determined based on a characteristic map. In particular, to determine the first and second values, the electrical voltage or the electrical current is suitably measured, in particular directly.

[0015] As a first condition, a further step checks whether the first value changes by more than a first limit value within a first time window. This check only checks whether the first value increases by more than the first limit value. It checks whether the electrical voltage used to create the first value increases by a specific value corresponding to the first limit value within the first time window or a time period corresponding to the first time window. Preferably, the check for the first condition is carried out essentially continuously, as long as the string is supplied with power via the DC-DC converter.

[0016] Furthermore, as a second condition, it is checked whether the second value changes by more than a second limit value within a second time window. In this case, it is only checked whether the second value decreases by more than the second limit value. It is checked whether, within a time period corresponding to the second time window, the electrical current, on the basis of which the second value is created, decreases by more than a certain value that corresponds to the second limit value. In summary, it is thus implicitly checked in particular whether the electrical voltage applied to the string increases within a certain time period and whether the electrical current decreases within a further time period. For example, the existence of the second condition is checked essentially continuously, at least as long as the string is fed by the DC-DC converter.Alternatively, the existence of the second condition is only checked if the first condition is met.

[0017] To determine whether the first value or the second value changes by more than the first or second limit value, the derivative of the time course of the first value or the second value is suitably created, which simplifies checking for both conditions. The arc is detected if the first condition and the second condition are present within a third time window. For example, the arc is only detected if the second condition occurs after the first condition. Suitably, the second condition is only checked if the first condition is present.

[0018] In summary, it is checked in particular whether the electrical voltage applied to the string increases and the electrical current flowing through the string decreases. When an arc occurs, it is formed, for example, between an electrical contact of one of the battery systems and a cable or the like connecting the battery systems, so that the arc is connected in series with the battery systems. Due to the arc and the (equivalent) capacitances provided by the individual battery systems as well as components of the DC-DC converter and other components, such as in particular a transformer feeding the DC-DC converter, the electrical voltage across the entire string increases and the electrical current decreases.After a certain transient and readjustment phase, which is particularly longer than the first, second and third time windows, the electrical voltage and current applied to or flowing through the string are normalized.

[0019] Arc detection makes it possible, for example, to switch off the DC-DC converter or to disconnect the string from the DC-DC converter or at least to interrupt it electrically so that the arc is extinguished. This means that any loose electrical connections can be detected relatively quickly and the arc can be switched off before any explosive gases that arise when charging the battery systems reach the area of ​​the arc / loose connection. This reduces the risk of explosion when charging the battery systems and thus increases reliability. It also prevents a pole of the respective battery system from burning down or at least becoming deformed due to an existing arc during charging and the increased temperature. This also reduces the number of rejects during charging of the battery systems.By checking for both conditions, the method is comparatively robust and independent of the number of arcs occurring and, essentially, their respective duration. In particular, the method for detecting arc faults is only executed when the battery systems, which are particularly rechargeable battery systems such as lead-acid batteries, are being charged for the first time. Alternatively, the method for detecting arc faults is executed when the battery systems that have already been charged at least once are being recharged and / or recharged.

[0020] Preferably, the first value is created by averaging the electrical voltage applied to the string over a fourth time window. In other words, the electrical voltage applied to the string during the fourth time window is recorded. During the fourth time window, the electrical voltage is determined several times, in particular more than twice, more than five times, or more than ten times. For example, the electrical voltage is determined continuously or, for example, as a function of further processing, which is suitably carried out digitally. In particular, the determination is coordinated with any A / D converter, for example with its sampling rate. For example, the electrical voltage is recorded again every microsecond, every second microsecond, every fifth microsecond, every tenth microsecond, every twentieth microsecond, every fiftieth microsecond, or every hundredth microsecond.In particular, the arithmetic mean is used for averaging, so that first the sum of the values ​​for the applied electrical voltage is calculated and divided by the number of values ​​(measured values). Suitably, as soon as a new value for the electrical voltage is determined, i.e. in particular every 10 microseconds, the first value is newly created. The electrical voltage values ​​determined during the fourth time window are averaged and this value is used, for example, as the first value. In other words, the mean value itself is used as the first value. However, it is particularly preferred to use the derivative of the time course of the mean value, in particular approximately determined by means of numerical calculation, as the first value and to compare it with the first limit value.

[0021] Alternatively, or particularly preferably in combination with this, the second value is created by averaging the electrical current flowing through the string over a fifth time window. In this case, the electrical current is expediently recorded, for example measured, several times during the fifth time window and these values ​​(measured values) are used for averaging. During the fifth time window, the electrical current is determined several times, in particular more than twice, more than five times or more than ten times. For example, the electrical current is determined (recorded) continuously or, for example, as a function of further processing, which is suitably carried out digitally. In particular, the determination is coordinated with any A / D converter, for example with its sampling rate. Suitably, the arithmetic mean is created. For example, the mean itself is used as the second value.However, it is particularly preferred to use the derivative of the time course of the mean as the second value and to compare it with the second limit value.

