I looked for a way to get the cell voltages into HA and had to put in some effort to get the code working for my model BMS. I am running on a PiZero with wireless and Bluetooth. The code fetches the values over bluetooth and send them to my MQTT server. Maybe this can help someone.
And from studying the image, cell7 has issues. The I made some DIY busbars and tapped some screws into them and battery 7’s screw is stripped so it is not torqued similar to the others. This graph confirmed to me that I need to put some effort into fixing the connection
Well done!
I have been logging my BMS data (also via a Pi-Zero but not BT) to Grafana and it is well worth the effort to set it up.
Here is a couple of Graphs that I find quite useful
Daily Delta Deviation ( The difference between the Maximum and Minimum Delta for each day)
You can quickly pick up a trend if one of the cells starts misbehaving
The result of me fixing the cell 7 connection as mentioned above. To the left side on the graph the wild swings of cell 7(light green) reported more than 3.65V when charging with 60A. Which seems to stop the inverter’s charging. I also checked and the connection was slightly warm to the touch.
Towards the right of the graph cell 7’s connection was fixed and never reported above 3.44V. Interesting to note that this cells experienced daily “balancing” as it reported a high voltage and is clearly the cell with the most voltage sag. Clearly it is lacking charge compared to the other cells. The hot connection is also cool to the touch so the connection is fixed.
The issue of the reported SOC jumping to 100% is also gone. Today my battery pack charged without misreporting the SOC. For anyone finding the Daly BMS jumps to 100% this is likely the issue. A poor busbar connection causing voltage drop resulting in elevated ghost voltage readings. Once the BMS sees 3.6V it resets the SOC smart shunt to 100%.
I hope this post can help someone with their DIY battery
