BMS Reader and Interface Box Deployment
Now that all the BMS units have been built up and the last ones are being installed, it was time to install the BMS Reader in the car. Shown in the first image below is where I mounted the BMS Reader in the trunk, in front of the trunk battery box. I installed the BMS Reader there because the 12V battery is in the trunk and the first BMS units to be deployed are in the trunk battery box. To be able to connect all the BMS units to the Reader and have +12V and GND connected to the BMS units I had to build the Interface Box. Below the image of the BMS Reader is the Interface Box. The connection design was detailed in a blog from 08/2023 when I first started designing the deployment of the BMS units (link). The actual layout of the BMS units is not like the drawing shown in the image. Instead of connecting each of the BMS units to the Interface Box, the BMS units are connected together in parallel with M12 tees and one connection to the Interface Box is made. Only one Interface box is used, with the two parallel strings of BMS units connecting to it.
The M12 tee connectors where first described in the blog on deploying the BMS units in the trunk battery box (link)and recently the blog on deploying the BMS units on the rear seat batteries (link). The image of the BMS connections from that blog showing the parallel connection is shown again here, below the Interface Box image (jump). That image is the way the BMS units will be arranged in the rear seat. The four rear seat BMS and the four trunk BMS units are connected together in parallel and each of those parallel strings are connected to the BMS Reader, via the Interface Box. They are all wired to create the correct topography for CAN BUS as detailed in this article (link). The topography diagram from that article is shown at the bottom of this blog. For the BMS interface the two BMS units at the end of each parallel string have the 120 ohm termination resistors soldered on the BMS PCA. Each node on my BMS system has a CAN Transceiver and Controller.
UPDATE: BMS measurements on 6-5-2026
BMS Reader deployment in the trunk.
Interface Box deployment in trunk.
The two cable connections on the top are from the the two sets of batteries and the cable connection on the right is to the BMS Reader. The box is switched because it is directly connected to the battery and the BMS Reader and BMS units do not need to be on at all times. The power for the box is fused. Nine BMS units plus the BMS Reader draw about 500mA at 13V. All the BMS units and the BMS reader individually draw about 50mA, so ten connected together draw 500mA. Eventually I will make a cable connection to the LEVID controller for the Instrument Cluster, so the battery voltages can be displayed on the 7-segment display.
Layout
BMS parallel connection using M12 tees.
CAN Topography Diagram
UPDATE: Above is the average cell voltage for 8 of the battery modules, measured recently with the new BMS system. Compare the plots shown in the blog from 5-3-2023. The batteries are at a different state of charge (SOC) but the cell voltage distribution is still just as small, less than +/- 2mV. No adjustment of the individual battery modules was done. This is after over three years of charge and discharge cycles since the data presented on 5-3-2023 and now is nearly 6 years of using the batteries without any adjustment.
One issue that has not been highlighted since I made the blog post about the problem, is that every battery, except one, has one or more cells not reading by the BMS. This is caused by a problem in the connection from the battery cells to the BMS connectors on the batteries, that was detailed in the blog from 10-27-2020. The images from that blog are shown below.
Unfortunately as detailed in that blog, the only way I found to fix this issue is to add wires to connect to the cells not reading by the BMS and that can only be done with the battery on a workbench. No way to accomplish that with the battery in the car. I found that it really is not clear where the problem lies. The soldering to the BMS PCB that connects the battery cells to the BMS connector look fine. It could be the PCB has some defect or the connector to the BMS on the end of the battery is defective. What is very strange is that the PCB does not have a conformal coating on it to protect the PCB. That is usually done on any automotive PCB because it protects the PCB from damage and corrosion.
Update 6/13/2026
Since I had all the hardware to test both the BMS units and batteries I thought would test the battery that has been sitting on my bench. Below is a screen shot of software I developed to test my BMS units (see blog 04-26-2026). The code graphically displays the battery voltage of each cell. The software does a lot of other calculations but what is most striking about this image is that cells 6 and 7 are measuring the wrong voltage. Cell 7 is not even measuring a voltage. This problem is exactly due to what I describe above - there is some broken connection in this battery between the battery cells and the BMS connection. The open connection also causes the next cell to read wrong because the cells are wired in series and the way the BMS IC circuit is designed. What is troubling is that this battery has never been in my car. It has been sitting on the bench for the past 6 years. This particular battery was found to have the first and second cell not returning the correct voltage, when it was first received and tested. I repaired those connections over 5 years ago. Now a new set of cell connections are broken. The battery has only been exposed to the temperature environment of the garage. But that temperature does not vary widely. The garage is heated in the winter and cooled in the summer. Not anywhere near the temperature variation a battery in a car parked outside would see.
