I ran into an amazing opportunity to add a lot of capacity to my solar system. There is a local company that does EV conversions to classic cars. Apparently a customer had an issue and they replaced the entire battery bank which left around 75 100Ah 3.6V cells as surplus. These are the same cells that are in my 6 LiFePO4 Lisuateli batteries. So I made a deal to pick up the whole lot.
The cells had been sitting a the bottom of the state of charge so I was concerned if they would fully recover. First thing I did was to charge up 4 of the cells making a 12V battery. I used my HP variable power supply to charge the cells in parallel to 3.65V. The cells took the charge and after doing a discharge test it appeared the cells were in good condition. The discharge curve is very flat and came out as I had expected.
The charge/discharge test took 3 days. While that was going on I was also working on a prototype compression system. The wood allows for mounting the BMS to the side. The plan is to build 9 24V 100Ah batteries. I believe this design will be time and cost consuming so I’m not sure ultimately this is how I will go. Whatever the end result there needs to be some compression on the cells to keep them from bloating.
The next thing to do was to add a BMS. I decided to go with the very popular JK B2A8S20P. This BMS can charge/discharge at 200A, has a 2A active balancer and has Bluetooth and RS485 ports for data. It can be configured for 12 or 24V.
I had an extra Raspberry Pi so I loaded VenusOS and now I have both 12 and 24V systems. I’ll continue to use the 12V setup for testing of cell groups. I was even able to get the battery to connect to the system via Bluetooth. Something the JK BMS seems to do better than the JBD BMS in my other batteries. However it’s not perfect and I’m waiting on an interface box from JK to allow it to physically plug into the Pi. This will allow me to monitor all the aspects of the battery when out of Bluetooth range.
Finally, I put together a 24V battery and added it into the existing system. It performed as expected and I’m excited to get a few more built. However the cost for more BMSes, wire, etc. is kind of prohibitive. So I may try to parallel a second set of cells. This would make an 8S2P (16 cells) configuration. I’m waiting on another BMS, bus bars and other items to get this going.
The system is really shaping up. There is still a lot to do, but it’s been a great experience and I now have a system that in a power outage could run the refrigerators and power the well without having to fire up the generator. Looking forward to getting more backup power stored! As always there will be more to come.
In the meantime if you are interested in how the BMS controls the cells, Andy from the Off Grid Garage has an excellent series of videos using the same BMS starting here