[0022] In particular, between 100 and 400 measured values, for example 300 measured values, are used to determine the first and second values, respectively. The length of the fourth time window is preferably the same as the length of the fifth time window, which simplifies processing. Due to the averaging, short-term current or voltage peaks are smoothed, which prevents incorrect identification of arcs. Furthermore, any fluctuations in the supply voltage of the DC-DC converter are taken into account. Alternatively, for example, instead of averaging, the maximum or minimum of the electrical voltage present during the fourth time window or the electrical current flowing during the fifth is used.

[0023] Suitably, a moving average is calculated in each case so that, if further measured values ​​are provided, older measured values ​​are no longer used for averaging. The length of the fourth and / or fifth time window is expediently constant. In particular, each time a new measured value of the electrical voltage or electrical current is available, the first or second value is recalculated. For this purpose, a so-called FIFO memory ("First in, First out") is used, for example, to simplify the calculation. In particular, only the values ​​used for averaging are stored in the memory, and these are used to create the first or second value. For example, to calculate the average, a fraction of a value to be deleted from the memory is subtracted from an already calculated average, and a fraction of a new value to be saved is added.

[0024] In particular, the length of the fourth time window is equal to the inverse of the product of twice the number of phases of the DC-DC converter and a supply frequency of the DC-DC converter. In other words, twice the number of phases by which the DC-DC converter is operated is first determined. For example, the DC-DC converter is connected to a three-phase AC network. In this case, the number of phases is three ("3"). Twice the number of phases corresponds, in particular, to the number of sub-bridges of the DC-DC converter. If the DC-DC converter has a bridge circuit, for example, constructed as a B6 circuit and thus has six semiconductors, twice the number of phases is six ("6"). If the DC-DC converter is connected to a single-phase AC network, twice the number is two ("2").

[0025] The supply frequency is the frequency of each individual phase of the DC-DC converter, for example 50 Hz or 60 Hz. For a three-phase DC-DC converter with a supply frequency of 60 Hz, the length of the fourth time window is 1 / 360 s, or approximately 2.8 ms. For a three-phase DC-DC converter with a supply frequency of 50 Hz, the length of the fourth time window is 1 / 300 s, or approximately 3.3 ms. Alternatively, an integer multiple of this is used. In other words, the length of the fourth time window corresponds to double, triple, quadruple, etc. of the inverse. Thus, the length of the fourth time window is a multiple of the inverse, i.e., single, double, triple, etc. of the inverse.Due to the length of the fourth time window and the averaging used to determine the first value, fluctuations in the electrical voltage caused by the AC voltage used to power the DC-DC converter are essentially completely smoothed out. This prevents erroneous identification of the arc.

[0026] Alternatively, or particularly preferably in combination with this, the length of the fifth time window is equal to the inverse of the product of twice the number of phases of the DC-DC converter and a supply frequency of the DC-DC converter, or an integer multiple thereof. In particular, the length of the fifth time window is thus 1 / 360 s, 2 / 360 s, 3 / 360 s, 4 / 360 s, 5 / 360 s, etc., provided the DC-DC converter has a B6 circuit and is operated at a supply frequency of 60 Hz. Consequently, fluctuations in the electrical current that are caused by the design of the DC-DC converter and not by an arc are smoothed out, which prevents erroneous identification of arcs. Suitably, the length of the fourth time window is equal to the length of the fifth time window. If the DC-DC converter only has a single phase, the length of the fourth orfifth time window is equal to the inverse of twice the frequency by which the DC-DC converter is supplied.

[0027] For example, the length of the third time window is chosen to be equal to the length of the fourth time window. Alternatively, the length of the third time window is chosen to be equal to the length of the fifth time window. Particularly preferably, the length of the third time window is chosen to be equal to the length of the fourth time window and equal to the length of the fifth time window. In other words, the length of the fourth time window is equal to the length of the fifth time window. Such a selection avoids artifacts and any arc is detected comparatively quickly and reliably. For example, between 1 millisecond and 65 milliseconds, for example 3 milliseconds, is used for the length of the third time window. Alternatively, 2 milliseconds is chosen as the length for the third time window.

[0028] The first limit value is selected, for example, to be greater than or equal to 15 volts. To maintain the arc, an electrical voltage of at least 15 volts is required, so that if the arc occurs and is connected in series with the battery system, an increase in the electrical voltage of at least 15 volts occurs. As a result, the arc is reliably detected if the electrical voltage applied to the string increases by more than 15 volts. However, the first limit value is particularly preferably selected to be less than 12 volts. Alternatively, or particularly preferably in combination with this, the first limit value is greater than 6 volts. For example, the first limit value is between 10 volts and 8 volts. Consequently, if the first value increases by more than 6 volts, 8 volts, 10 volts or 12 volts within the first time window, the first condition is met.With this type of selection for the first limit value, erroneous detection of the arc is essentially ruled out due to the lower limit. The upper limit is selected to be lower than the electrical voltage required for the arc. In particular, if the electrical voltage is averaged to create the first value, the arc is detected relatively quickly after it occurs. Furthermore, it is not necessary for the arc to exist essentially continuously. This method also detects arcs that occur periodically or erratically, for example due to a loose electrical connection between the battery systems, or whose duration is shorter than the first and / or second time window.

