3-wire to 'sub panel'

[coulter]

Charlie, There are a few places where I'm not understanding the physics you are using.

... But that time (the current) spent outside the house is not relevant to the safety of the person standing inside the house. It cannot shock the person, if for example the person touched a section of conduit, because it is heading towards the ?real source,? the transformer, and away from the point at which the person is touching the conduit. ...What changes is that the direction of current flow is away from the house.

This logic completely baffles me. My understanding, following jon's reasoning, is if one were to touch a wire with a difference of potential twix the hand and foot, one will get shocked. The direction of current flow or power flow is, ahhh, "not relevent". I recall a couple of guys Thevinen (sp) and Norton that would agree with me.

Suppose, by way of counter-example, you bond neutral to ground at a sub-panel. ...

Excellent example and absolutely true. But nothing to do with my point. Just to insure I am clear: I do not advocate three wire feeds to sub-panels. I do not even advocate three wire services. I really do not advocate bonding the service xfm neutral to the utility neutral.

Following my current understanding, (yuck - minor pun) the system used in the US - in particular:
1. multiple grounding on the utility side of the service point; and
2. connection of the service xfm neutral to the utility neutral;
- contribute to the relatively poor US electrical safety record.

We are barely one step above some third world countries.

carl

 

[charlie b]

My understanding is if one were to touch a wire with a difference of potential twix the hand and foot, on will get shocked. The direction of current flow or power flow is, ahhh, "not relevent".

What I tried to describe (apparently without being clear) had to do with the relationship between the point at which a person touches something metal and the point at which an energized wire touches something metal. Back to my example, in which conduit performs the role of equipment grounding. That means that at each load, there is some type of connection (e.g., green wire screwed down to an outlet box) through which the external metal parts of the load are electrically joined to the conduit that houses the branch circuit wires. That conduit run is (eventually, piece by piece, joint by joint) electrically connected to the main service panel?s enclosure. Bonding jumpers inside that enclosure connect it to the neutral bar, and from there to the utility?s neutral wire.

Under ?normal? conditions (no ground faults), and under ?proper wiring? conditions (i.e., neutral and ground separated at the sub-panel), why is it safe to touch the conduit? For that matter, why is it safe to touch the metal case of the load (e.g., the light fixture)? Why does current that passes through a load, and that makes its way to the service panel?s neutral bar, not flow from that point back along the conduit run to the hand of the person touching the conduit? Because of ?Kirchhoff?s Voltage Law,? that?s why. There is no ?driving force? to cause current to flow from the N-G bond point, along the conduit run, through the person?s body, through dirt, through the ground rod, and back to the N-G bond point. That is what I meant, when I said that the direction of current flow (i.e., out towards the utility pole) is relevant. Current that gets to the N-G bond point is going to head towards the utility transformer, and not back into the house, because there is nothing that is going to cause it to flow the other way (towards the person touching the conduit).

 

[electrofelon]

Stickboy1375 wrote:

The reason we run a 3 wire to a service is because the NEUTRAL is suppose to carry a fault... just not in a sub-panel application...

My question isn't really about the NEC at all. I have two angles: 1) is there something I am forgetting or not taking into account that makes a N-G bond at a 'sub panel' a safety risk, but NOT a safety risk when done at the service. 2) is the Iwire answer mostly it - utility's never coordinated with the NEC and now the infrastructure is in, so the NEC does what it can, but it would be safer to have a quadplex service drop and non N-G bonded service.

Charlie b wrote:

...neutral and ground separated at the sub-panel), why is it safe to touch the conduit? For that matter, why is it safe to touch the metal case of the load (e.g., the light fixture)? Why does current that passes through a load, and that makes its way to the service panel?s neutral bar, not flow from that point back along the conduit run to the hand of the person touching the conduit? Because of ?Kirchhoff?s Voltage Law,? that?s why. There is no ?driving force? to cause current to flow from the N-G bond point, along the conduit run, through the person?s body, through dirt, through the ground rod, and back to the N-G bond point.

