Grounding ESD mat

[Antlab]

Hi all.
I recently bought my first ESD mat. It is a complete kit, with rubber mat, cord and wrist band.
Now the question is where to connect the cord to the ground. I read many and contrasting suggestions on the net.
Some say to connect it directly to the ground hole of the wall outlet. Other reply that this is a dangerous method, if something goes wrong with the house wiring. Other still suggest to connect the cord to the chassis of an instrument or PC.
Could you please try to clarify to me the subject, in order to avoid frying some components or worse?
Thanks

A.

[Ajahn Lambda]

The PC (and most instrumentation) chassis are both, almost always, connected to earth ground.


I agree that connecting it to the neutral wire in a household wiring system could be dangerous, albeit rare and unlikely to cause a problem, but an earth ground should be fine.  It's the way almost all of the ESD mats I've used have been connected, and the only line in a mains system guaranteed to always be connected to earth (literally, the Earth).


You could go out of your way to connect it to another, separate earth ground, such as a cold water pipe or a new embedded copper grounding rod, but I think that would be a bit of overkill.

[ejeffrey]

The ground pin of your ESD mat should be connected to the ground pin on the electrical circuit that feeds your workbench equipment.  You can do this connection any way you like: by plugging it into a provided ground terminal on an equipment chassis, or by directly connecting it to the ground wire.  Several manufacturers make 'plug in' ground terminals that plug into an electrical outlet and have several 1 megaohm terminals to ground for connecting ESD mats, wrist straps, and anything else.  These aren't necessary for functionality or safety, but you may find them convenient.

[cybergibbons]

The 1M? resistor thing is bollocks IMO. Half of the equipment on my bench has a grounded case. If I'm going to be playing with mains equipment, I keep one hand hand "behind my back" (really I just don't use it, rather than actually put it behind my back). If I feel like killing myself, I'll touch ground with one hand and line with the other...

[IanB]

It is as well to check that the ground or earth connection you are relying on is really grounded. In my house I have one power socket where the ground pin is floating at 20-40 V higher than true ground. It's become a little puzzle to solve at some point to determine where the fault lies. In the mean time it is not a socket I can trust with 3-pin plugs.

[Antlab]

It is as well to check that the ground or earth connection you are relying on is really grounded. In my house I have one power socket where the ground pin is floating at 20-40 V higher than true ground.

Hi IanB,
how did you measure those voltage values?
And do you use some ESD device?

A.

[Ajahn Lambda]

It is as well to check that the ground or earth connection you are relying on is really grounded. In my house I have one power socket where the ground pin is floating at 20-40 V higher than true ground. It's become a little puzzle to solve at some point to determine where the fault lies. In the mean time it is not a socket I can trust with 3-pin plugs.

Check the socket's connections first, then look to the neutral bus in the breaker box.  You're looking at some potentially serious problems, no pun intended.

[ejeffrey]

The 1M? resistor thing is bollocks IMO. Half of the equipment on my bench has a grounded case. If I'm going to be playing with mains equipment, I keep one hand hand "behind my back" (really I just don't use it, rather than actually put it behind my back). If I feel like killing myself, I'll touch ground with one hand and line with the other...

Safety is only part of the reason for the 1 MOhm resistor.  While it eliminates one possible mechanism for injury it is in no way sufficient to protect you from live equipment.  The other purpose of the 1 Mohm resistor is to limit the discharge current and thus reduce the risk of ESD damage.

[Monkeh]

It is as well to check that the ground or earth connection you are relying on is really grounded. In my house I have one power socket where the ground pin is floating at 20-40 V higher than true ground. It's become a little puzzle to solve at some point to determine where the fault lies. In the mean time it is not a socket I can trust with 3-pin plugs.

Most likely you've got a damaged cable or a bad connection at one end and what you're seeing is capacitive coupling.

[IanB]

Hi IanB,
how did you measure those voltage values?
And do you use some ESD device?