[0029] For example, the quotient of the first limit and twice the system inductance is chosen as the second limit. The system inductance is determined, calculated, or estimated, for example, by measuring. The system inductance describes the inductance of the system used to charge the battery systems and is determined primarily by any transformer used to feed the DC-DC converter. In addition, the system inductance depends, in particular to a small extent, on the string, i.e. the cabling and the battery systems. The system inductance is specifically tailored to the DC-DC converter and / or the battery systems used, as well as their number. For example, the second limit is greater or smaller than this quotient. Consequently, if the second value decreases by more than the quotient, the second condition is met.

[0030] In particular, the length of the first time window is equal to the inverse of the product of twice the number of phases of the DC-DC converter and a supply frequency of the DC-DC converter. In other words, twice the number of phases by which the DC-DC converter is operated is first determined. For example, the DC-DC converter is connected to a three-phase AC network. In this case, the number of phases is three ("3"). Twice the number of phases corresponds, in particular, to the number of sub-bridges of the DC-DC converter. If the DC-DC converter has a bridge circuit, for example, constructed as a B6 circuit and thus has six semiconductors, twice the number of phases is six ("6"). If the DC-DC converter is connected to a single-phase AC network, twice the number is two ("2").

[0031] The supply frequency is the frequency of each of the individual phases of the DC-DC converter, for example 50 Hz or 60 Hz. For a three-phase DC-DC converter with a supply frequency of 60 Hz, the length of the first time window is therefore 1 / 360 s, or approximately 2.8 ms. For a three-phase DC-DC converter with a supply frequency of 50 Hz, the length of the first time window is therefore 1 / 300 s, or approximately 3.3 ms. In other words, a check is carried out to determine whether the first value changes by more than the first limit value within 2.8 ms or 3.3 ms, and in particular whether it increases by more than the first limit value. Alternatively, an integer multiple of this is used in each case. In other words, the length of the first time window corresponds to twice, three times, four times, etc., the inverse value. Thus, the length of the first time window is equal to a multiple of the inverse, i.e., one, two, three, etc. of the inverse.

[0032] Alternatively, or particularly preferably in combination with this, the length of the second time window is equal to the inverse of the product of twice the number of phases of the DC-DC converter and a supply frequency of the DC-DC converter, or an integer multiple thereof. In particular, the length of the second time window is thus 1 / 360 s, 2 / 360 s, 3 / 360 s, 4 / 360 s, 5 / 360 s, etc., provided the DC-DC converter has a B6 circuit and is operated at a supply frequency of 60 Hz.

[0033] If the DC-DC converter has only a single phase, the length of the first or second time window is equal to the inverse of twice the frequency used to power the DC-DC converter. Preferably, the length of the first time window is equal to the length of the second time window. Such a selection of the length of the first or second time window enables relatively reliable detection of the arc fault, since the changes in electrical voltage and current follow one another relatively quickly when the arc fault occurs. However, relatively short-term fluctuations, for example due to manufacturing tolerances of the DC-DC converter, are not incorrectly identified as a flashover.Suitably, the first and / or second time window is determined based on a sampling frequency, in particular a frequency at which any A / D converter is operated. For example, the length of the first time window is an integer multiple of a time period corresponding to the sampling frequency.

[0034] Suitably, the arc is only detected if the electrical voltage applied to the string prior to the third time window, within which both the first and second conditions occur, is greater than a third threshold. In other words, if both the first and second conditions exist within the third time window, a check is carried out to determine whether the electrical voltage was greater than the third threshold immediately prior to this third time window. For example, the third threshold is greater than 100 volts, greater than 120 volts, and in particular less than 200 volts. For example, the third threshold is substantially equal to 150 volts. For example, the third threshold is between 50% and 75% of the product of the number of cells, if any, in the string and 2 V.Alternatively, or in combination with this, the arc is only detected if the electrical current flowing through the string before the third time window is greater than a fourth limit value, whereby both the first and second conditions are met within this third time window. Thus, it is checked whether the electrical current flowing through the string is greater than a fourth limit value immediately before the third time window, within which both the first and second conditions are met. The fourth limit value depends, for example, on the battery systems and / or is greater than 1 ampere (A), 2 amperes or 3 amperes and, for example, less than 10 amperes. For example, 5 amperes is used as the fourth limit value.Alternatively, or particularly preferably in combination with the two previous conditions, the arc is only detected if, during the third time window, during which the first and second conditions are also met, the electrical current flowing through the string is greater than a fifth threshold. The fifth threshold is, for example, greater than 0.5 amperes, 1 ampere, and, for example, less than 3 amperes or 2 amperes.

