Usually, a battery pack has at least one Battery Control Unit (BCU) to measure and to control currents, voltages, and temperatures of the cells or the modules in the battery pack. Normally, the BCU works when specified conditions are detected, then, safety devices work. Therefore, the cell does not go into abnormal use conditions. On the other hand, the abnormal events caused by an internal short circuit arise from the cell itself. Therefore, if an abnormal event is detected, the BCU will not be able to control the event, and the internal short circuit will continue, which could result in smoke or a fire.

IEC 62660-3¹⁾ states that forced internal short circuit (FISC) test should be conducted to evaluate the safety of the internal short circuit failure of lithium-ion cells for automobile applications. As one of the procedures for the FISC test, disassembling of the cell is necessary. And, some countries pointed out that conducting the FISC test is difficult because of both safety concerns and technical issues. Therefore, ceramic nail and ceramic nail with Ni tip tests are proposed in the IEC/TR 62660-4²⁾ as the alternative tests for conducting the FISC test easier and safer. Here, cylindrical cells are not included in the TR. However, the ceramic nail testing has an issue that the nail descending movement is stopped when an internal short circuit occurs and the nail is kept at that position till the end of the test, and that the cell swelling can occur after the internal short circuit of laminate cells. The swelling could stumble into multiple layer internal short circuits that are larger than the FISC test.

Therefore, in this study, the purpose is to evaluate an alternative test method which is easier to conduct than the FISC test and which can reject the effect of the cell swelling of laminate cells. We conducted the FISC test and developed a new alternative test, release a ceramic nail with Ni tip 30 seconds after the internal short occurred, to evaluate the validity of the alternative test.

The FISC test was conducted in compliance with IEC 62660-3. Two ceramic nail tests were conducted using a nail with a 0.3 mm Ni tip which has a diameter of 3 mm and angular degree of 45º. One is to keep the nail in place during the test, the other is to release the nail by an air chuck 30 seconds after the internal short circuit. Here, the 30 seconds is exactly the same as the pressing time in the FISC test. As evaluation candidates, six types of single cells, which have a capacity of approximately 2 to 10 Ah, were used. The tests were conducted at least 3 times for each cell.

The descending speed of the indenter or the nail was set at 0.05 mm/s, the stop condition was when over 2 mV drop was detected, measurement frequency for voltage was 1 kHz, the measurement time was set for 1 hour after the voltage drop is detected. The cell surface temperature, ambient temperature, nail descending speed, and the load of the nail were also measured. During the test, the events were investigated. After the test, the cells were completely discharged, disassembled, then, the number of the internal short circuit layers was counted.

The average numbers of short circuit layers are shown in Fig. 1. All of the events for the FISC tests are heat generation. The average number of the internal short circuit layers is around one layer, and it is confirmed that there is almost no differences in the number of the internal short circuit layers in cell types. In case of the nail released method, the events are almost heat generation. The average number of the internal short circuit layers and the dispersion of laminate and non-laminate cells are about one layer larger than the FISC ones. On the other hand, in case of the nail held method, the events are heat generation. The average number of the internal short circuit layers and dispersions for laminate cells in the held method are about three layers larger than the FISC ones, and those for non-laminate cells are one layer larger.

Therefore, the previous nail test can be improved for laminate cells by applying the new method of releasing indenter.

Acknowledgment

This paper is based on results obtained from a project commissioned by the New Energy and Industrial Technology Development Organization (NEDO).

References

1) IEC 62660-3 (2016)

2) IEC/TR 62660-4 (2017)