Micromobility BMS architecture

According to the different positions of the charging FET (CFET) and the discharging FET (DFET), the BMS architecture of electric two-wheeled vehicles can be divided into the following four types: high-side series architecture/high-side parallel architecture/low-side series architecture/low-side parallel architecture ;The main difference lies in two points: one difference is whether CFET and DFET are placed on the high side or low side; the other difference is whether CFET and DFET are connected in series or in parallel.

1.1 High or Low-side

According to different application scenarios, an appropriate BMS architecture should be selected. The low-side solution is currently relatively mature and easy to implement. Most two-wheelers are also designed based on the low-side solution. At the same time, most current analog front-ends also integrate low-side driver capabilities.

However, there is a shortcoming in the low-side protection scheme: when CFET and DFET are turned off, the ground of the battery pack and the system end are no longer the same ground. Therefore, once protection is triggered to turn off the charge and discharge FETs, the battery end and system end are no longer connected to the same ground. Able to communicate directly. If you want to continue to communicate, you need to use isolated communication, which will not only increase the cost, but also increase the power consumption, especially during under-voltage protection. Excessive communication power consumption is even worse for the already under-voltage battery pack. Therefore, low-side solutions are mainly used in products that are more cost-sensitive and do not have complex communications.

Compared with low-side protection, in the high-side protection scheme, even after the protection is triggered, the battery pack and the system are still grounded, so they can still communicate with each other without increasing isolation communication, and they are disconnected after the protection is triggered. The positive terminal of the battery makes the system safer.

1.2 Series or Parallel

The charging port and discharging port of the series architecture share one port. The disadvantage is that the number of CFET and DFET needs to be selected according to the maximum charge and discharge current. If the difference between the charging current and the discharge current is relatively large, such as a general electric vehicle lithium battery pack The charging current is smaller than the discharging current. If you choose a series architecture, you need to choose more CFETs than actually needed, causing unnecessary waste. And whether it is charging or discharging, all current needs to pass through CFET and DFET, which will generate more losses and heat, and also reduce the effective capacity of the battery to a certain extent. The advantage is that there is no need to worry about reverse current, because the back-to-back connection of CFET and DFET can block reverse current. In addition, the series architecture saves one power line and one terminal block.

Compared with the series architecture, the parallel architecture can select the number and type of CFETs and DFETs according to the actual charge and discharge current needs. And whether it is charging or discharging, it only passes through the first-level FET, so there is less loss and heat. The disadvantage is that reverse current needs to be considered, such as flowing through the body diode of the CFET to the charging port, or flowing through the body diode of the DFET to the cell. To block these current paths, additional circuits are required to assist the implementation, and the parallel architecture requires an extra wire. The power cord and an extra port are not suitable for use in some situations.

Note: The series connection scheme is also called the same-port scheme, and the parallel scheme is called the different-port or split-port scheme.

Other two-wheeler BMS architecture

In addition to the above classification according to the location of CFET and DFET, the two-wheeler BMS architecture can also be classified according to the number of analog front ends, the presence or absence of MCU, etc.

According to the number of simulation front ends, two-wheeler BMS can be divided into cascade architecture and non-cascade architecture.

2.1 Cascade architecture

The current mainstream electric two-wheeled BMS, such as electric bicycles, scooters, balance bikes, etc., generally use 10S, 14S or 16S battery packs. Therefore, for the current mainstream electric two-wheeled BMS, the above single AFE solution can be used. Figure 1 ~ Figure 4 are all non-cascade architecture. However, for some applications that require relatively large power, such as electric light motorcycles or electric motorcycles, the voltage is usually higher than 60V, and a battery pack of more than 16 strings is needed to achieve greater power, and two AFEs are needed. Use in cascade, that is, adopt a cascade architecture.

2.2 Non-Cascade architecture

According to the presence or absence of MCU, two-wheeler BMS can be divided into independent architecture and non-independent architecture. Figure 1 ~ Figure 4 all work together with MCU, so they are all non-independent architectures. When the AFE works in independent mode, it can still monitor the battery status and control the charging and discharging FETs. When the protection condition is triggered, it automatically controls the FETs to turn off for protection. When the protection condition is removed, it automatically restores the FETs to turn on.

The advantage of the independent architecture is that it can save one MCU and is suitable for applications with more stringent cost requirements. However, due to the lack of MCU, there is a loss in flexibility. Users need to choose independent or non-independent BMS architecture according to actual needs.