Venue outlets have open ground. What does that mean?

[blastoff99]

Correct on the GFIC comment. My understanding is that some local electrical codes specify using a GFCI outlet where a ground is not available in the outlet box. It's not ideal, but provides more protection than a standard grounding style outlet with no ground connected.

Two things (from a non-electrician): installing a 3-prong outlet in an ungrounded box may or may not be safe (the open ground thing), but I can’t imagine it’s legal anywhere that inspections occur.

....

Am not remotely familiar with electrical codes for buildings that are open to the public, but I can tell you that installing GFCI breakers in the box and then using "grounded" outlets with a 2 wire system is enough to keep the city inspector happy here. My insurance company was also totally happy with it. Huh.

That said, i have real electricity in the laundry room, so that's where my computer printer is. If I want to plug in my real bass gear, that's where it's going too.

For whatever that's worth.

 

[Old Garage-Bander]

I'm good. The grounding is correct. But I appreciate your adding your comment, as I made no mention of the grounding. Thanks.

Meaning then that your iso transformer breaks the ground?

 

[Old Garage-Bander]

If you plug the amp into a GFCI then anything plugged into that amp is protected from power getting out of it.
Ditto if the PA is in a protected circuit anything coming from a mic would be protected.
An unprotected PA with a fault could shock you via the mic through your guitar but not trip a GFCI your amp was plugged into though.
This is why it seems so odd to me to put a whole lot of GFCI's throughout a house on selected outlets instead of a single unit on the mains feed into the house. I guess the benefit is when you have a dodgy toaster it only trips that circuit instead of your whole house though.

My read (OP please correct me if I'm wrong) was that he was asking about a GFCI to plug his GUITAR into, he didn't say amp. We can assume he meant amp but have no way of knowing without a confirmation. Of course there are GFCI's that he can plug his amp into, that's pretty much been the theme of this whole thread. This is why I went the other way.
I envisioned something with 1/4" jacks and the ability to detect an A.C. power imbalance on the instrument cable to protect from shock by shutting down equipment. The answer to that is no, and no. There is no such device, (that I am aware of) and it doesn't make sense (IMHO).

 

[TomB]

Am not remotely familiar with electrical codes for buildings that are open to the public, but I can tell you that installing GFCI breakers in the box and then using "grounded" outlets with a 2 wire system is enough to keep the city inspector happy here. My insurance company was also totally happy with it. Huh.

That said, i have real electricity in the laundry room, so that's where my computer printer is. If I want to plug in my real bass gear, that's where it's going too.

For whatever that's worth.

Yes, same here. I should have specified “non-GFCI” 3-prong outlets where the green screw is just left unconnected is not allowed here. …but it is done DIY, I’ve found.

 

[blastoff99]

Yes, same here. I should have specified “non-GFCI” 3-prong outlets where the green screw is just left unconnected is not allowed here. …but it is done DIY, I’ve found.

It meets code here - as long as there's GFCI breakers. As for the outlets, the green screw is just chillin'.

 

[TomB]

It meets code here - as long as there's GFCI breakers. As for the outlets, the green screw is just chillin'.

Agreed - as long as the circuit contains a GFCI breaker, and as of this year’s local code update that GFCI breaker is now to be labeled “no mechanical ground”. I’ve encountered them in non-protected circuits as well. No electrician would do that.

 

[Lostgallifreyan]

I am not familiar with the Rane document you mention. But if you are referring to the ground lift switch for XLR cable connections that has nothing to do with shock protection. Signal cable drains should only have connection to earth at one end. From mixer to microphone, there is no issue because the mic is not grounded. But mixer to bass amp DI could cause a ground loop due to both devices being connected to ground via the mains.

True, but the normal chain of events I have heard of many times in the past is this one:

Something hums when it shouldn't.
Someone thinks it's too complex to understand audio wiring...
...but they have heard of grounding.
Normal off-the-shelf audio cables aren't open-ended at one end of shield.
Said someone hasn't got a soldering iron, may not ever have used one.
They bought cheap gear, so no signal ground lift.
SO THEY OPEN THE MAINS PLUG GROUND, because they think that's the most easy, familiar and accessible ground available to open, and the hum goes away.
If they don't immediately start getting shocks, they assume all is well! >
They may even assume that the signal ground will help make things safe again at this point!

It's as horrible a sequence as I make it sound, but I have repaired enough stuff in my time, I learned to always open the plug to see if the ground was opened. A surprising number of times, it was. The ground screw was usually loose suggesting this was a deliberate (but lazy) action.

I didn't mention earlier about the one-end-open shield link, but you're right, and if more firms made them like that, things would be better. They'd need to make it clear which end, and why, so people could find the safest and best results by their own efforts easily.

