Battery storage systems are an important source for powering emerging clean energy applications. The Battery Management System (BMS) is a critical component of modern battery storage, essential for efficient system monitoring, reducing run-time failures, prolonging charge-discharge lifecycle, and preventing battery stress or catastrophic situations. The BMS performs functionalities such as data acquisition and monitoring, battery state estimation, cell equalization, and charge protection, making it computationally intensive to manage large scale battery storage. This necessitates the development of a System-on-chip (SoC) solution which can perform real-time and high speed battery management with improved accuracy and efficiency.
The proposal envisages the development of energy-efficient and lightweight mixed signal Battery Management SoC architecture. The proposed SoC architecture comprises an Analogue front-end (AFE) for voltage, current, and temperature sensing, Analogue to digital converters, a charge protection circuit (overvoltage, undervoltage and overcurrent conditions), a digital signal processing unit, and a master controller with required communication interfaces as shown in the Fig. The AFE architecture also incorporates charge equalization using active cell balancing. Active balancing redistributes charge among cells with imbalanced characteristics as opposed to dissipating energy in the form of heat (as in the case of passive balancing solutions), leading to improved thermal management. The mixed signal SoC offers improved hardware and software components optimization, leading to higher system performance with low latency for data handling and processing.
The SoC architecture can handle multiple cells by integrating low speed sensors with a high speed processor, enhancing the computational capability of edge devices. Additionally, the SoC offers access to cloud resources through the radio transceiver and multiple SoCs can be integrated to form a BMS cluster to provide cloud-based management capabilities. The scalable architecture also promotes ease of integration and reduces the development time for different battery configurations which varies across applications. With modern battery technology evolving rapidly, the SoC provides reconfigurability for adapting to newer technologies without the need to change the underlying processing hardware. The resulting scalable and flexible BMSoC makes it applicable across applications from IoT devices to automotive and renewable energy grid storage.

Functional Specifications of the BMSoC

1. Cell Measurement: Upto 16 cells (Rechargeable Li-ion cells)–  Combination of 16S1P, 8S2P, and 4S4P (S-series, P-Parallel connection of Energy storage cells)

• Individual cell voltages
• Every parallel string current.
• Accuracy +/-1mV
• Conversion time 37μs/cell (16 bit ADC)

2. Battery monitoring - Digital logic implementation of the Battery State Estimation (State-of-Charge, State-of-Health, State-of-Power) methodology.

• ARM Cortex (M-series) based System-on-chip (SoC) for low power computation.
• Battery Management Acceleration unit (BMS Accelerator) to be interfaced with Cortex M-core for high speed throughput.

3. Cell Balancing: Topology for equalization of charge among healthy and weak cells based on State-of-charge computation for high speed charge balancing  (requires string current  & individual cell voltage).

• High voltage analog differential MUX.

4. Internal Regulator :-

• 14.4V/28.8V/57.6 , 9A to 12 A current , Ripple 14mV , external switch, Buck boost configuration.

5. Clocking of all subsystems.

6. Charge and Load detection.

7. Temperature sensor integration.

8. OTP (One-time Programmable) Calibration  for whole system.

Design Data Inputs:

• Design specifications
• System level modelling and Simulations
• Target CDK - 130/90/65 nm (for high voltage applications)
• Third-Party IPs (Digital and Analog)