Check the socket's connections first, then look to the neutral bus in the breaker box.  You're looking at some potentially serious problems, no pun intended.

Most likely you've got a damaged cable or a bad connection at one end and what you're seeing is capacitive coupling.

I measured the voltage with the normal AC voltage range on a multimeter. The ground on this outlet is at higher voltage than the neutral on the same outlet, and higher than the ground on other good outlets.

This particular outlet is on a circuit with just one outlet and one canned light in the ceiling. The ground on all other outlets in the house is good.

The ground connection to the terminal on the socket is good, so the problem lies upstream. It is hard to check that connection at the neutral bus in the breaker box because of the difficulty identifying the exact cable entering the box, and because the breaker box is in any case a rat's nest of cables. However, I can probably figure it out given a little time and patience as I know which breaker the hot wire is attached to. I just need to trace back through the tangle of wires and see where it enters the box. Before I do that I will check out the wiring at the canned light.

It could well be capacitive coupling that I am seeing. But I don't have a meter with a low-Z range that might help. Possibly I will shut off the main service breaker and then do a resistance test between ground and neutral on the problem outlet. That at least will tell me if the ground is open circuit or not.

[cybergibbons]

The 1M? resistor thing is bollocks IMO. Half of the equipment on my bench has a grounded case. If I'm going to be playing with mains equipment, I keep one hand hand "behind my back" (really I just don't use it, rather than actually put it behind my back). If I feel like killing myself, I'll touch ground with one hand and line with the other...

Safety is only part of the reason for the 1 MOhm resistor.  While it eliminates one possible mechanism for injury it is in no way sufficient to protect you from live equipment.  The other purpose of the 1 Mohm resistor is to limit the discharge current and thus reduce the risk of ESD damage.

When I last looked into this, I couldn't find any research to back that up the "limit the discharge current" reasoning. The 1MOhm resistor theoretically limits the current at mains to less than 0.5mA, but I can't find a single death attributed to a wrist strap.

[Monkeh]

Hi IanB,
how did you measure those voltage values?
And do you use some ESD device?

Check the socket's connections first, then look to the neutral bus in the breaker box.  You're looking at some potentially serious problems, no pun intended.

Most likely you've got a damaged cable or a bad connection at one end and what you're seeing is capacitive coupling.

I measured the voltage with the normal AC voltage range on a multimeter. The ground on this outlet is at higher voltage than the neutral on the same outlet, and higher than the ground on other good outlets.

This particular outlet is on a circuit with just one outlet and one canned light in the ceiling. The ground on all other outlets in the house is good.

The ground connection to the terminal on the socket is good, so the problem lies upstream. It is hard to check that connection at the neutral bus in the breaker box because of the difficulty identifying the exact cable entering the box, and because the breaker box is in any case a rat's nest of cables. However, I can probably figure it out given a little time and patience as I know which breaker the hot wire is attached to. I just need to trace back through the tangle of wires and see where it enters the box. Before I do that I will check out the wiring at the canned light.

It could well be capacitive coupling that I am seeing. But I don't have a meter with a low-Z range that might help. Possibly I will shut off the main service breaker and then do a resistance test between ground and neutral on the problem outlet. That at least will tell me if the ground is open circuit or not.

You are assuming that you have a TN system. That can easily be a faulty assumption.

[SeanB]

The ground connection to the terminal on the socket is good, so the problem lies upstream. It is hard to check that connection at the neutral bus in the breaker box because of the difficulty identifying the exact cable entering the box, and because the breaker box is in any case a rat's nest of cables. However, I can probably figure it out given a little time and patience as I know which breaker the hot wire is attached to. I just need to trace back through the tangle of wires and see where it enters the box. Before I do that I will check out the wiring at the canned light.