-------------------------------------------------------------------------------------------------

Another point (not in reply, just don't want to waste a post)...
I don't think anyone's fully emphasised yet that the GFCI not only won't work without a ground (hence not being a substitute for LACK of ground), it must have one, and ONLY one, ground, because this is the one that forms a current path if current leaks to ground anywhere else. For this reason, unlike the older RCD, it can detect a leak upstream, not only after its connected location. This is why GFCI must be used for solar panel arrays, it detects trouble with the panels as well as the load side of the charge controller. Not very relevant here but it helps show why GFCI is replacing RCD for fixed installs...

EDIT: About that... This single-ground for GFCI has an important implication. It means that there should only be ONE GFCI in the system, grounded to the main 'bonding' ground for the building. What is safe to add on demand downstream is RCD, 5 mA trip, for any amp or mains-powered mixer whose owner thinks might be less safe without one.

 

[reverendboom]

Another point (not in reply, just don't want to waste a post)...
I don't think anyone's fully emphasised yet that the GFCI not only won't work without a ground (hence not being a substitute for LACK of ground), it must have one, and ONLY one, ground, because this is the one that forms a current path if current leaks to ground anywhere else. For this reason, unlike the older RCD, it can detect a leak upstream, not only after its connected location. This is why GFCI must be used for solar panel arrays, it detects trouble with the panels as well as the load side of the charge controller. Not very relevant here but it helps show why GFCI is replacing RCD for fixed installs...

EDIT: About that... This single-ground for GFCI has an important implication. It means that there should only be ONE GFCI in the system, grounded to the main 'bonding' ground for the building. What is safe to add on demand downstream is RCD, 5 mA trip, for any amp or mains-powered mixer whose owner thinks might be less safe without one.

Please give me your source as I have done, because a GFCI will most certainly work without a ground.

ummm. I'm still on the first page but GfCI's do not require a ground to work. What they do is monitor the current between the hot and neutral wires and if there is a drop on the neutral (indicating a fault to ground) it trips. BTW, ground means ground not just the bare copper wire in the house.

How does a GFCI outlet work? | HowStuffWorks

 

[Old Garage-Bander]

If you plug the amp into a GFCI then anything plugged into that amp is protected from power getting out of it.
Ditto if the PA is in a protected circuit anything coming from a mic would be protected.
An unprotected PA with a fault could shock you via the mic through your guitar but not trip a GFCI your amp was plugged into though.
This is why it seems so odd to me to put a whole lot of GFCI's throughout a house on selected outlets instead of a single unit on the mains feed into the house. I guess the benefit is when you have a dodgy toaster it only trips that circuit instead of your whole house though.

The partial answer is that you only need a GFCI on the first outlet in the circuit. All other outlets on the circuit are thus protected. The rest of the outlets can be regular ones.
It might also be that if you "whole-house GFCI, when it trips you lose power to everything in the house, not just on the one circuit. Imagine all the digital clocks in the house blinking 12:00, and needing to be reset.

 

[Old Garage-Bander]

I think the main reason is to prevent fire in unsupervised fixed installs, because it often takes only a few tens of mA at AC line voltage to do that if water and wood can be heated enough to support an arc, causing carbonisation of the wood. Also, if there is a sudden high current fault, the RCD or GFCI will act a lot faster than a standard breaker that is only intended to replace a fuse, and it will open both live and neutral too.

In my post just above this one, I mention the 5mA trip type of RCD, which is what is best used as a spot fix when prevention of electrocution is the main need.

It takes a lot more than a few tens of milliamps to start a fire.
Fire starts when wire heats to the point of melting the insulation. A properly sized breaker will trip well before enough current flows to melt the insulation. In a 15 amp wall outlet circuit, it will take a lot more than 15 amps (15,000 mA) of current to melt insulation.
GFCI's do not act on over-current situations. They act on an imbalance between hot and neutral. This imbalance can be just a few mA.
Circuit breakers and fuses protect equipment/infrastructure. GFCI's protect people. Having is the best plan.

 

[Lostgallifreyan]

Please give me your source as I have done, because a GFCI will most certainly work without a ground.

Ok, here's one. It's a nice document too:
http://https..pdhonline.com/courses/e321/e321content.pdf

The problem is that terminology is a bit iffy, and the doc makes clear that GFCI and RCD and RCCB are considered synonymous. The detailed schematics and text are clearly what Brits will recognise as an RCD or RCCB. They don't refer to ground.

In the case of solar installs, and the main GF detector in a building, they do refer to one single ground, as an extension of the normal RCD business. Well, they probably should because it's important for a case where monitoring of a ground flow return can detect a fault outside the local subcircuit.

Where the term GFCI really is a synonym for RCD any number can be used, they won't upset a real ground fault detector because they don't have a ground link that might share the leakage current between all such devices, likely compromising the ability of the main one to detect the fault current.

