4
\$\begingroup\$

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

  • wrist strap
  • antistatic mat
  • common point ground

Wrist strap and antistatic mat both should grounded to the common point ground.

ANSI/ESD S6.1 recommends a non-resistor ground cord [...] to ground worksurfaces. However, the cord may have a 1 megohm resistor for non ESD purposes.

You might think: just put a resistor between Common Ground and mains earth...

ANSI/ESD S6.1, section 6.4.2 recommends: The resistance [...] from the common point ground to the AC equipment ground shall not be greater than 1 ohm.

Both plans below don't follow ANSI/ESD S6.1, because i think the path between antistatic mat and mains earth should NOT be a low impedance path.

And if there should be a resistor between an dissipative surface (like an antistatic mat) and mains earth for user safety, then there should be a resistor between a conductive surface (unpainted inside of a PC case) and mains earth too.

enter image description here

In theory, is one plan more preferable over the other?

Plan A: only 1MΩ between wrist strap and common ground. Thus discharging takes less time. Therefore (maybe?) less chance of a potential difference between wrist strap and the PC (with ESD sensitive hardware inside) and less chance of an ESD event.

Possible disadvantage of plan A: because the user is constantly charging himself (by moving) and discharging to the common ground, the common ground will be bouncing more than it would in plan B. (Please correct me if i'm wrong.)

Both plans are easy to execute.

  • Plan A
  • Plan B The yellow marked resistors are inside the grounding plug.
\$\endgroup\$
  • 3
    \$\begingroup\$ Please tell my how the case can become a shock hazard when connected to safety earth? Who told you to add a resistor in series? \$\endgroup\$ – pipe Mar 29 '17 at 20:37
  • \$\begingroup\$ Every conductor connected directly to mains earth can become a shock hazard. That's why a wrist strap should be grounded via a resistor. To protect the user. Plan B just has 2 resistors in series for extra protection. \$\endgroup\$ – Marty Mar 29 '17 at 20:52
  • 2
    \$\begingroup\$ Think about that for a moment. It means that every appliance you own that is connected to earth is a hazard. This is obviously not the case, or else it would not be legal. Your PC is connected to earth during normal operation. So is everything else with a metal case. \$\endgroup\$ – pipe Mar 29 '17 at 21:11
  • 3
    \$\begingroup\$ Short answer. To protect yourself if you accidentally touch anything with a high voltage. \$\endgroup\$ – pipe Mar 29 '17 at 21:17
  • 2
    \$\begingroup\$ How, exactly, does a faulty GFCI cause a grounded case to become a shock hazard? \$\endgroup\$ – Finbarr Mar 29 '17 at 22:27
2
+50
\$\begingroup\$

The reason to ground the PC case has nothing to do with ESD: it's a measure against PC case going live. If the case is grounded, a live wire touching the case will create a prompt short circuit, which will (hopefully) trigger the circuit breaker and instantly remove the power.

Grounding yourself via wrist strap has nothing to do with your safety: it actually makes things worse for you. If you touch a live wire while not wearing the strap, the current you'll get will be limited by your body's capacitance and leakage via your shoes etc. Should that happen while you're wearing a grounded wrist strap (without a resistor), that current will only be limited by your body's resistance, which is not that big at high voltage. That's why you need a resistance in series with your strap.

Now, it is true that your PC case which is grounded without any resistance remains a shock hazard: if you manage to touch it with one hand, and a live wire with the other one, you'll get a bad shock. But this is quite improbable and requires you to do something very bad (touch a live wire) in the first place. A case going live, on the other hand, is not an impossible event: a simple bump on the side may deform it hard enough to touch something on the power supply PCB. Should that happen, you'll have that same very bad stuff right in front of you and without any warning: a 1MOhm resistor can withstand 230V indefinitely.


Edit: after the exchange in comments it became clear that you don't need protective Earth. In that case, I would go for Plan A, as it has more wiring behind the resistor and less wiring connected directly to Earth. Those resistors are installed for your safety as explained above, so it's clearly better to have more touch-safe wiring on your desk.

