7

I'm trying to wrap my head around what the benefit is of having to use a 4-prong outlet for newer appliances when it appears the neutral and ground both end up in the same spot.

In a dual pole circuit, you have red (hot), black (hot), white (neutral), green (ground).

That makes sense but...since the neutral and green both end up being grounded, given the neutral bus bar is grounded at the panel, what is the actual benefit in this case?

  • 1
    This is the same question as why normal 3-conductor outlets have 3 conductors instead of 2, right? – user253751 Jun 10 '20 at 14:40
22

Yeah, the root problem is, you see they are both tied together in the panel, but you don't know what that thing is... or you don't even realize that is a thing.

Its name is the Neutral-Ground Equipotential Bond. Since we don't know what it is, let's get rid of it. SNIP!

Life without an Equipotential Bond

Now, your 2 hots are 240V apart, and neutral is in the middle, giving you 120V from neutral to either hot. None of these have any relationship to ground at all. We cut that, remember? So our old assumption that "Neutral is always near ground" is now void.

  • Neutral is probably near ground.

  • But it could be 120V from ground (I actually had that happen once, when the "snip" happened for real in a panel and I didn't know it).

  • There could be leakage in the 9600V supply transformer, and L1, N and L2 could be 9600, 9720 and 9840V from ground, respectively.

See, having the hots and neutral "rattling around out there" at random voltages isn't good. We want to do something to peg neutral at a relationship to earth. Hey, it could be through a 1 volt 2000A transformer - now neutral is pegged 1 volt AC from ground. But it's pegged. It could be a car battery - so now neutral has a 0 volt AC and 12 volt DC bias from ground. (both of which would be cool for diagnostic purposes). But the point is, it's no longer floating. We've pegged it.

2kA transformers and car batteries are neat in theory, but if you have 100 houses to wire this week, you want a simple/cheap equipotential bond - one with 0V bias -a piece of copper.

Actually, you can go cheaper still, simply by using the neutral bar as the equipotential bond, i.e. just putting neutrals at one end of the neutral bar, and grounds at the other end. Actually, we could just inter-mingle them.

And this is where you show up. And see a bunch of neutrals and grounds spammed onto the exact same bar, and go "WTH? Why 2 wires to the same thing?" The whole logic of how we got here is not apparent to you, and it just looks redundant.

Well, you can do your part to fix that. Go out and get accessory ground bars for your panel. Move all your grounds to the ground bar. (and while you're at it, fix any illegal double-taps, especially the deadly "put a circuit's N and G under the same screw" nightmare... because if that loses connection, you have the bootleg scenario below). For more style points, remove the bonding screw/strap and replace it with a nice big fat copper wire between N bus and G bus. Route it so you can put a clamp ammeter around it (why would you want to do that, haha! ;)

Now we can return to our assumption that neutral is relatively safe.

Fault current return

The other thing is, electricity travels in loops. Power doesn't want to get back to ground, it wants to get back to source. Source being e.g. the neutral at the transformer.

So suppose our safety grounding actually worked, and it caught a bolted ground fault. Do we want that fault to just be "caught" indefinitely? No, we want a breaker trip to occur, and that means we need to flow enough current back to source (neutral) to trip the breaker for sure. Again, the neutral-ground equipotential bond gives us that path: Down the ground wire, over the N-G bond, and over the neutral bus back to the transformer. A satisfying 500A of bolted fault flows, and the breaker goes SNAP!

Or if we're in GFCI-land, a satisfying 0.007 amps of current flows, and the GFCI goes "Unequal current! Snap!"

Let's bootleg ground everywhere! What could go wrong?

When you say "Hey, we don't need 2 wires", and just attach ground to neutral since they're the same thing in the panel, that is called bootlegging ground.

Now, in ranges and dryers, appliance manufacturers got a special exception cut so they wouldn't lose appliance sales to people who had 3-prong dryer/range outlets. They said "For grounding the dryer and range, let us bootleg ground legally. The socket is behind equipment and is rarely disturbed."

Of course you can do the same at any receptacle you please, just by jumpering ground to neutral.

So what happens if the neutral wire breaks? Think about it.

Normally, the appliance quits, but its conductance pulls the neutral wire up to 120V, since the power can't get back.

