Wiring a 3rd garage stall. Do I need to start over?

[kevin]

Sorry, I think my point was misunderstood. I was trying to point out that a GFCI has no special circuitry to examine the neutral current and the hot current and compare them. That is, the "toroidal coil" doesn't distinguish between hot, neutral, phases, whatever. Instead, it simply detects the net sum of all current passing through the coil. As you know, a one pole GFCI has two wires going through the coil, and a two-pole GFCI has three wires going through. Or, rather, in fact they would have an extra wire each for the test button, which leaks a bit of current through in one direction, then loops it around back to the supply side without going back through the coil. But the coil could care less about neutrals, hots, phases, or whatever. Talking about "balances", "comparing", etc, can be misleading. Talking about total current seems to me a much more useful way of thinking about a GFCI (since it avoids the one-pole two-pole confusion), and is a description much closer to what the circuit actually implements.

Yup.

[Chris Lewis]

We're both saying the same thing, but I think the "must balance" terminology is more understandable than "net sum".

Balance says much more directly what the device is looking for, which in these sorts of discussions is more important than implementation details.

"Balance" meaning "discrete measurement" is more a digital concept.

Obviously, there's no need to have a microcomputer in a GFCI.

People forget about analog.

"Total current" is only going to make sense to people who can think in phases (or worse, vector math, and remember the left-hand-rule ;-) Otherwise, most people are going to get confused about the "zero sum" methodology in the toroid and come out with "How can it be zero (and not trip) when you're drawing current, you moron!" ;-)

[Mark Lloyd]

They do. Analog can work quite well for special-purpose devices.

The total current (through a properly wired and non-tripping GFCI) will be zero, although I expect a lot of people to fail to understand that.

In the same way, the total current in a 3-phase wye is zero.

[Chris Lewis]

According to Mark Lloyd :

Including me. Unless I had prewarning you were speaking in net sum terms. Otherwise, obviously, the total current is what is passing through a wire.

[Mark Lloyd]

Of course, there's more than 1 wire in there, and I did say "total".

[Chris Lewis]

According to Mark Lloyd :

Well, yes. But you're wearing your electronics hat, not your electrical hat.

Unless you're used to thinking of current in terms of instantaneous _direction_, and think that the hot current is balancing (and in opposite direction) to the neutral, the total isn't going to be the electronics notion of "net current flow", it's going to be the electrical notion of how much current the circuit is delivering to the load at the far end.

Which isn't the same thing. Of course.

[Mark Lloyd]

I know the difference. A GFCI is affected by the current flowing through it, not the current flowing through the load.

BTW, it is true that I've had more electronics experience.

[Chris Lewis]

According to Mark Lloyd :

I know you know. The problem is with people who don't have an electronics hat ;-)

[The "you" in the quoted sentence wasn't supposed to be _you_, Mark, but a generic "you". Oops. Sorry about that.]

Sorry to belabor it, but you did it _again_, Mark ;-)

Most electrical tradespeople (and most DIYs) would be thoroughly confused by that statement. I think an electronics person would also be. [Unless they've been following this conversation. But, everybody's probably ignoring us now ;-) You (generic you ;-) may catcall and boo now ;-)]

Let's try something a little more precise - something that expresses a electronics notion in terms understandable to "pure" electricians:

A GFCI is affected by the net sum of current flowing through the current carrying conductors, remembering that in a circuit without leakage, the net current flow will sum up to zero.

[Eg: in 120V circuit, the hot and neutral have equal and opposite direction currents. In a pure 240 circuit, the same is true of the two hots. In a 240/120V circuit, if the hots differ, the neutral makes up the difference, and it still sums up to zero.]

A GFCI is an elegantly simple circuit.

It shows ;-)