Once again, the Universal Serial Bus comes to the rescue in the ecological and economic battles confronting the consumer-electronics markets.
USB first changed the electronics world by enabling near-universal connectivity among the myriad of computer and portable devices. Now, in response to a rising need for ecological charging strategies, USB is once again being called upon to unify the consumer electronics industry as a universal charging technology.
The arbitrary hardware and connector differences between chargers for all sorts of consumer electronics devices require users to discard fully functional chargers, simply because next-generation devices often aren’t compatible with them. To eliminate this waste stream, governments worldwide have begun to introduce local regulations that impact how portable devices can be charged. For example, the Chinese government and the European Commission require all new handsets to charge through a micro USB connector.
The near-complete penetration of USB into computers and electronic devices positions it well as a platform upon which to replace the plethora of brands of chargers. In addition, USB’s economies of scale combined with a device’s ability to connect to a charger with a standard USB cable, which most consumers already own, means that USB provides a greener charging alternative compared to individual charging units.
The most effective green strategies, however, are those that not only reduce the waste stream, but also lower cost. Moving to USB-based charging technology allows devices to be charged with the data cable they’re shipped with, eliminating the need for a home wall-based charger and separate aftermarket cigarette lighter/car charger, each costing $20 to $40. USB also uses a sleek, aesthetically pleasing connector that can sit flush with the casing of a device or the dash of a vehicle, compared to bulky standalone chargers.
Making the shift to USB-based charging poses a variety of design challenges, though. The current draw limitations of Standard Downstream USB ports can significantly increase charging time compared to the chargers they’re intended to replace, causing them to fall short of consumer expectations.
Universal chargers must also address compatibility issues arising from the use of intelligent communication protocols. Such protocols authenticate chargers to prevent non-compliant chargers from providing too much current and, thus, damaging devices. Finally, given the cost-conscious mindset of the consumer electronics market, any USB-based charger must be implemented at the lowest possible expense.
Dedicated Charging Ports
For all of its advantages as a charging technology, USB is first and foremost a data-communications interface. As such, a full USB controller comprises complex circuitry to support high-speed data transfers. The Standard Downstream Port (SDP) definition within the USB specification also limits powering and charging device—it only supplies up to 500-mA charging current per attached device. This 500-mA limitation is due, in part, to the SDP’s need to simultaneously support high-bandwidth data transfers.
For applications that require data transfers, a full USB controller provides a reliable, ubiquitous interface as well as charging capabilities. For many applications, however, data-transfer capabilities are unnecessary. For example, a wall charger only supplies power and has no need to receive data. In these cases, a USB port optimized solely for charging will provide a more efficient and cost-effective approach.
Such dedicated charging ports (DCPs) reduce the cost and complexity of USB-based chargers by implementing only those components required for charging. The savings are significant: a DCP-based charger controller can be less than half the cost of an SDP-based implementation requiring a full USB controller and physical layer (PHY).
DCPs are appropriate even for applications requiring USB-based data-transfer capabilities that, depending on the application, may need several USB ports. In a car, for example, the driver and passengers may wish to charge multiple devices. Developers should consider the true value of implementing each port as a full USB port since only one device connects to the console at a time.
While the primary USB port can be used to connect devices, implementing additional ports with full USB capabilities is unnecessary given that they only will be used to charge devices. Secondary ports, then, are better implemented as DCPs offering fast charging capabilities, since DCPs can increase the current draw limit from 500 mA to 5 A.
Is Data Transfer Always Necessary?
The need for data-transfer capabilities is worth a closer look. For example, streaming audio from a portable device to an entertainment system may be better implemented using a standard audio jack than a USB port. From a system perspective, using a full USB port to charge both a device and stream audio seems to be the most cost-effective approach.
However, the architecture required to support USB-based playback is fairly complex. Not only must the system include a USB port with full functionality, it also must support a wide range of audio codecs to ensure compatibility (Fig. 1a).
The issue of universal audio compatibility should not be overlooked. Home-entertainment equipment and automotive manufacturers don’t want to lose customers because a prospective customer’s digital device isn’t supported.