It may well be that solar installs, which can have a thousand volts of DC in an 'upstream' array of panels, makes the single ground link mandatory so nothing compromises it, but a similar thing might make large venues safer because it could reduce the risk of false triggers caused by differences due to insulation leaks in long conductors.

So long as the extra devices are all RCD, truly having no ground link of their own, then all's good, and you get local indication of a fault too which could reduce downtime for a show. If a local RCD trips fast enough, a main GFCI with a ground link might not open the circuit, but in this case it wouldn't have to, which further increases reliable indication of where the fault is, so the more RCD's the better.

 

[Lostgallifreyan]

It takes a lot more than a few tens of milliamps to start a fire.

Well, probably true in any likely situation, but I was going by the raw numbers, assuming only 110VAC at 30mA that could be 3.3W, rising to 6.9W on EU 230V mains. Given a small enough volume, those powers will burn metal film and flameproof glazing off a ceramic substrate.

If a small arc in a drop of water arose, the heat dissipation might be enough to disassociate H2 from O in a bit of damp dust, and the dry dust nearby might cause all sorts of serious unrest after that. It's probably an unlikely chain of events, but not impossible, and whoever set the limit for fire prevention at 30 mA trip likely considered these odds. So it's probably safe, but a 5mA trip would be a lot safer...

EDIT: Actually, if a very small arc happened on wood, in a partition wall, it may heat gently till local carbonisation happens, then it might glow like a small and overloaded carbon resistor (wood is a very poor thermal conductor), and if sustained long enough, the heat might cause convection in air, and while the event will be slow, it will definitely ignite and start moving very fast. It all depends on whether the detector senses the current quickly. If it's just below threshold it might not, so there could still be enough sustained localised heat to ignite something this way.

 

[TomB]

This is what we typically call a GFCI outlet over here. Are you talking about something else with that name in your parlance?

Invalid Link Removed

Because there are also breaker panel-mounted ground-connected devices here that go by a similar name, except breaker vs outlet, to protect a single circuit.

 

[Old Garage-Bander]

True, but the normal chain of events I have heard of many times in the past is this one:

Something hums when it shouldn't.
Someone thinks it's too complex to understand audio wiring...
...but they have heard of grounding.
Normal off-the-shelf audio cables aren't open-ended at one end of shield.
Said someone hasn't got a soldering iron, may not ever have used one.
They bought cheap gear, so no signal ground lift.
SO THEY OPEN THE MAINS PLUG GROUND, because they think that's the most easy, familiar and accessible ground available to open, and the hum goes away.
If they don't immediately start getting shocks, they assume all is well! >
They may even assume that the signal ground will help make things safe again at this point!

It's as horrible a sequence as I make it sound, but I have repaired enough stuff in my time, I learned to always open the plug to see if the ground was opened. A surprising number of times, it was. The ground screw was usually loose suggesting this was a deliberate (but lazy) action.

I didn't mention earlier about the one-end-open shield link, but you're right, and if more firms made them like that, things would be better. They'd need to make it clear which end, and why, so people could find the safest and best results by their own efforts easily.

-------------------------------------------------------------------------------------------------

Another point (not in reply, just don't want to waste a post)...
I don't think anyone's fully emphasised yet that the GFCI not only won't work without a ground (hence not being a substitute for LACK of ground), it must have one, and ONLY one, ground, because this is the one that forms a current path if current leaks to ground anywhere else. For this reason, unlike the older RCD, it can detect a leak upstream, not only after its connected location. This is why GFCI must be used for solar panel arrays, it detects trouble with the panels as well as the load side of the charge controller. Not very relevant here but it helps show why GFCI is replacing RCD for fixed installs...

EDIT: About that... This single-ground for GFCI has an important implication. It means that there should only be ONE GFCI in the system, grounded to the main 'bonding' ground for the building. What is safe to add on demand downstream is RCD, 5 mA trip, for any amp or mains-powered mixer whose owner thinks might be less safe without one.

Regarding balanced (usually XLR) audio cables have the shield connected at both ends... The only time you need to break the shield at one end of the cable is when you have multiple pieces of equipment power from wall outlets. In the case where you only have one powered device. You may want the shield connected at both ends. In cutting the ground at one end, you may be allowing other forms of unwanted signal to get on the audio pairs. You also have option to break the shield at one end or the other.
This is why they make cables this way. It is up to the user then to break the shield if and where they need.

 

[Lostgallifreyan]

Regarding balanced (usually XLR) audio cables have the shield connected at both ends... The only time you need to break the shield at one end of the cable is when you have multiple pieces of equipment power from wall outlets. In the case where you only have one powered device. You may want the shield connected at both ends. In cutting the ground at one end, you may be allowing other forms of unwanted signal to get on the audio pairs. You also have option to break the shield at one end or the other.
This is why they make cables this way. It is up to the user then to break the shield if and where they need.

