042-043_EETE-VF

EETE SEPT 2013

FUEL CELLS & ENERGY STORAGE Battery lab testing and the limits of the datasheet By Achim Loesch Unlike many other kinds of electronic component, a battery pack is highly sensitive to changes in operating conditions. In most cases, the battery specifications that design engineers pay close attention to include cycle life, capacity and peak power output. The values of all these parameters vary with variations in operating conditions such as ambient temperature, discharge profile and charging rate. As this article shows, the engineer can infer a certain amount of information about the expected performance of a battery system in a real application from specifications contained in the cell’s datasheet. But in many cases the actual usage conditions will deviate substantially from the ideal conditions specified in a cell datasheet. This means that, if a design engineer Laboratory testing is an excellent and helpful complement to the design process needs to be certain about the way a battery pack will perform in the field, additional methods of measuring or calculating expected performance will be required. The value of the datasheet When designing or specifying a battery pack, components such as the charging circuit, protection circuit and housing all have a measurable effect on the performance of the system. But the primary limiting factor on the electrical performance of the battery pack is the performance of the cell or cells. This is why the datasheet of a cell is – or at least should be – an important resource for the battery pack designer. In practice, however, a typical cell datasheet tells the engineer about the performance of the cell only under a specific set of conditions. Let us take as an example a widely used product, the NCR18650, a cylindrical lithium-ion cell supplied by Panasonic. The product’s datasheet (dated February 2010) specifies a minimum nominal capacity of 2,750mAh. This specification, however, only applies at a constant discharge current of 550mA and at a temperature of 25°C. The same datasheet provides discharge curves, showing the voltage and total power output at three discharge rates – but again at constant current and 25°C. The effect of variations in temperature on output voltage and discharge capacity is shown in a separate graph, but now only at a single – again constant – output current. Cycle life – the number of charge/discharge cycles that the cell can sustain – is also shown, but again at a single constant-current output and at a constant 25°C. These data are useful for comparing one cell with another, and making a relative judgement about different cells’ performance. For instance, if the designer judges that discharge capacity is an important parameter in an application that will normally operate at an average ambient temperature of 5°C, a comparison showing that cell A has a greater discharge capacity than cell B at both 0°C and 25°C is useful, and suggests that cell A will probably also have a greater capacity at 5°C. This kind of relative judgement is important in component selection. But to know how the selected cell will actually perform in an end product, absolute performance data are required – and here, the designer runs up against the limits of the datasheet. True, by applying common sense the engineer can extrapolate a certain amount from the datasheet specifications. Such extrapolation is not, however, normally supported by de-rating models supplied by the cell manufacturer. It is impossible in any case to extrapolate from datasheet specifications for usage conditions that vary markedly from the datasheet conditions. And in practice, this will apply to many different kinds of end product. For example, a pedelec (electrically assisted bicycle) typically has a discharge profile that is completely different from the datasheet’s neat, constant-current output. On a trip through hilly terrain, the rider might draw the peak power output when climbing a hill, then switch off the electric motor as the pedelec freewheels downhill, in a repeating pattern of high-discharge/ zero-discharge episodes. In many cases, the temperature inside the battery pack’s housing will be much higher than 25°C, as high currents flow through the power circuit and generate waste heat. Equally, however, the pedelec might also need to be rated for operation in cold northern climates in which operation and storage at temperatures far below 25°C can be expected. Clearly, for the designer of a pedelec a cell’s datasheet specifications showing performance at a constant current and a constant mild temperature are not adequate as an indication of the performance of the battery pack in the field. The effect of abuse on a battery Other conditions are, in any case, explicitly not specified in the datasheet of the cell or any other component of the battery pack. Abuse takes many forms, but its effects can be hard to predict. Manufacturers of portable data loggers, for instance, know that their devices will occasionally be accidentally dropped by users, normally from around waist-high to the ground. Other abuse conditions suffered by battery-powered devices include over- and under-temperature, use of nonapproved chargers, and vibration. A battery pack can potentially fail for mechanical or electrical reasons when subject to abuse. But how much abuse leads to failure? How is the data logger manufacturer to know whether the battery can withstand multiple drops? Achim Loesch is General Manager for the Power Pack Solutions at Varta Microbattery - www.varta-microbattery.com 42 Electronic Engineering Times Europe September 2013 www.electronics-eetimes.com


EETE SEPT 2013
To see the actual publication please follow the link above