Further complicating compatibility is the proprietary codec employed by Apple devices, which prevents devices from streaming unprotected content over USB. No hardware-based codecs currently exist for Apple’s content format. Therefore, supporting playback from Apple devices requires a subsystem that can run iTunes (Fig. 1b).
The cost of components, e.g., a high-performance processor, operating system, and additional memory, clearly exceed the cost of the alternative—a simple audio jack for playback and a low-cost USB charger controller for power (Fig. 1c). Stereo and car console manufacturers worldwide are using this approach, as it guarantees 100% connectivity with all digital devices while minimizing cost and system complexity.
Since devices must be tethered when charging, the more consumers find a device useful, the more inconvenient the device seemingly becomes to charge. Thus, charging time is a high-profile feature of any portable consumer-electronics device. Ideally, a device can be completely charged within a short period, such as during the drive home.
Charging time is a function of charging current—the more current a device can draw, the faster it can charge. Charge current depends on the power source and the method of power supply. For example, devices can draw substantially more current from a traditional wall charger than a USB-based wall charger using an SDP, which is limited to 500 mA maximum.
By supporting a higher charging current than that specified by the SDP standard, a DCP combines the cost advantage of a universal USB-based charger with the higher current of a wall-based charger. With the ability to supply above 1.8-A charging current, a DCP can dramatically reduce device charging time compared to SDP-based ports.
Recently, the drafters of the USB spec introduced the Charging Downstream Port (CDP) as an alternative to the SDP. A CDP can supply higher current—1.5 A to 5 A—even when transferring data through the USB2.0 protocol.
However, not all USB controllers support CDPs, which creates significant compatibility issues because a CDP may require external circuitry to operate with legacy devices. In any case, a CDP still requires a full USB controller, resulting in higher system cost compared to a DCP for a charge-only port in applications that don’t require data transfers.
Exceeding USB port and hub current draw limits can potentially damage electronics, from the portable device itself to the charger and USB hub. SDP-based chargers can draw up to 500 mA per device, which requires the hub to manage the maximum draw current it can support among multiple devices: a four-port hub must be able to safely deliver 2 A (i.e., four ports at 500 mA each).
While the USB standard defines the amount of current drawn by devices, overcurrent issues can still arise due to non-compliant equipment or operation. Overcurrent damage can also occur if a device or charger has a short.
To avoid overcurrent failures and the associated costly service calls and returns, many device manufacturers design their power-management blocks to draw less than the maximum allowed. While providing a margin of safety, it also results in longer charge times for devices.
Looking for a competitive advantage, many aftermarket charger manufacturers allow their chargers to source up to 700 mA to enable faster charging. Such non-compliant chargers can exceed a device’s or hub’s rated limits, which may cause device printed circuit board (PCB) and component damage.
Specifically, overcurrent is one of the top reasons for handset returns, ultimately reducing manufacturer profits due to higher return rates and increased liability (i.e., user injury). Multi-port hubs are also vulnerable to overcurrent damage if several non-compliant chargers are connected, and their total draw exceeds the hub’s limits.
To design a robust USB-based charging port, developers must consider how to protect the system against non-compliant chargers. In applications that introduce USB-based charging to, say, an automobile, overcurrent damage to internal vehicle circuitry can be difficult and expensive to repair. To protect against shorts and non-compliant chargers that draw more than the allowed current, a power switching circuit can be introduced to shut down the USB port.
A power switching circuit monitors both the current being drawn and the dynamic temperature across the power-supply path. For applications in which circuit protection is also important, power-policing capabilities can protect the system from faulty chargers or phones attempting to draw a high, continuous current that will melt delicate electronics.
Even though a fast charger may be responsible for damaging a portable device, consumers often assume failure is the fault of the device manufacturer. To that end, manufacturers (particularly in the handset market) are adding intelligence to the charging process via communication protocols to prevent non-compliant chargers from providing too much current.
