Research and implementation of 12v 18650 battery pack management system - Hardware implementation of 12v 18650 battery pack management system

　　The design of the hardware must achieve reasonable management of the power battery pack. First, it must ensure the accuracy of the collected data; second, reliable and stable system control; and last, very important, is anti-interference. In the specific implementation process, the type of data that needs to be collected from the power battery pack is determined based on the design requirements; the hardware design is determined based on the collection volume and accuracy requirements. The 12v 18650 battery pack management system control board designed in this article is attached to the 12v 18650 battery pack pack and powered by the 12v 18650 battery pack pack. As an online management system, the basic functions to achieve real-time management functions are as follows:

　　1. Real-time collection and display of battery information, including cell voltage, total battery pack voltage, temperature, and charge and discharge current;

　　2. Remaining power estimation function and display;

　　3. 8-cell 12v 18650 battery pack charge/discharge intelligent management, when the battery voltage, charge and discharge current, battery temperature and remaining capacity are too low or too high, an alarm signal is sent;

　　4. Balance and diagnose the single battery during the charging and discharging process;

　　5. Intelligent detection/alarm of faulty battery;

　　6. Store the charge and discharge information of lithium batteries, communicate with the host computer, and check the use process of lithium batteries on the host computer. 3.1 Hardware structure of 12v 18650 battery pack management system

　　This 12v 18650 battery pack management system is mainly composed of a charging module, a data acquisition module (including voltage, current, and temperature data acquisition), a balancing module, a power calculation module, a data display module, and a storage communication module. The system block diagram is as follows: the data acquisition module is responsible for collecting various status parameters of the battery, such as current, voltage, and temperature; the charging control module automatically charges according to the three stages of precharging, constant current charging, and constant voltage charging, and based on the collected data Control the charging and discharging process; the balancing module uses a 15W switching power supply to balance charge a single battery at the appropriate time to make the batteries in the battery pack more balanced and consistent; the power estimation unit mainly analyzes the collected status parameters and based on research The power estimation algorithm derived from experiments estimates the current power of the battery; the data display unit adopts a graphic LCD screen, which can display each battery voltage, charge and discharge current, remaining power, battery temperature and charging time; storage communication The unit regularly stores charge and discharge information (voltage, current, charge and discharge time, etc.) through the memory chip. It can communicate with the computer through the serial port and display the charge and discharge information on the computer. 3.2 Introduction to ATMEGA8L processor

　　ATMEGA8L is a high-performance 8-bit microcontroller launched by ATMEL Corporation of the United States. It has made significant improvements based on absorbing the advantages of pIC and 8051 microcontrollers. It is fast, cost-effective, can be programmed online, and can be programmed in one chip. The combination of enhanced performance RISC 8-bit CPU and downloadable FLASH makes it a highly flexible and low-cost embedded high-efficiency microcontroller suitable for many requirements. It has the following features:

　　●High performance, low power consumption 8-bit AVR microprocessor

　　●Advanced RISC structure

　　-130 instructions - most instructions execute in a single clock cycle

　　-32 8-bit general-purpose working registers - fully static operation

　　-Performance up to 16MipS when operating at 16MHz

　　-Hardware multiplier requiring only two clock cycles

　　●Non-volatile program and data memory

　　-8K bytes of programmable Flash in the system, erasing and writing life: 10,000 times

　　-The optional Boot code area with independent locking bits enables in-system programming through the on-chip Boot program, with true simultaneous read and write operations

　　-512 bytes of EEpROM erasable, lifespan: 100,000 times

　　- 1K bytes of on-chip SRAM - lock bits can be programmed to enable encryption of user programs

　　●Peripheral features

　　-Two 8-bit timers/counters with independent prescalers, one of which has a compare function

