By Keysight Technologies
Whether you are an engineer in design or manufacturing, Li-Ion cell and battery performance testing is both a priority and a challenge for you. This is especially true for evaluating cells for self-discharge. Cells exhibiting high levels of self-discharge have higher likelihood of failure and must be sorted out and the cause identified. Unfortunately, this has traditionally been a long and tedious process to perform.
What is a cell’s self-discharge? Self-discharge of an electrical cell is the loss of charge over time while not connected to any load. Some amount of self-discharge is a normal attribute resulting from chemical reactions taking place within the cell. Compared to other types of rechargeable cell chemistries, lithium ion cells have rather low self-discharge. On their own they may typically lose about 0.5 to 1% of their charge per month.
Additional self-discharge can result from leakage current paths existing within the cell. Particulate contaminants and dendrite growths produce internal “micro-shorts”, creating such leakage current paths. These are not normal attributes and they can lead to catastrophic failure of the cell. Because of this it is a top priority during the design of the cell to eliminate possible causes of high self-discharge. In manufacturing, it’s critical to screen out any cells exhibiting abnormally high self-discharge as early as possible in the process.
Traditionally self-discharge is evaluated by measuring the decrease of a cell’s open-circuit voltage (OCV) over time. While it is not challenging to measure a cell’s OCV, the challenge is that it is very time-consuming. Because lithium ion cells have very little change in OCV as they discharge, it takes weeks to months to detect a significant loss of a cell’s state of charge (SOC), and to discern a good cell from one having high self-discharge.
An alternate means to determine a cell’s self-discharge is to instead measure its self-discharge current. When such a measurement is correctly implemented, cells exhibiting excessively high self-discharge can be identified and isolated in a small fraction of the time required by the traditional OCV approach. This mitigates the associated expenses, complexities and hazards of a large amount of work-in-progress (WIP). Now the challenge you are faced with is having a suitable solution for this task. Equipment possessing required stability and resolution has not previously existed to make it practical.
In this application note, details of these two methods for determining a cell’s self-discharge will be examined and compared, along with the approach Keysight has taken to create a cell performance test solution that greatly reduces measurement time.
Open-Circuit Voltage (OCV) Method
The open-circuit voltage (OCV) method for measuring a cell’s self-discharge is depicted in Figure 1. The self-discharge behavior is represented as a resistance, RSD, whose value is equal to the cell’s OCV divided by the self-discharge current ISD. ISD Is simply the rate of charge loss in coulombs per second. As would be expected, ISD is very small, typically a few to a few hundred microamperes, depending on cell size. As charge is slowly lost, the cell’s voltage very slowly drops. A high accuracy, high resolution voltmeter is required to measure the cell’s voltage loss over an extended period.
Looking at a representative example of a lithium ion cell’s discharge characteristics in Figure 2 it can be seen there is very little change of voltage over most of the range of the cell’s state of charge (SOC). It can also be seen that the voltage change varies considerably depending on the exact % SOC. For this cell by itself without any load, assuming a self-discharge of 1% per month equates to a voltage loss of about 3 to 12 mV per month, depending on its % SOC. The decrease of a cell’s OCV over time is an indirect and imprecise indicator of what the cell’s self-discharge…