USB battery charging: it’s harder than it looks

As the USB port becomes increasingly ubiquitous, it is also becoming accepted as a universal charging port. Unfortunately, this concept of universal is easier to say than it is to do. This article is an introduction into the common challenges a designer runs into in creating this highly desirable, omnipresent USB charging port.

Why is this taking so long?

So, what does it mean to provide a "fast" charge?  This usually boils down to customer expectations. The common example is, “I charge my phone, MP3 player …fill in blank, in x hours at home, but at work, with my laptop, with my monitor, with my new adaptor…fill in blank, it takes all day to charge!” 

So we start with the "native" charger that comes with any device. This charging experience is the baseline for customer satisfaction.

The native wall charger for a device will very often have a special signature on the data pins to let a device know it is safe to charge with more current. In some cases, it also prevents the device from charging at all if the host is unknown. This signature may come in the form of a specific voltage placed on D+, or D-, or both.

Refer to Figure 1, which illustrates a common architecture for a wall charger using this methodology. Note that these configurations are implemented so the manufacturer can sell more accessories.

Figure 1 - Common architecture for a wall charger

(Click here to enlarge schematic.)

Make no mistake: selling specialized accessories is definitely in the business plan for a portable product. For every chargeable product purchased, about 50% of us will go out and buy another charger. The reason is simple: we do not like carrying them around, so we leave a charger in the other places we frequent, such as in our office or in the car.

What is the "right" charging current? (Hint:  There may be three!)

To begin an analysis of USB charging, you first need a system to help measure the current on Vbus and to measure and apply voltages on D+ and D-. This can be done by creating a board that both the peripheral and the host can plug into while exposing their D+, D+ and Vbus lines for analysis.

Jumping ahead, it is time to evaluate the charging current with a device connected via your interposer board. So let’s assume we are all smart enough to determine what voltage the native charge places on D+ and D- and we recreate a discrete charging circuit to confirm our suspicions. We then apply the right voltages, just like the native charger on D+ and D-, but the charging current is not matching our previous results.

It is time to check your power. No, not just whether things are plugged in, but the level of power. Battery-power level plays a key role in charging. Many of us who have worked on cell-phone designs can tell you that a deeply discharged lithium-ion battery needs to be trickle charged before the real charging can start.

This, too, complicates knowing whether you have an optimal charging current. The peripheral that gets plugged into a USB port may have several different points of charging before it is full. It most likely has a low-charging mode for the aforementioned trickle charging. It also may have a different charging state for when the battery is nominally charged. Finally, it may have a charging state for a fully charged battery.

As a result, you will need to observe what the charging current is when a) a given device’s battery is empty, b) when it is midway charged, and c) when it is fully charged. Sound time consuming?  You bet it is, but it is a necessary evil for complete characterization.

Can I have the kitchen sink, too?

We now have a growing understanding for customer charger configurations and what we would like to be seeing for charging current. For many applications (such as PC, monitor, docking station), you may want fast charging and the ability to transfer data at the same time.

In this regard, there has been a lot of confusion as to what is possible. The reason for this goes back to the fact that many native chargers place a voltage on the D+ and D- pins of the USB port. Since traditional data communication on the USB is based on 3.3V for USB1.1 and 300mV for USB 2.0, putting a different voltage on these lines eliminates the possibility for enumeration and communication.

There are some exceptions to this rule. For instance, there are devices which require you to download device-specific software to your host when you first plug the device into the USB port. Some cell phones are like this for syncing purposes, and now some facilitate charging at a higher current while communicating. So for device communication and charging, we may be limited by what the device will allow given a specific software driver.

But all is not lost and there is help on the way. A recent specification has been created to help with this data plus charging challenge: it is the USB-IF Battery Charging Specification revision 1.2 (BC1.2) and the full specification can be found at http://www.usb.org/developers/devclass_docs.

This specification was created to try to unify battery-charging attributes for USB 2.0 in the future. The idea was to minimize the number of cell-phone chargers ending up in landfills, by converging on one USB-charging specification. The European Union has been an early adopter to the notion of less waste. Specifically, they have committed using the same microUSB connectors on data-enabled cell phones, but they have yet to fully adopt the BC1.2 specification.

In the BC 1.2 specification, there is a mode referred to as Charging Downstream Port (CDP) that allows for data and higher charging currents. If a voltage between 0.4V and 0.8V is sensed on D+ of a host or hub device, then D- should respond with 0.5V to 0.7V.

More details of timing associated with this specification that can be found in the specification. Once CDP has been established, peripheral devices are allowed to draw up to 1.5A and simultaneously communicate data. Devices with this technology, including cell phones, should be showing up this year.

Far from simple

In summary, the USB port has infiltrated our life for providing power and we need to be intelligent if we want to be the provider of this power. Hopefully, this primer on USB charging should help you from going down many of the dark alleys in which our colleagues get stuck.

About the Author

Mitch Polonksy is the director of product marketing for analog products and technology at SMSC (Hauppauge, New York). Prior to his role at SMSC, Mitch was in marketing at Motorola. He holds a BA in mathematics form Emory University, an MSEE from Georgia Institute of Technology and an MBA from Arizona State University’s W.P. Carey School of Business.

Editor's note: Liked this? Want more?

If you are interested in "power" issues such as components; efficiency; thermal concerns; AC/DC and DC/DC supply topologies; batteries; supply ICs; complete supplies; single- and multi-rail management; and supply monitoring: then go to the Power Management Designline home page for the latest in design, technology, trends, products, and news. Also, sign up for our weekly Power Management Designline Newsletter.