Lithium-ion battery SOC state of charge measurement

qwqwqw2088

Lithium-ion battery SOC state of charge measurement [Copy link]

Battery SOC is defined as the percentage of charge remaining in the battery, so it ranges from 0% to 100%. Because SOC measurement serves the same purpose as the gas gauge in a car, ICs that provide SOC measurement are often called "fuel gauges" or "fuel gauge" ICs. SOC measurement emphasizes intelligent battery management systems. As SOC changes, the battery management system calculates the optimal charging voltage and current values. Therefore, SOC measurement ICs are often paired with battery charger ICs in a design or included as a feature in a more comprehensive charge management and battery protection design. SOC measurements are used by host systems to manage power usage and by applications to notify the user when the battery charge is getting low. In electric vehicles, for example, SOC measurements are important for the estimation of the remaining range available to the vehicle and appear on the driver’s display panel as the familiar fuel gauge and range estimate. In fact, automotive applications require reliable SOC measurements to reduce “range anxiety” as drivers begin to accept these newer vehicles. In fact, reliable SOC measurements are critical to ensuring the safety of rechargeable batteries in general and maximizing battery life, especially for lithium-ion batteries. Poor estimates of SOC can lead to overcharging and overdischarging, which can result in reduced battery performance and life. Worse, uncontrolled charging can even cause battery failure, thermal runaway, or even uncontrolled venting and explosion. However, for lithium-ion batteries, accurate measurement of SOC is difficult at best. Lithium-ion batteries maintain a near-constant voltage output over most of their discharge range (Figure 1). As a result, the common approach of simply relating voltage measurements to the charge remaining in the cell cannot be used for these batteries.

Figure 1: For a typical lithium-ion battery, the voltage remains relatively flat over a wide range of cell capacity, complicating the traditional approach of reporting state of charge based simply on the cell terminal voltage. (Courtesy of Infinite Power Solutions.) Therefore, determining the SOC of a lithium-ion battery is primarily an estimation process and remains the subject of active research to find better SOC estimation methods. For large, complex battery packs such as those used in electric vehicles, the need to maximize the cost efficiency of the battery requires very sophisticated SOC estimation methods based on neural networks, fuzzy logic, and adaptive filters. However, for many other energy harvesting applications, less sophisticated methods based on current measurement, voltage measurement, or model-based approaches provide sufficient information about the battery SOC required for the application. Current-Based Methods Current-based methods track changes in the remaining charge in a battery by measuring discharge and charge currents. In this method, called coulomb counting, the battery management system estimates SOC by calculating the net increase and decrease based on the current measurements. Although the method is highly accurate in theory, the practical characteristics of the circuit make it prone to error, especially over time. Uncertainties in current sensor accuracy, parasitics, and battery aging introduce errors that accumulate over time, requiring periodic recalibration. Due to the simplicity and relative accuracy of this approach, engineers will find support for current-based SOC measurement in a variety of ICs. For example, the LTC2941 and LTC2942 from Linear Technology Corp. feature dedicated coulomb counting circuitry. The LTC2942 infers charge flow by integrating the voltage measurement of the battery current across a sense resistor. The IC applies the differential voltage between SENSE+ and SENSE- to an auto-zero differential analog integrator to convert the measured current into charge (Figure 2).

Figure 2: Fuel gauge ICs often include dedicated circuitry for coulomb counting. In the Linear LTC2942, for example, a differential analog integrator measures current by tracking the voltage across an external sense resistor. (Courtesy of Linear Technology.)

