Real-Time Battery Health Monitoring with Open-Source SDK
Introduction
A recent article on the DEV Community (source: DEV Community) discusses a method to recognize when a battery is only 73% healthy. This technique involves monitoring the voltage terminal of a battery, which decreases as it approaches full charge and then returns to its original state once it's fully charged.
While this method can help identify whether a cell is still functional or needs replacement, it doesn't provide an accurate assessment of the health of the battery based on instantaneous SoH values. The concept of SoH (State of Health) follows how much capacity a cell can still provide between being full and empty. As cells age, they lose their remaining capacity over time, while their SoH value decreases with time.
By using this method to monitor the SoH of a battery in real-time without relying on cloud models or complex calculations, it allows users to visualize battery health metrics without needing external servers or complex algorithms. The open-source SDK described here follows an Exponential Moving Average (EMA) algorithm entirely numerically, ensuring precision by filtering out noise and rejecting outliers that might indicate glitches rather than actual data.
This method requires only two key steps: when the cell reaches its maximum tension point, it declares "this is full" and sets the coulomb counter to the nominal capacity; slowly charging with each milliampere-hour released from the battery, reducing the counter gradually; and marking a cell as empty once it reaches its minimum tension. Using these measurements, you can determine the true usable capacity of your battery.
Additionally, this SDK includes a plausibility check that rejects all SoH values outside the 30-120% range of nominal capacity, ensuring accuracy by rejecting false positives. The process is entirely digital, using only ~200 bytes of flash memory and no static RAM. It integrates seamlessly with existing coulomb counters on devices.
This method allows you to visualize battery health in real-time without needing external servers or complex calculations. Even if you have limited resources, this open-source application can provide an interesting view into the efficiency and longevity of your device's performance.
Source: DEV Community