USB Charging Details

Q. Why a Smart charger, it’s all just plug and go isn’t it?

A. Simple answer, no.  
More complex answer, there are three main ‘types’ of USB port;

  • SDP; Standard Downstream Port, which supplies up to 500mA for powering an attached device with data
  • CDP; Charging Downstream Port, can supply/charge up to 1500mA and still have data running
  • DCP; Dedicated Charging Port, which doesn’t pass data but can supply power over 1500mA

The signalling of the various types of port is by using the D- and D+ pins, in a variety of combinations, which are often manufacturer specific.  For example, the following are just some of the combinations in use;

D+ Voltage D- Voltage
2.0 2.0
2.0 2.8
2.8 2.0
1.2 1.2
2.7 2.7
3.3 3.3


In addition to the various voltages there are also the possibilities of a short between the two pins or open circuit.   Each manufacturer simply chooses a value to use, there is no common standard within the DCP world.  Some devices even want to see an ‘active’ charger, where they perform a handshake too.

When you plug your device in, it also has to work out what type of port it is connected to, which is complex in itself too.  A snippet from the specification over how they do that is given below.


Q. My charger is rated in Watts, but you talk about Amps, whats the difference.

A. Nothing really, they are just different ways of giving the same information.  

Watts = Volts * Amps

Amps = Watts / Volts

To save you thinking too hard about it, the common charger profiles used are;

2.5 Watt = 500 mA, or 0.5 Amps

5 Watt = 1000 mA, or 1.0 Amps

7.5 Watt = 1500 mA, or 1.5 Amps

10 Watt = 2000 mA, or 2.0 Amps

12 Watt = 2400 mA, or 2.4 Amps


Q. What is the difference between ’rapid’, ‘standard’, and ’trickle’ USB charging?

A. Nothing, they are a myth.  The USB charging standards dictate that a device can only draw;

  • 100 mA prior to charger detection
  • 500 mA following charger detection
  • up to 1500 mA following the detection of a USB BC1.2 standards compliant charger

If the device detects its own dedicated chargers particular signature it can draw power above 1500 mA.  So if your device is ‘trickle’ charging it is because it doesn’t believe it is plugged into the right charger, the charger does not support BC1.2, or the charger cannot cope with the demands of the device.

It’s important to note that regardless of the stated power output of any charger, it is the device that determines how much current it can draw not the charger.   So if your device doesn’t believe it is See the question below on how the device detects the charger type.


Q. A generic basic charger and a ‘cheater cable’ is just the same, is it not?

A. Well, no, not really.  Cheater cables try to emulate the a specific companies charger signature, so fool the device into thinking it is plugged into its own charger.  Which seems great, unless the charger cannot actually handle the needs of the device, let alone the quality of the cable itself.  Then you can get all sorts of interesting things happen.

  • The charger may not be able to provide the current the device wants, so normally the voltage drops.  This makes charging slower and puts stress on the device being charged.  There are a few companies in the UK now specialising in repairing devices with broken charge controllers, as a result of the influx of low quality chargers.
  • as well as damaging the device, it can also lead to problems with the charger, overheating and lots of RF interference are the prime examples
  • cheater cables are often poor quality, which doesn’t help when you want to pull as much power through them as possible, they are cheap for a reason
  • you will need to buy a cheater cable for each device you own

The starting point for the design of the Charge2 and Charge4 was to address the problems that generic cheap USB chargers caused in a light aircraft, details are available here.


Q. How does the device detect what the charger is?

A. The ‘communication’ is done by a handshake via changing the resistance / voltage levels on the two data pins within the USB connection, in the right sequence and to a strict timeline (measured in milliseconds).  A passive charger (such as a cheap plug car style) will probably a hardwired ‘profile’ which is normally a short of fixed set of resistors on the data pins, some chargers do have ICs to support some devices on the market at the moment they rolled off the production line.  There are in excess of seven different types of bespoke identification out there at the moment, if you fancy some bedtime reading, pop over to the USB Developers website, it’s all on there.  http://www.usb.org/developers/

Charge2 and Charge4 take this a little further by using a micro controller to make emulations extensible, so for future device requirements new devices can be supported.  Once we get hold of one of the new devices a custom profile can be created and entered, extending the life of the charger.  The problem is that whilst a USB standard exists, anything which wants to charge over 1.5 Amps isn’t supported by the standard, and all the manufactures like to do things differently.

As an example, taken from the specification, this diagram below describes what your device does when plugged in;

detection-www


Assuming the charger is correctly built to act just as a charger and nothing else, then your device has to decide if it is plugged into a Charging Downstream Port (CDP) or a Dedicated Charging Port (DCP), the description of what happens is a bit longwinded;

During Primary Detection the PD shall turn on VDP_SRC  and IDM_SINK . Since a DCP is required to short D+ to D- through a resistance of RDCP_DAT , the PD will detect a voltage on D- that is close to VDP_SRC.

A PD shall compare the voltage on D- with VDAT_REF . If D- is greater than VDAT_REF , then the PD is allowed to detect that it is attached to either a DCP or CDP. A PD is optionally allowed to compare D- with VLGC  as well, and only determine that it is attached to a DCP or CDP if D- is greater than VDAT_REF , but less than VLGC . The reason for this option is as follows.

PS2 ports pull D+/- high. If a PD is attached to a PS2 port, and the PD only checks for D- greater than VDAT_REF , then a PD attached to a PS2 port would determine that it is attached to a DCP or CDP and proceed to draw IDEV_CHG . This much current could potentially damage a PS2 port. By only determining it isattached to DCP or CDP if D- is less than VLGC , the PD can avoid causing damage to a PS2 port.

On the other hand, some proprietary chargers also pull D+/- high. If a PD is attached to one of these chargers, and it determined it was not attached to a charger because D- was greater than VLGC , then the PD would determine that it was attached to an SDP, and only be able to draw ISUSP.

This actually all means that your device drains then puts some voltages onto the Data pins, looking to see what is there.  Depending on what it sees it will make a decision over what type of device it thinks it is plugged into and act accordingly.


Q. So there is USB 1, USB 2, USB 3, and USB 3.1 - what is the difference?

A. These are all variations of USB protocols, which are faster the higher the number.  In terms of power delivery, or charging ability, they are comparable with one another.  Any device which has a USB Type A connector (the one found on laptops and PCs worldwide) is compatible with the power delivery of Charge2 and Charge4.

Things will change with the announcement of the new Type C USB connector in August 2014 intended to go with USB 3.1, which is reversible and the same at both ends.  A first for the USB specification, it also means that you won’t be able to use a device with a type C connector until your laptops and PC’s are also all fitted with type C connectors.  No doubt at some point that will happen, but it will take a while.  In the mean time it is expected that devices will be supplied with adapter leads, to go from type C to type A, and normality returns in terms of power delivery.  Much like Apple moving from the 30 pin connector to lighting, the device side changes by the host/charger end has remained the same.

There are changes within USB 3.1 for higher power delivery, but for power up to 12 Watts (2.4 Amps) everything is backwards compatible with the previous versions.  If you go over 12 Watts then it’s a new ballgame in the charger market, as the voltages go up from 5v to 20v for full 100 Watt power delivery.  Of note is that the USB 3.1 introduces a new power delivery scheme ‘USB Power Delivery 2.0’, which co-exists on devices along with the current BC 1.2 standard that Charge2 and Charge4 support, so new devices will still charge fine with a suitable USB lead.