　　-A 16-bit timer/counter with prescaler, compare function and capture function

　　-Real-time counter RTC with independent oscillator

　　-Three-channel pWM-TQFp and MLF packaged 8-channel ADC 10-bit ADC

　　-pDIp packaged 6-channel ADC10-bit ADC

　　- Byte-oriented two-wire interface - Two programmable serial USARTs

　　-SpI serial interface that can work in master/slave mode

　　-Programmable watchdog timer with independent on-chip oscillator

　　-On-chip analog comparator

　　●Special processor features

　　-Power-on reset and programmable power-down detection

　　-On-chip calibrated RC oscillator

　　-On-chip/off-chip interrupt source-

　　5 sleep modes: idle mode, ADC noise suppression mode, power saving mode, power-down mode and Standby mode

　　3.3 Sampling module circuit design

　　3.3.1 Implementation of voltage sampling

　　In order to prevent overcharging and damaging the battery during charging, the lithium-ion battery requires its terminal voltage to be strictly controlled below 4.2V. This battery management system adopts the voltage detection scheme shown in Figure 3-1.

　　Its working principle is:

　　First, the multi-way switches Kn-1 and Kn-2 (n=1, 2, 3, 4, 5, 6, 7) controlled by the MCU synchronously connect the capacitors to both ends of each unit battery, so that the capacitors are charged and Make the capacitor voltage equal to the voltage of the cell under test; then the MCU controls the multi-way switches n1K and n2K to open, and at the same time closes the switches K1 and K2 to connect to the A/D for measurement. This solution can directly use the 10-bit common ground A/D in the microprocessor without adding a separate A/D, saving design costs [4]. In actual circuits, analog switches are implemented using relays. 3.3.2 Implementation of current sampling

　　Current is a key parameter for battery capacity estimation. Therefore, the requirements for its current sampling accuracy, anti-interference ability, zero drift, temperature drift and linearity error are very high. In practice, LEM's closed-loop current sensor LTSR25-Np is used, which has excellent accuracy, good linearity and optimal response time. Its rated current is 25A and can measure up to 80A current, meeting the system design requirements. The current sensor can convert the charge and discharge current into a voltage signal of 0-5V, and the A/D conversion sent to the microcontroller can measure the charge and discharge current. Its working principle is shown in Figure 3-3: Its operating characteristic curve is shown in Figure 3-4: VREF in the figure is the reference point voltage, the default is 2.5V; Ip is the measured current. 3.3.3 Implementation of temperature sampling

　　3.3.3.1 Introduction to temperature sensor DS18B20

　　Battery temperature is a key parameter for the system to evaluate the SOC of the battery and determine whether the battery can be used normally. Temperature affects the charging efficiency of the battery. At the same time, if the temperature of the battery exceeds a certain value, it may cause irreversible damage to the battery. The temperature difference between battery packs causes imbalance between battery pack cells, thereby reducing battery life. The temperature detection in this battery management system uses the digital temperature sensor DS18B20 produced by DALLAS Semiconductor Company in the United States. It has a single-chip structure and can output 9-12-bit digital quantities without the need for an external A/D. The communication adopts a single bus protocol, and various operations on the DS18B20 can be completed through a data line. At the same time, the data line can also be used as a power line, that is, it has a parasitic power mode. Because each DS18B20 contains a unique serial code, multiple DS18B20s can be connected to each bus at the same time. This makes the DS18B20 wiring simple and the system design flexible. It is suitable for a variety of temperature measurement systems, especially when combined with a microcontroller. Temperature detection and control system. The interior of DS18B20 mainly includes a parasitic power supply, a temperature sensor, a 64-bit laser ROM single-wire interface, a high-speed temporary register for storing intermediate data, TH and TL flip-flops and control logic for storing user-set upper and lower temperature limits, 8-bit Cyclic redundancy check code generator and other seven parts. The internal memory of the DS18B20 temperature sensor includes a high-speed temporary RAM and a non-volatile electrically erasable EEpROM. The latter is used to store the user-set temperature alarm upper and lower limit values TH and TL. The former's internal configuration register can be used for Determine the digital conversion resolution of the temperature value. The higher the resolution is set, the longer the temperature data conversion time is required. Therefore, there is a trade-off between resolution and conversion time in practical applications. In addition to the configuration register, the high-speed scratchpad memory also consists of 8 other bytes. Among them, the 1st and 2nd bytes are temperature information, the 3rd and 4th bytes are TH and TL values, the 6th to 8th bytes are unused, showing all logical 1; the 9th byte reads out all the previous 8 A byte CRC code can be used to ensure correct communication. DS18B20 stores the converted temperature value in the 1st and 2nd bytes of the high-speed temporary storage memory in a 16-bit sign-extended two's complement form. Corresponding temperature calculation: when the sign bit S=0, directly convert the binary bit to decimal; when S=1, first convert the complement to the original code, and then calculate the decimal value. During operation, the system operates the DS18B20 in the form of ROM commands and memory commands. The ROM operation instructions are: read ROM (33H), match ROM (55H), skip ROM (CCH), search ROM (FOH) and alarm search (ECH) commands. The scratchpad instructions are: write scratchpad memory (4EH), read scratchpad memory (BEH), copy scratchpad memory (48H), temperature conversion (44H) and read power supply mode (B4H). 3.3.3.2 Temperature detection circuit design and working principle

