We have a lot of sites all across the country, so this one was out of range for me to get to but I did assist another of our technicians with rectifying this site.
This particular set up is a Deye 50kW inverter with 2 separate battery stacks running in dual channel on the inverter (this particular inverter has 2 battery channels, each limited to 50A. Using both of them allows a max of 100A on the inverter). However the inverter was periodically giving out an F58 (BMS Communication Fault), while this was never interpreted as a critical fault (the inverter stayed online), the one battery did disengage from the system (likely the battery disengaging from the system causing the fault).
I spent some time analysing data from the app, and as we can see below, both from the graph as well as the spreadsheet, that we’re not getting communication from the second batteries BMS.
It’s important to note that if you’re running dual channel on your batteries, the communication link from each set of batteries needs to run in dual channel as well (ie battery 1 needs to be connected to battery 1 with its BMS link on BMS 1, battery 2 needs to be connected to battery 2 with its BMS link on BMS 2). The 2 battery channels on the inverter are handled independently.
So it’s evident in this case that only BMS 1 is connected to the inverter, but even more strangely, it somehow seems to know what is actually happening on battery 2’s BMS (as you can see by the spreadsheet).
With our technician on site, I had him check all the battery communication cable on site. Below image shows the initial suspicion was correct with only 1 BMS cable plugged in to BMS 1, as well as the BMS information pages where BMS 2 is not getting any information.
The reason why the inverter was somehow still getting information from both batteries is that the two batteries communication links were running in parallel to only the one batteries port. With the two battery channels runnning independently this was causing issues where one of the battery banks was being overcharged (as the inverter never really knew what was happening with that bank), this was also presenting the F58 fault on the Deye inverter.
The next images below show that we connected a second cable from battery bank 2’s BMS to BMS 2 as well as the new updated BMS info screens on the inverter.
With the inverter now running in true dual channel (battery power and BMS communication in dual channel), it knows exactly what is happening with each battery bank and properly manage its charging on each independent channel.