There has been a great deal of speculation in this forum about why some people seem to be having trouble with dying 12 volt batteries. This is true of both the hybrid Pacificas and the gas versions (especially the ones with the start/stop feature). In the thread named "Dead battery adventures" there was some discussion of adding a datalogger to monitor the 12 volt battery.
installed a pair of commercial voltage and current meters on his battery, one to measure the charging current and one for the discharging current. But unless he stood there and took notes he wasn't able to see what was going on over time, especially when he was driving the van
took his van into his shop and tested the battery with an automatic battery monitor and was able to post a graph of the current and voltage over a period of a couple of hours.
In my case I wanted to find out more and I thought this would be a good opportunity to combine my skills in computer software (I am a retired software developer) and a chance to add to my rather skimpy knowledge of electronics. So I decided to build a battery monitor that would record what was going on with the 12 volt battery in my hybrid Pacifica over a long period. These were my goals:
1. Collect minute by minute observations of voltage and charge/discharge current over a period of weeks or months.
2. There should be no modification of the van's wiring. Everything should be easily disconnected to avoid having the dealer pointing accusing fingers at me.
3. The monitor should not present any load on the 12 volt battery. It should be self powered, except possibly when the van is running when it would be acceptable to draw a small amount of power in order to recharge the monitoring device's battery. More on this later.
4. The monitor should periodically transmit its collected data from the van to my home computer whenever it is parked in the garage. When it is away from home it should record the data on a data card.
After several months I think I have succeeded. It took that long mostly because of my lack of electronics expertise that I previously mentioned. I knew that I needed a small low power microprocessor, a real-time clock, an SD card interface for the data and a wi-fi module for transmitting the data to my home computer. I quickly settled on an Adafruit Feather Huzzah with ESP8266 microprocessor board https://www.adafruit.com/product/2821
. That took care of the microprocessor and the wi-fi. To that I added a RTC and SD add-on board https://www.adafruit.com/product/2922
. This stacks on top of the Feather Huzzah and takes care of the real-time clock and the SD card.
But I was stumped on how to actually sense the voltage of the 12v battery and the current going in and out. The voltage was easy but the current was more difficult. I knew that Chrysler had kindly put a battery sensor of their own on the negative post of the battery and that I could use the shunt that is a part of that to sense the current. I measured the resistance of the shunt by placing a load of several automotive lightbulbs on it and measuring the voltage across the shunt. I came up with a shunt resistance value of 0.00015 ohms. That is pretty small and would require detecting very small voltages, on the order of a few millivolts, or even less if I wanted to be able to measure the current draw when the van was sitting idle. I spent quite a bit of time trying to devise a circuit involving op-amps that could amplify this small voltage into something my microprocessor could read. I think this can be done, but not by me. I then discovered by chance that what I needed was readily available in the form of an INA226 current sensor chip on a breakout board. Just $10 from Amazon: https://www.amazon.com/Bolsen-CJMCU-.../dp/B07HQQ1SKY
The INA226 chip will detect current in both directions by measuring the voltage across the shunt. It will also measure the voltage of the 12 volt battery too. It even has an I2C interface for connecting to my microprocessor.
So what I have come up with is a small monitoring module (see the first picture) that sits in the compartment in the back left, just above the 12 volt battery where the emergency tire pump lives (see the second photo). The monitor is connected to the 12 volt battery by three wires, one to the positive terminal and one to each side of the shunt that is clamped to the negative terminal (see the third picture). Another cable runs from the monitoring module down under the carpet and across to the USB power outlet on the right side of the third row seats. This was supposed to provide power to recharge the small LiPo battery in the monitoring module but it turned out that the cable interferes with my measurements so I am not using it for the moment. As a result the monitoring module is entirely self-powered and is only connected to the van by the sensing wires.
What the monitoring module does is spend most of its time asleep in a low power state. Once a minute it wakes up and takes a snapshot of the 12 volt battery's voltage and the current going in or out at that instant. Once an hour, on the hour, it saves its data to the SD memory card. Also on the hour it tries to connect to my home computer over the wi-fi network and upload the contents of the SD memory card to a program that runs on my desktop computer 24 hours a day. If it can't connect because the van is not at home then it just keeps saving the data on the SD card. If it can connect then the data gets saved on my computer in the form of comma delimited text files that are readily readable by Excel for the purpose of analysis and charting.
For an example of the charting that can be done, see the fourth picture. This represents 4 hours in the life of the battery. The blue line is the voltage of the battery. The green bars represent charging current. The red bars are discharging current. It starts with the van parked and idle, with a small current draw of around 20 milliamps. That is right at the lower limit of what the monitoring module can measure. At around 2:20pm I started out on a shopping expedition. As soon as I started the van it immediately began charging the 12 volt battery at a rate of around 18 amps. This was not surprising since I had spent several hours the previous day attaching assorted loads to the 12 volt battery and measuring the current. At 2:40 pm I made my first stop. You can see the voltage drop as the van shuts off and the battery is no longer getting charged. But the van does not go completely idle right away -- it continues to draw a current of around 4 amps. I am not sure what for but I can definitely say that it does take a while for the van to go fully asleep. This pattern goes on for three more driving sessions followed by stops. The battery continues to be charged while driving but at a declining rate. At 4pm I return home, stop the van and immediately plug in the level 2 charger. For the next hour and twenty minutes you can see that the 12 volt battery is being charged at the same time as the motive battery. The 12v battery charging rate continues to taper off, ending up at around 0.5 amps. At 5:23 pm the level 2 charger finishes its job and the charging of the 12v battery also stops. But the van does not go idle yet. It continues to draw current from the 12v battery for another 10 or 15 minutes until it finally goes back to its idle current draw of 20 milliamps at 5:34pm.