Since we bootlegged ground, the ground is also pulled up to 120V. We have electrified the chassis of the appliance. And yeah, this has killed people. It's always written up as "defective wiring", implying it was wired wrong... and that's true, except in the case of dryers and ranges which were legally bootlegged. There, the wiring was "correct" by Code (certainly not correct in any real world evaluation), and a simple wire failure caused it.

The electrical system ought to endure simple failures like a loose wire without it becoming hazardous.

  • 2
    "Neutral-Ground Equipotential Bond" is one of my favorite bands! :) That bad joke out of the way...thank you! This is an incredibly good explanation and helped me wrap my head around the whole thing. I really appreciate this! – DA01 Jun 10 '20 at 15:55
4

Safety: Neutral may be tied to ground at the street transformer, many meters away, and is a current-carrying conductor. A true ground is connected to the earth locally, e.g. a conductive water pipe.

With a broken neutral, as can happen if the entrance cable is damaged (I've had this happen after a branch hit the cable), or with a heavy fault current on one side of a split-phase line, the neutral is no longer at ground potential. That means an appliance chassis connected to the neutral could deliver a possibly fatal shock to someone touching the ground, e.g. one hand on a clothes-washer, the other on the shut-off valve. It also causes unbalanced voltage between the two lines of a split-phase AC mains and the neutral. For example, with 120-0-120 VAC split power, if a refrigerator drawing 600 W is on L1, and a laptop power supply drawing 60 W is on L2, the voltage on the 'fridge would be about 22 VAC, and the PC supply gets fried with ~220 VAC.

If a true ground is connected to the chassis,

  • The appliance should still be safe to touch and
  • The difference in voltage between ground and neutral can be sensed, to trip a breaker, before too many appliances go up in smoke.
  • 1
    Thanks. This does help. The last bullet point still confused me thought. If L1 and L2 are going to the breaker, and neutral is going to the neutral bus, and ground to ground (and neutral bus is grounded)...how would the breaker detect anything in the neutral line? – DA01 Jun 10 '20 at 3:54
  • 1
    Your observation is correct - simple overcurrent-protection breakers do not monitor neutral, only hot(s). This is one reason why it's important to follow code - not mixing circuits across cables, using proper two-pole breakers for multi-wire branch circuits (MWBCs), etc. - so the unmonitored neutral doesn't accidentally get overloaded. – nobody Jun 10 '20 at 14:26
  • @DA01, Neutral should be close to ground potential. If the neutral is not connected to the outside transformer, or if it has high resistance or too high unbalanced current, it would not be at ground. The difference between N and G is easy to detect in the house. – DrMoishe Pippik Jun 10 '20 at 17:05
0

I believe DrMoishe over complicated the issue. While the neutral and ground are connected electrically in a main panel there is an important difference.

The term for a neutral is a Grounded Neutral Conductor which is intended to conduct power during normal operation. It is intended to be close to ground potential (no or very little voltage between neutral and ground), but it is intended to carry current.

The other conductor called an "Equipment Grounding Conductor" is intended only to carry current in the event of an equipment fault. Like a motor winding burning out and shorting to the chassis or frame of the equipment.

In summary, It's OK for the neutral conductor to carry current during normal operations. The grounding conductor is a safety device to carry errant currents safely away from people when there is a failure.

The more recent codes are requiring Ground Fault breakers in more and more places which also helps protect ppl against equipment failures. Hope this helps.

  • I think so...so it's really about just having an extra path for errant current? In other words, it's not so much that they're both grounded in the panel, but offers current another way to get there. – DA01 Jun 10 '20 at 3:52
  • Since the neutral wire is carrying current, and its resistance isn't quite zero (as it's not a superconductor), there must be a potential difference between its ends. So while the earth-bonded end will be at the same potential as earth, the appliance end won't be. Is that significant? – gidds Jun 10 '20 at 17:04
  • @gidds I might get snipped from other "purists" here, but I think this business about 240v appliances not being "grounded" is over thinking things. That the neutral and ground were (not anymore) the same conductor it was bonded to the chassis and there was little if any actual current on it. The vast majority of the load is across the 2 hots. So functionally it served as a ground maybe carrying a tiny bit of current from a transformer, clock, etc. but nothing to worry about. OK. Let the snipping and down-voting begin. If you all really strongly disagree I'll delete this. – George Anderson Jun 10 '20 at 17:40

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.