Editor's
note: In this article, the author makes a
strong case for the versatility and broad
applicability of USB On-the-Go. Read it and see
if you are convinced, or perhaps not. Even if
not, you'll end up knowing a lot about USB OTG
interface, connectors, applications, and
protocol.
Abstract
When we consider today's
small portable devices, they really are just
small-form-factor PCs. Typically, they have
multi-gigabyte storage capacity for rich
content. Their functionality is enabled by
multiple processors and co-processors. Most
offer multiple wireless interfaces. They enable
both digital still and camcorder functionality.
Most are multi-media players, for both music and
video. They utilize full-featured operating
systems. They offer desktop equivalent
applications such as spreadsheets, documents,
PDFs, e-mail, and more.
So, when we consider these devices, we really
need to think of them as more than just a mobile
phone, MP3 player, or digital still camera. Just
as PCs have thrived with USB as the standard
interface, so can these portable devices.
However, given the usage scenarios for these
devices, standard USB just doesn't offer
everything they need. On the other hand, USB
On-the-Go (USB-OTG) is a near-ideal, wired
interface that enables all of these situations
over a single, small-profile receptacle that can
eliminate the need for any other wired
interface.
What is USB On-the-Go?
To understand what USB OTG is, we first must
understand some basic principles of stand USB.
To understand the basic principles of USB, we
really need to think about what the PC looked
like back in the early 1990's prior to the
releases of the USB 1.0 specification:
- More and more users wanted to connect more
and more peripherals to their PCs
- The various peripheral ports (parallel,
serial, P/S-2) to choose from were starting to
present the user with many problems
- Additionally, the simple variety of the
connections was causing problems for the average
user.
- Most did not know which port to use
- They did not understand the limitations that
the port presented
USB-OTG
was released in July 2003 as an addendum to the
USB 2.0 Specification that defines a new class
of devices. This class of devices is intended to
extend the functionality of a peripheral product
to include limited host capabilities. As the
name implies, the original target of the
specification was consumer portable devices with
which end-users may have wanted to share data
when a computer was not available.
Usage examples include sharing contact
information between two PDAs or cell phones,
sharing pictures from one DSC or camera phone
with another, or printing directly from a DSC or
PDA. Like standard USB, OTG is a point-to-point,
host-centric bus and is not intended as a
peer-to-peer networking connection. OTG Products
must act as a standard peripheral when connected
to standard USB host such as a PC. The OTG
addendum mainly addresses how a device must act
when it is in host mode.
Just like a standard USB host port, an OTG
host must supply power. However, the required
supply current is limited to 8 mA. This is not a
maximum; as a matter of fact there is not a
maximum that can be supplied. However, since the
majority of the potential OTG-enabled products
are battery-powered, there is always a trade-off
between extending battery life and providing
enough power for a bus-powered USB peripheral.
If an OTG product will provide more than the
minimum 8 mA, there are two obvious choices for
this increased supply current. The first would
be 100 mA; this is the maximum current that any
low-power USB bus-powered peripheral will
require. This will enable the OTG device to
serve most flash drives as well as many other
low-power peripherals such as keyboards or mice.
The next choice would be to go to the 500 mA
level of a standard USB host port. This would
allow the OTG device to satisfy any bus-powered
USB peripheral's power demands.
A second major difference between standard
USB hosts and OTG products is that an OTG device
may not have a simple way to add drivers for
"unrecognized" devices. This may be for one of
two reasons. First, there just may not be a
method for loading new software on to the OTG
device, especially if the OTG port has the "new"
peripheral connected to it.
That said, some OTG devices may have wireless
capability that may enable network connectivity
for downloading a driver for the unrecognized
device. This would require a very sophisticated
operating system to make this possible.
The second potential limitation would be the
internal memory of the OTG device. These devices
allocate a fixed amount of storage for the
operating system. Adding additional drivers into
this space may not be possible due to the lack
of open memory. Therefore, an OTG device must
supply what is called a Targeted Peripheral List
(TPL). This allows the device manufacturer to
specify exactly what peripherals they will
support.
The USB specification uses what is called the
USB Header information to identify each USB
device. The "first" two entries of any device's
header are the Vendor Identification number
(VID) and Product Identification number (PID).
The TPL is a list of VIDs and PIDs that will be
known by the OTG host and that drivers will be
included automatically to support. This list
ends up as a file that is part of the operating
system. In addition to this file embedded in the
device, the OTG product must ship with a text
version of the TPL for the end"user to
reference.
Since adding support for new devices may not
be possible on an OTG device, it is required
that there be no "silent" failures. In other
words, if an end-user plugs in an unsupported
device (not on the TPL), there must be error
messaging provided for the end-user. The
preferred method is to display a graphical or
text message on the LCD or screen of the OTG
device.
