4
\$\begingroup\$

A simple EPA (ESD Protected Area) consists of 3 things:

  • wrist strap
  • antistatic mat
  • common point ground

When doing field service (installing/replacing hardware parts) the ESD sensitive part goes straight into a static shielding bag and the replacement part comes out of a static shielding bag and goes straight into the PC. You only need a wrist strap to bond yourself to the PC.

enter image description here

On a desktop PC you can connect to the conductive case. Laptops on the other hand often have a non-conductive plastic housing.

Bonding to the ground pin of an USB port is an alternative option. This means: reaching equipotential with the motherboard's ground-plane via the USB port.

enter image description here

  • Can the USB port and/or motherboard handle the current spike? ESD damage?
  • What about user safety?
  • Should there be a resistor between PC and wrist strap?

an USB port
(source: usb3.com)

\$\endgroup\$
5
  • \$\begingroup\$ Does your wrist strap have a a series resistor built in (1 Meg for example)? \$\endgroup\$
    – jagjordi
    Apr 2, 2017 at 9:44
  • 1
    \$\begingroup\$ Many laptops aren't grounded. There's nowhere for the discharge to go - you are just equalizing your charge with the charge from the laptop. \$\endgroup\$
    – JRE
    Apr 2, 2017 at 9:48
  • \$\begingroup\$ @jagjordi No, my wrist strap does not have a series R built in. Often, the R is inside the coil cord between wrist strap and whatever you connect it to. \$\endgroup\$
    – user127725
    Apr 2, 2017 at 10:04
  • \$\begingroup\$ @JRE You're right. I edited my question. You are discharging yourself to the laptop until all charge is distributed evenly over all things that are bonded together. Without resistor, this happens almost instantly, resulting in a current spike. I would guess an USB port can handle this, but i really want to be sure before blowing out the USB port and/or causing ESD damage to the motherboard. \$\endgroup\$
    – user127725
    Apr 2, 2017 at 10:07
  • \$\begingroup\$ USB ports will have TVS protection diodes, specifically because the normal user is more than likely to touch one of the ports during use probably causing an ESD strike in the process. Same goes for plugging in any device - e.g. a USB memory stick will have some static charge which needs to be equalised when it is plugged in. \$\endgroup\$
    – Tom Carpenter
    Apr 2, 2017 at 14:09

3 Answers 3

Reset to default
7
\$\begingroup\$

You should NOT connect your anti-static gears to USB ground pin. The USB ground is SIGNAL GROUND. An ESD event can surge along the signal ground and elevate its potential relative to other ICs inputs (called sometimes as "ground bounce"), and it can damage internal ICs. If you don't have Earth ground on your workspace, at most you should use SHIELD of the USB connector, but not GND. The SHIELD usually is properly designed to re-direct and absorb the surge without exposing internal low-votage CMOS ICs to dangerous voltages.

\$\endgroup\$
10
  • \$\begingroup\$ +1 Good point about causing damage to the system ground. If you get two ESD spikes in a row then who knows what will happen on the second one?! \$\endgroup\$
    – user98663
    Apr 4, 2017 at 7:07
  • \$\begingroup\$ @ali-chen Thanks. I edited my question. Is the GND pin not just the ground for the power? Then D- should be the signal ground? You can even buy special USB Ground Adapters, which only connect to the GND pin, if i'm correct. \$\endgroup\$
    – user127725
    Apr 4, 2017 at 14:09
  • \$\begingroup\$ @Marty, D- is not any kind of ground at all. Go read up about how USB works. \$\endgroup\$
    – user98663
    Apr 4, 2017 at 14:30
  • \$\begingroup\$ I fail to see how an ESD into the USB ground can damage anything. The discharge will essentially go through the ground plane and the laptop chassis. \$\endgroup\$
    – Dmitry Grigoryev
    Apr 4, 2017 at 14:41
  • \$\begingroup\$ @DmitryGrigoryev You think it's perfectly safe? \$\endgroup\$
    – user127725
    Apr 4, 2017 at 14:53
0
\$\begingroup\$

Your laptop probably has an isolated power supply. Does it have this symbol?

enter image description here

This means everything inside the laptop is isolated from the power outlet, including USB ground. Connecting your wrist strap to will not connect it to Earth, so you will still be able to produce ESD with equipment not connected to the laptop.

This is in contrast with desktop PCs whose cases are connected to protective Earth:

enter image description here

Connecting your strap to their cases will work as expected: you will be at the same potential with all grounded equipment in your lab.

\$\endgroup\$
6
  • \$\begingroup\$ Indeed. Connecting your wrist strap to the (GND pin of an) USB-port will not connect it to earth. The wrist strap gets bonded to the laptop, correct? The charge on my body will distribute over my body and the laptop. Result: equipotential between me and the laptop. \$\endgroup\$
    – user127725
    Apr 4, 2017 at 14:51
  • \$\begingroup\$ @Marty Yes, that is correct. You won't be able to shock your laptop if you're connected to it via USB GND. \$\endgroup\$
    – Dmitry Grigoryev
    Apr 4, 2017 at 14:59
  • \$\begingroup\$ Can an USB port handle the charge coming from the wrist strap? The moment you plug in your wrist strap into an USB port there will be a current spike. \$\endgroup\$
    – user127725
    Apr 4, 2017 at 15:10
  • \$\begingroup\$ Obviously yes, since it can handle those spikes when you plug regular USB devices. If you're pedantic, use USB shield instead of USB GND, but a continuity test will almost certainly tell you they are the same thing. \$\endgroup\$
    – Dmitry Grigoryev
    Apr 4, 2017 at 15:22
  • \$\begingroup\$ Thanks. But, generally, there isn't a 4kV (or more) potential difference between these regular USB devices and the USB port, right? \$\endgroup\$
    – user127725
    Apr 4, 2017 at 16:45
0
\$\begingroup\$

ESD will take the path of least impedance, not simple DC resistance. It is, after all, a transient high speed event. So while the ground pin on the USB port may have DC continuity, it is not the ideal path to dissipate electric charge.

The metal shroud on a USB A connector provides a low impedance path to the chassis and/or chassis plane. The chassis connections are much more tolerant to ESD events, it's what they are designed for.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.