Couldn't the same thing be said of conduits and metallic frames connected to a bonded 'sub panel' though?

 

[George Stolz]

Charlie, I must take this opportunity to disagree, for these opportunities are few and far between and must be seized as they surface.

I submit that you're giving the electrons a greater consciousness than they possess, and that it all simply comes down to resistance. So long as the resistance of the unbalanced return path is low, then any system is safe, because the resistance of people will be higher, plain and simple.

When the resistance increases, people will become a proportionately more attractive path, regardless of their relative position to the bonding point. Therefore, electricity does not care about the bonding point at all, it's simply taking all available paths to the source, favoring the easier ones.

In the case of contact with an EGC during a ground fault, the EGC is a low resistance path (hopefully). The person is a high resistance path. All paths will be taken, regardless of their geographical location, IMO.

Now, I will go back and sit on my rock, and pretend I said nothing.

 

[rcampbell]

Aside from the obvious safety concerns which seem to have already been covered (which DO exist; as i was trying to reconnect a coupling on an older building once, the two ends arced together like you wouldn't believe; turned out the noodle was bonded again downstream- I've been a believer since!) another concern is harmonics. If you're running dedicated neutrals to cut down on harmonics, and you bond your neutral more than once, you just lost your isolated neutral. That's putting it simply, and there's a whole lot of fun stuff to read about on how these two relate, but I'm not an ace at harmonics issues. As far as why can you run a 3 wire (or 4wire for 3phase installations) for you're service entrance, you really dont want "your ground" to be common with the rest of the city's ground.

 

[stickboy1375]

Is calling the neutral a noodle just in certain parts of the country? and if so.... WHY????? :grin:

 

[electrofelon]

rcampbell wrote:

As far as why can you run a 3 wire (or 4wire for 3phase installations) for you're service entrance, you really dont want "your ground" to be common with the rest of the city's ground.

I dont quite follow, when you say "your ground" do you mean the system earthing system (rods, cee, etc), of the fault return path? What would make the systems common?

Charlie,

I was about to agree with george's last post, and I suppose I do, but I feel this is not addressing the original question (although I am interested, feel free to address it). I am not trying to show that three wire feeders to subs ( or ranges or dryers) are safe or should be allowed. The question is, do the same safety concerns surrounding three wire feeders also apply to services?

Ethan

 

[George Stolz]

The question is, do the same safety concerns surrounding three wire feeders also apply to services?

Yes, they do. If a neutral opens on a feeder to a detached structure, two hazards result. The EGCs will begin trying to get rid of unbalanced neutral current by any means necessary, and the effective ground-fault-current-path is missing, which could be deadly if a ground fault were to occur while the neutral is missing.

 

[coulter]

...I submit that you're giving the electrons a greater consciousness than they possess, ... So long as the resistance of the unbalanced return path is low, then any system is safe, because the resistance of people will be higher, plain and simple.

...electricity does not care about the bonding point at all, it's simply taking all available paths to the source, favoring the easier ones. ...

George -

I think you are absolutely right on.

...Now, I will go back and sit on my rock, and pretend I said nothing.

That is my plan. I wasn't getting anywhere with reasonable application of physics.

carl

 

[coulter]

... is there something I am forgetting or not taking into account that makes a N-G bond at a 'sub panel' a safety risk, but NOT a safety risk when done at the service. ...

As several others have also said, I don't think you're forgetting anything.

... is the Iwire answer mostly it ... it would be safer to have a quadplex service drop and non N-G bonded service.

Iwire's answer is pretty good, and I also think it has to do with money.

(JAO)About the time we were figuring out how screwed up we were - I'm guessing right after the war - the cost to change was high and would have to be born by the consumer. Couple that with 99.99% of the city folk only care if the lights don't come on. That only leaves the insurance companies. And their only concern is paying out less in claims than they can charge.

... it would be safer to have a quadplex service drop and non N-G bonded service.

I think so. And would also delete the solid connection between the xfm primary neutral and the secondary neutral.

carl