A protocol uses secure mechanisms to allow devices to authenticate a battery charger as a known and trusted source. If the charger isn’t recognized, the user is given a message that the accessory isn’t supported. This prevents the user from accidentally connecting to an unauthorized fast charger, which may deliver too much current and damage the device.
Another advantage of using a protocol to authenticate chargers is that different devices can draw varying amounts of charge current safely. To supply current in a safe and reliable way, the universal charger must be able to operate within the safe limits of the device currently being charged, even if it can supply more current.
During the handshaking process, the device and charger can negotiate these limits. This also allows the device and charger to act together to work with the ideal charging profile of the device, minimizing charge time while maximizing the battery charge lifetime.
Implementing an authentication protocol introduces complexity, additional cost, and potential interoperability issues. To serve as a charger for a single device, the DCP must support a single charging protocol. To serve as a universal charger, however, the charger must support the variety of charging protocols implemented across the industry.
BlackBerry handsets, for example, follow the 1.0 version of the USB Battery Charging specification, while Nokia and Motorola devices follow the 1.1 spec. The Chinese government has mandated that all phones sold follow the YDT-1591 spec. Apple, which owns a large share of the market, utilizes its own proprietary protocol. To further complicate issues, version 1.2 of the USB Battery Charging specification has been released too.
Automatically identifying the protocol being used for a particular device requires an evaluation layer above the actual protocol layer. Thus, before charging can begin, the DCP needs to determine the brand and type of device that’s been plugged in before it can select the proper charging protocol. Next, the DCP offers the device the different current draw levels it supports. The device then acknowledges how much current it wants to draw and begins to draw that current.
These protocols can be implemented in software using a MCU, but it introduces additional components and cost to the system, as well as significant programming complexity. When the protocols are integrated into the USB charger controller, the overall system architecture makes it possible to implement a universal charger in an efficient, self-contained, and straightforward manner.
DCP Or CDP?
Developers can take several different approaches to create an efficient and cost-effective USB-based charging port that safely provides substantially more draw current than the SDP’s 500-mA limitation:
• Dedicated Charging Port (DCP): The primary function of a DCP is to provide charging current. Since the charger controller will not need to support data transfers, its internal circuitry can be significantly less complicated than that of a full USB controller (Fig. 2). Charger controllers, like Pericom’s PI5USBxxx family, are self-contained ICs with full support for today’s various communication protocols and optional power policing capabilities. Compared to an SDP using a standard USB controller, a DCP-based charger controller provides higher current and faster device charging at lower cost. The level of integration also enables developers to implement a universal charger with a single chip.
• Charging Downstream Port (CDP): Some charger applications, such as an automotive entertainment system, may also need to support data transfers. A CDP can be implemented by placing a charger controller inline between a standard USB controller and the USB port connector (Fig. 3). In this configuration, the charger controller transparently passes all data between the USB controller and port. At the same time, the controller boosts the charging current beyond 1.5 A for all data rates to convert the SDP into a CDP. Because the charger controller manages the charging communication protocol, there’s no need to modify the USB protocol stack on the host side, and it eliminates any legacy concerns.
• Integrated USB Controller: The inline charger-controller approach also works in applications where the host application processor has an integrated USB controller, but a higher charging current is desired. Placing a charger controller between the application processor’s USB controller and the USB port provides full data capabilities with charging current beyond 1.5 A for all data rates using the CDP protcol.
Weighing all the factors, USB is positioned to become both the universal data connection and universal charging technology worldwide. The ability to replace individual charger units with a standard USB cable is compelling both ecologically and economically.
DCPs overcome the limitations imposed by the low 0.5-mA draw current supplied by an SDP. That’s because they enable significantly faster charging with currents beyond 2.0 A per device.
DCP-based charger controllers also simplify system design by delivering key universal charger capabilities in a single chip: They eliminate data-transfer circuitry where it’s not needed; offer full support of charging communication protocols; guarantee interoperability across all devices; and lower system cost. In addition, optional power switch capabilities can be implemented to ensure reliable operation and prevent damage to system electronics due to overcurrent or shorts.