qwqwqw2088

In turn, the host controller reads the accumulated charge register (ACR) provided by the programmable prescaler. The ACR is incremented or decremented by 1 each time the prescaler underflows or overflows, so the integration time is effectively scaled by a factor M, programmable from 1 to 128. The device also includes a 14-bit Σ-Δ ADC for monitoring the battery voltage at SENSE-, enabling engineers to employ a voltage-based approach to SOC estimation. Voltage-Based Methods The voltage-based method measures the battery voltage and relates that value to the charge level. Here, the battery management system measures the voltage of the battery when it is either connected to an external load or open circuit. Measurement of the open circuit voltage (OCV) can provide sufficiently accurate results, but requires special considerations. For example, the flow of current in a lithium-ion battery causes an uneven distribution of ions in the electrolyte. This phenomenon, called the diffusion effect, can introduce errors in SOC estimates. As a result, battery management systems can improve SOC estimates by measuring the battery OCV after the battery chemistry has had a chance to equilibrate, thereby reducing the diffusion effect. Therefore, using OCV for SOC estimation can be problematic in dynamic applications where fluctuating load currents result in voltage variations and associated diffusion effects. While each approach has some limitations, semiconductor manufacturers are offering solutions that include on-chip hardware designed to support a combination of these approaches for improved SOC estimates. During fluctuating current or high current states, coulomb counting methods track the net change in SOC. During quiet periods, voltage-based methods including OCV measurement help correct for errors accumulated by coulomb counting. Along with Linear Technology’s LTC2941/LTC2942 devices, engineers can find devices including the Texas Instruments BQ2700 and STMicroelectronics STC3100 and STC3105 that include hardware features that enable engineers to use coulomb counting and voltage measurements to more accurately estimate SOC. The TI BQ2700 includes a dedicated fully differential delta-sigma coulomb counter circuit for measuring charge and discharge currents and an ADC for voltage and temperature measurements. The BQ2700 automatically compensates for offsets in the coulomb counter and ADC, so no user calibration or compensation is required. At the heart of the STMicroelectronics STC3100 and STC3105, a coulomb counting circuit tracks the SOC as the battery is charging or discharging at high rates (Figure 3).

Figure 3: The STMicroelectronics ST3105 features a dedicated digital coulomb counter that includes a 28-bit accumulator that holds the result of the current conversion. (Courtesy of STMicroelectronics.) To enable power consumption to be managed, the STC3105 provides two power modes: an operating mode that measures current every cycle, and a power-save mode that measures current only every other cycle. The ST device also includes a 14-bit sigma-delta A/D converter for voltage and current measurement. The STC3105 measures the battery voltage every four seconds with an accuracy of +/- 0.5%, allowing engineers to use these results to calculate SOC using the OCV method. To mitigate the diffusion effect, the device includes a battery voltage relaxation timer. The host processor can check this timer to ensure that the battery has been at rest long enough to ensure a more accurate SOC measurement. In addition to a 10-channel, 12-bit Σ-Δ ADC for cell voltage measurement, the Atmel ATmega406 offers a dedicated Σ-Δ ADC for coulomb counting and provides different measurement modes that enable engineers to trade measurement accuracy for power consumption. In the device’s Instantaneous Current Conversion (ICC) mode, the ADC produces a 13-bit signed result in approximately 3.9 ms, providing a method to measure the battery voltage and discharge current at approximately the same time while calculating impedance. The device’s Accumulated Current Conversion (ACC) mode is designed to provide high-accuracy results, even when the target application is running and drawing current from the battery. Although the conversion time is longer (128-1000ms), the device provides 18-bit results. Finally, the device’s Regular Current Condition (RCC) mode provides the same accuracy and conversion time as the ICC mode, but works in conjunction with the MCU’s Sleep mode to provide results during application Sleep mode and only when the current level exceeds a selectable threshold. You can also find ICs such as the Maxim DS2786 that estimate SOC based on an internal algorithm that combines coulomb counting and OCV measurements. The DS2786B measures the net charge flow while the battery is charging or discharging. During quiescence, the DS2786B waits for a programmed relaxation time and then uses the OCV model stored in the device’s EEPROM, along with the battery characteristics and application parameters, to adjust its SOC estimate from coulomb counting. The device’s EEPROM is constructed from SRAM shadows, allowing the host to override the OCV voltage curve and scaling factors to accommodate different battery types (Figure 4). Figure 4: The DS2786 stores battery characteristics and other parameters in on-chip EEPROM using shadow SRAM—enabling a host controller to change battery parameters through a serial interface. (Courtesy of Maxim Integrated Products.)

