USB device cable shield connection - grounding it or not?

Discussion in 'General Electronics Chat' started by colinb, Aug 30, 2011.

  1. colinb

    colinb Thread Starter Member

    Joined:
    Jun 15, 2011
    351
    35
    I have been trying to understand how to ideally handle the cable shield
    on a USB device. (Full Speed USB, in this particular instance.)

    As seems to be the case with many signal integrity issues,
    contradictory recommendations abound, each with its own unsupported
    claims. Even authoritative-sounding sources such as Texas Instruments,
    Intel, FTDI, and Cypress Semiconductor seem to disagree on the correct
    way to handle the cable shield on USB devices.

    Contrary to my initial supposition, the purpose of the USB cable shield
    is not to protect the USB data lines from outside interference, but
    rather to prevent the USB device from radiating EMI.

    Here are some of the options that have been recommended.
    Note that (2)—series capacitor to pass high frequencies only—seems to
    directly contradict (3)—series ferrite bead to block high frequencies
    only.

    (1) Connect shield directly to signal ground.

    - “Full speed devices use a shielded cable which requires that the
    connector shell be tied to the ground plane.”
    Intel. EMI Design Guidelines for USB Components. Sec 5.4 (p. 9).​

    (2) Connect shield to signal ground through a capacitor.
    (Possibly with high-value parallel resistor approximately 1 Mohm.)

    - Connect shield to signal ground with 0.01 µF to 0.47 µF capacitor.
    FTDI. Debugging FT232BM and FT245BM Designs. Section 3.2 (p. 11).​

    - Cypress recommends a 1 Mohm resistor in parallel with a 4.7 nF capacitor.
    Steve Kolokowsky & Trevor Davis (Cypress Semiconductor). Common USB Development Mistakes – You Don’t Have To Make Them All Yourself! Figure 7 (p. 7).​

    - “Tying the shield directly to ground would create a direct path
    from the ground plane to the shield, turning the USB cable into
    an antenna. To limit the USB cable antenna effect, it is
    recommended to connect the shield and ground through an RC
    filter. Typically, R = 1MΩ and C = 4.7nF in Figure 3-5.”
    Atmel AVR1017: XMEGA - USB Hardware Design Recommendations. Section 3.3.3 (p. 8).​

    (3) Connect shield to signal ground through a ferrite bead.

    - “Place a ferrite in series with the cable shield pins near the
    USB connector socket to keep EMI from getting onto the cable
    shield.”
    Texas Instruments Application Report. USB 2.0 Board Design and Layout Guidelines. Sec 2.2.4 (p. 3). SPRAAR7 – December 2007.​

    (4) Do not connect cable shield to ground on the device at all.

    - As referenced in the AAC thread where to terminate usb cable shield?, Hardware Book says USB devices must
    not connect the shield to their own ground.
    Hardware Book. Universal Serial Bus: Shielding.

    Whether or not the device has a metal chassis, and the handling
    of chassis ground and signal grounds, (as well as how the USB cable
    ground is connected to either one) is certainly important as well, but
    this isn't clearly discussed in most of the writings on USB cable shield
    grounding.

    The device I'm developing is a bus-powered device which will likely be
    in an unshielded plastic enclosure.

    Thanks in advance for any bits of wisdom on this topic full of
    contradictory information. I recently posted this question on si-list,
    and even there I got little in the way of answers.

    Colin
     
    #1
    endolith likes this.
  2. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2

    according to the specifications laid out by the usb-if
    A metal braid is required to enclose all the wires in the USB 3.0 cable. The braid is to be​
    terminated to the plug metal shells, as close to 360
    ° as possible, to contain EMI.

    apart from that there is no other requirement the same goes for usb 2.0

    it appears that this is a grey area but the sheild should be connected to both sides but not connected to ground in any way.

    will keep looking for you on the specs and see if i find something
     
    #2
  3. colinb

    colinb Thread Starter Member

    Joined:
    Jun 15, 2011
    351
    35
    Interesting. However, I think that the shield (cable braid and plug/socket shell) must be connected to ground on at least one end. If the shield were left floating, then other signals would be able to couple to it and use the shield as a nice long antenna to radiate EMI.
     
    #3
    endolith likes this.
  4. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2
    6.8 USB Grounding​
    The shield must be terminated to the connector plug for completed assemblies. The shield and chassis are
    bonded together. The user selected grounding scheme for USB devices, and cables must be consistent with​
    accepted industry practices and regulatory agency standards for safety and EMI/ESD/RFI.
     
    #4
    colinb likes this.
  5. colinb

    colinb Thread Starter Member

    Joined:
    Jun 15, 2011
    351
    35
    Thanks for posting that snippet. So the shield is connected the the chassis ground, but beyond that, handling of chassis ground is left up to the designer.

    That means most any of the suggestions in my prior cited sources are acceptable according to the USB specification; but how this shield/chassis connection is best handled as far as connecting to other grounds in the system is still a point of contention.
     
