Group Actions
Engagement
  • Author Author: Catwell
  • Date Created: Date Created
  • Views 578 views
  • Likes 1 like
  • Comments 4 comments
Related
Recommended

USB 3.0 cables to deliver 100 watts of power

Catwell

image

(via USB.org)

 

Since the launch of the Thunderbolt cables over a year ago, there has been a subtle competition with the Universal Serial Bus (USB) cables. Thunderbolt cables offer 10Gbps of transfer rate while SuperSpeed USB cables only offer 5Gbps. In addition, Thunderbolt offers 10W of power through the cables to a device. However, the cables are much more expensive, $49.00 compared to $4.49, USB being the later. Recently the USB 3.0 promoter group has stepped up the competition and approved the power delivery spec for up to 100W over a USB cable.

 

 

The power delivery system has been designed to work with new and older cables alike. To accomplish this, the cables include an intelligent system to locate and check for USB wires and determine their capabilities. For example, a full sized cable would be able to handle 5A of current through it while a micro USB cable would safely be able to handle 3A. The system will also know whether it is working with a low quality wire to keep things as safe as possible.

 

 

In addition to the smart check system, the cables would be capable of switching direction of power flow. This can allow laptops to charge a device and also use it for a power source if necessary. While delivering power it will be delivered in steps between 10W and 100W, each one ideally doubling the one before. As a result, they can ramp power up or down between profiles. With much more power to work with now, USB ports on laptops can possibly double as a charging port. USB is looking to stay ahead of the competition and remain one of the most successful interfaces in the history of personal computing.

 

image

USB architecture (via USB.org)

 

Cabe

http://twitter.com/Cabe_e14

  • It seems to me that power and ground are both grounds from an AC standpoint, and not an issue for EMI.  The digital transients on the power lines would not increase.  (If this were a project I were working on I'd knock on wood b/c EMI lends itself to famous last words.)

     

    I think higher power is a great idea.  I'm confident many devices could use the power.

     

    5A is a lot if those are little 26-28 guage wires.  At first I thought better to reduce the current and go to 48V, but I think some NEC or UL rules kick in at 35V; so it's probably right to stay at 24.

  • in reply to Catwell

    I think that even 30W is enough for wide range of applications.

  • in reply to DAB

    I'm sure the early USB 3 specs are subject to change.

     

    Only certain systems will be able to supply the 100W. In this case USB 3 profile 5, 20V@5A is available.

    Many PCs will be limited to 12V@3A, due to the 12V rails in power supplies, and the current carry of a micro USB cable (3A). So a max of ~30W or so.

     

    Since each port negotiates with the connected device for power needed, and available. It should be ok. These numbers do not seem like a major EMI sources.

     

    Cabe

  • I think that there will be a lot of problems pushing that much current down the cables for any but short lengths.

    I also worry about ground loops being formed between devices, especially with high frequency data.  Unless the developers are very careful, they could generate a lot of EM interference with even short distances between devices.

     

    I would caution all device manufactures to plan on an internal power source and limit any drain pulled from the main device to USB3 peripherals.  Otherwise I think that we may create some nasty problems and fire hazards.

     

    Just my opinion,

    DAB