qwqwqw2088

Model-Based Approaches IC manufacturers also offer SOC measurements based on proprietary methods built around models of Li-ion battery performance. For example, Maxim stand-alone fuel-gauge devices such as the DS2780 estimate SOC based on a variety of battery and circuit characteristics, including temperature, load current, and charge termination point. The device measures battery temperature with an integrated temperature sensor with a resolution of 0.125°C every 440 milliseconds. For current measurement, the DS2780 continuously measures the current flowing into and out of the battery in active mode by measuring the voltage drop across a low-value current-sense resistor. Engineers can correct for temperature variations in the current-sense resistor by programming the value of the sense-resistor temperature coefficient. In addition, engineers can program the accumulation bias register to account for typical error sources in coulomb counting due to battery self-discharge and static offsets. Finally, these measurements are combined with the stored characteristics as input to on-chip algorithms that estimate SOC and related results (Figure 5). Figure 5; The Maxim DS2780 fuel gauge IC algorithm combines a variety of real-time measurements with stored parameters to calculate a variety of battery measurements, including SOC. (Courtesy of Maxim Integrated Products.) At the basis of these algorithms, the DS2780 stores cell characteristics using a piecewise linear model that includes three curves—full empty, efficient empty, and spare empty—each constructed from four line segments (Figure 6). Each curve defines the change in state of charge at a respective point due to temperature. The DS2780 processes the measurements and battery characteristics every 440 ms and stores the results as a function of the current temperature in on-chip registers.

Figure 6: In the Maxim DS2780, the battery cell is modeled using a piecewise linear model that includes three curves, each constructed from four line segments. (Courtesy of Maxim Integrated Products.) Maxim’s MAX17040, MAX17048, and MAX17049 employ a lithium-ion battery modeling scheme called ModelGauge that continuously measures SOC during various charge and discharge operations. Here, the algorithm determines SOC by simulating the internal nonlinear dynamics of a lithium-ion battery based on the battery’s impedance and the chemical reaction rates in the cell. The Maxim algorithm eliminates the need for external current-sense resistors and battery relearn cycles. In this approach, the device employs a custom model built by characterizing the battery at multiple discharge currents and temperatures. MAXIM ModelGauge ICs come with preloaded models that are sufficient in many cases. Texas Instruments (TI) uses a model-based SOC algorithm called Impedance Track with its BQ27541, BQ20Z75, and BQ34Z100 fuel-gauge ICs. To improve SOC accuracy, the TI Impedance Track algorithm leverages three types of information: chemical depth of discharge (DOD), battery resistance, and external factors, including load and temperature. The algorithm uses OCV measurements in a relaxed voltage state and calculates DOD based on battery-technology-specific tables stored in flash memory. Engineers can set up Impedance Track devices for specific battery chemistries using TI-supplied firmware. The battery resistance value is updated during discharge, but the algorithm delays the resistance update to reduce distortion in response to load-related transients. Temperature values are critical to the algorithm, but temperature measurements are difficult to obtain because temperature changes significantly during discharge. Therefore, the algorithm predicts future temperatures to allow for temperature correction of the battery impedance near the end of discharge. Here, the algorithm collects temperature dependency data during discharge and uses the results, including heat exchange coefficients and thermal time constants, to update its thermal model parameters. During relaxation, the algorithm also measures the external temperature and uses that result to define a temperature distribution model based on the current temperature and extending to the end of discharge. Conclusion Determining the SOC in lithium-ion batteries is challenging at best, often relying on highly sophisticated methods to ensure accuracy in the complex battery management systems used in electric vehicles. However, for many applications, simple algorithms based on current measurements, voltage measurements, cell characterization models, or a combination of each provide sufficiently accurate results. Specialized fuel gauge ICs implement these methods in many variations, giving engineers the broad range of options they need to easily implement designs with battery management systems that can accurately estimate lithium-ion battery SOC.