    #5
  6. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2
    according to the emi standards the connector requires a low impedance to ground
     
    #6
  7. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2
    The V​
    BUS and ground in a PC are subject to noise both from on-board sources and also from the switching
    power supply. Tests have shown that it is necessary to carefully decouple both V
    BUS and Gnd. This is done
    with ferrite beads. Separate ferrite beads may be used on each V
    BUS line to each downstream USB
    connector. Each ferrite bead on the V
    CC lines should be rated at 500 ma. Separate ferrites are useful, not
    only for EMI suppression, but also for their series DC resistance which limits the inrush current during a
    hot plug event. For a discussion on hot plugging and voltage droop see the
    USB Voltage Drop and Droop
    Measurements
    white paper. Ferrites should also be used on the Gnd lines to USB connectors.
    If bypassing V
    CC with ferrite beads still does not provide sufficient attenuation then small (1000 pF)
    capacitors can be connected on the connector side of V
    CC to the connector shell through very short traces.
    If this approach is used care must be taken to insure that the capacitors do not provide a path for ground
    plane noise to enter via the capacitors.

     
    #7
  8. colinb

    colinb Thread Starter Member

    Joined:
    Jun 15, 2011
    351
    35
    But what is ground when two pieces of equipment are involved? For mains-powered equipment, it must be earth ground, but what about portable battery-powered equipment? Obviously in USB there is the GND wire, which the USB Host supplies, so that could be the ground. What if the USB Device is self-powered and does not share ground with the USB Host? I think the answer is the USB requires a common ground from Host to Device and for the special situations where ground must be isolated between these pieces of equipment, special isolation circuitry must be used (optocouplers?).
     
    #8
  9. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2
    The shell should be provided with a low impedance connection to signal ground and, in cases where
    a metal chassis is present, a means of accepting a metal EMI gasket clip.

    thats all I have on the subject

    seems it need to be connected to ground to avoid emi​
     
    #9
  10. colinb

    colinb Thread Starter Member

    Joined:
    Jun 15, 2011
    351
    35
    ian123,
    Thanks for all the info you posted!
     
    #10
  11. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2
    ok there are two things we are talking about one is the sheild and one is the ground.

    the sheild needs to be jioned to the ground through a low impedance but the ground is supplied by the host.

    so therefor the way i see it is the sheild and the host ground will be joined through a low impedance on the host side but on the device side if they dont share a common ground you would need to isolate the grounds.
     
    #11
  12. ian123

    ian123 Member

    Joined:
    Aug 24, 2011
    73
    2
    my pleasure
     
    #12
  13. KG1321

    KG1321 New Member

    Joined:
    Jun 17, 2015
    1
    1
    The old USB adapter cable for the IPOD used the shield as the ground (-) conductor.

    The ground is connected to the chassis ground through the "host" device.

    If lighting were to strike your host device would the device be "ungrounded?"
     
    #13
  14. ErnieM

    ErnieM AAC Fanatic!

    Joined:
    Apr 24, 2011
    6,511
    1,323
    There are always two devices involved when using the USB, least you just have one unconnected device and no bus. o_O

    In my PC (I assume a Dell is industry standard) the USB shield goes to the case of the PC. This is also common to power ground.

    On a battery powered device just ground the shield, you have no other choice of connections. My inclination is not to use any other components but solder, no caps or resistors, but I have no standard to quote on this practice.

    Should the case of two devices be of different but common potentials a USB galvonic isolator is necessary. Such are available. Such may not be cheap.
     
    #14
  15. jwd217

    jwd217 New Member

    Joined:
    Jun 19, 2012
    5
    1
    To be an effective EMI shield, it must be connected to ground and/or the negative power supply. The thing I see missing here is it must only be connected on one end. If it is connected on both ends, then current could flow through it and cause EMI. One of the wires inside the cable should be the negative power supply return. Most devices will have the power supply negative and ground connected to the chassi. If both devices are this way and both ends of the shield are connected to ground, then the shield will become a negative power current conductor.
     
    #15
  16. cowades

    cowades New Member

    Joined:
    Feb 6, 2012
    7
    1
     
    #16
  17. cowades

    cowades New Member

    Joined:
    Feb 6, 2012
    7
    1
    Since it's powered by the USB and not a separate power supply, some capacitor from your signal ground to the shield is probably the best choice. This should keep your circuit moving around with the shield, as it's potential moves, so nothing couples in or radiates out. The reason for so many different recommendations on how to connect the shield has to do with there being so many things a USB can connect to. An unshielded, bus powered plastic box is going to be different than a wall powered printer. There is no one answer, so we've got to think about ground and shield currents for each application.

    If you ever have to test this at an EMI test facility, take a pocket full of various capacitors.

    ground.jpg
     
    #17
    ErnieM likes this.
  18. elix-cz

    elix-cz New Member

    Joined:
    Jul 28, 2015
    2
    0
    About mine mind there is only one right solution for this question - connect shield and ground with this network:

    100nF || 100nF || (220nF + 3.3R) || (220nF + 3.3R) || 50R

    || … parallel + … serial

    Of course use SMD. X7R 50 V ceramics is ok. Don’t matter if you have or not something as conductive case. If yes, connect USB connector shield with case by spring (quite expensive) or by something else. A question of why you have to answer along : )) Regarding EMC obviously can be found much nonsense : ))
     
    #18
  19. elix-cz

    elix-cz New Member

    Joined:
    Jul 28, 2015
    2
    0
    USB.png USB.png Here is it with solved ESD protection - for high speed : ))
     
    Last edited: Jul 29, 2015
    #19
Loading...