　　The temperature detection system adopts direct power supply mode. When DS18B20 is in memory writing operation and temperature A/D conversion operation, there must be a strong pull-up on the bus, and the maximum pull-up turn-on time is 10 μs. Since the single-wire system has only one wire, the sending and receiving ports must be three-state. At the same time, since reading and writing are separated in operation, there is no signal competition problem. Before system installation and operation, the hosts should be connected to DS18B20 one by one and their serial numbers should be read out. The working process is: the host sends a pulse, and after the "0" level is greater than 480μs, it resets the DS18B20. After the response pulse sent by the DS18B20 is received by the host, the host then sends the read ROM command code 33H (lower bit first), and then sends A pulse (15μs) followed by reading one digit of the DS18B20 serial number. Use the same method to read the 56 bits of the serial number. For the overall flow chart of the system's DS18B20 operation, it is completed in three steps: 1. The system searches for the DS18B20 serial number through repeated operations; 2. Starts all online DS18B20s for temperature A/D conversion; 3. Reads out the online DS18B20 converted temperature data. The host starts the temperature conversion and reads the temperature value; the host writes the memory data. When more detection points need temperature measurement, use other ports of ATMEGABL for expansion. The specific circuit diagram is as follows: 3.4 Charging control module design

　　The conventional charging method is carried out in three stages: precharge, constant current and constant voltage. The timing diagram is shown in Figure 3-6: In order to improve the charging efficiency, the intermittent charging method is used in the precharge and constant voltage charging stages of this battery management system. As shown in Figure 3-7: When charging a lithium-ion battery pack equipped with a battery management system, a matching constant-voltage current-limiting power adapter must be connected externally. Its constant voltage value U is

　　U=4.2*N+loss voltage

　　In the formula: N is the number of battery cells. The current limit value is the conventional charging current of the power 12v 18650 battery pack of 0.3C (C is the battery capacity). Before charging, the system must be initialized first, and then automatically proceed in the three stages of precharging, constant current charging and constant voltage charging. Charge. 1.Initialization

　　Although the initialization phase does not start charging the battery, it is an important step in the entire charging process. The intelligent energy management module initializes and self-tests itself at this stage to determine whether it is working properly, and at the same time detects whether the charging conditions meet the charging requirements:

　　(1) Whether the polarity of the external charging power supply is correct;

　　(2) Whether the external charging voltage is within the specified range;

　　(3) Whether the temperature at that time was within the allowable range;

　　(4) Whether the terminal voltage of the lithium-ion battery (each cell) is above the minimum allowable charging voltage;

　　(5) Whether the lithium-ion battery terminal voltage (each unit) is higher than the overcharge detection voltage;

