# Battery Management System

A Battery Management System (BMS) is an electronic circuit that can manage a rechargeable device. Like most electronics, accumulators are limited in the voltage and current they can handle. While some are quite robust in terms of e.g. overvoltage or deep-discharge, it is vital especially for Li-on batteries, to monitor charge, discharge and charge cycles to ensure a long lifetime. In most cases, the battery comes in a pack, which consists of multiple modules, which each again consists of multiple cells. A single cell is the atomic unit of a battery and comes in different shapes ([1]). The pouch and prismatic cell shapes are most used in consumer electronics (mobile phones, laptops). In applications with higher energy demand, prismatic, cylindrical and pouch shaped cells are common. Pouch cells need to be packed into modules, which includes a frame, sensors and sometimes features for cooling. Multiple modules are than stacked into a pack, which includes the BMS, cooling, fuses and necessary wiring amongst other components ([1]).

# Main Functions

# State of Charge

The BMS will optimize the charging rate and determine when to stop dis-/charging based on the state of charge (SOC).

The SOC of the cells is calculated by combining current and voltage measurements, sometimes temperature.

The most simple SOC determination is done with an open circuit voltage (OCV) lookup table. It has a significant deviation due to the OCV not being accurately measurable during current flow. A more sophisticated algorithm consists of combining the voltage measurement with a current measurement, accumulating the current into or out of a cell, also known as coulomb counting.

The current research on SOC estimation suggests that a favorable algorithm to perform the combination of current and voltage is the extended Kalman filter (EKF) or variations of it. This method is known as sensor fusion and is well established. More advanced methods use an equivalent circuit model of the cell (ECM) in order to account for changing cell characteristics like internal resistance. These methods not only provide a better SOC, but also the state of health (SOH) of the battery. Good literature by Dr. Gregory L. Plett on implementation of such algorithms can be found here (opens new window).

# Electric Cell Protection

As described in chapter "Battery", different cells have different minimum and maximum voltage levels as well as different phases while charging/discharging.

The BMS is responsible to measure the voltage, current and temperature and stop or reduce dis-/charging in order to stay within defined safety limits of the cells.

To disconnect the battery from the load/charger, different types of switches can be used. For low current systems, a MOSFET switch is often easiest to use, while a mechanical or solid state relais can be necessary to switch higher voltages and currents.

# Thermal Protection

In order to keep the battery in a safe operational state, the thermal managmenent is vital. The system can either be active or passive, using mostly air or another liquid coolant. The necessary hardware like fans and their electronics are mostly not considered part of the BMS but are controlled by it.

# Balancing

To optimize performance and lifetime of battery packs, balancing is used to distribute load and prevent localized over- or under-charging of cells. Since cells differ slightly in capacity depending on the quality of the manufacturing process and used materials or by cell-aging, cells connected in series can show different voltages. To normalize the SOC for each cell, a BMS can either actively transfer energy from higher charged cells to those with lower SOC or passivly, mostly by wasting energy from cells with 100% SOC until every cell is fully charged.

Passive balancing can be done by converting excessive charge to heat using a resistor. The resistor is connected in parallel to the cell and switched on with a transistor for the cells with higher voltage, thus having excessive charge in comparisons to the other cells in series. Passive balancing slightly increases the thermal energy dissipated by the pack and decreases the overall efficiency.

Active balancing aims to distribute the energy better while charging and discharging, but need much more circuitry.

# Communication

A BMS can also feature communication capabilities to distribute information to other BMS, charge-controllers or logging systems. Commonly used for standalone setups is a serial connection while BUS connections like CAN are preferably used in systems with several components. See chapter Communication for more.

# Bus precharge

A common feature may also be a precharge system to protect a connected load. Precharging protects the large capacitors inside the load (inverter) against high inrush currents on battery connection. To slowly charge these capacitors, the load is connected over a resistor for a short period of time. Afterwards, the resistor is removed and the load connected normally.

# Topologies

# Centralised

The BMS is directly connected to every single cell of the battery, monitoring several states in parallel. For standalone setups, where size is fixed, this topology is easiest to implement. The systems developed by Libre Solar follow this centralized approach.

Centralized BMS layout
Figure 1. Centralized BMS layout.

# Distributed

A monitoring unit is connected to each cell, reporting information about the cell to a central controller. A good Open Source Project using this approach is Stuart Pittaways diyBMS (opens new window).

Distributed BMS layout
Figure 2. Distributed BMS layout.

# Modular

A modular setup is designed for larger battery packs. A specialized board is connected to 8 to 16 cells, measuring cell voltages and temperatures. Each of those units is than connected to a main BMS controller, which is responsible for the analysis of the data. An Open Source Project can be found here (opens new window).

Modular BMS layout
Figure 3. Modular BMS layout.

References

[1] Automotive Batteries 101, David Greenwood, University of Warwick, 2018 Link (opens new window)