If the OTG device does not have a
text/graphical display, some other means of
messaging must be provided. Examples would be to
illuminate a specific LED, flash an LED in a
specific pattern, change the color of an LED,
etc. Whatever the messaging system is chosen,
this must be conveyed to the end-user.
In standard USB, the receptacle on a host is
called an "A" connector and the receptacle on a
peripheral is referred to as "B". Since the
target end equipment units are primarily small,
portable consumer electronics, a standard "A"
receptacle was too large for these products.
With the original release of the OTG
specification, these new connectors were called
"mini" connectors.
The specification has subsequently been
modified to mandate "micro" connectors for OTG
devices. Specifically, OTG devices must
implement a micro-AB receptacle. The AB
nomenclature indicates it is both an "A" and "B"
receptacle rolled into a single unit. The
micro-AB receptacle can have either a micro-B
plug or a micro-A plug inserted.
It seems obvious that if you insert a micro-A
plug, then the device would be a host and, if
you insert a micro-B plug, it would be a
peripheral. But how does the micro-AB based
product know what has been inserted? The OTG
specification added a fifth contact, called ID,
to the four contacts present in standard USB
systems.
Since the ID pin is not present in standard
USB Connections, the specification carried this
forward and has defined a no connect on the ID
pin as the initial peripheral. Conversely, if
the ID pin is grounded, then that device will
assume the initial host role. This is shown
graphically in Figure 1.
Figure 1: Cable
orientation will determine initial host/target
roles.
(Click on image to
enlarge)
As described above, the cable orientation
will determine the roles the two OTG devices
will assume at connection. Host Negotiation
Protocol (HNP) defines a method for dynamic
switching between host and device roles.
Why switch roles? The need for this can be
understood if we look at Figure 2 and
consider that every OTG device must include a
TPL.
The device on the left has the printer on its
TPL, but it is NOT on the printer's TPL. If the
user plugs in the cable "backwards" as shown in
Figure 2, then communication between the two
devices is not possible without the roles
reversing. This could be accomplished be
removing the cable from both devices and then
starting over with the ends reversed.
The OTG specification built in a protocol to
automatically address this situation. HNP allows
the roles to be reversed silently and
automatically enabling the communication. This
enhances the end-user experience as opposed to
mandating that they disconnect the cable and
reverse it. For this reason, it is important to
note that while the host is usually the
A-device, it is not always the case.
Figure 2: Host
negotiation protocol example.
(Click
on image to
enlarge)
Session Request Protocol (SRP) enables a
method for bus power to be turned off/on at the
discretion of the "host" device to save power
when communication is not needed. Many of the
target end-equipment for OTG are battery-powered
devices. Extending battery life is of utmost
importance to both the manufacturers as well as
the end user.
With this in mind, the A-device (as indicated
by cable orientation) in an OTG connection can
"power off" the bus and go into a sleep mode
extending battery life. This will also allow the
B-device to go to sleep if it so desires.
However, the end-user may desire for the
communication to start-up again and initiates
this request on the B-device. SRP allows the
B-Device to request the A-Device to turn on VBUS
power and start a session.
An OTG session is defined as the time that
the A-Device is furnishing VBUS power. To
wake-up the A-device, the B-device pulses first
the D+ wire, and then the VBUS wire to wake up
the A-Device. The A-Device detects the pulse
(A-Device can respond to either pulsing),
causing it to switch on VBUS and start a
session.
The SRP protocol is more complex than this
simple illustration. The B-Device, for example,
must first measure VBUS to ensure that a session
is not in progress. It must also be able to
differentiate between a classic PC or an OTG
device at the other end of the cable. It does
this by delivering measured amounts of current
to the VBUS wire and noting the resulting
voltage. Note that in the example shown in
Figure 2, the A-device that has to supply the
power is the printer which is actually the
peripheral in the session.
Why should USB On-the-Go be implemented on
Portable Devices?
As previously discussed, the USB OTG
specification has been released for over five
years and the adoption rate has lagged.
Therefore, the questions many people have are:
- What is the real benefit of OTG?
- Why would I want it in my device?
First, let's revisit the original
target devices: portable consumer products such
as cell phones, PDAs, digital still cameras, MP3
players, and similar. Nearly all of these device
use USB as their method to interface to the PC
today and share data. All of these devices also
have other wired interfaces today, including
power for battery charging, audio input/output,
video output, capability to talk to non-PC
devices, and so on. However, as form factors
become smaller and smaller, the ability to
remove/reduce the required external
receptacle(s) space is of great interest.