　　2.Precharge

　　Precharging is not necessary every time. Its purpose is that when the battery is over-discharged, stored for too long, or the battery is damaged, and the battery terminal voltage is lower than the minimum charging voltage allowed by the lithium-ion battery, it must be charged with a small current (approximately Precharge (1/10 of the normal charging current), so that the terminal voltage of the lithium-ion battery rises above the minimum allowable charging voltage before switching to the next charging process - constant current charging. The principle of pre-charging is that the power adapter applies a relatively small charging current (about 1/10 of the normal charging current) to the battery through MCU control, so that the battery below the minimum allowed charging voltage reaches the minimum allowed charging within a fixed period of time. voltage value to avoid considering a deeply discharged battery as a non-rechargeable battery. As shown in Figure 3-8, the precharge of this module is that the power adapter precharges the battery through the precharge switch S1, resistors R4, and S3. At this time, the MCU controls the discharge switch S3 to be fully turned on through the program, and the precharge switch Tube S1 conducts intermittent conduction, using a short conduction time and a long off time (equivalent average current is small) to precharge the battery until the terminal voltage of the battery rises to the minimum charge allowed by the lithium-ion battery. voltage (between 2.5-2.7V, related to temperature), and then enter the next charging stage - constant current charging; if the precharge battery terminal voltage cannot reach the minimum allowable charging voltage for a long time, it means that the battery has been damaged, and the program Enter charging prohibited state.

　　3. Constant current charging

　　This battery management system requires the external charging power supply to be constant current for constant current charging of lithium batteries. Its constant current value should be less than the maximum allowable charging current of the lithium-ion battery. This system is set to 0.3C. The MCU controls the charging switch S2 through the program. , the discharge switch S3 is fully turned on, and the power adapter charges the battery pack with constant current through the charging switch S2 and the discharge switch S3. The battery voltage will rise slowly. Generally, the charging time is 2-3 hours. At this time, the battery power reaches 70%-80% of the full power. When the voltage of a single battery unit reaches the set termination voltage, constant current charging is terminated. The charging current decreases rapidly and charging enters the maintenance charging process. 4. Stay charged

　　This battery management system adopts the pulse charging method in the maintenance charging stage. In this stage, the pulse charging method intermittently charges the battery with a constant current at the same current value as the constant current charging stage for a fixed time t, and then turns off the charging. loop. Due to the existence of charging current, the battery voltage will continue to rise beyond the charging termination voltage. After the charging circuit is cut off, the battery voltage will slowly decrease again. When the battery voltage returns to the charging end voltage, the charging circuit is reopened, the battery is still charged at a constant current value, and then the charging circuit is closed to wait for the battery voltage to increase.decline. Under the action of pulse charging current, the battery will be gradually filled, and the rate of decrease of battery terminal voltage will also gradually slow down. This process continues until the battery voltage recovery time reaches a certain preset value, and the battery can be considered to be nearly full. High-power electronic switches are used for various protections and working states of this system. This system uses high-power, low-on-resistance MOSFETs. The reason for using MOSFETs is that this circuit has both a charging circuit and a discharging circuit. For this reason Electronic switching devices should have bidirectional conduction capabilities, and MOSFETs have this capability. The MOSFET in the actual circuit uses IRF4905, its typical on-resistance is 20mΩ, Vds=55V, Id=74A.

　　3.5 Balanced module principle and scheme design

　　3.5.1 Overview of balancing methods for 12v 18650 battery pack packs at home and abroad

　　When a lithium-ion battery pack is used in series with multiple single cells, no matter how good the performance and quality of the single cell are, if the characteristics of the single cells used in the set are inconsistent, it will cause the failure of each single cell in the battery pack. There is serious inconsistency between charging and over-discharging. As far as internal single batteries are concerned, over-charging and over-discharging are more likely to occur when used in series than when used individually, and are difficult to detect. When the characteristics of any one battery deteriorate, it will cause cascading and aggravated damage to other batteries in the battery pack, resulting in a domino effect. The quality of a battery pack is determined by the worst-quality battery among them. The poor quality of one battery not only affects the performance of the entire battery pack, but also causes a vicious chain reaction, making the poor ones worse and the good ones quickly getting worse. . In order to solve the above problems, the current common practice is to select and pair single cells to form a high-quality battery pack to minimize the differences between single cells. Even if the power lithium-ion battery pack solves the early technical problems of assembly, the characteristics of the battery pack will also change during use. Currently, due to the change in characteristics of the battery pack during use, the overall characteristics of the battery pack will decline sharply and some batteries will fail. There is no effective solution to the phenomenon of accelerated damage. Only when one battery is detected to be overcharged or overdischarged during the charging and discharging process of the battery pack, the protection circuit will shut down the entire charging and discharging circuit. Due to the above reasons, it is extremely important to solve the balance problem of each single-cell 12v 18650 battery pack in the battery pack in actual use of the power lithium-ion battery pack (especially when charging). At present, the balancing methods adopted abroad mainly include: energy consumption method and no energy consumption method. 3.5.1.1 Energy consumption balancing method

