The latest developments in this battery technology focus primarily on two areas: higher float voltages and higher battery capacities. Both of these industry trends attempt to address the increasing need for longer system operation at a time when the convergence (integration of multiple functions on one device) and performance of everyday consumer gadgets is draining batteries faster than ever.

Advancements in battery technology over the last few years have allowed a 10% average annual increase in capacity for a given battery pack form-factor. At the same time, batteries with higher capacities in absolute terms have also entered the market to enable acceptable system run-times in the myriad of new, high-complexity handheld devices that have entered our lives over the last decade. Having more energy available to the system is great news for the consumer, since it prolongs the device's usable operation. However, batteries that provide more energy to the system also require more energy to be fully charged. With traditional charging methods, this translates to significantly longer battery charging times, a non-desirable feature for the consumer market. Let's examine where this limitation is coming from.

The rate of charge or discharge of a battery is commonly expressed in relation to the actual battery capacity. The term used in the industry for describing charge or discharge current is the C-rate. For example, for a 1000mAh battery, a charge current of 1C (1000mA) will theoretically charge the battery in one hour.

According to most battery manufacturers, a typical battery is safely charged at a rate of 0.7C to 1.2C during the fast-charge phase. This is the charging phase at which the battery voltage is high enough (usually higher than 2.8V or 3.0V) for it to accept a high charge current level (see Figure 1). When the battery voltage is below the so-called pre-conditioning level, only a fraction of the charge current (typically 0.1C) should be provided to the cell for a safe replenishment of the deeply-discharged cell. Hence, in theory, as soon as battery voltage is higher than approximately 3.0V, system engineers would want to provide the desired 0.7C to 1.2C rate, which safely charges the battery in the shortest time possible. In practice, achieving this "optimization" poses a great challenge and — as described in the following sections — accomplishing this task requires that traditional charging circuit design undergo radical changes.

Figure 1: Typical Li-Ion charging profile

Linear chargers
In the vast majority of today's applications, which utilize single-cell Li-Ion or Li-Polymer battery packs, the preferred type of charging is the linear mode one. Similar to linear DC/DC regulators, linear battery chargers provide an easy way of doing the job; they require a minimum number of external components and allow for easy design implementation given their simple mode of operation. As battery packs with a capacity of 800mAh or more are widely adopted in hand-held consumer devices, as is the case today, linear battery chargers are unable to provide the short charging times demanded by system designers and consumers.

One of the drawbacks of linear chargers with integrated power elements is the prohibitive high level of power dissipated when charge current is delivered to the battery. This is especially true when the device has transitioned from the pre-charge to the fast-charge phase. At that point, the P-channel pass transistor has to dissipate the maximum power because of the high input-to-output voltage differential (see Figure 1). The first linear charging solutions on the market required the use of heat sinks to prevent overheating during normal operation.

Most of the second-generation linear chargers are addressing this issue by incorporating thermal foldback current limiting circuit. This function allows the battery charger IC to deliver as much current as possible into the battery as long as its junction temperature (i.e. temperature in the silicon) stays below a certain level. This is great news for the designers, since they don't have to design the battery-charging circuitry assuming a worst-case scenario anymore. The IC will automatically scale back charging current once the junction temperature reaches a certain level, thereby eliminating thermal run-away issues. Yet such an algorithm does not address the need for shorter charging times, since the foldback current operation limits the available charge current, thus extending charging time.

There are two potential alternatives for overcoming this limitation of linear battery charging solutions. The first one is the use of linear charging ICs that work with external power devices (see Figure 2). In such an implementation, power is no longer dissipated by the IC itself, but by the external FET. There is a good number of power FETs in the industry that can provide a good balance for cost, current capability and thermal behavior. The downside of this implementation is the fact that the system still has to deal with heat generated in the FET. Furthermore, the number of necessary components vs. the fully-integrated linear charger has significantly increased (sense resistor, power FET, blocking diode, etc.), generating concerns regarding reliability (more components subject to failure) and board space.

Figure 2: Typical linear battery charger implementation with external FET

Switch-mode chargers
An ideal solution would reduce or eliminate the power dissipation altogether. This can be accomplished via a switch-mode charging topology, which is very efficient and allows a significantly higher charging-current level while maintaining a very low IC temperature. The ability to fast-charge new, high-capacity battery packs at approximately a 1C rate extends battery life while maintaining a short charging period. As demonstrated in Table 1, the combination of power dissipation and current capability of a switch-mode battery charger compares very favorably with linear charging solutions, both with and without thermal foldback limit.

Assumptions:
Input (Adapter) Voltage: VIN = 5.5V
Battery Voltage: VBATT = 3.3V
Package: 3x3 DFN
Ambient Temperature: TA = 65°C
Max. Junction Temperature: TJ = 105°C

Lower average heat dissipation results in more system benefits than just higher charging current and shorter charging time. One of these benefits is the elimination of "hot" spots in the device. The concentration of heat in one place can be a major concern for the consumer, since the electronic equipment might be perceived as malfunctioning or, even worse, as a danger to the user's health. Hence, it is in the manufacturer's best interest to eliminate the possibility of "hot" spots in the device by implementing the most efficient power design possible. Lower average ambient temperature in the electronic equipment enclosure also results in higher reliability. Based on the Arrhenius equation widely used for predicting failure rate and operating life in electronic systems, a 10°C temperature rise in the IC will reduce the operating life of the IC by more than 50%.

Figure 3 demonstrates a switch-mode implementation for battery charging from an AC power source. In this application the battery charger IC can deliver a continuous charging current of up to 900mA and maintain a very low temperature level (that is, power dissipated) in the system. Such an implementation enables users to take full advantage of new, high-capacity batteries since operating times can be extended while charging times remain well within the acceptable range.

The SMB135 battery charger used in this topology offers an additional benefit: unlike linear battery chargers, it allows a charging current that is higher than the input (adaptor) current. This is possible because of the fundamental operation of a switch-mode, step-down regulator to provide an output voltage lower than the input voltage, and an output current higher than the input current. This capability of the SMB135 enables reduced system cost, since the wall adaptor can be designed to output a lower power level, thereby resulting in cost savings associated with the power transistor, transformer and/or the required LC filtering.

Figure 3: Typical switch-mode battery charging from AC adapter

Portable electronic devices are continuously integrating more features and achieving new levels of performance. The popularity of all new gadgets is highly dependent upon the ability to sustain small form factors and ease-of-use, thereby preserving an excellent user-experience. These trends have placed increasing pressure on battery manufacturers and system designers, since increased functions and higher performance require more available energy for acceptable system run times. One of the major challenges is to be able to sustain short charging times while using new, high-capacity batteries. Conventional battery charging methods have limitations that do not allow safe and fast battery charging without major compromises. However, new switch-mode battery charging solutions are available that address the need for short charging time, while eliminating additional issues of system cost, hot spots and system reliability.

About the author
George Paparrizos, is the Senior Product Marketing Manager at Summit Microelectronics. Before joining Summit, he was a product marketing manager at Microchip Technology, specializing in the battery, power and thermal management product lines. Mr. Paparrizos has authored numerous articles for industry publications. He holds a MSEE degree from the Technical University of Aachen, Germany (RWTH) and an MBA from the Haas School of Business at UC Berkeley. gpaparrizos@summitmicro.com