[0035] The additional conditions ensure that the battery systems are actually being charged and that the system is already in a steady and / or stabilized state. In other words, the first and second conditions do not arise, for example, due to switch-on effects of the DC-DC converter or other components. Arcs that occur during a switch-on process are also not identified. When switched on, and essentially shortly after switching on and the start of charging of the battery systems, no explosive gases have yet formed, so there is no risk of explosion. As an alternative to checking for the additional conditions, a check is carried out, for example, to determine whether the battery systems have already been charged for a certain period of time. This information is transmitted, for example, via a cable, a line, a digital signal, or a bus system.The arc is only detected if the battery systems have already been charged for a certain period of time, in particular 10 seconds or 1 minute.

[0036] The method for producing electrical battery systems provides for first providing a number of uncharged battery systems. Examples of battery systems used include accumulators based on the electrochemical principle. In particular, rechargeable battery systems comprising an electrolyte are provided. For this purpose, electrodes and an electrolyte are first mechanically placed in a housing. Lead-acid batteries are particularly preferred as battery systems. In uncharged lead-acid batteries, a number of electrodes made of lead, which are suitably interconnected, are arranged in a sulfuric acid solution.

[0037] The uncharged battery systems are electrically connected in series to form a string. For example, a cable or the like is placed or attached to the poles of each battery system and electrically contacted with a further pole of each of the further uncharged battery systems. The string is electrically contacted with a DC-DC converter, for example directly or particularly preferably via a disconnection device. In other words, the string is electrically contacted with the disconnection device and the disconnection device is electrically contacted with the DC-DC converter. The disconnection device enables, in particular, an electrical separation or at least an electrical interruption of the string from the DC-DC converter. The connection of the uncharged battery systems to the string and the contacting with the DC-DC converter takes place, for example, in a single method step.

[0038] In a further process step, an electrical voltage is applied to the string using the DC-DC converter, creating an electrical current flow through the string. The disconnection device is preferably controlled appropriately for this purpose. The electrical voltage applied to the string is equal to the product of the number of battery systems and the charging voltage required for one of the battery systems, and is, for example, essentially equal to 400 volts. The electrical current is, in particular, between 5 amperes and 100 amperes, between 10 amperes and 90 amperes, and, for example, between 30 amperes and 80 amperes.

[0039] In addition, a method for detecting arc faults when charging electrical battery systems is implemented. A first value corresponding to an electrical voltage applied to the string is created. Furthermore, a second value corresponding to an electrical current flowing through the string is created. The first condition is checked whether the first value changes by more than a first limit value within a first time window. This checks whether the first value increases by more than the first limit value within the first time window. The second condition is checked whether the second value changes by more than a second limit value within a second time window. This checks whether the second value decreases by more than the second limit value. To determine the conditions, in particular the derivative of the time course of the first and / or second value is determined.

[0040] The arc is detected if the first condition and the second condition occur within a third time window. The length of the third time window is, in particular, substantially equal to 3 milliseconds or, for example, less than 3 milliseconds, in particular equal to 2 milliseconds or 1 millisecond. Thus, the occurrence of an arc during the manufacture of the electrical battery systems can be relatively reliably detected.

[0041] For example, if a certain amount of time has passed since the electrical voltage was applied and the electrical current flow was established, and / or a certain electrical current and / or a certain electrical voltage is present across the string, the battery systems are fully charged and thus created. In this case, the current supply is preferably interrupted and / or the application of the electrical voltage is interrupted. In particular, the interconnection of the battery systems is removed. In other words, the string is separated again, so that the individual battery systems are separate. These are then loaded and / or packaged, for example.

[0042] For example, after the electrical voltage is applied and the electrical current flow is established, the electrical current is initially regulated to a specific value. After a certain period of time has elapsed, regulation to a specific electrical voltage is suitably carried out. The battery systems are preferably cooled during charging, i.e., as long as the electrical voltage is applied to the string and / or as long as the electrical current flows. For this purpose, the battery systems are exposed to cooling air, for example. Alternatively, or in combination with this, the battery systems are arranged, for example, in a cooling basin that is at least partially filled with a liquid, in particular water.

[0043] Conveniently, the string is electrically disconnected from the DC-DC converter or at least electrically interrupted when the arc is detected. In other words, the current flow to the string, and thus the charging of the battery systems, is stopped. The application of the electrical voltage to the string is also stopped. As a result, the arc collapses, and the plasma, which serves to maintain the arc, cools. Therefore, there are no sparks in the vicinity of the string, so that any oxyhydrogen cloud that may arise, for example, due to the charging, particularly due to the dissociation of water, cannot ignite.