With balanced lines, sure, but there's the quagmire of handling mixed types, and I let Rane do that because it's a big wheel and I didn't want to reinvent it.

 

[Lostgallifreyan]

This is what we typically call a GFCI outlet over here. Are you talking about something else with that name in your parlance

In my case, yes, there can be a monitor of the ground link as well as the normal line differential monitor and trigger. These devices are rarer, but they do exist. Solar arrays mandate their use.

In some district mains wiring the neutral may be linked to ground at a substation or elsewhere, with a big fat earthing connector. Copper plate or similar. Since RCD's got popular in the UK, especially in cities, they stopped doing that, relying instead on a single earthing rod and a firm bonding point very close to it, for each building.

In the older arrangement the ground monitor form of GFCI would be useless because there was always another link! It is more feasible now to have a true local ground monitor in a building because the same local ground that is intended to reduce false triggers by localising the earthing reference, can do the same thing for a ground monitoring GFCI. If a sufficiently nasty fault arose in a neighbouring property on the same phase, the nearby properties might also get disconnected on that phase if they detected the surge in their own links, and how well they did this would depend on how good the property's own earth rod worked, to a large extent, but in the case of a serious event this disconnection might be desirable. I doubt anything but a high current short direct to ground would do it, so there'd not be a lot of false alarms.

As to refinement of terminology, I think some standards organisations are going to have to duke that one out for a while..

 

[Old Garage-Bander]

Well, probably true in any likely situation, but I was going by the raw numbers, assuming only 110VAC at 30mA that could be 3.3W, rising to 6.9W on EU 230V mains. Given a small enough volume, those powers will burn metal film and flameproof glazing off a ceramic substrate.

If a small arc in a drop of water arose, the heat dissipation might be enough to disassociate H2 from O in a bit of damp dust, and the dry dust nearby might cause all sorts of serious unrest after that. It's probably an unlikely chain of events, but not impossible, and whoever set the limit for fire prevention at 30 mA trip likely considered these odds. So it's probably safe, but a 5mA trip would be a lot safer...

EDIT: Actually, if a very small arc happened on wood, in a partition wall, it may heat gently till local carbonisation happens, then it might glow like a small and overloaded carbon resistor (wood is a very poor thermal conductor), and if sustained long enough, the heat might cause convection in air, and while the event will be slow, it will definitely ignite and start moving very fast. It all depends on whether the detector senses the current quickly. If it's just below threshold it might not, so there could still be enough sustained localised heat to ignite something this way.

OK... you're talking about 7 watts, 30 mA (worst case). The minimum electrical circuit in the US uses 15 amp breakers. The wire has to have a lot more current than that running through it to even start to get warm. Even if you have bare wire attached along a dry, wooden stud, a load that uses 7 watts of energy will not set the wood on fire. Not even paper.

We can talk about what might happen under very specific, and very rare circumstances until we are blue in the face. This discussion is about practical applications seen in everyday life. Your descriptions seem based on an insulated wire attached directly to combustable material. I don't know how they do things in the UK, but over here that would not be up to code going way back. Even our oldest "knob and tube" wiring used standoffs to separate the wire from the wall/Studs/Plaster and lath. At the very minimum all three wires; Hot, Neutral, and ground are contained within an overall plastic sheath. And in many cases in conduit. Throw water into the equation and restrictions get even tighter.

Codes do not guarantee, 100% protection. They are a compromise between what is practical, reasonable and safe. And this isn't even the real discussion here. What we are talking about is what happens when codes are not followed.

 

[Smallmouth_Bass]

My read (OP please correct me if I'm wrong) was that he was asking about a GFCI to plug his GUITAR into, he didn't say amp. We can assume he meant amp but have no way of knowing without a confirmation. Of course there are GFCI's that he can plug his amp into, that's pretty much been the theme of this whole thread. This is why I went the other way.
I envisioned something with 1/4" jacks and the ability to detect an A.C. power imbalance on the instrument cable to protect from shock by shutting down equipment. The answer to that is no, and no. There is no such device, (that I am aware of) and it doesn't make sense (IMHO).

Um, if I understand the question properly, any protective equipment would be at the outlet. So yes, the amp. But, this conversation has evolved beyond my comprehension.

 

[Lostgallifreyan]

We can talk about what might happen under very specific, and very rare circumstances until we are blue in the face. This discussion is about practical applications seen in everyday life. Your descriptions seem based on an insulated wire attached directly to combustable material.

Well, we have a lot of old houses, and mice are incontinent, and cause dust by making nests. It really does happen... Ideal methods based on codes can define some acceptable threshold, but thresholds can be crossed.

 

[FrenchBassQC]

Currently renting, from a plumber.
You guessed it... we've had lots of plumbing issues.

:shrug:

My dentist doesn't have a single tooth in his mouth...