　　A typical method is to use a heating resistor to bypass and shunt the current. The schematic diagram of the bypass shunt equalization method is shown in Figure 3-9. B1, B2...Bn are the unit cells that make up the lithium-ion battery pack, K1, K2...Kn are the multi-way switches controlled by the MCU, and R1, R2...Rn are the discharge balancing resistors. When the battery pack is charged, the charging current I is equal in each battery. When the battery voltage of a certain section (for example: B2) is higher than other batteries and exceeds a certain value, the multi-way switch K2 controlled by the MCU is closed, and B2 is shunted through R2, causing the voltage of B2 to drop. This cycle is repeated n times so that each of the lithium-ion battery packs The unit cells can be balanced charged. This solution is simple and reliable, but the resistor will consume electric energy and generate heat. When using it, you need to pay attention to the resistance value and power of the resistor. Its biggest disadvantage is that during discharge (work), the balance of each unit consumes the electric energy of the lithium-ion battery pack in vain. . 3.5.1.2 Balancing method without energy consumption The balancing method without energy consumption is to use an active shunt element or a voltage or current conversion device to transfer energy from one cell to another cell. These devices can be analog or digital. The two main methods are capacitive balancing and energy conversion. The schematic diagram of capacitor balancing is shown in Figure 3-9. B1, B2...Bn are the unit cells that make up the lithium-ion battery pack, K1, K2...Kn are the multi-way switches controlled by the MCU, and C is the balancing capacitor. When the battery pack is charging, if the battery voltage of a certain cell (for example: B2) is higher than other cells by more than a certain value, and B3 is the lowest, the multi-way switches K2 and K3 controlled by the MCU are closed, and KA and KB are switched to point a, B2 C is charged through K2, K3, KA, and KB. After C is fully charged, the multi-way switches K3 and K4 controlled by the MCU are closed. KA and KB are switched at point b. The capacitor C is charged to C through K4, K3, KA, and KB. B3 releases electric energy, causing the voltage of B2 to drop and the voltage of B3 to rise. This cycle is repeated n times so that each unit of the lithium-ion battery pack can be charged in a balanced manner. This solution is also relatively simple and reliable, but you should pay attention to the charging and discharging time of the capacitor during use. Its biggest advantage is that it can balance the functions of each unit battery during charging and discharging (working), and does not consume the lithium-ion battery pack. of electrical energy. Cell balancing using energy conversion uses an inductor or transformer to transfer energy from one battery or group of batteries to another. Two active energy conversion methods are the switching transformer method and the shared transformer method. The switched transformer approach shares the same switch as the flying capacitor from before. The topology is shown in Figure 3-10. The current I of the entire battery pack flows into the transformer T, and the output of the transformer flows into the cell Bn after being corrected by the diode D. This is determined by the setting of the switch S. In addition, an electronic control device is needed to select the target battery. and setting switch S. This method of equalization is faster, but consumes the energy of the entire battery pack. In addition, there are disadvantages including complex design and numerous components, including controllers, electromagnetic induction coils and switches. The efficiency is low due to switching losses and electromagnetic losses. The shared transformer method has only one magnetic core, and the secondary coils are connected to each unit. As shown in Figure 3-11, the current I of the battery pack flows into the primary of the transformer, and each secondary generates an induced current. The lower the voltage of a single cell, the smaller its reactance, and therefore the greater the induced current. In this way, the balanced current obtained by each cell is inversely proportional to its SOC. In a shared transformer, the only active component is the switching transistor on the secondary side. Without the need for closed coils, the shared transformer method can balance the battery pack quickly and with low loss. Its disadvantage is also complicated and has many components, because each secondary requires a rectifier device.