[0044] The disconnection device serves to electrically interrupt (disconnect), for example, separate, a string from a DC voltage source, in particular a DC-DC converter. In particular, the disconnection device serves to galvanically isolate the string from the DC-DC converter. The string itself has a number of battery systems electrically connected in series, such as, in particular, lead-acid batteries. The disconnection device comprises an electrical current sensor, by means of which it is possible to detect an electrical current. Furthermore, the disconnection device has an electrical voltage sensor, by means of which an electrical voltage is detected during operation. The disconnection device preferably has a switch, such as, for example, a relay or a semiconductor switch, such as a power transistor, in particular a MOSFET, IGBT, or thyristor.For example, the shutdown device has an interface to a bus system or the like. The shutdown device also has a control unit which is coupled, in particular in terms of signaling, to the electrical current sensor and / or the electrical voltage sensor. For example, the control unit comprises an A / D converter by means of which an input signal from the current sensor or the voltage sensor is converted into a digital word during operation. The switch and / or the interface is expediently also coupled to the control unit so that it can be operated by means of the control unit. The electrical current sensor, the electrical voltage sensor and the switch are expediently arranged such that, provided the shutdown device is electrically connected between the DC-DC converter and the string, an electrical current flowing through the string is detected by means of the electrical current sensor during operation.The electrical voltage sensor detects the electrical voltage present in the string during operation. The switch can be used to establish or interrupt the flow of current from the DC / DC converter to the string. The interface allows for the output of a control signal or telegram, particularly to the bus system (data bus).

[0045] The control unit is operated according to a method in which a first value corresponding to an electrical voltage applied to the string is created. Furthermore, a second value corresponding to an electrical current flowing through the string is created. The first condition is checked whether the first value increases by more than a first limit value within a first time window, and the second condition is checked whether the second value decreases by more than a second limit value within a second time window. An arc is detected if the first condition and the second condition are present within a third time window. If the arc is detected, in particular the switch is controlled such that the string is no longer electrically supplied by the DC-DC converter. In other words, the switch is in particular opened.Alternatively, or in combination, a control command or telegram is issued via the interface. The control command or telegram is preferably used to shut down the DC-DC converter. The control unit is suitable, in particular designed and configured, to carry out this process.

[0046] The developments and advantages described in connection with the method for detecting arc faults during the charging of electrical battery systems are to be transferred analogously to the method for producing electrical battery systems and / or the shutdown device and vice versa as well as to each other.

[0047] An embodiment of the invention is explained in more detail below with reference to a drawing. In the drawings: Fig. 1 shows a method for producing electrical battery systems, with a method for detecting arc faults when charging electrical battery systems, Fig. 2 shows a schematically simplified view of a number of battery systems that are electrically connected in series to form a string that is fed by a DC-DC converter, Fig. 3 shows a shutdown device for electrically interrupting the string from the DC-DC converter, Fig. 4 shows a time profile of an electrical voltage applied to the string when the arc fault occurs, Fig. 5 shows a time profile of an electrical current flowing through the string when the arc fault occurs, Fig. 6 shows a time profile of a first value when the arc fault occurs, and Fig. 7 shows a time profile of a second value when the arc fault occurs.

[0048] Corresponding parts are provided with the same reference numerals in all figures.

[0049] In Fig. 1 A method 2 for producing electrical battery systems 4 in the form of lead-acid batteries is shown. In a first step 6, the uncharged battery systems 4 are provided. In a second step 8, the uncharged battery systems 4 are connected in series to form a string 10 and to a disconnection device 12, as shown in Fig. 2 Here, a negative pole of each of the battery systems 4 is electrically connected to a positive pole of another of the battery systems 4 by means of a bridge 14 in the form of a cable to form the string 10. The two connection ends of the string 10 are electrically connected to the shutdown device 12.

[0050] In a third work step 16, which, for example, occurs simultaneously with, before, or after the second work step, the shutdown device 12 is electrically contacted with a DC-DC converter 18. The DC-DC converter 18 has a bridge circuit 20 with six semiconductor switches 22 in the form of MOSFETs or thyristors. Alternatively, the semiconductor switches 22 are conduction semiconductor diodes. The bridge circuit 20 is thus a B6 circuit and consequently has six partial bridges. The DC-DC converter 18 is fed by means of a three-phase transformer 24, which is electrically contacted with a three-phase AC network 26. The AC network 26 thus has three phases, each phase being sinusoidal and having a frequency of 60 Hertz.

[0051] In a fourth step 28, which occurs subsequently, an electrical voltage 30 is applied to the disconnecting device 12 and thus to the string 10 by means of the DC-DC converter 18. Furthermore, due to the electrical voltage 30, an electrical current 32 is created via the disconnecting device 12 and via the string 10. An electrical voltage of 392 volts and an electrical current flow of 50 amperes are provided by the DC-DC converter 18, with 28 battery systems 4 being provided. Furthermore, the voltage is initially regulated to the desired electrical current 32. After a certain period of time, the voltage is regulated to the electrical voltage 30.

[0052] The shutdown device 12 is in Fig. 2 shown in a simplified schematic and has a forward conductor 34 and a return conductor 36. The forward conductor 34 extends between a first connection 38 and a second connection 40 and is formed, for example, by means of a busbar. One of the connections of the DC-DC converter 38 is electrically connected to the first connection 38 in the assembled state. One of the connections of the string 10 is connected to the second connection 40 in the assembled state. The return conductor 36 extends between a third connection 42 and a fourth connection 44, with the remaining connection of the DC-DC converter 18 being connected to the third connection 42 and the remaining connection of the string 10 being connected to the fourth connection 44 in the assembled state. A switch 46 in the form of a relay is incorporated in the return conductor 36, so that an electrical current flow between the third connection 42 and the fourth connection 44 can be interrupted by means of the switch 46.The switch 46 is signal-connected to a control unit 48, so that the switch 46 is operated by means of the control unit 48.

