How to extend the life and reliability of car batteries?

One in five car failures are caused by batteries, a problem that will become more serious in the coming years as electric-by-wire, start/stop engine management and hybrid (electric/gas) vehicles become more common.

To reduce failures, it is necessary to accurately detect the battery voltage, current, and temperature, pre-process the results, calculate the charging status and operating status, send the results to the engine control unit (ECU), and control the charging function.

The modern automobile was born in the early 20th century. The first cars relied on manual starting, which required a lot of strength and was very risky. This "hand crank" of the car caused many deaths. In 1902, the first battery starter motor was successfully developed, and by 1920, all cars were powered by electric starting.

Initially, dry-cell batteries were used, which had to be replaced when their charge ran out. Soon, liquid batteries (the old lead-acid batteries) replaced dry-cell batteries. The advantage of lead-acid batteries is that they can be charged while the engine is running.

Lead-acid batteries have changed little over the last century, with the last major improvement being sealing them. What has changed is the demand on them. Initially, batteries were used only to start the car, honk the horn, and power the lights. Today, they power all of the car's electrical systems before the ignition.

The proliferation of new electronic devices is not just about consumer electronics such as GPS and DVD players. Today, body electronics such as engine control units (ECUs), power windows and power seats are standard equipment in many basic models. The exponentially increasing loads have had a serious impact, as evidenced by the increasing number of faults caused by electrical systems. According to ADAC and RAC, almost 36% of all car faults can be attributed to electrical faults. If this figure is analyzed, it can be seen that more than 50% of the faults are caused by one component, the lead-acid battery.

Assessing the health of the battery There are two key characteristics that reflect the health of a lead-acid battery:

(1) State of Charge (SoC): SoC indicates how much charge a battery can deliver, expressed as a percentage of the battery’s rated capacity (i.e., SoC for a new battery).

(2) State of Health (SoH): SoH indicates how much charge the battery can store.

State of Charge The State of Charge indicator is like a battery's “fuel gauge.” There are many methods for calculating SoC, but two of the most common methods are open circuit voltage measurement and coulomb determination (also called coulomb counting).

(1) Open circuit voltage (VOC) measurement method: There is a linear relationship between the open circuit voltage of the battery when it is unloaded and its charge state. This calculation method has two basic limitations:

First, in order to calculate the SoC, the battery must be open circuit with no load connected; second, this measurement is only accurate after a fairly long stabilization period.

These limitations make the VOC method unsuitable for online calculation of SoC. This method is commonly used in car repair shops, where the battery is removed and the voltage between the positive and negative terminals of the battery can be measured with a voltmeter.

(2) Coulometric method: This method uses coulomb counting to calculate the integral of current over time to determine the SoC. This method can calculate the SoC in real time, even when the battery is under load. However, the error of the coulometric method increases over time.

Generally, the open circuit voltage and coulomb counting method are used in combination to calculate the battery's state of charge.

State of Health The State of Health reflects the general condition of the battery and its ability to store a charge compared to a new battery. Due to the nature of batteries, SoH calculations are complex and rely on knowledge of the battery chemistry and environment. A battery's SoH is affected by many factors, including charge acceptance, internal impedance, voltage, self-discharge, and temperature.

It is generally considered difficult to measure these factors in real time in an environment such as an automobile. During the cranking phase (engine starting), the battery is under the greatest load and this is the most reflective of the battery's SoH.

The actual SoC and SoH calculation methods used by leading automotive battery sensor developers such as Bosch and Hella are highly confidential and often protected by patents. As owners of the intellectual property, they often develop these algorithms in close cooperation with battery manufacturers such as Varta and Moll.

Figure 1 shows a commonly used discrete circuit for battery detection.

Figure 1. Discrete battery detection solution

The circuit can be divided into three parts:

(1) Battery detection

The battery voltage is sensed through a resistive attenuator tapped directly from the positive terminal of the battery. To sense the current, a sense resistor (100mΩ is typical for 12V applications) is placed between the negative terminal of the battery and ground. In this configuration, the metal chassis of the car is typically ground, and the sense resistor is installed in the current loop of the battery. In other configurations, the negative terminal of the battery is ground. For SoH calculations, the temperature of the battery must also be sensed.

(2) Microcontroller

Microcontrollers or MCUs perform two main tasks. The first is to process the results of the analog-to-digital converter (ADC). This can be as simple as performing only basic filtering or as complex as calculating SoCs and SoHs. The actual functionality depends on the processing power of the MCU and the needs of the car manufacturer. The second task is to send the processed data to the ECU via a communication interface.

(3) Communication interface

Currently, the Local Interconnect Network (LIN) interface is the most commonly used communication interface between battery sensors and ECUs. LIN is a single-wire, low-cost alternative to the well-known CAN protocol.

This is the simplest configuration for battery detection. However, most precision battery detection algorithms require simultaneous sampling of the battery voltage and current, or the battery voltage, current, and temperature.

For simultaneous sampling, up to two additional analog-to-digital converters are required. In addition, the ADC and MCU require regulated power supplies for proper operation, resulting in increased circuit complexity. This has been addressed by LIN transceiver manufacturers by integrating regulated power supplies.

The next step in automotive precision battery sensing is an integrated ADC, MCU, and LIN transceiver, such as the ADuC703x family of precision analog microcontrollers from Analog Devices.

The ADuC703x offers two or three 8 ksps, 16-bit Σ-Δ ADCs, a 20.48MHz ARM7TDMI MCU, and an integrated LIN v2.0-compliant transceiver.

The ADuC703x series integrates a low dropout regulator on chip and can be powered directly from a lead-acid battery.

To meet the needs of automotive battery sensing, the front end includes the following components: a voltage attenuator to monitor the battery voltage; a programmable gain amplifier that supports measuring full-scale currents from less than 1A to 1500A when used with a 100mΩ resistor; an accumulator that supports coulomb counting without software monitoring; and an on-chip temperature sensor.

Figure 2 shows a solution using this integrated device.

Figure 2. Example of a solution using integrated devices.

A few years ago, only high-end cars were equipped with battery sensors. Today, more and more low- and mid-range cars are equipped with small electronic devices, which ten years ago were only found in high-end models. As a result, the number of failures caused by lead-acid batteries is increasing. In a few years, every car will be equipped with a battery sensor, thus reducing the risk of failures caused by the increasing number of electronic devices.