sfuphantom / bms-fw

Battery Management System - Firmware written for TMS570LS1227
3 stars 1 forks source link

BMS Slaves Test Setup #63

Closed danrobren closed 5 months ago

danrobren commented 5 months ago

The BMS Slave chip is the LTC6811.

LTC6811 Data Sheet: https://www.analog.com/media/en/technical-documentation/data-sheets/LTC6811-1-6811-2.pdf

The battery is 96S 5P lithium ion. This means there are 96 cells in series and 5 in parallel. Therefore the battery votlage is 96 times the individual cell voltage and the capacity in amp*hours (Ah) is five times that of an individual cell. The cells we use have a minimum votlage of 3.1 volts when fully empty and 4.2 volts when full. Therefore the battery is 297.6 V empty and 403.2 V fully charged.

We cannot allow the cells to go above or below these maximum/mimimum voltages for safety reasons. It is the BMS master's job to measure the voltage of the entire battery, but the BMS slaves measure the votlage of the individual cells. Not every cell of the 96S 5P battery is being measured; each slave channel will connect to a bank of 5 cells in parallel and measure their temperature and voltage collectively.

Not every cell has the same internal resistance due to manufacturing variance, therefore they will discharge at different rates. Over multiple charge cycles, this effect can compound such that cell voltages are very different from eachother, even though the total battey voltage appears normal.

This is why we do balance charging. Every charge cycle, it is the BMS slaves' job to make sure all the cell voltages are equal. The BMS slaves report these cell votlages (and temperatures) back to the BMS master such that it can make decisions about whether to trigger an HV shutdown.

The BMS salve board is not finished as of 23/05/2024, but its functionality can be emulated much like the ADS7044 ADC which measures the total battery voltage This ADC communicates over SPI and its functionality was emulated using an SPI module on the launchpad itself. The goal is to do the same for the BMS slaves such that the code can be tested and ready for integration when the BMS slave board is complete.

danrobren commented 5 months ago

General Notes from the Datasheet

danrobren commented 5 months ago

SPI Interface Specification Altium Schematic:https://americas-team-phantom-simon-fraser-university.365.altium.com/designs/7BDB24F0-FA72-44F9-A03A-2E26A19A2FF2?variant=[No+Variations]&activeDocumentId=BMS_SLAVE_SHEET.SchDoc#design

danrobren commented 5 months ago

Questions for Faizan (BMS Slave PCB Designer)

  1. LTC6811-1_Daisy_Chain Why does the
danrobren commented 5 months ago

A new Accumulator Box team member was hired to replace Faizan, who is MIA. The new member was to take over development of the BMS slave PCBs. Through some discussions with Tractive System team, it was determined that the previous slave chip and board will work. BMS is switching back from the LTC6811 to the BQ76PL455A. The next step is to determine how much progress had been made on the BQ76PL455A firmware.