[0053] The shutdown device 12 further comprises an electrical voltage sensor 50, which includes an A / D converter. During operation, the electrical voltage sensor 50 detects an electrical voltage present between the second terminal 40 and the fourth terminal 44 and converts it into a digital word, which is forwarded to the control unit 48. During operation, the electrical voltage sensor 50 detects the electrical voltage again every 10 microseconds. The shutdown device 12 further comprises an electrical current sensor 52 for measuring an electrical current flowing between the third terminal 42 and the fourth terminal 44. The electrical current sensor 52 comprises a sensor probe 54. The sensor probe 54 detects the electrical current flowing on the return conductor 36.In addition, the electrical current sensor 52 has an A / D converter 56, by means of which the measured values ​​of the sensor probe 54 are converted into a digital word and forwarded to the control unit 48. During operation, the A / D converter 56 creates a new digital word every 10 microseconds.

[0054] During method 2 for producing the electrical battery systems 4, a method 58 for detecting arc faults 60 is further carried out, wherein the arc fault 60 forms, for example, between a pole of one of the battery systems 4 and the associated bridge 14, for example because the bridge 14 becomes detached from the associated pole due to faulty fastening. The method 58 for detecting arc faults 60 provides that, in a fifth work step 62, a first value 64 is created based on the electrical voltage 30 detected by the electrical voltage sensor 50 and applied to the string 10. For this purpose, a new value (measured value) of the electrical voltage is detected every 10 microseconds by the electrical voltage sensor 50. These measured values ​​are averaged over a fourth time window 66, and thus the arithmetic mean of these measured values ​​is created.The first value 64 is the time derivative of the mean value of the electrical voltage 30.

[0055] The time course of the mean value of the electrical voltage 30 is shown in Fig. 4 and the time course of the first value 64 is in Fig. 6 shown. The length of the fourth time window 66 is equal to the inverse of the product of twice the number of phases of the DC-DC converter 18 and a supply frequency of the DC-DC converter 18. The number of phases of the DC-DC converter 18 is three and the supply frequency is 60 Hertz, so that approximately 270 measured values ​​each contribute to the creation of one of the mean values ​​and thus approximately 560 contribute to one of the first values ​​64. To calculate the mean value, for example, all measured values ​​are added together again or an already created mean value is used and from this the oldest first value 64 is subtracted and the most recent first value 64 is added.

[0056] In a sixth work step 68, a second value 70 is created based on the electrical current 32 detected by the electrical current sensor 52.

[0057] For this purpose, the arithmetic mean of the electric current 32 flowing through the string 10 is first created over a fifth time window 72. The temporal course of the mean value of the electric current 32 is shown in Fig. 5 The second value 70 is the time derivative of the mean value and in Fig. 7The length of the fifth time window 72 is constant and equal to the length of the fourth time window 66 and thus equal to the inverse of the product of twice the number of phases of the DC-DC converter 18 and the supply frequency of the DC-DC converter 18. Due to the choice of the length of the fourth and fifth time windows 66, 72, an averaging is carried out over any ripples in the electrical voltage 30 and the electrical current 32 that are caused by the individual phases of the DC-DC converter 18, so that the average value is essentially constant unless the arc 60 occurs.

[0058] The control unit 48 has two memories that operate according to the FIFO principle. The measured values ​​of the electrical current sensor 52 are stored in one of the memories, and the measured values ​​of the electrical voltage sensor 50 are stored in the other of the two memories. Both memories are identical in design and have a total of 270 memory locations. All measured values ​​stored in the memory are used to calculate the respective mean value of the electrical voltage 30 or the electrical current 32.

[0059] In a seventh work step 74, a check is performed to determine whether the first value 64 changes by more than a first limit value 78 within a first time window 76. The first limit value 78 is selected to be less than 12 volts and greater than 6 volts and is equal to 8 volts. 10 microseconds are selected as the length for the first time window 76. Consequently, a check is performed to determine whether the first value 64 changes by more than the first limit value 78 within the time period between the renewed determination of the first value 64. If the control arc 60 ignites at an ignition time 79, the electrical voltage 30 increases by the value required to maintain the fault arc 60 connected in series with the battery systems 4. This electrical voltage is at least 15 volts.

[0060] In an eighth work step 80, the second condition is checked to determine whether the second value 70 changes by more than a second limit value 84 within a second time window 82. The length of the second time window 82 is equal to the length of the first time window 76 and is equal to 10 microseconds, so that each time the second value 70 is determined again, a check is also carried out to determine whether it has changed by more than the second limit value 84. The second limit value 84 is the quotient of the first limit value 78, i.e., 10 volts, and twice the system inductance. The system inductance depends on the transformer 24, the cable length of the bridges 14, other cable lengths, and the number of individual battery systems 4.

