misunderstanding about gfci and grounding?.

[rong111]

hi all,

someone told me yesterday that a gfci does not open when there is a small leakage from h-n. a gfci requires a return path(earth) to be able to operate.

the concern here is that aquarists count on gfci's(foolish) when reaching into the aquarium. there are electric devices submerged in the aquarium. they are double potted and ul approved as submersible devices. however they can and do occasionaly break. since the aquarium is isolated from any ground when only polarized (2 wire)h-n cords enter the aquarium. he tells me that a gfci is useless here untill the aquarist reaches into the aquarium and completes the return path. and that the ciccuit can draw up to the rating of the mains breaker without opening the gfci under this condition!

now the pet stores would love to sell everyone a "grounding rod". a rod that sits in the water and connects to the earth terminal of a gfci recept.

i have heard this argument from aquarists for years that know nothing about electricty. this is the first time i heard it from a fellow electrical worker.

i am hoping this guy is wrong and that a gfci CAN measure the difference between h-n and open if there is a difference. i always thought they could, so this is big news to me.

so, do these aquarium installations require the water to be grounded to the gfci to provide maximum protection for the fish keepers or not?

thank you,
ron g.

 

[bphgravity]

Re: misunderstanding about gfci and grounding?.

Your friend is wrong. The equipment grounding conductor plays no role in the operation of a GFCI device. The sensor monitors the L-N loads. At any point there is up to a 5mA difference between the two conductors (ground-fault) the unit will trip. A two wire utilization equipment will operate just fine on a GFCI.

Some GFCI testers require the connection of the equipmewnt grounding conductor for proper operation. As far as the ground rod deal, I have never heard of bonding water before!

 

[rong111]

Re: misunderstanding about gfci and grounding?.

thank you so much. i knew he was wrong!
someone always comes along and tells me i am all wrong. then i beleive him when i was right in the first place.

i told him that a gfci monitors the difference between line(that i always call "hot") and neutral. it can also detect leakage from l-g and n-g as far as i know. that is why the gfci was invented. to open on very small v/c leakage. before gfci it pretty much required a dead short(lethal) to open a mains breaker. well i will show him this. i am sure he will say we are all wrong. some people are just that way.

as for bonding the water, i said it will actually introduce a small capacitive load to the water from emf carried by submerged motors. not enough to hurt humans (but the fish however).

the truth be told, water is actually a very good insulator. the average aquarium installed on a carpeted floor is about as likely to electrocute someone as birds on high tension wires. i.e, it probably won't happen. but, if this grounding rod is installed and there is a fault then the fishkeeper reaches in the water while touching a good path to earth(light reflector etc) that would be very bad.

the problem is comercial aquarium electric installers are doing this all over the place.


thank you,
ron g.

 

[charlie b]

Re: misunderstanding about gfci and grounding?.

I still think you have an inaccurate picture in your mind. First, let me re-emphasize Bryan?s comment in this way. Let us give you the task of providing GFCI protection. Stand at the hot leg of the receptacle, and count electrons as they leave. Have a buddy stand at the neutral leg, and count electrons as they return. If you and your buddy get a different count (i.e., different by as much as 1/200th of an amp), then trip the circuit. You do not have authority to investigate why the two counts are different, nor where the extra electrons went to or came from. Your only assignment is to trip the circuit if the counts are different.

Now back to the fish tank. Suppose that there is a breakdown in the pump, such that the water comes into contact with the bare metal of the hot conductor. Suppose further, as you have already said, that the tank is well insulated from planet Earth. At this point, you have correctly asserted that no additional current paths exist, and that the ?GFCI Team? (you and your buddy) have no idea that anything is wrong. Finally, suppose that a person reaches into the tank. Now you do get an additional current path. Current will flow into the person?s hand, through their body to their feet, into the floor, into dirt, along dirt to the nearest ground rod, up the ground rod to the main panel, and thus back to the source. Even though the water is not a good conductor, and even though there is significant resistance in the dirt, there could be enough current flow to kill the poor aquarist. Here is where the ?GFCI Team? steps in to save the day. You and your buddy will quickly see that something is wrong, that your count is higher than your buddy?s count, and you will terminate the event by tripping the circuit. Here?s my point: A GFCI does not prevent shock. Rather, it terminates the event in a short enough time so as to prevent injury.

