Accurate Fuel Gauging Through State-of-Charge Measurement

Portable electronic devices rely on batteries as their only source of power. Whether you use smartphones, fitness trackers, action cameras, outdoor navigation devices, cameras, or handheld transceivers, we have all experienced an unexpected low-battery warning. In most cases, such a warning message is no more than an inconvenience, but for safety and emergency equipment, the consequences could be dire.

Because we rely so heavily on electronics running on battery power, it is essential that devices can accurately gauge power levels. Early battery-powered equipment attempted to determine battery status through a combination of cell voltage and load current but failed to consider the battery’s state-of-charge condition. The device’s circuit protection features typically include an under-voltage cut-off function, which might be invoked too early, resulting in an unexpected shutdown, all because the battery’s state-of-charge was not determined.

This blog examines how you can accurately gauge a battery’s state-of-charge condition in your battery-powered designs.

Fuel gauging by measuring battery voltage is unreliable due to the variables introduced by the cell materials, their chemistry, and the ambient temperature. Also, battery impedance changes with state-of-charge and with the battery’s age, further complicating accurate measurement. The chemistry of each battery type creates a unique discharge signature, with some suiting a voltage-based state-of-charge more than others. Some voltage and load current discharge curves are incredibly shallow, making voltage fuel gauging only able to indicate either 100 percent or flat.

Measurement of the charge and discharge current of a battery, termed coulomb counting, is another method of estimating state-of-charge. Together with accounting for battery age and the self-discharge characteristics, the coulomb counting method works well.

With an increasingly tech-savvy consumer audience, determining and presenting a battery’s state-of-charge with a high-degree-of-accuracy is a critical success factor today for any handheld consumer electronics device. The pursuit of implementing a truly accurate method of predicting a battery’s state-of-charge has led many manufacturers to characterize each and every battery against specific application and use-case scenarios. This introduces a time-to-market delay, and, if not performed in-house, requires shipment to a third-party vendor. However, some batteries, such as lithium-ion, are subject to increasing regulation relating to transporting them. The regulation includes not only the logistical nature of handling the batteries but the amount of charge they contain when shipped.

In addition to the legislation covering battery transportation, maintaining a relatively volatile battery—with many capable of delivering hundreds of amps—within safe operating parameters during routine use and storage requires further monitoring electronics to be incorporated into a design. The threat posed by after-market battery suppliers who, to save costs, do not pay attention to battery safety, has led device manufacturers to add cryptographic battery authentication techniques to their batteries and end-product designs.

As mentioned above, being able to measure a battery’s state-of-charge accurately and display it on the fuel gauge needs to take into account not only the operating power consumption but consumption during standby too. The quiescent current drawn when the device is sitting in a cupboard or “in the box” during transportation and warehousing also needs consideration. Fuel gauging naturally consumes energy itself, so this needs to be factored into the calculation with regard to how much charge the battery can be shipped with, and whether there is enough charge available for the consumer to operate the device once they receive it. As consumers, we like to use things as soon as we’ve opened the box rather than having to charge them before use. To maintain an accurate fuel gauge indication out of the box requires the associated circuitry to be always on. Turning the fuel gauge off until the consumer opens the box and turns on the product means the state-of-charge indication is likely to be inaccurate. Note that for safety during shipping and warehousing, it is prudent to keep the battery protection features (temperature, current, and voltage monitoring) enabled should excessive temperatures or component failure cause a short-circuit condition to occur.

An example IC that encompasses fuel gauge, protection, and authentication functions in a single 3mm x 3mm package is the low-power Maxim Integrated MAX1730x series. Suitable for measuring the state-of-charge of lithium-ion or lithium-polymer batteries, the MAX17301 has an extremely low quiescent current, 24µA, when the output FETs are active, down to as low as 18µA during active hibernation. By disabling the load FETs, the current can drop to 0.1µA. The IC provides a comprehensive range of battery health and safety protection features, including over-voltage (temperature dependent), over-charge current, battery under/over-temperature, under-voltage, and over-discharge/short-circuit. Other features of the MAX1730x include one-wire and I2C peripheral interfaces for communication to the host microcontroller so it can read the MAX1730x’s data and control registers (Figure 1).

Figure 1: Functional block diagram of the Maxim Integrated MAX1730x. (Source: Maxim Integrated)

The health status and protection requirements are determined by the battery voltage, current, and temperature. The battery’s state-of-charge is calculated using Maxim’s ModelGauge m5 algorithm. This algorithm combines the long-term stability of the battery’s open-circuit voltage measurement with the linearity and accuracy of coulomb counting. The additional input to the algorithm of temperature compensation yields an accurate state-of-charge reading (Figure 2). The algorithm calculates the open-circuit voltage of the battery even when it is delivering a load.

Figure 2: Calculation of the corrected state-of-charge by the Maxim MAX1730x using the ModelGauge m5 algorithm. (Source: Maxim Integrated)

The MAX1730x compensates the fuel gauge value with battery aging characteristics and discharge rate. It provides a reading of either the state-of-charge percentage or milliampere-hours (mAh) over a wide range of operating conditions. The algorithm also provides the time-to-full-charge and an age-forecasting feature to predict when the battery may start losing capacity due to age and use. A data-logging function uses non-volatile memory to record up to 13 parameters over the lifetime of the battery, including the time since the first power-up. Figure 3 illustrates the operation of the ModelGauge m5 algorithm.

Figure 3: The flowchart details the Maxim Integrated ModelGauge m5 algorithm operation. (Source: Maxim Integrated)

Being able to accurately present a battery’s state-of-charge in a fuel gauge without the need for time-consuming battery characterization is vital to the success of any product. Maxim’s MAX1730x ModelGauge m5 algorithm-based fuel gauge IC not only makes this possible but also saves valuable board space and bill of materials (BOM) by incorporating protection and authentication functions.