[0061] In a ninth work step 86, it is checked whether the first condition and the second condition are present within a third time window 88. In other words, it is checked whether less than the length of the third time window 88 has elapsed between the change in the first value 74 by more than the first limit value 78 and the change in the second value 70 by more than the second limit value 84 within the respective time window 76, 82. The length of the third time window 88 is chosen to be equal to the length of the fourth and fifth time windows 66, 72 and is thus equal to 1 / 360 s.

[0062] If the first and second conditions are met, a tenth work step 90 checks whether, before the third time window 88, within which both the first and second conditions are met, the electrical voltage 30 applied to the string 10 was greater than a third limit value 92, which is 150 volts. Furthermore, a check is made as to whether, before the third time window 88, within which both the first and second conditions are met, the electrical current 32 flowing through the string 10 was greater than a fourth limit value 94, which is 5 amperes. In addition, a check is made as to whether, during the entire third time window 88, within which both the first and second conditions are met, the electrical current 32 flowing through the string 10 is greater than a fifth limit value 96. The fifth limit value 96 is used as 2 amperes.

[0063] If these three additional conditions and the first and second conditions are met, the arc 60 is detected in an eleventh work step 98. In this case, a message is output, for example. In addition, a twelfth work step 100 is subsequently carried out, in which the switch 46 is actuated and the string 10 is thus electrically disconnected from the DC-DC converter 58. If the first and second conditions are either not met at all or do not occur within the third time window 88, no arc 60 is detected and, after a certain period of time has elapsed, depending on a characteristic curve specification and / or depending on an electrical voltage / electrical current provided / applied by the string 10, the current supply to the string 10 by means of the DC-DC converter 18 is terminated in a thirteenth work step 102. In addition, the bridges 14 are released and the now charged battery systems 4 are removed.Also, no (false) detection of the arc 60 occurs if the first and second conditions are met within the third time window 88, but the electrical voltage applied to the string before the third time window 88 was not greater than the third limit value, or the electrical current 32 flowing through the string 10 before the third time window was not greater than the fourth limit value, or the electrical current 32 flowing through the string 10 during the third time window 88 is not greater than the fifth limit value 96.

[0064] In summary, the charging of the individual battery systems 4 is interrupted when the arc 60 is detected. This prevents any oxyhydrogen cloud that may arise during charging from igniting. Furthermore, the individual bridges 14 and / or the individual poles of the battery systems 4 can be examined after the arc 60 is detected and replaced if necessary. Reattachment of the bridges 14 is also possible. The electrical current sensor 52 and the electrical voltage sensor 50 of the shutdown device 12 are used for this purpose, with the analog measurement data being converted into a digital word by means of the respective A / D converters 56. In particular, typical characteristics of the arc fault 60 are identified by means of the control unit 48. If the arc fault 60 is identified, the switch 46 is actuated, so that the string 10 is electrically disconnected from the DC-DC converter 18.

[0065] By averaging the fourth and fifth time windows 66, 72, high-frequency elements are filtered out, which arise in particular due to the operation of the semiconductor elements 22. In other words, a smoothing process is performed. The length of the fourth and fifth time windows 76, 72 is selected such that at least 20 measured values ​​are used to determine the first and second values ​​64, 70, respectively, thus achieving sufficient accuracy.

[0066] Due to the comparatively high battery capacity, the electrical voltage 30 increases by at least 15 volts when the arc fault 60 occurs, whereas the electrical current 32 decreases exponentially, with the exponent initially equal to the negative quotient of the initial voltage of the arc 60 and the system inductance. After a certain period of time, these effects subside due to the discharge of the individual battery systems 4 and the subsequent regulation.

[0067] The first condition is to check whether the first value 64 changes by more than the first limit value 78 within the first time window 46. Due to the averaging, this corresponds to checking whether the difference between the moving average of the electrical voltage coming from the right and that coming from the left is above the first limit value 78. Furthermore, the second condition is to check whether the second value 70 changes by more than the second limit value 84 within the second time window 82, i.e. whether the difference between the moving average of the electrical current 32 determined from the left and from the right is greater than the second limit value 84, wherein the second limit value 84 is selected, for example, as the negative quotient of the first limit value 78 and twice the system inductance.

[0068] Furthermore, it is checked whether the first condition occurs essentially at the same time as the second condition or at least within the third time window 88, wherein, for example, only the arc 60 is detected if both conditions are present within the third time window 88. For example, it is additionally checked whether the second condition only occurs after the first condition. Otherwise, the arc 60 is not detected. Furthermore, in the tenth work step 90, it is checked whether the electrical voltage 30 was greater than the third limit value 92 before the assumed time of the arc 60, and whether the minimum electrical current 32 directly before the assumed arc 60 was greater than the fourth limit value. It is also checked whether the flowing electrical current 32 is greater than the fifth limit value during the assumed arc 60.In this way, false detection of the arc 60 during the start of a current supply, i.e. during switching on of the transformer 24 and / or the DC-DC converter 18, is avoided.

