Actually that should be CURRENT instead of POWER.
The GFI looks for a case where the current in the line and neutral are not equal.
If they are not, it is assumed that current is going somewhere it is not supposed to, like a person who is touching a hot connection.
IIRC from my expert (Holmes on Homes, thank you Home and garden channel) installing 2 GFI on the same circuit is a guarantee that the circuit will go dead. As mentioned above, the two interfere with each other from testing and trip.
Presumably the best protection would be at the panel if the outlet is outdoors and possibly subject to weather damage?
presumably one GFI anywhere on the circuit will protect evrything on that circuit? Even zombies won’t be electrocuted.
Do this.
zombie or no
having any electricity protected by GFI before it gets outside the house is a good thing, either a breaker or a GFI receptacle.
having it upgraded if needed and maintained is a good thing. there were just two cases that made news in the USA of death to people in the water near boats and docks.
True. In fact, the NEC allows you to install a GFCI receptacle in an enclosure that does not contain a ground wire as long as 1) local code does not forbid it, and 2) you place a sticker on the GFCI receptacle that says “no equipment ground.” Downstream receptacles can be standard, three-terminal type (i.e. with a ground socket but no ground wire going to it) as long as each receptacle also has the sticker. This is often done when replacing two-terminal receptacles w/ three-terminal receptacles in old homes, since it’s a lot easier than running ground wires to each outlet.
No, the push-to-test button will still work on an ungrounded GFCI receptacle. (Schematic.)
No, not quite.
A GFCI outlet or breaker will protect anything further down the circuit. But not anything before it.
It isn’t clear from your statement if that is what you meant.
One consideration between a breaker and an outlet (besides the fact that outlets are a lot cheaper than breakers) is the length of your total run beyond the point of GFCI protection. Too long will give nuisance trips. Your documentation will give you their recommended maximum length.
As an opinion, I think that breakers are more durable. You might consider mounting either a breaker in a weather tight enclosure just before your over water run if that will cut down on your total protected run length. You can also do the same thing with an outlet, but in my experience outlets don’t last as long as a breaker.
Are breakers designed to trip less easily than outlets to avoid this? If so, are they less safe than an outlet?
Or is it just that in this application (for a dock), there’s presumably a much longer than typical run, where it might become a problem?
Unknown, you figure 100’ total run length, for say the GFCI in your bathroom
For convenience outlets on a dock, I doubt the sensitivity would be lower in the code (yes I am too lazy too look it up).
But lets say you need to run out to a T-shaped dock and it would be, to each arm, 100’ out, then 30’ on each arm.
Answer would be to run a feeder out to the end of the Tee, non GFCI protected, to a small breaker panel, then feed outlets from that. You only need GFCI protection at the point a human interacts with it, in a nutshell.
That is, if you shock yourself by putting yourself between the actual hot wire and the actual neutral wire, you will NOT be protected, you may be killed! The reasoning behind the GCFI is that, most commonly, accidental shock comes from contacting the hot wire and the GROUND, which is basically a higher-resistance path in parallel with the neutral wire (i.e. the neutral wire is grounded at the fuse/breaker box and at the receptacle, if possible). That way, if even a very small current flows between the hot wire and ground WITHOUT flowing through the neutral wire at the GFCI, the device trips. You can get 15, 30, or whatever the current rating is flowing indefinitely as long as ALL the current flows through both wires; if the current detours through ground, even a small amount (typically 0.001 to 0.005 amperes) trips the breaker. And it can do so in a small fraction of an AC cycle!
How does it do it? Basically, both wires are wound TOGETHER around the core of an electromagnet. Since they are wound in opposite directions, their magnetic field cancels out, unless there is a “ground fault”, that is current flowing to ground and not through the neutral wire on the magnet core. The unbalanced magnetic field, even if very small, trips the breaker (in some units, it “induces” a voltage in another winding on the magnet, acting as a transformer, which triggers various electronic circuits; this may be used to record the event for computer monitoring, but all you really need is for the magnet to trip a latch to break the circuit).