USB Current Limiting

Discussion in 'The Projects Forum' started by cldudley, Jan 6, 2012.

  1. cldudley

    Thread Starter New Member

    Oct 7, 2009
    11
    0
    Hello there,

    I am in the finishing stages of a board design where I am using an FT232 from FTDI to provide a serial interface to the outside world for communication.

    The board has a DC barrel jack to plug in a 5V center-positive wall wart to power the board, but kind of on a lark I decided to add a jumper to allow the use of the buss power from the USB jack when there is no wall wart plugged in.

    The +5V from USB and the center pin of the barrel jack goes to a 3-pin jumper connector, where you can select the 5V from either (but not both!) to go to an ADP1706 3.3V LDO regulator, which provides all of the power to the rest of the board and any offboard experiments I might design.

    This is all fine and good, but I am wondering if I should provide overcurrent limiting with a PTC from the USB, or will the USB hub limit the current itself? The regulator is a 1A device, and has sufficient thermal relief via copper pours to handle the full current, but the USB port certainly can't provide that, even though my wall wart can. If my board/experiements draw more than 500mA, will the USB port clamp it to 500mA, or will I damage my laptop/computer?

    I am aware that devices are required to start with a 100mA limit, the FTDI chip allows you configure it's EEPROM to request the upgrade to 500mA. Also I have the "sleep mode" pin hooked up to the microcontroller, so if it is being powered from USB it can go to sleep when the USB bus does, but that is neither here nor there I suppose.

    Really I am curious as to whether I should current limit the USB buss power on my board, or am I just duplicating protection that already exists in the computer?

    EDIT: Adding the schematic.
     
    Last edited: Jan 6, 2012
  2. Robert.Adams

    Active Member

    Feb 16, 2010
    112
    5
  3. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    Not absolutely sure about it but since I managed to short / overload my laptop's USB driver IC and it got destroyed I guess they don't have a true current limiting feature.

    You could post your schematic so we can see what could be done.
     
  4. Robert.Adams

    Active Member

    Feb 16, 2010
    112
    5
    I read only poor designs should allow that to happen...Could you post what type of laptop it was?
     
  5. praondevou

    AAC Fanatic!

    Jul 9, 2011
    2,936
    488
    I think it was a Dell 800. It's been a few years. I managed to repair it though.
     
  6. Robert.Adams

    Active Member

    Feb 16, 2010
    112
    5
    I was hoping you'd say something like HP so I could have a good laugh. Darn. I'm starting to suspect that you can burn out USB ports on most computers.
     
  7. kubeek

    AAC Fanatic!

    Sep 20, 2005
    4,670
    804
    From what I know, a typical hub has a 2A polyfuse on the port side, which protects the power supply and is shared by the 4 ports. According to specification the hub is supposed to monitor the current and shutdown the port if it crosses the 500mA limit, but that is rarely the case.

    Also, don´t forget to set the "fuses" in the FTDI to set the device as bus powered, because the port is also supposed to shut down if you draw above 100mA and didn´t introduce the device as bus powered.
     
  8. cldudley

    Thread Starter New Member

    Oct 7, 2009
    11
    0
    There is actually more than that settable through the EEPROM, the control bus can be configured to give different status information, including the "USB suspend" state, so my uC knows to close up shop and wait for power to become available again.

    The FTDI datasheet is formatted very differently from most other manufacturers, but is surprisingly readable, and they provide a pre-designed circuit for most situations one may be interested in.
     
Loading...