[0069] The invention is not limited to the exemplary embodiment described above. Rather, other variants of the invention can also be derived therefrom by those skilled in the art without departing from the scope of the claims. In particular, all individual features described in connection with the exemplary embodiment can also be combined with one another in other ways without departing from the scope of the claims. List of reference symbols

[0070] 2Process for manufacturing electrical battery systems 4Battery system 6First work step 8Second work step 10String 12Shutdown device 14Bridge 16Third work step 18DC converter 20Bridge circuit 22Semiconductor switch 24Transformer 26AC network 28Fourth work step 30Electrical voltage 32Electrical current 34Forward conductor 36Return conductor 38First connection 40Second connection 42Third connection 44Fourth connection 46Switch 48Control unit 50Electrical voltage sensor 52Electrical current sensor 54Sensor probe 56A / D converter 58Process for detecting arc faults 60Arc fault 62Fifth work step 64First value 66Fourth time window 68Sixth work step 70Second value 72Fifth time window 74Seventh work step 76first time window 78first limit 79ignition point 80eighth work step 82second time window 84second limit 86ninth work step 88third time window 90tenth work step 92thirdLimit value 94fourth limit value 96fifth limit value 98eleventh work step 100twelfth work step 102thirteenth work step

Claims

1. Method (58) for detecting arcs (60) when charging electric battery systems (4), which are electrically connected in series to form a string (10), which is supplied by means of a DC voltage converter (18), in which method - a first value (64) corresponding to an electrical voltage (30) applied to the string (10) is generated, - a second value (70) corresponding to an electric current (32) flowing through the string (10) is generated, - a check is carried out, as a first condition, to determine whether the first value (64) increases by more than a first limit value (78) within a first time window (76), - a check is carried out, as a second condition, to determine whether the second value (70) decreases by more than a second limit value (84) within a second time window (82), and - an arc (60) is detected when the first condition and the second condition are met within a third time window (88).

2. Method (58) according to Claim 1, characterized in that the first value (64) is generated by averaging the electrical voltage (30) applied to the string (10) over a fourth time window (66), and / or in that the second value (70) is generated by averaging the electric current (32) flowing through the string (10) over a fifth time window (72).

3. Method (58) according to Claim 2, characterized in that the length of the fourth or fifth time window (66, 72) is selected to be constant and corresponds to the reciprocal of the product of twice the number of phases of the DC voltage converter (18) and a supply frequency of the DC voltage converter (18) or to an integer multiple thereof.

4. Method (58) according to Claim 2 or 3, characterized in that the length of the third time window (88) is selected to be equal to the length of the fourth or fifth time window (66, 72).

5. Method (58) according to any one of Claims 1 to 4, characterized in that the first limit value (78) is selected to be lower than 12 V and / or greater than 6 V.

6. Method (58) according to any one of Claims 1 to 5, characterized in that the second limit value (84) is selected to be equal to the quotient of the first limit value (78) and twice a system inductance.

7. Method (58) according to any one of Claims 1 to 6, characterized in that the length of the first and / or second time window (66, 72) is selected to be constant and corresponds to the reciprocal of the product of twice the number of phases of the DC voltage converter (18) and a supply frequency of the DC voltage converter (18) or to an integer multiple thereof.

8. Method (58) according to any one of Claims 1 to 7, characterized in that the arc (60) is detected only - if the electrical voltage (30) applied to the string (10) prior to the third time window (88) is greater than a third limit value (92), - if the electric current (32) flowing through the string (10) prior to the third time window (88) is greater than a fourth limit value (94), and / or - if the electric current (32) flowing through the string (10) during the third time window (88) is greater than a fifth limit value (96).

9. Method (2) for manufacturing electric battery systems (4), in particular lead-acid storage batteries, in which method - a number of uncharged battery systems (4) is provided, - the uncharged battery systems (4) are electrically connected in series to form a string (10), - the string (10) is electrically contacted by a DC voltage converter (18), - an electrical voltage (30) is applied to the string (10) by means of the DC voltage converter (18) and an electric current flow (32) through the string (10) is generated, and - a method (58) for detecting arcs (60) is performed, in which method - a first value (64) corresponding to an electrical voltage (30) applied to the string (10) is generated, - a second value (70) corresponding to an electric current (32) flowing through the string (10) is generated, - a check is carried out, as a first condition, to determine whether the first value (64) increases by more than a first limit value (78) within a first time window (76), - a check is carried out, as a second condition, to determine whether the second value (70) decreases by more than a second limit value (84) within a second time window (82), and - an arc (60) is detected when the first condition and the second condition are met within a third time window (88).

10. Method (2) according to Claim 9, characterized in that the string (10) is electrically disconnected from the DC voltage converter (18) when the arc (60) is detected.

11. Cut-off device (12) for the electrical interruption of a string (10), which has a number of battery systems (4) electrically connected in series, from a DC voltage source, in particular a DC voltage converter (18), having an electric current sensor (52), and having an electric voltage sensor (50), and also having a control unit (48), which is operated in accordance with a method (58) according to any one of Claims 1 to 8.