27
\$\begingroup\$

I have gone through this answer on Superuser, while it provides good information, I still have a specific question from electrical/electronics engineering point of view.

Is it wrong and unsafe to connect the other end of the ESD wrist strap to the ground point on the wall socket?

What harm can it do and if it’s wrong, why?

I was watching a Youtube video about assembling computers by a user who seems to be a professional and he says he has been in the business for over 2 decades.

The video discourages connecting ESD strap to wall socket ground, it also says that when you connect ESD strap to metallic part of cabinet, remove the power plug from the PSU. However except a reasoning that you could get electrocuted with such a setup, no further technical explanation is given.

I have worked in an IT company that was TL9000 quality certified. This standard demands ESD protection deployment wherever necessary. All relevant test benches there had ESD mats connected to common ground. All terminals had a point to plug the ESD strap’s other end and all those were connected to ground and eventually going to the earth pit. So discouraging ESD strap connection to wall socket ground (that ultimately goes into the Earth pit) is a totally new concept to me.

\$\endgroup\$
  • \$\begingroup\$ It's probably not good for the ground or zero voltage sink to have essentially endless capacitance. As they are delighted to discharge safely large volt or amps at low rate. Analogously to amp hours of battery, the battery can provide large voltage or current for a certain amount of time, the wrist strap is like a small low amp hr battery but for ground. So your wrist strap shouldn't handle high wattage. \$\endgroup\$ – marshal craft Apr 1 at 14:39
  • 5
    \$\begingroup\$ Please clarify your comment - I don't understand it at all. \$\endgroup\$ – Daniel Apr 1 at 16:32
  • 1
    \$\begingroup\$ The purpose of the strap is not strictly to ground you to what you are working on but to control the speed of the static energy to ground. You want the semi truck to drive slowly through the residential neighborhood, you dont just want it to get to the other side as fast as it can. There is a resistance for this. Your question also applies to the chassis of the computer or product, it can be as incorrect to ground the strap to the product as it is to the ground on a wall socket. \$\endgroup\$ – old_timer Apr 1 at 17:04
  • \$\begingroup\$ Mine is connected to the socket ground- indirectly. It has a series resistor in the lead, and connects to a terminal on the mat on the bench, which is in turn grounded to the socket via another series resistor. \$\endgroup\$ – Phil G Apr 1 at 19:02
53
\$\begingroup\$

If your ground strap has a proper current limiting resistor, then you would be perfectly fine connecting it directly to ground.

If your ground strap does not have a resistor, then it greatly increases your chances of dying if you connect it directly to ground. If you don't have a resistor in your ground strap, and you touch a live wire while wearing your ground strap, then you will get full line voltage and current applied to your body - and you won't be able to break the ground connection.


Ground straps are there to drain and prevent a build up of static charge. You don't need a fast discharge.

A ground strap should have a certain resistance (usually 1 megohm or more.) That drains any charge on your body, but limits the current in the event you accidentally touch a live wire.

A 1 megohm resistance will drain the charge on your body quickly enough - your body's capacitance is only some few 10s of picofarads. If you were charged to 10000 V (can easily happen) then it would take less than 1millisecond to discharge your body through a 1 megohm resistor. Fast enough.

If you are grounded through a 1 megohm resistor, and touch a live wire (say, 220V ac) then a maximum current of 0.2 milliamperes will flow through your body. It might tingle if you are exceptionally sensitive, but it won't injure or kill you.

\$\endgroup\$
  • 15
    \$\begingroup\$ I might point out that if you grab a live wire, there still may be other, less resistive paths to ground, such as through your shoes/butt/chair, or hand resting on the metal table, etc. ESD straps are in no means a substitute for shock prevention/safety devices like GFCIs or isolation transformers. \$\endgroup\$ – shenles Apr 1 at 19:01
  • 3
    \$\begingroup\$ @Shenles: Nobody said they were a substitute for a GFCI or an isolation transformer. You use an ESD strap to discharge static electricity. There's safe ways to use them, and unsafe ways to use them. Using an ESD strap involves connecting your body to ground, potentially making you part of the electric circuit. There is no comparison between an ESD strap and safety equipment. It takes a considerable misunderstanding to mix the two. \$\endgroup\$ – JRE Apr 1 at 19:07
  • 6
    \$\begingroup\$ I get that, just wanted to clarify a bit since the last paragraph states that "It might tingle if you are exceptionally sensitive, but it won't injure or kill you.", which is definitely true, assuming no other paths to ground exist. Just wanted to help someone that might accidentally interpret that the wrong way. Apologies if it sounded like I was disagreeing with you in any way. \$\endgroup\$ – shenles Apr 1 at 19:09
  • 4
    \$\begingroup\$ Do they sell grounding wristbands without a resistor in them? That sounds pretty dangerous! Going to go double check that mine all have resistors now.... \$\endgroup\$ – Hearth Apr 1 at 21:06
  • \$\begingroup\$ "Using an ESD strap involves connecting your body to ground..." Depends upon your definition of ESD strap. The ESD wrist strap we bought from 3M has a very high resistance built-into the cord that is grounded at the far end, placing one's body very safely away from ground. But +1 for your answer. :-) \$\endgroup\$ – Mike Waters Apr 2 at 18:58
10
\$\begingroup\$

If your wrist strap has built-in resistance, there's nothing wrong with connecting it to mains ground or PSU chassis.

However, if your wrist strap is just a conductor and you happen to touch a live wire, the strap will provide a low-resistance path to ground. This will result in a very intense shock, compared to cases where the current has to go through significant resistance along its path (like your shoes/clothes) and thus is much smaller.