\$\endgroup\$
  • \$\begingroup\$ Thanks. How can a case go live, if the PSU is not connected to mains power? Completely disconnected from hot and neutral. \$\endgroup\$ – Marty Apr 4 '17 at 18:07
  • \$\begingroup\$ If the PSU is not connected to mains power there's no need to connect the case to Earth, at least from security perspective. Is you PC powered by a SELV? \$\endgroup\$ – Dmitry Grigoryev Apr 5 '17 at 8:19
  • \$\begingroup\$ Laptop: class II power supply with connection to mains earth. There is a 1MΩ resistor between the GND of the power jack and mains earth. That connection to mains earth is not a return path for current. It's a reference and electrostatic bleeder. Desktop PSUs often are not class II. \$\endgroup\$ – Marty Apr 5 '17 at 16:01
  • \$\begingroup\$ I want to make a connection between case and earth for ESD protection. Not when the PC is powered on. Use case: maintenance (installing/replacing hardware parts) inside an ATX case. Power cable disconnected. The hot and neutral prongs from the ATX PSU are completely disconnected from the wall outlet. \$\endgroup\$ – Marty Apr 5 '17 at 16:14
  • \$\begingroup\$ Ok, I see now. When you said "PC", I really pictured a 220V appliance with ATX case. You don't need protective Earth then. I have updated my answer. \$\endgroup\$ – Dmitry Grigoryev Apr 5 '17 at 16:15
7
\$\begingroup\$

If by plan B you mean to add a resistor to the mains ground, then no plan B is not an option. If you add a resistor in series you will have two problems:

1) It affects the saftey of the earth mains, if you size the resistor wrong in the event of a fault you could have an issue with the resistor blowing out or creating common mode voltage in the case

2) You create common mode voltages and cause the whole chassis ground to bounce depending on the amount of current flowing throw chassis ground which could cause problems for the supply and circuitry (if your ground bounces up and down by 10's of mV it might create digital noise, which could be possible depending on the resistor on earth ground.) You would also increase the potential between the PC and other devices creating noise problems.

You can put resistances in series with the wrist strap and the anti static mat, but not the PC. Anti static mats (all the mats I've seen already have MΩ's of resistance in the mat material itself. Most wrist straps also have a 1MegΩ resistor built in.

If your not playing around with anything over 30V you probably don't need to worry about any of this anyway. Keep the mains ground as low resistance as possible as required, and put resistors (that are probably already there, you can tell with a good DMM (voltmeter)) in between the mat and ground and the strap and ground. Shown here:

enter image description here

If look at this question the diagram would look like this, the grounds would be connected in the building wiring:

enter image description here

In the above image the wrist strap is connected to the mat, if it's the kind with a 1MΩ resistor in it, it doesn't matter where you connect it, if you connect it to the chassis (shown as wrist strap1) , it still gets grounded. If you connect the wrist strap to the mat it gets grounded, and if you connected it to the mats ground (shown as wrist strap2) it still gets connected, each time with at least 1MΩ to ground or any other potential that might affect those who might grab onto earth mains. If you take a 1MΩ resistor and poke it into an electrical outlet it's not going to kill you, but a fork might.