Finally, let?s discuss placing a ?conductive material? (I absolutely refuse to call it a ?ground rod?) into the tank, and connecting it to the receptacle?s ground terminal. Three questions arise: (1) Will it change the description of events, as I have written in the second paragraph above? (2) Will it create a new hazard? (3) Will it provide an additional level of safety? The answers are, (1) Maybe, (2) No, and (3) No.

As to question (1), a leakage current from the motor?s damaged insulation into the water might find a path back to the source via the ?conductive material.? That leakage current will be detected by the ?GFCI Team,? and might be high enough (depending on the resistance of the water and the locations of components within the tank) to result in a trip. In this case, then the event will have been terminated before the aquarist has a chance to place a hand in the tank.

As to question (2), if the leakage current from the motor to the water, and from there to the ?conductive material? and back to the source is not high enough to cause a trip, then it will not be high enough to harm an aquarist who places a hand into the tank.

As to question (3), a person who places a hand into an ?energized tank,? if I may use that description, will receive a shock until the ?GFCI Team? does its thing. That would be true with or without that added ?conductive material? being in the tank.

 

[rong111]

Re: misunderstanding about gfci and grounding?.

thank you charlie b.

the only thing i would like clarification on is:
if the tank becomes "energized" while protected by a gfci, the gfci will not trip untill someone reaches into the tank and provides a path to earth? or the gfci does not require a path to earth to operate and will trip as soon as there is a difference in current between the hot and neutral legs regardless if a path to earth is present?


my buddies original point was that if a blowdryer drops into a sink full of water the gfci operates because the sink has a drain conected to earth. if a blowdryer was dropped into a plastic bowl of
water sitting on a wood table on a carpeted floor,
the gfci will not trip untill someone or something completes a path to earth he tells me.

the reason for the "ground rod" or conducting material if placed close enough to the source of leakage is to trip the gfci before someone reaches in the tank. supposedly.

i guess you brought up a good point though charlie, that the current may not find the "ground rod" before someone reaches in anyways. being that water is a pretty good insulator. however i would imagine that if it takes reaching in to trip the gfci, that a 1/200 of an amp shock really doesnt hurt much at all. if that is where these devices trip. i still maintain the best thing to do is always shut the mains before sticking ones hand into possibly energized water. with large comercial installations this is not really feasible however. these large tanks actually have a "ground" rod next to each electric device.
if there is current leakage the fish certainly won't let you know, just like birds on a wire i mentioned before.


thank you,
ron g.

 

[dereckbc]

Re: misunderstanding about gfci and grounding?.

Originally posted by rong111:
If the tank becomes "energized" while protected by a gfci, the gfci will not trip until someone reaches into the tank and provides a path to earth? or the gfci does not require a path to earth to operate and will trip as soon as there is a difference in current between the hot and neutral legs regardless if a path to earth is present?

however i would imagine that if it takes reaching in to trip the gfci, that a 1/200 of an amp shock really doesn't hurt much at all

1. Provided there are no grounded parts inside the tank. And it may not trip when someone reaches in, they would have to be grounded to initiate a trip.

2. A GFCI does not need a EGC wired to the receptacle, only needs EGC or ground to enable the test switch.

3. I think you have the 1/200 of an amp wrong. A GFCI will not limit current, only the time in which it will operate. It takes only minimum of 1/200 of an amp to initiate the trip in 5 milli-seconds. In reality, the fault can go much higher say 10-to-40-amps or higher depending on the conditions.

 

[rong111]

Re: misunderstanding about gfci and grounding?.

thanks guys you have explained it fine.

from what i have learned here i think the first line of safety in this situation is to remove the power before reaching in.


ron g.