\$\endgroup\$
4
\$\begingroup\$

In addition to the caveat of having a resistor built-in so that you don't electrocute yourself if you happen to find a live wire, there's also the question of whether the grounding point that you've found is actually ground.

Consider old construction that's been upgraded a few times and grandfathered into the electrical code. It's grossly illegal to build anything new like that, but it's allowed to stay because it was to code when it was built. So you may have anything inside the walls depending on its age, some of which don't have a ground wire at all, and may be difficult to tell at a glance which is live and which is neutral.

Technically, it's illegal in most codes to create a "bootlegged ground" - that is, to connect the neutral and ground terminals of an outlet both to neutral - but it's frighteningly common anyway, and a cheap 3-prong tester won't show it as anything wrong. But the load current makes the neutral wire not ground anymore, and if it were to break, then the end attached to the outlet would suddenly become live via the load. And if it's bootlegged, then the "ground" also becomes live along with it.

Now consider a polarity-reversed bootlegged ground. If it's difficult to tell which wire is what and it takes more effort to test them than the sparky can give, then you have a 50/50 chance of having it backwards. Now the bootlegged "ground" is connected directly to live! And a cheap 3-prong tester will still think that it's okay! (because the neutral and ground terminals are still connected to each other, which is all it cares about)

Then you plug your grounding strap into what you think is ground...

\$\endgroup\$
  • \$\begingroup\$ TN-C mains is the term you are looking for... indeed infamous for failing with the ground live! \$\endgroup\$ – rackandboneman Apr 3 at 0:26
3
\$\begingroup\$

The issue is your mains ground may be rubbish.

Mains grounding starts at the Ufer-in-foundation, grounding rods, or water pipe connection (Oh, the things that can go wrong with water pipe connections!!)

From there, it goes to your Grounding Electrode System, the above plus the wiring which carries the grounding to the main service panel.

In the main panel is the Neutral-Ground equipotential bond, which clamps neutral near earth and assures the transformer secondary doesn't float at high voltages due to capacitive coupling or leakage. It also gives fault current on ground a way back to neutral. This thing breaking can cause problems.

From the main panel, safety ground is carried onward to any subpanels and to the individual branch circuits. Then, up each branch circuit one by one, each connection daisychaining off the next.

Where can this go wrong? Oh, so many places.

  • The grounding source may have an issue because the water company installed a plastic smart meter.
  • The grounding electrode cables may be disconnected or corroded.
  • The ground wire from the panel may be broken.
  • The ground wire for a subpanel circuit may have been landed on a neutral bar, because that's allowed in the main panel (since it's just a funny way to do the equipotential bond).
  • The ground wire may be broken.
  • A group of receptacles may have islanded grounds, when a 2-wire circuit is extended with 3-wire (w/ground) cable; in the addition area, people typically connect grounds to each other, but don't bring the grounds back to the main panel. In this island of grounds, any ground fault lights up all the grounds in the island, because it can't carry current back to the main panel and the N-G equipotential bond.

  • The ground may not even be present at that receptacle. Pre-1960s wiring didn't have ground wires. Sometimes people just slap a 3-prong outlet to replace a 2-prong, so they can plug in their equipment, and leave the ground to dangle. It's the "islanded grounds" problem but just within that receptacle.

  • Ditto, but protected by a GFCI, which is a legal way to leave 3-prong outlets ungrounded. This requires a sticker stating "no equipment ground", but people tear off the sticker because it's ugly. Anyway, this "ground" is useless, but at least the circuit is less likely to kill you.

  • Bootlegged ground. Sometimes for lack of ground, people connect ground to neutral so that 3-lamp testers will "pass" the receptacle. This is nasty, because if the neutral wire breaks, the ground is energized with mains voltage.

  • Someone installing a workshop subpanel off an old NEMA 10 (hot-hot-neutral) circuit that went to a dryer or range. No grounds here, so the whole subpanel's grounds are either islanded or bootlegged!

But yeah, if your grounding is tip-top, then grounding your ESD mat/strap is the right thing provided you do it through a 1-megaohm resistor. The point of that is so if the worst happens and either the ground is energized or you touch mains, current flow is impeded below dangerous levels.

\$\endgroup\$
  • \$\begingroup\$ The OP didn't say which country he/she was in, and the rest of the world doesn't follow the USA's codes on electrical wiring. None of your description in the first paragraphs applies in the UK, for example. \$\endgroup\$ – alephzero Apr 3 at 15:12
  • \$\begingroup\$ @alephzero you scared me and made me go look! It's a bit unfair since UK has a lot of legacy work out there that's a little different, and you may have seen some of that, or if they underground your power they may let you use the metal supply conduit as earth. But yeah, modern best practice in both UK and EU conforms pretty exactly to what I said. They just call it different things or conceputalize it differently. \$\endgroup\$ – Harper Apr 3 at 16:44
1
\$\begingroup\$

ESD Straps have to be connected to ground. The electrostatic energy of your body shall be discharged to ground instead to your device under test. Be sure to use correct (certified) ESD straps as they have a defined resistance. Inappropriate ESD straps with too less or zero resistance are a safety risk. If you would touch a voltage source a current would flow through your body to ground.

\$\endgroup\$
  • 9
    \$\begingroup\$ They don't have to be connected to ground. That's the easiest (and often best) way. You can connect your ESD strap to the ground of an unplugged PC to work on it. That keeps you and the PC at one level. You still have to be careful when you pick up parts to be installed. People who work on PCs often do it this way because they don't have an ESD protected work bench. You don't have that if you are fixing a PC in an office workspace, for example. \$\endgroup\$ – JRE Apr 1 at 11:10

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.