It could well be capacitive coupling that I am seeing. But I don't have a meter with a low-Z range that might help. Possibly I will shut off the main service breaker and then do a resistance test between ground and neutral on the problem outlet. That at least will tell me if the ground is open circuit or not.

If you want to identify the breaker put a radio on it and turn up the volume to where you can hear it at the breaker box. Then turn off breakers until the radio turns off. If the ground is floating, but it is connected at the plug socket end then it is broken or not connected in the wall at some junction box. Easiest to trace with a tone generator and probe to find the wire in the wall.

[alm]

When I last looked into this, I couldn't find any research to back that up the "limit the discharge current" reasoning. The 1MOhm resistor theoretically limits the current at mains to less than 0.5mA, but I can't find a single death attributed to a wrist strap.

So there were zero fatalities out of how many instances of people using a wrist strap with a low-impedance connection to ground? I've never seen an ESD wrist strap without resistor, so it's not surprising to me that you won't find many fatalities caused by a lack of resistor.

[IanB]

You are assuming that you have a TN system. That can easily be a faulty assumption.

No, not assuming anything. I am in the USA where homes are wired according to the diagram on this page (confirmed by inspection) -- scroll down to section headed "The Ground Wire".

In short, the neutral and ground wires from every receptacle are taken back and connected to the neutral bus bar in the main service panel where the circuit breakers are. If the circuit is wired correctly there should be a near zero resistance path between the ground and neutral at every point in the system via the bus bar.

I may for interest take a picture of the panel to show what it looks like inside. It is rather foreign to anyone coming from the UK.

Evidently in this one particular faulty circuit the ground wire has a broken or poor connection somewhere on the way back to the service panel. Since there are only two receptacles on this circuit (a wall outlet and a canned ceiling light) my suspicion lies with the previously mentioned canned ceiling light.

[IanB]

If you want to identify the breaker put a radio on it and turn up the volume to where you can hear it at the breaker box. Then turn off breakers until the radio turns off. If the ground is floating, but it is connected at the plug socket end then it is broken or not connected in the wall at some junction box. Easiest to trace with a tone generator and probe to find the wire in the wall.

I do know which breaker it is. I have mapped out all the breakers and the receptacles they feed for future reference. But the breaker in the service panel identifies the hot wire. It is not trivial to trace that hot wire back to the entering cable and then follow the ground conductor to the neutral bus bar, given the dense jumble of wires in the cabinet.

That said, I suspect the grounding problem lies somewhere between the wall outlet and the service panel. That tone generator sounds nifty, but I don't think it is worth it for a single use. I will continue with visual inspection of the parts I have not investigated yet.

[Monkeh]

You are assuming that you have a TN system. That can easily be a faulty assumption.

No, not assuming anything. I am in the USA where homes are wired according to the diagram on this page (confirmed by inspection) -- scroll down to section headed "The Ground Wire".

Okay, if it's been inspected and confirmed it's a TN, fine. Just don't assume that every property is wired in that manner. There are a multitude of different systems.

I may for interest take a picture of the panel to show what it looks like inside. It is rather foreign to anyone coming from the UK.

I doubt there's anything unusual except the short-sighted antiquity of the design..

I do know which breaker it is. I have mapped out all the breakers and the receptacles they feed for future reference. But the breaker in the service panel identifies the hot wire. It is not trivial to trace that hot wire back to the entering cable and then follow the ground conductor to the neutral bus bar, given the dense jumble of wires in the cabinet.

Quality.

[SeanB]

[

That said, I suspect the grounding problem lies somewhere between the wall outlet and the service panel. That tone generator sounds nifty, but I don't think it is worth it for a single use. I will continue with visual inspection of the parts I have not investigated yet.

Luckily I have the tone set for tracing network and phone cabling. Quite useful, even though the tracer has needed repair a few times, soldering is not the best in these, and they do tend to get a lot of flexing on the board in use. Not supposed to be serviced, but possible to open eventually. Even replaced the little speaker when that died with an open coil.