Using SEPLOS V3 CVL in config.ini seems to be ignored #1056
Comments
HF-SPSler
changed the title
|
|
Hello Manuel, first, thank you for your fast response. From the way you answered me, it´s obvisously not the whole config file :) I changed the posted config.ini and the actual log-file. I´m sorry for not reading your instructions to post a log file completely. For the first logfile I used Second command shows this: Yes I will try to reinstall the driver, maybe it´s a problem with the nightly build dev version I installed? Can you recommend a version of the dbus.serialbattery driver for use with the SEPLOS V3? Can you give me any hint which communication protocoll has to be chosen in SEPLOS BMS studio config? Which kind of changes do you mean, changes in code or config.ini only? Because, in code I did not change anything. Before I connected the new SEPLOS V3 Battery I used the whole system with a 5kWh DIY battery with JK BMS. So basicly the system was able to run very satisfying. Thank you very much in advance. |
|
@marcelrv maybe you know why this is the case that CVL is not the calculated one? |
|
In the seplos v3 integration I update the max battery voltage based on your log for reference, I have a very ' simple' config.ini, where the voltage prob comes from the config.default.ini as default config has 55.20 this is what I see in the screen
Maybe go first with a simple config and then 'enhancing it' may help narrowing down the issue. |
If I recall right, the protocol you set in BMS Studio refers to the canbus protocol the BMS is using. |
|
@marcelrv thank you for helping out. @HF-SPSler comment out these lines and it should work: dbus-serialbattery/etc/dbus-serialbattery/bms/seplosv3.py Lines 210 to 211 in 59a6d00 |
|
Hello together, @marcelrv The batterypack is connected with one of the RS485 ports to a isolated RS485 to USB converter. The CAN port (on the very left side) is unused. This is because I want to gain some experiences regarding the behavior of cell voltages while charging/discharging and some other things. As far as I know, this is unfortunately not possible via CAN. Today early in the monring there was some time to connect directly to the battery via BMS Studio. I found out, that for CVL in VenusOS the following parameter from the battery is used:
After changing this value from 57,6V to 56,8V the value of CVL in VenusOS changes too: In my opionion, this should be changed in your code, shouldn´t it? Actually I still need a recommendation from you, which version of the driver is the best for using a SEPLOS V3 BMS? Kind regards and many thanks! |
|
Aleays the latest version is recommended.
Yes, marcelrv wrote this driver and I missed it. Please try what I wrote above. |
|
Okay, I will try to comment out the lines you posted above today in the evening. |
|
if the self.max_battery_voltage is representing the charge voltage request, then indeed the code should be changed to use parameter 102
|
|
But is the BMS adapting the CVL or is it fixed to this value? In case we could add a BMS specific setting where you can select to use the parameters of the BMS or of the driver. Default would lead to the driver values to reduce confusion. |
I don't really know. I have not looked that detailed in the behaviour of the BMS. (screenshot from the CAN bus interface, which is reporting the BMS CVL) |
|
Further informations how you can (static) set CVL, CCL, DCL via BMS.. but thats should not the usecase while using the dynamic settings of the serialbattery-driver. https://docs.google.com/spreadsheets/d/1E7zX6uXqczvpEEJKD8jER31LWkNqWMgvUytVCVfMvlg/edit#gid=0 thxs to andy @offgridgarage |
|
I think the real question is @mr-manuel : should a BMS implentation set the self.max_battery_voltage. (most implementations do set this value) |
|
In my opinion, sorry if it´s sound hard, the BMS has not to tell anyone a maximum voltage request. I suggest that only one part in a system gives setpoints and this part is in my eyes the driver, or for example the parameters in a MPPT charger if there is no VenusOS/Cerbo. Or am I wrong? I can only display my point of view but I am truely no professional :) On the other hand, you can make it optional in config.ini as @mr-manuel suggested it:
So, should I comment out the two lines which are marked above or better not at this moment? |
|
Yes, comment out the two lines. I have already fixed it in the code, but not uploaded it yet. Have to wait to finish some other tests before. |
|
Seems that I´m no linux specialist or too stupid to comment out correctly... every time I try to comment out, the BMS connection to VenusOS failed and I get the Error 67 in Venus Remote Console. It does not matter if I type in #self.max_battery_voltage = spa[0x05] / 100 with a space between # and tags.... Sorry for that :( |
|
I think you have to replace the two lines with something like this? |
|
Putting a |
Describe the problem
Hello everybody,
since two days I´m testing the dbus-serialbattery driver in use with a SEPLOS V3 BMS. Before that, more than one of a year there was a JK BMS connected and used successfully. With SEPLOS V3 charging, discharging and so on wokrs fine but there is one issue with CVL value in remote console. In config.ini I set the following values (which worked fine with the last used JK BMS):
MAX_CELL_VOLTAGE = 3.460
FLOAT_CELL_VOLTAGE = 3.375
Although the max_cell_voltage is set to 16 x 3.46 = 55,36V, the venus remote console shows a value of 57.6V which means that die parameter in config.ini might be ignored? I tried to find out why this happened, but I can´t find the reason. I would be very happy about any help, tanks.
Driver version
1.3.20240428dev
Venus OS device type
Raspberry Pi
Venus OS version
3.20
BMS type
Seplos
Cell count
16
Battery count
1
Connection type
Serial USB adapter to RS485
Config file
Relevant log output
Any other information that may be helpful
No response
The text was updated successfully, but these errors were encountered: