119
Dynamic Pulse Charging Scheme for Series Connected Cells

Tuesday, 15 May 2018: 08:40
Room 607 (Washington State Convention Center)
J. F. Stephens (FAMU-FSU College of Engineering), P. L. Moss (Florida A&M University - Florida State University), L. Morris Jr. (Department of Electrical and Computer Engineering), and M. H. Weatherspoon (Florida A&M University - Florida State University)
Imbalance of cells in a battery pack is a crucial issue that has numerous drawbacks on the performance of the pack. The disregard of balancing cells can lead to cells drifting apart from each other and a reduction in pack capacity [1]. Cell balancing strategies usually take part during charge because of variation in voltages and state of charge (SoC). The more conventional way of charging batteries in series requires a battery balancing strategy to achieve a more efficient full charge for each battery. In this study, we propose a current pulse charge topology that can charge batteries in a series connected stack independently. Improving the efficiency and performance of the stack by providing each battery with its own charge current, allowing batteries to charge up fully despite their SoC.

Methodology

This system uses a current source that precisely maintains a constant current injected into a switching network, 2N low resistance solid state switches are used to control the pulse current for each cell as seen in Figure 1. Where N is the total number of cells. During the charging process, the individual batteries in series are connected to the power and ground bus via a microcontroller or signal block, to receive a charge for a specified period. When a battery is fully charged, the signal block will detach the battery from the system and allow the others to continue charging. This configuration provides a means of isolation or bypassing capabilities for each cell in the stack without the need of a solid-state transformer (SST). The pulses are precisely controlled by varying the duty cycle for a specific switching period which is based on the average current represented by equation (1).

Eq (1)

Where Iavg is the average current, T is the period of the pulse and i(τ) is the pulse current induced in battery [2,3]. Determining the duty cycle for a fixed period can be derived from equation (2) and (3).

Eq (2)

And

Eq (3)

Where Δt is the duration the on time and D is the duty cycle. Using this approach constant current/constant voltage can be achieved by using pulse charging simply by varying the duty cycle. Furthermore, this approach allows each battery to be charged independently.

Results

To test this strategy, three 1.1Ah LiFePO4 batteries from A123 Systems were used. The desired average current that was induced into the batteries was obtained by applying a 1A maximum current pulse, with a period of 300 ms and a duty cycle of 33%, equating to 0.333A. Figure 2 validates the proposed topology’s effectiveness as a possible new way of charging series connected cells using pulse current by demonstrating each battery reaching its maximum potential for three distinct initial states of charge. Additionally, the switching network demonstrated isolation while charging with disconnect times (DTN) as shown.

Figure 1. Proposed charging circuit

Figure 2. Results for charging batteries at three different states of charge using the proposed charging circuit.

References

[1] Daowd, Mohamed, Noshin Omar, Peter Van Den Bossche, and Joeri Van Mierlo. "Passive and Active Battery Balancing Comparison Based on MATLAB Simulation." IEEE Xplore (2011). Print.

[2] [1] Beh, Hui Zhi, Grant A. Covic, and John T. Boys. "Effects of Pulse and DC Charging on Lithium Iron Phosphate (LiFePO4) Batteries." IEEE 315.(2013): 315-20. Print.

[3] Yin, Meng Di, Jeonghun Cho, and Daejin Park. "Pulsed-Based Fast Battery IoT Charger Using Dynamic Frequency and Duty Control Techniques Based on Multi-Sensing of Polarization Curve." Energies (2016): 1-20. Print