Folks far more experienced and knowledgeable than I (some in both this forum and Mike Holt's forum) have said that the equipment grounding conductor is a sham; that the safest circuit would be one without a EGC that was GFCI protected at the panel.

I can certainly see how a GFCI protected branch circuit would be safer than one without GFCI protection, but I've actually heard people claim that an EGC can create additional hazards, and should be removed completely. It seems to me that proper grounding and GFCI protection just compliment each other to create a safer circuit overall, I fail to see any hazard created by proper grounding.

What are some thoughts on this?

-John

[This message has been edited by BigJohn (edited 03-22-2004).]

The shortest answer to that IMO is this.

If you make certain there are no grounded objects in the home, it would be almost imposable to receive a shock.

A line to line shock is rare, you would need to stick your finger firmly in a light socket or some other unusual activity, and if you do that a grounding conductor only hurts you.

To do this would mean isolating the plumbing, gas piping etc.

We are getting close to that already with the use of plastic plumbing.

It took me some time to think in this way but it does make sense.

Outdoor outlets would still need grounding.

JMO, Bob

In the long run, I would trust an equipment ground more than a GFI. GFI's can fail (I know, so can EGC's). Very few people actually test their GFI's on anything resembling a regular basis (this includes me). Therefore, GFI's can create additional hazards also.

Bob,

I don't know if I agree with your position:

Equipment grounds can also fail, but I think most would agree with Toms position that they are far less likely to fail than a GFCI, which isn't even guaranteed to work right out of the box, let alone ten years from the installation time.

It seems like it'd be almost impossible to remove the potential for a L-G shock: Even if all strong paths to ground were removed [metal pipes, equipment grounds], if there is still enough conduction to trip a GFCI, then there is still enough current flow to shock someone. And the severity of that shock would increase the closer someone got to the grounding point for the service equipment.

It just seems kinda uncertain.

-John

Evening John.

First it is not really my position, it is IMO something to think about, there is more than one way to accomplish electrical safety.

I was brought up with grounding and expect it, but it does have some drawbacks.


I agree 100% with that, GFCIs do not open until current flows.


In the commercial buildings I work in I would agree with that, a wood framed house, I think it would be very possible to eliminate all grounded surfaces.

With out grounded surfaces there is no possibility of a L to G shock.

This would be a major change in many house systems, phone, cable TV, plumbing etc.

This is not a route I am going to try at this point in my house but I do find it interesting.

Think about the shocks people get in their house.

I think it is safe to say the large majority of these shocks are L to G.

Take away all the grounds and you will have less people getting shocks.

All food for thought, I still will use my grounding type extension cord plugged into a GFCI.

'Props' to Bennie.

I don't think that it would be possible to eliminate _all_ grounded surfaces in a home...even wood conducts to some extent, especially if wet.

But what about taking the approach of using GFCIs and no equipment ground to the logical extreme, and use a _non grounded_ electrical system. A L-G fault does not carry much current if the electrical system is not bonded to ground

You could even have your EGC, to provide a good solid fault current path and make certain that the potential of exposed metal remains the same as Earth potential...but rather than having to carry the trip current for an OCPD, the EGC would be called upon to carry the detection current of the GFCI.

Note: I believe that exactly this issue was debated at the turn of the 20th century, and the consensus was that grounded systems were safer, and most of the reasons are still valid. On the other hand, technology has changed significantly. Ground fault detection is _much_ more sensitive than it was 100 years ago, and power consumption is higher, to the point where one could consider a transformer SDS in every home. I don't 'support' this point of view, but I believe that it has some merit...certainly a useful thought exercise.

-Jon

A reason for the green ground wire on things like window air conditioners is that, if the line should short to the metal case, the electricity would get shunted to ground before it can get at the user. There's other cases such as a toaster, which are not grounded. There is a high likelyhood that a user will use a metal knife to unjam a piece of bread, and touch the heater wires. And complete a shocking circuit if the toaster case was grounded.

There's cases like someone using a power sanding tool (with grounded metal case) reaching to move a defective metal floor lamp for better lighting while still holding the sander. And getting zapped.

At what point is it determined that things are "safe enough" and that going further doesn't get you any further or costs way too much? A defective GFCI can be a more dangerous situation than if GFCI's were never invented (an outlet wouldn't be there if GFCIs didn't exist).

The POCO grounds the neutrals feeding customers to shunt any capacitence leakage of the high voltage distribution voltage thru the transformers up on the telephone pole. So floating lines are not really an option.

Winnie,

Norway does it this way.

BigJohn and Bob,

By using both an EGC and a GFCI you have eliminated the additional risk introduced by the EGC but at the same time retained the increased safety it offers.

[This message has been edited by C-H (edited 03-23-2004).]

I don't recall ever seeing statistics on L-G vs. L-N shocks (they have probably never been compiled), but my guess would be that L-G accounts for 90% plus.

Removing all grounded conductive objects from the house would certainly minimize the shock risk, but it would be difficult to maintain this ground-free condition, and additions and modifications would have to be scrutinized carefully to avoid introducing a ground somewhere (e.g. running a copper pipe outside to a faucet). You would also have to consider the possible effects of grounds on other services (e.g. telephone, cable TV).

Use of an ungrounded supply for safety-isolation is practical in small, tightly controlled areas, but becomes increasingly harder as the area served grows bigger.

Look at a British bathroom as an example. The only receptacle permitted by our IEE Wiring Regulations is a low-power type for an electric shaver, fed through a 1:1 isolation transformer with a floating secondary. (And yes, all the pipework is required to be bonded to ground.)

In that situation, the isolated circuit is very short and easily maintained. Extend that to a whole house, and it becomes quite a bit harder. Don't forget that there could still be a considerable "touch current" even with an isolated system due to stray capacitance. At 120V 60Hz, it takes little more than 0.1uF to allow a current of 5mA to flow.

Just to add in my 2 cents (tuppence, whatever...)

The code was changed specifically for 240V clothes dryer circuits such that a 4-wire plug must be used (2 "hots" a neutral and an EGC).

Does anybody have information on the "behind the scenes" thought that went into the code change?

I have no idea of what deliberations the code panels went into, but the biggest problem as I see it is that with the 3-wire system it takes only a bad neutral connection to result in the frame becoming energized at 120V via the control circuits.

As the dryer/range could be very close to a securely bonded sink (or other grounded object) in the kitchen, it would be quite easy for someone to sustain a hand-to-hand 120V shock.

At least with a separate EGC, you need two faults before that could happen (broken EGC and short to frame).

Good afternoon all:

The whole concept of eliminating strong paths to ground seems like it would be very limited in it's effect, just because of the scope of the project: Not only would all plumbing have to be insulated, but the proximity to any concrete (especially reenforced) would be a concern; the proximity to the service equipment and GE would be a concern; every house would have to be built to incredibly stringent standards and even then it seems like there is a lot of room for error.

I once recieved a bad shock while standing on the wooden floor on the third story of a wood frame house. I'm sure the EGCs and plumbing did their part to conduct the fault current, but obviously the wood itself had to be significantly conductive to even allow the fault current to get to the EGCs and pipes.

Unless what I experienced was somehow a special circumstance, it leads me to believe that in wood-frame houses, if you were standing near a foundation wall or there was no basement, you would be at significant risk for a severe shock regardless of whether there was an EGC or copper piping somewhere near-by: The conductivity of the materials used in the house is sufficient enough to allow dangerous current levels to pass.

People ask me if it's safe to shower during a lightning storm if their bath-tubs are fiberglass. My default response is: "If it wasn't designed with electrical insulation in mind, you can't expect it to preform in that capacity."

Is that really that far off base?

-John