enter image description here

\$\endgroup\$
  • \$\begingroup\$ We are talking about static charge. The PC case (and thus also the internal electronics) will be grounded to the common ground, which is grounded to mains earth. Line and neutral are nowhere in this story. The PC case is conductive, so a hard link to mains earth, while you're installing hardware parts, could be dangerous. You'll be touching that case. That seems like a potential shock hazard to me. \$\endgroup\$ – Marty Mar 29 '17 at 21:07
  • 1
    \$\begingroup\$ The PC case is already grounded to mains earth, in fact every product with AC mains is required by law to have chassis ground connected to mains ground to shunt potential shorting currents. The idea is, if you do have a short, it doesn't start a fire and the current returns to earth not to a user that is touching the chassis so the chassis is always at 0V, if you install a resistor (by splicing the cord and installing a resistor) you will have much bigger problems in the event of a short. \$\endgroup\$ – Voltage Spike Mar 29 '17 at 21:14
  • \$\begingroup\$ Yes, but you're not constantly touching these devices. When you're working on a device (whatever product it may be) you disconnect it from AC mains and thus from mains earth. When you're working on it, while it's still connected to mains earth, you should protect yourself. \$\endgroup\$ – Marty Mar 29 '17 at 21:18
  • 1
    \$\begingroup\$ You already have, if you have a 1MOhm resistor on the strap and you touch AC mains the maximum current you'll have is 1.2mA. \$\endgroup\$ – Voltage Spike Mar 29 '17 at 21:19
  • 1
    \$\begingroup\$ None of the diagrams in question 274820 suggest that you should put a resistor in series with a device's ground to mains ground. They do suggest a different grounding strategy than the one I've suggested in 'plan C', because they have the strap attached to the mat which would give additional protection between the mat and the strap. The most important thing is: between the mat and ground there needs to be 1MΩ and between the strap and ground there needs to be 1MΩ. If the PC is sitting on the mat, you have 1MΩ between the chassis and ground. \$\endgroup\$ – Voltage Spike Mar 29 '17 at 21:35
1
\$\begingroup\$

Without using the mat case:

enter image description here

There is no meaning for adding the R2 resistor. It is just going to make the discharge slower and it does not add any additional protection function.

When using the mat case:

According to that the discharge resistors of the mat and the wrist strap are already built in, the point of discussion should be around the PC's discharge resistor connected to the ground.

Refereing to plan B, in case you have the PC on your mat (which is the normal thing to do) the PC is already connected to ground through the mat and you dont have to add the discharge resistor between the PC and Ground. Look at the following figure.

enter image description here

If your PC is not on the mat you should not touch the PC's hardware unless the PC is grounded which leads us to the first case.

Additional info:

The equivalent circuit of the setup in this video with roughly estimated values of the resistors is as following:

enter image description here

\$\endgroup\$
  • \$\begingroup\$ Thanks. Why not plan A? Plan A: only 1MΩ between wrist strap and ground-plane of the ESDS (ESD sensitive) hardware... So, less time to reach equipotential. So, less change for ESD on the ESDS hardware. \$\endgroup\$ – Marty Apr 1 '17 at 10:09
  • \$\begingroup\$ Like you mentioned at the beginning of your question, there are two concepts to protect equipments from ESDs. The first one is called bonding. Here, no grounding is required and rather than eliminating the static charge, it equelizes the charges between the user and the chassis of the PC thus reducing the risk of static electricity. The second one (which we are discussing here) is called grounding. In this case grounding is essential to discharge ESD to ground. Thus you should care about the resistance between the used equipments and ground, not between the wrist strap and other equipments. \$\endgroup\$ – Macit Apr 1 '17 at 10:38
  • \$\begingroup\$ Indeed, according to ANSI/ESD S6.1 (i could only find the 2005 revision) there a 2 concepts. (1) Grounding a Common Point Ground to earth. (2) Equipotential bonding. \$\endgroup\$ – Marty Apr 1 '17 at 15:01
  • \$\begingroup\$ Regarding to the 1st concept, ANSI/ESD S6.1, section 6.4.2 recommends: "The resistance of the conductor from the common point ground to the AC equipment ground shall not be greater than 1 ohm." Like plan B. On the other hand, this is exactly like plan A. \$\endgroup\$ – Marty Apr 1 '17 at 15:01
  • \$\begingroup\$ Please look at my edited answer to see the equivalent circuit of the setup that you have shared. \$\endgroup\$ – Macit Apr 1 '17 at 18:10
-1
\$\begingroup\$

An Anti-static mat and wrist strap when used for ESD have the job to not allow a potential to be develop across you body or the mat. For example by rubbing against the chair during the winter.

\$\endgroup\$
  • \$\begingroup\$ After your edit the answer now does not answer the question in any way. \$\endgroup\$ – Transistor Dec 30 '17 at 10:04

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.