Power Train
Large Li-ion battery packs: Active balancing improves many parameters – part 2
In Part 1 one of this article it was shown that an active cell balancing Battery Management System (BMS) is the best solution for large Li-Ion battery packs for automotive applications. A conclusion was reached that active cell balancing is a key enabler of the EV revolution. The 2nd part of this article reviews the popular active balancing methods and presents a unique active cell balancing BMS solution from National Semiconductor.
The first Li-Ion battery cell balancing methods addressed only the main issue - prevention of cells overcharging. The simplest solution was to bypass the over-charged cells by providing an alternative path for the charging current and dissipate the excessive energy. Passive balancing methods were developed based on this principle.
The next step was to improve the efficiency of the over-charge protection by redistributing the excessive charge to other cells instead of dissipating it. This led to the introduction of active cell balancing.
The evolution of BMS's was affected by the fact that the first BMS topologies were designed to manage small number of low capacity cells – typically 3 to 12. Integrated circuits (ICs) developed for small battery packs are ineffective in dealing with the challenges posed by large EV battery packs. Another solution is needed.
Active cell balancing
Fig. 1 Charge “shuttling” techniques examples. For full resolution, click here.
In battery packs consisting of less than approximately 12 cells, “charge shuttling” methods based on the principle of transferring the charge from one cell to a capacitive and inductive storage element and then to the under-charged cell are used. The storage element is being connected to various cells by a transistor switch matrix. In order to limit the number of switches the charge transfer is typically limited to the adjacent cells. This method becomes very inefficient as the number of cells grows.
For example, if the efficiency of one transfer is 85% and the charge needs to be moved to the 12th cell, the overall efficiency of such transfer becomes 0.8511 = 0.167 = 16.7%. More than 83% of the energy is lost as heat.
In capacitive shuttling the process becomes very inefficient when the cell voltage differential is small. The flat voltage characteristics of some cell chemistries exacerbate the problem.
In general, the charge shuttling methods are characterized by low balancing currents adequate only for low number of small capacity cells.
Transformer isolated charge transfer utilizes a bidirectional DCDC converter. A switch matrix connects the converter to a cell requiring balancing. Charge can be transferred between an individual cell and the entire group of cells (battery module) in either direction.
This method is much more efficient. Only two steps are needed for cell-to-cell transfer regardless of the location of the cells in the module. The balancing current magnitude is easily scalable by choosing transformer size and switches. This topology is more suitable for large automotive battery packs.
The number of cells that can be served by one converter is still limited to practically no more than 12 due to complexity of the switch matrix and in some topologies the maximum number of transformer windings.
Fig. 2 Transformer isolated balancing and module balancing examples. Forf full resolution, click here.
Module balancing
The EV battery packs contain up to several 100's of cells and are divided into modules. The need for module balancing is as important as the need for cell balancing due to module parameters mismatch, unequal temperatures affecting performance and causing uneven ageing or the need to replace bad modules with new ones for maintenance purposes.
Transferring of the charge between modules is not addressed by the previously described balancing methods.
One of the methods providing module balancing capability involves connecting the circuitry of each module to one cell of the adjacent module to provide a path for charge transfer. This method is inefficient because the charge has to be transferred to one cell and redistributed among all cells of the module. Transferring the charge to a far away module requires several steps and the efficiency drops further.
Monitoring
A key to superior BMS performance lies in the accuracy and speed of the cell voltage measurement.
The cell voltages are affected by the fast changing currents. It is critical to measure the entire pack cell voltages almost simultaneously to guarantee the precision of the state-of-charge (SOC) and state-of-health (SOH) estimation.
Active balancing BMS solution from National Semiconductor
The National Semiconductor's active balancing BMS is a system level solution optimized for large Li-Ion battery packs. It is a printed circuit board assembly product featuring a set of application specific ICs providing a complete active cell balancing, precise data acquisition, protection and robust battery management solution.