　　3.5.2 Balancing scheme adopted by this battery management system

　　The balancing scheme used in this battery management system is based on the principle of the switching transformer balancing method in the above method. It directly uses DC/DC switching power supply. During the charging and discharging process, based on the detected voltage value of each single battery, it is judged that balanced charging is required. For a single battery, use the power of the battery pack to perform additional balanced charging on the battery. The schematic diagram is as follows: the DC/DC switching power supply uses Xinxing DOM-24D15S5, the input voltage is 18-36V, and the output voltage is 4.6-5.5 V is adjustable, the output current is 3A, the actual picture is as follows: 3.6 Implementation of LCD display

　　This battery management system uses the DM12864M Chinese character graphic dot matrix LCD module, which can display Chinese characters and graphics. It has built-in 8192 Chinese characters (16*16 dot matrix), 128 characters (8*16 dot matrix) and 64*256 dot matrix display. RAM. Main technical parameters and display characteristics:

　　Power supply: VDD3.3V-5V (built-in boost circuit, no negative voltage required);

　　Display content: 128 columns * 64 rows; LCD type: STN;

　　Interface with MCU: 8-bit or 4-bit parallel/3-bit serial;

　　Various software functions: cursor display, screen shift, custom characters, sleep mode, etc. This system uses a serial interface, and the circuit schematic is shown in Figure 3-14. The LCD module can display the total voltage of the battery pack, the voltage of each single cell, charge and discharge current, charge and discharge time, operating temperature, and remaining power. 3.7 Implementation of serial communication

　　During the 12v 18650 battery pack charging and discharging process, the 12v 18650 battery pack management system writes the charging and discharging information, including the voltage of each battery, charging and discharging current, operating temperature, battery power, etc., into the Flash memory chip SST25VF020 in real time and saves it during the 12v 18650 battery pack charging process. After the discharge is completed, you can communicate with the host PC through the serial port, check the charging and discharging process of the 12v 18650 battery pack on the host computer, and save the usage information of the 12v 18650 battery pack on the host computer for future reference. SST25VF020 is the SST25VF series Flash memory chip. Its chip has the following characteristics: total capacity is 2M; single power supply read and write operations, working voltage is 2.7-3.3V; low power consumption, working current is 7mA, waiting current is 3uA; clock frequency is up to 33MHz, fast programming, fast Erase, fast reading; data storage for 100 years; CMOSI/O compatible, etc. Its packaging is SOIC and small size WSON packaging. After the MCU samples the required data, it stores the collected data into FLASH through SpI serial communication. When the system is connected to the PC through an asynchronous serial port, the data stored in FLASH is read into the PC through SpI serial communication, so that the collected data can be analyzed and processed. After saving the collected data, you can erase FLASH to prepare for the next collection. The storage circuit is shown in Figure 3-15: The system’s serial interface UART requires two-way communication with the PC’s serial interface RS232, but the PC’s RS232 interface level uses negative logic, that is, logic 1:-3—- 15V, logic 0: +3—+15V. Among the TTL levels used by microcontrollers, high level (3.5—5V) is logic 1, and low level (0—0.8V) is logic 0. The microcontroller needs to be connected externally. The level conversion circuit chip converts the 1 represented by the TTL high level into the negative voltage signal of RS232, and converts the 0 represented by the low level into the positive voltage signal of RS232. It is to solve the level conversion problem. This system uses the MAX232C level conversion chip of MAXIM Company. This chip is highly integrated and powered by +5V power supply (built-in voltage multiplication circuit and negative power supply circuit). It only needs to connect a few external capacitors. The conversion from TTL level to RS232 level can be completed. The hardware connection circuit is shown in Figure 3-16:

Read recommendations:

Hanging neck fan

Lithium battery positive pole material

Lithium ion battery usage requirements

lithium battery energy storage direct sales

9V rechargeable battery