BMS values are not shown correctly in log messages during startup

[hoschult]

Describe the bug During startup of serialbattery the values printed seems not to be correct. However, viewing cell voltage level via console seems to be right

To Reproduce Use Jbd bms and connect via UART to raspberry

Expected behavior right values to be shown - not sure if this is a general issue or only wrongly stated in log file

Screenshots

VenusOS (please complete the following information):

• Device type: Raspberry Pi
• Firmware Version 2.90 Venus os / serialbattery 0.141

Battery/BMS (please complete the following information):

• BMS/Battery type: JBD
• Cells: 16
• Interface: See above

[mr-manuel]

@iLeeeZi could you check, if you have the same behaviour with the nightly build?

[iLeeeZi]

2023-05-01 16:11:40.291232500 INFO:SerialBattery:Starting dbus-serialbattery
2023-05-01 16:11:40.292203500 INFO:SerialBattery:dbus-serialbattery v1.0.0-jkbms_ble (20230501)
2023-05-01 16:11:40.292652500 INFO:SerialBattery:Testing LltJbd
2023-05-01 16:11:40.454924500 INFO:SerialBattery:Connection established to LltJbd
2023-05-01 16:11:40.455442500 INFO:SerialBattery:Battery LLT/JBD connected to dbus from /dev/ttyUSB3
2023-05-01 16:11:40.455913500 INFO:SerialBattery:========== Settings ==========
2023-05-01 16:11:40.456426500 INFO:SerialBattery:> Connection voltage: NoneV | Current: NoneA | SoC: None%
2023-05-01 16:11:40.456909500 INFO:SerialBattery:> Cell count: None | Cells populated: 0
2023-05-01 16:11:40.457395500 INFO:SerialBattery:> LINEAR LIMITATION ENABLE: True
2023-05-01 16:11:40.457938500 INFO:SerialBattery:> MAX BATTERY CHARGE CURRENT: 75.0A | MAX BATTERY DISCHARGE CURRENT: 150.0A
2023-05-01 16:11:40.458411500 INFO:SerialBattery:> CVCM:     True
2023-05-01 16:11:40.458974500 INFO:SerialBattery:> MIN CELL VOLTAGE: 2.9V | MAX CELL VOLTAGE: 3.5V
2023-05-01 16:11:40.459685500 INFO:SerialBattery:> CCCM CV:  True  | DCCM CV:  True
2023-05-01 16:11:40.460210500 INFO:SerialBattery:> CCCM T:   True  | DCCM T:   True
2023-05-01 16:11:40.460751500 INFO:SerialBattery:> CCCM SOC: True  | DCCM SOC: True
2023-05-01 16:11:40.506500500 INFO:SerialBattery:DeviceInstance = 3
2023-05-01 16:11:40.507177500 INFO:SerialBattery:com.victronenergy.battery.ttyUSB3
2023-05-01 16:11:40.666449500 INFO:SerialBattery:publish config values = 0

Seems to be still the same

[mr-manuel]

Thanks, I will check it

[ramack]

While working on the Heltec BMS I currently see the same and I guess the output is written to the log, before the data is read by the corresponding methods.

[mr-manuel]

Later today I release a fix. This comes from, that some BMS already read the data with def test_connection(self): and others not.

[mr-manuel]

This is fixed with the latest jkbms_ble nightly for all BMS. Since I could not test all BMS, please leave a comment with your BMS name, if it's not fixed for you.

[iLeeeZi]

2023-05-02 14:33:28.460361500 INFO:SerialBattery:========== Settings ==========
2023-05-02 14:33:28.460874500 INFO:SerialBattery:> Connection voltage: 13.18V | Current: -2.12A | SoC: 55.893382352941174%
2023-05-02 14:33:28.461542500 INFO:SerialBattery:> Cell count: 4 | Cells populated: 4
2023-05-02 14:33:28.462006500 INFO:SerialBattery:> LINEAR LIMITATION ENABLE: True
2023-05-02 14:33:28.462489500 INFO:SerialBattery:> MAX BATTERY CHARGE CURRENT: 75.0A | MAX BATTERY DISCHARGE CURRENT: 150.0A
2023-05-02 14:33:28.463013500 INFO:SerialBattery:> MAX BATTERY CHARGE CURRENT: NoneA | MAX BATTERY DISCHARGE CURRENT: NoneA (read from BMS)

Works with JBD. SOC might need some rounding :smile:

[mr-manuel]

Ok, I added it. Will be in the next PR.

[iLeeeZi]

@mr-manuel Can you change this: https://github.com/Louisvdw/dbus-serialbattery/blob/jkbms_ble/etc/dbus-serialbattery/bms/lltjbd.py#L161

To

self.soc = round(100 * capacity_remain / capacity, 1)

Or

self.soc = round(100 * capacity_remain / capacity, 2)

To not lose improved resolution from #439

[mr-manuel]

Sure

[ramack]

Wouldn't it be best to round only on the output printing?

[mr-manuel]

There is no advantage in having more then one/two decimals in the SoC, since the SoC is already only an estimation and never exact in this case. The decimals are only for graph smoothing.