The USB Implementer's Forum (USB-IF, the
trade association for USB) has many initiatives
in place or in development to address the needs
of these portable devices that have exceeded the
initial capabilities/scope of the original USB
specification. These include the ability to
charge a battery over a USB connection, the
smaller form-factor micro connectors, and
headset audio, to name just three key ones. When
you combine these initiatives and wrap them with
the USB OTG specification, you can enable all of
this functionality while only requiring a single
small form-factor receptacle for all the wired
connectivity needs of a portable device.
First, let's look at some of the work going
on in the area of battery charging. The initial
USB specification did not plan for battery
charging as part of the normal use case
scenario. However, by putting a definition on
how peripherals can charge off the 5 V/500 mA
available, there is no reason that a device can
not use this for charging. The device must meet
all standard USB power specifications, never
exceed 500 mA of current draw during operation,
less than 100 mA when not configured, and less
than 500 μA when the PC is in a suspend mode.
The Chinese government has taken this to the
legislative level by mandating that all new
handsets sold in China must charge over their
USB receptacle. This can be via a standard USB
port connection or via a dedicated USB charger.
A USB Charger is a power supply that plugs into
the wall, but has a USB plug on the output side.
The data lines are left open so the device will
not try and communicate.
bviously there are not the same current
limitations, since it is no longer a USB host
that is providing the power. They have developed
a specification for what these chargers look
like so that any charger will work with any
handset. This does eliminate the need of the
handset manufacturer to provide a new charger
every time you buy a new device, as any charger
will work with any USB devices.
In addition, there has been a recent
development by the USB-IF to define a method for
the battery-powered device (whether OTG-capable
or a standard peripheral) to detect if the
connection is a standard USB host connection or
a dedicated USB charger. This is in order to
make sure that the USB device is not damaged by
a charger that was developed by a different
vendor.
One of the key issues is how the OTG product
will manage VBUS current draw when its battery
is dead. Typically, recharging dead batteries
demands a large current spike upon first
connection to get the battery to a base
threshold, to manage the balance of the device.
However, if the charging is happening over a
standard USB host connection, this is not
possible while meeting the USB specification.
Therefore, the end-equipment manufacturer must
account for dead battery during their power
management design phase.
As discussed above, the OTG specification
originally released with mini-connectors
defined. While the mini-B has established itself
as a standard connector on many small
form-factor peripheral devices, the mini-AB was
found to be lacking to meet all the demands for
these devices. It was not be small enough for
many of these space-conscious applications.
Of even more concern was the robustness of
the receptacle. There was concern that the
insertion/removal life cycle was not adequate
for the number of possible uses to which these
portable devices would be exposed.
The final improvement was to add a latching
mechanism to prevent accidental cable removal.
This receptacle can replace all the other wired
interfaces on portable devices,saving space and
cost.
The key new development is the initiative to
define a standard interface between a USB-OTG
handset and a USB headset. This is a new device
class that will define a standard interface to
allow any OTG-capable phone/MP3 player/Video
player that adds headset support to host a
digital USB headset for both playback as well as
mic input.
Since the headset will be plugging into a
battery-powered device and will most likely
receive all of its power from the device, it is
critical to focus on power consumption for the
headset. This will allow manufacturers of these
type of audio centric products to remove the
2.5mm audio jack from their products and use the
already available USB receptacle.
When we think about today's portable consumer
devices, they really are just small form-factor
PCs with many of the same characteristics of
PCs, including storage capability, multiple
processors, multiple wireless (and wired)
interfaces, multi-media functionality,
full-featured operating systems, and many
desktop-like applications. Given the real
benefit that PC users have derived from USB, it
makes sense that the portable devices can also
extend these benefits to their users.
References
- For more information on USB-OTG, go to http://www.usb.org/developers/onthego/
- For more USB solutions, go to http://www.ti.com/usb
Other USB-related articles
- "Optimization of current-limiting solutions
for USB 3.0," click here
- "PPTC devices help protect electric motors,
transformers and industrial controllers," click
here
- "Maintain USB signal integrity when adding
ESD protection," click here
- "Which USB is Right for Your Application?
(Part 2)," click here
About the Author
Dan
Harmon has held the position of Product
Line Marketing Manager for Digital Interface
Products at TI for the last seven years. During
his over 20 year career at TI, Dan engineered
night vision FLIR systems and was a camera
design engineer on CCD Imaging Products before
becoming the CCD Product Marketing Engineer. He
earned a BSEE from the University of Dayton and
a MSEE from the University of Texas in
Arlington. Dan can be reached at ti_danharmon@ti.com.
.gif){kind=small}
.gif){kind=small}
.gif){kind=small}
.gif){kind=small}
.gif){kind=small}
.gif){kind=small}
.gif){kind=small}