Battery stacks and battery management systems (BMS) are already widely used around us, from power tools, robotic vacuum cleaners, and drones to micro-mobility applications like e-motorcycles and e-bikes. Less conspicuous projects such as uninterruptible power supplies (UPS) and renewable energy storage require large numbers of battery cells. Each battery stack needs to be monitored to ensure it can be safely charged and discharged and to measure the overall health of the batteries. Rechargeable batteries present several challenges, requiring testing to very precise voltage levels. Furthermore, batteries are tested in a stack, necessitating accurate measurements at high common-mode voltages. The future trend is to add more cells to battery stacks as a way to power higher-voltage systems.
How does a battery management system work?
A BMS device typically consists of multiple cell measurement pins (12 to 24 or more channels) that input to a front-end analog-to-digital converter (ADC). This ADC measures the voltage of the battery cell, allowing for precise measurement of individual cells. Each cell has an additional pair of cell balancing pins, which also have ADC inputs. These pins are intended to help balance the voltage between cells in the stack. The remaining pins are power pins, analog and digital control lines. Each stack of cells in a battery requires a BMS, so in some cases, an electric vehicle may have 6 to 12 or more BMS devices, not including those used for redundancy. These devices are typically powered by the lower and upper rails of each cell module, so each BMS device floats on the BMS or cell module below it. This means that all these devices require digitally isolated, daisy-chained communication and outputs to a main controller, typically a microcontroller unit (MCU).
Electric vehicles will drive the advancement of BMS.
The largest market for future battery systems, and the main focus of this article, is automotive applications. These vehicles include (see Figure 1) all-battery electric vehicles (BEVs, 400 volts and above), as well as internal combustion engine (ICE) vehicles with start/stop technology (typically 48-volt systems), mild hybrid (48V battery-driven), and hybrid electric vehicles. In 2022, electric vehicles accounted for less than 5% of new car sales, but many automakers expect sales to rise to 50% by 2030. Given this, electric vehicle technology represents one of the fastest-growing markets for many semiconductor manufacturers.
Driving electric vehicle adoption requires several key deliverables, the first being a higher-quality Battery Management System (BMS), which impacts driving range. With a more accurate BMS, consumers can get more range from the same battery pack. For example, if the BMS can sense charge levels with 1% or higher accuracy, the battery can be charged closer to its maximum storage capacity. Think of it as a protection zone—with a 5% margin of error, the battery should only be using between 15% and 85% of its capacity. If the BMS is more accurate, then there's no need to protect the available charge in the battery. So, going from a 5% error to a 1% error allows for an 8% increase in storage capacity, translating to more miles per charge.
Secondly, in terms of safety and reliability, accurate state of charge (SOC) can improve battery utilization with greater distance projection accuracy while maintaining battery safety (avoiding catastrophic failures). Greater battery utilization and efficiency also allow for smaller, lighter battery packs, reducing vehicle costs.
Several trends are driving changes in electric vehicle batteries. The first, as mentioned above, is improved accuracy, which directly translates to more mileage between charging and gentler battery aging. Better "fuel gauge" accuracy also enhances driver safety and confidence.
The second major trend is that battery stacks are moving towards higher voltages, requiring more cells in the stack. This drives the demand for more front-end ADCs and battery balancing pins in each BMS device. Today's mainstream batteries operate at around 400 volts, with some performance electric vehicles already using 800-volt systems. Looking ahead, these levels are expected to reach 1000+ volts within just a few years, accelerating charging and helping electric vehicles charge in time closer to that of internal combustion engines. These capabilities create a competitive advantage for semiconductor suppliers, who can now add value to battery packs. However, higher overall voltages mean electric vehicles need to accommodate more batteries. Currently, manufacturers assemble batteries into modules and then assemble these modules into battery packs. This modularity creates a need for more interconnects, increasing the cost and weight of battery packs. Battery-to-pack architecture is a novel approach that places batteries directly within the battery pack, avoiding the module carrier model. Moving to a cell-to-pack architecture means manufacturers can fit the same number of cells into smaller, lighter components, or add more cells to existing areas. This can result in more cells in the stack, meaning more front-end ADC channels in the BMS device.
ATE Test Challenge
These trends in battery management systems present new challenges for automated test equipment (ATE) companies. The first challenge revolves around improving the accuracy of the BMS. When measuring the discharge profile of a battery, most of the usable area falls along a very tight curve. The full lithium-ion state of charge (SOC) ranges from ~4.3V (fully charged) to 2.2V (discharged) from 100% to 0%. Observing the full range of lithium ions, this seems like an easy task to measure the changes (~2.1V voltage range or 21 mV/1% SOC change).
A typical lithium-ion discharge range is 80-20% or 90-10% of the full cell capacity. In the 80-20% range, the state of charge (SOC) voltage is fairly flat between 3.75-3.65V (total ~100mV or 1.7 mV/1% SOC variation). This explains why BMS suppliers study measurement accuracy of 100µV or 50µV within a 5V range.
This level of accuracy, below 100µV, directly translates to forced accuracy for the ADC in each cell of a BMS device driven by the ATE stimulation channel. While this should be easily achievable on low-voltage supplies, in certain test cases, it must be maintained within a 5-6V range. Equally important is providing very low noise and very low drift stimulation in order to have a fixed, known voltage bias at the ADC input.
The second challenge is that BMS devices can consist of 16 to 24 or more cells that must be biased towards common-mode voltage to simulate a cluster of cells. In some cases, this can exceed 120 volts for some upper cell pins. This drives the requirement for dense, high-precision, low-noise floating voltage/current sources (VIs) from the ATE. Due to the test characteristics of these devices, it is often necessary to multiplex high-precision measurement systems across all cell pins and corresponding discharge pins. If the ATE system does not have sufficient internal matrixing, this may require a large number of relays on the Device Interface Board (DIB), resulting in a very large application space requirement on the DIB. In most cases, semiconductor manufacturers want to test as many devices as possible simultaneously to reduce test costs. This is also important for BMS devices, as BEVs are expected to require very high volume in the near future. An ideal ATE system should include matrixing and multiplexing within the entire system to allow for the maximum site count.
Another challenge is the daisy-chain communication between each BMS device and the MCU. All BMS devices will send data to the bus. Since this bus is within the battery cell module or battery pack, it can be a very noisy environment. This also occurs under stacked voltages, requiring an isolated interface. This isolated communication typically operates asynchronously, requiring sufficient digital instrumentation capabilities to handle the unique pattern generation. In this case, some key features required by the ATE are dynamic switching of format and cycle, and the ability of digital instruments to read asynchronous data.
Battery and BMS equipment are constantly evolving to meet the demands of the automotive market. These improvements will drive the need for new testing methods. From 1000+ voltage systems to new battery chemistry setups, numerous challenges will arise in the short term. The explosive growth of electric vehicles will require new testing capabilities and shorter production ramps, demanding that ATE suppliers go beyond BMS requirements and provide cost-effective solutions at high field volumes.