The National Semiconductor's BMS provides >90% efficiency, high current cell balancing within a battery module as well as between modules using optimized isolated inductive topology. Charge can be bi-directionally transferred between any cells in the module as well as between modules with minimum number of transfer steps. Cell and module balancing can occur simultaneously. The performance is optimized by intelligent control algorithms selecting the optimum balancing strategy. The system is modular and scalable. Individual boards manage battery modules of up to 14 cells with 5V max per cell. Up to 32 modules can be stacked to manage high-voltage battery packs. The balancing current is typically on the order of several Amperes and can be scaled up or down by the choice of external components providing the ability to optimize the performance/cost tradeoff.
Along with charge balancing, the BMS system provides a complete monitoring of the battery pack. Each battery cell voltage is measured with unparalleled precision. The 2mV accuracy of the National Semiconductor's analog front end (AFE) IC plays a critical role in charge balancing but it has even more significant impact on the accuracy of the SOC and SOH estimation.
All the cells in the battery pack are measured within a narrow time window. Measurements of all cells are synchronized throughout the entire battery pack.
To improve the SOC estimation and minimize energy losses, the sleep mode current consumption of the ICs and the board is minimized to enable achieving <100uA per module.
The BMS is equipped with several layers of diagnostic and fault detection systems. Faults including cell under voltage, over voltage, communications failure, open sense lines, and cell over temperature are detected and reported to the main controller. A separate redundant fault detection circuitry reports the faults independently of the main hardware and firmware channel. Parameters are compared against programmable thresholds in firmware and are also monitored by independent comparator based fault detectors.
Each board is equipped with a multi-drop galvanically isolated CAN Bus interface allowing high-speed communications with other modules and the main controller.
Extensive diagnostics, programmability and configurability over CAN bus are provided.
Precise balancing and accurate SOC estimation made possible by the BMS allows full utilization of the energy stored in the battery pack, provides extended driving range and eliminates the “range anxiety” by enabling reliable fuel gauging function.
The safety, reliability and cycle life of battery packs is greatly increased thanks to precise control of the cells SOC during charging, discharging and idle state enabled by the sophisticated features of the National Semiconductor’s BMS.
The BMS provides an advanced and effective solution for large automotive batteries and stationary grid storage arrays as well as small battery packs. The uniqueness of National Semiconductor’s solution comes from the fact that it was optimized for large battery packs and doesn’t suffer from the performance limitations of other BMS solutions burdened by the legacy of small pack applications.
REFERENCES
[1] S. Moore, P. Schneider, “A Review of Cell Equalization Methods for Lithium Ion and Lithium Polymer Battery Systems.”, SAE 2001.
[2]Pablo Cassani et all, “Status Review and Suitability Analysis of Cell-Equalization Techniques for Hybrid Electric Vehicle Energy Storage Systems”, IEEE PELS Newsletter, 2nd Quarter 2008.
Jack Marcinkowski is a Senior Marketing Manager at National Semiconductor, based in the company's Headquarters in Santa Clara, California/USA. He is currently focusing on battery management systems. Mr. Marcinkowski holds MSEE degree and MBA degree from University of Southern California Los Angeles. His experience and expertise includes engineering, management and marketing in areas of power and automotive electronics and semiconductor industries.
- No news
- In Formula One, Freescale is in the pole position
- Volvo evaluates flywheel hybrid drive - fuel savings of up to 25%
- Bosch tests automatic driving on the Autobahn
- Bosch highlights radar technology for safety-relevant driver assistant systems
- Toyota utilizes SPARK Pro programming language in ultra-low-defect software
- Bosch stresses high costs for lower fuel consumption
- TRW succeeds with electrical power steering system in China market
- Universal charger connects plug-in hybrids globally to the grid - as long as it is a Porsche
- Students build electric racing car
- MOST Forum sketches 5-Gbps future
This site contains articles under license from EETimes Group , a division of United Business Media LLC.
